Mexico Policy Notes Mexico Policy Notes © 2019 International Bank for Reconstruction and Development / The World Bank 1818 H Street NW, Washington DC 20433 Telephone: 202-473-1000; Internet: www.worldbank.org Some rights reserved 1 2 3 4 21 20 19 18 This work is a product of the staff of The World Bank. The findings, interpretations, and conclu- sions expressed in this work do not necessarily reflect the views of The World Bank, its Board of Executive Directors, or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Nothing herein shall constitute or be considered to be a limitation upon or waiver of the privi- leges and immunities of The World Bank, all of which are specifically reserved. Rights and Permissions Attribution—Please cite the work as follows: World Bank. 2018. Mexico – Policy Notes (English). Washington, D.C. : World Bank Group. Translations—If you create a translation of this work, please add the following disclaimer along with the attribution: This translation was not created by The World Bank and should not be considered an official World Bank translation. The World Bank shall not be liable for any content or error in this translation. Adaptations—If you create an adaptation of this work, please add the following disclaimer along with the attribution: This is an adaptation of an original work by The World Bank. Views and opinions expressed in the adaptation are the sole responsibility of the author or authors of the adaptation and are not endorsed by The World Bank. Third-party content—The World Bank does not necessarily own each component of the con- tent contained within the work. The World Bank therefore does not warrant that the use of any third-party-owned individual component or part contained in the work will not infringe on the rights of those third parties. The risk of claims resulting from such infringement rests solely with you. If you wish to reuse a component of the work, it is your responsibility to determine wheth- er permission is needed for that reuse and to obtain permission from the copyright owner. Examples of components can include, but are not limited to, tables, figures, or images. All queries on rights and licenses should be addressed to the World Bank’s Senior Communica- tions Officer for Mexico; e-mail: bmmexico@worldbank.org. Cover design: www.sonideas.com / Alejandro Espinosa Contents 1. Fostering Productivity Growth•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 11 INTRODUCTION••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 12 KEY CHALLENGES•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 12 POLICY OPTIONS•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 16 Strengthening Institutions and Programs Working on Productivity Issues••••••••••••••••••••••••••••••••••••• 17 Implementing Interventions at the Sub-National Level••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 18 Developing an Integrated Strategy to Address Informality•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 18 References••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 19 2. Implementing Energy Sector Reform••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••21 INTRODUCTION AND REFORM PROGRESS•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 21 KEY CHALLENGES••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 23 Market Competition, Equity, and Efficiency•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••23 Natural Gas••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••25 POLICY OPTIONS••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 28 Creating Competitive, Equitable and Economically Efficient Markets••••••••••••••••••••••••••••••••••••••••••••••• 28 Decarbonizing the Energy Mix and Developing Sustainable Energy•••••••••••••••••••••••••••••••••••••••••••••••••• 28 Promoting the Adoption of New Technologies•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 29 References•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 30 3. Fostering Digital Development•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 33 CONTEXT AND REFORM PROGRESS••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••33 KEY CHALLENGES••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 34 Connectivity••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 34 Digital Inclusion and Digital Skills•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 36 Interoperability and Digital Identity•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 36 The Legal Framework and Open Data•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 36 POLICY OPTIONS••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 37 Connectivity••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••37 Digital Inclusion and Digital Skills•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••37 Interoperability and Digital Identity•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••37 The Legal Framework and Open Data•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 38 ANNEX 2. BROADBAND SUBSCRIPTIONS•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 39 CONTENTS 4. Developing a Deeper, More Inclusive Financial Sector•••••••••••••••••••••••••••••••••• 41 CONTEXT AND REFORM PROGRESS••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 42 KEY CHALLENGES••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 43 Financial Inclusion ••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 43 Financial Services for MSMEs•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 44 Financial-Sector Oversight••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 45 Development Finance Institutions (DFIs) Efficiency••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 45 Other Issues••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 45 POLICY OPTIONS••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 46 Financial Inclusion•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 46 Financial Services for MSMEs•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 46 Financial-Sector Stability••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 46 DFI Efficiency•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 47 Other Issues••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 47 5. Reforming Labor Policies to Encourage Formalization and Promote Productive Inclusion••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 49 CONTEXT AND REFORM PROGRESS••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 50 CHALLENGES•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 51 Lowering the Cost of Formalization for Firms and Workers•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 51 Integrating Women into the Paid Labor Force•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••52 Fostering Youth Inclusion••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••53 Addressing Skills Mismatches••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 54 POLICY OPTIONS••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 55 Lowering the Cost of Formalization for Firms and Workers••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••55 Facilitating the Entry of Female Workers into the Paid Labor Force••••••••••••••••••••••••••••••••••••••••••••••••••• 56 Fostering Youth Inclusion••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 56 Addressing Skills Mismatches and Skill Misallocation•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 56 Reviewing minimum wage legislation••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••57 References•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 58 6. Implementing and Expanding Competition Reforms•••••••••••••••••••••••••••••••••• 61 CONTEXT AND REFORM PROGRESS••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 62 KEY CHALLENGES••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 64 LOW LEVELS OF COMPETITION CONTINUE TO DISTORT INCENTIVES IN SEVERAL TRANSPORT AND LOGISTICS SECTORS.•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 66 POLICY OPTIONS••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 69 References••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 71 7. Improving the Education System’s Equity•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 73 INTRODUCTION•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 74 KEY CHALLENGES••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 75 MEXICO POLICY OPTIONS••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 77 POLICY NOTES Leveraging Performance Incentives to Improve Teacher Quality••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 77 4 Adopting a National Early Childhood Development Policy•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 77 Realign the National Education Budget to Provide an Equitable Distribution of Resources•• 78 CONTENTS 8. Strengthening the Healthcare System•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 81 CONTEXT AND REFORM PROGRESS••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 82 KEY CHALLENGES••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 83 Improving the Efficiency, Scope, and Equity of Health Services••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 83 Consolidating the Healthcare System•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 84 Developing a High-Quality, Equitable, Outcome-Driven Healthcare System••••••••••••••••••••••••••••••••• 85 POLICY OPTIONS••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 87 Improving the Efficiency, Scope, and Equity of Health Services••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 87 Consolidating the Healthcare System•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 87 Developing a High-Quality, Equitable, Outcome-Driven Healthcare System••••••••••••••••••••••••••••••••• 88 References•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 89 9. Enhancing Social Protection•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 91 CONTEXT AND REFORM PROGRESS••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 92 CHALLENGES••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 93 Efficiency of Social Assistance System••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 93 Expanding Access to Opportunities•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 94 Social Security••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 94 Infrastructure to Support a More Efficient Social Protection System••••••••••••••••••••••••••••••••••••••••••••••••• 96 POLICY OPTIONS••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 97 Improving the Efficiency in Social Assistance System and Reaching the Marginalized and Poorest••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 97 Strengthening and Expanding Access to Social and Economic Opportunities for the Poor•• 97 Providing a Sustainable, Equitable and an Integrated Social Insurance System for a Rapidly Aging Population••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 98 Building the Infrastructure to Support a More Sophisticated Social Protection System•••••••••• 99 ANNEX- FIGURES•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 101 References•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••103 10. Fostering Resilient, Productive and Low-Carbon Rural Territories•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 105 CONTEXT AND REFORM PROGRESS••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 106 CHALLENGES••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 107 Natural resource degradation and vulnerability to climate change in Mexico’s rural territories••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••107 The Unexploited Potential of the Forestry Sector••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••107 Ensuring a Robust and Sustained Agriculture Growth Pattern••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 108 POLICY OPTIONS•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 110 Aligning Productive and Environmental Objectives to support Growth, Conservation, and Resilience••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 110 Supporting Productivity and Competitiveness of Agriculture/Forestry Sectors and Sustainable Employment•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 112 MEXICO POLICY NOTES 11. Sustainably Managing Scarce Water Resources•••••••••••••••••••••••••••••••••••••••••••••••••••• 115 CONTEXT AND REFORM PROGRESS OF THE SECTOR•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••116 5 CONTENTS KEY CHALLENGES•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••117 POLICY OPTIONS•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 118 12. Fostering Efficient and Sustainable Urban Systems••••••••••••••••••••••••••••••••••••123 CONTEXT AND REFORM PROGRESS••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••124 MAIN CHALLENGES•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••124 Addressing the Lasting Effects of Expansionary Housing Policies•••••••••••••••••••••••••••••••••••••••••••••••••••••• 124 Coping with persistent Low-Quality Public Transportation and Limited Mobility Alternatives••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 125 Reverting the Neglect of Solid Waste Management•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 126 Addressing Inefficient and Costly Energy Use at the Municipal Level•••••••••••••••••••••••••••••••••••••••••••••• 127 POLICY OPTIONS•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 129 Renewing Government Efforts to Promote Compact and Inclusive Urban Development•••••• 129 Further Strengthening Transportation and Mobility Policies and Institutions••••••••••••••••••••••••••••• 129 Overhauling Solid Waste Management••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••130 Leveraging the Power of Local Governments to Create More Energy-Efficient Cities•••••••••••••••• 131 References:•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 132 13. Further strengthening fiscal sustainability with equity••••••••••••••••••••••••••••135 CONTEXT AND RECENT DEVELOPMENTS••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 136 KEY CHALLENGES•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 137 Domestic Revenue Mobilization.•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 137 EXPENDITURE POLICY•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••138 FISCAL FRAMEWORK••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 139 POLICY OPTIONS•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 139 Tax Policy•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 139 Expenditure policy•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 140 Fiscal Framework••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••141 References•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••142 14. Improving the Efficiency and Equalization Effects of the Fiscal Federalism Framework••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 145 CONTEXT AND REFORM PROGRESS••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 146 KEY CHALLENGES••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 146 Subnational Taxation•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 147 Intergovernmental transfers•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 147 Subnational Borrowing•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••150 POLICY OPTIONS••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 151 Increasing Own-Source Revenue Collection in Subnational Governments••••••••••••••••••••••••••••••••••••• 151 Improving the Efficiency and Enhancing the Equalization Effect of Intergovernmental Transfers•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 152 Strengthening Subnational Fiscal Discipline••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 152 MEXICO POLICY NOTES 6 CONTENTS List of Graphics Figure 1: Total Factor Productivity Index, Mexico, 1990-2016 (1990=100)••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 12 Figure 2: Growth Decomposition, Mexico, 1991-2016•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 12 Figure 3: Labor Productivity at the State Level•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 13 Figure 4: The Distribution of Sectoral Productivity vs. the Distribution of Firm-Level Productivity •••••••••••••••••••••••••• 14 Figure 5: Total Factor Productivity in the Manufacturing Sector by Firm Size, Mexico and the United States••••••• 14 Figure 6: Example of the Dispersion of Labor Productivity between Firms within Narrowly Defined Sectors: Cut-and-Sewn Apparel Manufacturing••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 15 Figure 7: Indicators of Corruption and the Ease of Doing Business, Mexico and Comparators, 2016••••••••••••••••••••••••• 15 Figure 8: Distribution of Management Practices, Mexico and the United States•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 16 Figure 9: The Composition of Wholesale Electricity Costs in Mexico••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 22 Figure 10: PPP Procurement Efficiency, Mexico and Comparators••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 23 Figure 11: The Recent and Projected Evolution of Natural Gas Production and Imports, 2005-2040•••••••••••••••••••••••• 25 Figure 12: Competitive Gas Markets, Mexico and a Hypothetical Best-Practices Frontier, 2017•••••••••••••••••••••••••••••••••• 26 Figure 13: Primary Energy Demand, by fuel, 2015•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 27 Figure 14: Drop in Prices in telecommunications services following the reforms•••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 35 Figure 15: Household Internet Access Rates among OECD Countries, 2010 and 2016••••••••••••••••••••••••••••••••••••••••••••••••••• 35 Figure 16: Fixed and Mobile Broadband Penetration in Mexico by State, 2016•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 35 Figure 17: Market Share by type of Market, 2017••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 36 Figure 18 Fixed Broadband subscriptions, OECD countries (June 2017)•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 39 Figure 19 Mobile Broadband subscription, OECD countries (June 2017)••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 39 Figure 20 Private Non-Financial Sector Credit to GDP••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 42 Figure 21 Share of Adults with an Account at a Formal Financial Institution, 2017 (%)•••••••••••••••••••••••••••••••••••••••••••••••••• 43 Figure 22: Financial Sector Structure••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 44 Figure 23: DFI-Backed Credit•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 45 Figure 24: The Share of Remunerated Workers without Access to Social Security Benefits by State (%) •••••••••••••••• 51 Figure 25: The Evolution of Real Labor Income, 2005-2017 (MX) ••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 52 Figure 26: The Evolution of the Female Labor-Force Participation Rate, 1992-2016••••••••••••••••••••••••••••••••••••••••••••••••••••••• 52 Figure 27: The Labor-Force Participation Rate by Education Level, Gender, and Age••••••••••••••••••••••••••••••••••••••••••••••••••• 53 Figure 28: Youth Not in Education, Employment or Training (NEET), Mexico and OECD Countries•••••••••••••••••••••••••• 54 Figure 29: The Perceived Effectiveness of Antimonopoly Policy••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 63 Figure 30: The Frequency with which COFECE Applies Competition Policy Tools*••••••••••••••••••••••••••••••••••••••••••••••••••••••• 63 Figure 31: Number of regulatory restrictions identified in key sectors (overall)•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 65 Figure 32: Subcategory: Number of rules that reinforce dominance or limit entry••••••••••••••••••••••••••••••••••••••••••••••••••••••• 65 Figure 33: Subcategory: Number of rules that are conducive to collusive outcomes or increase costs to compete in the market••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 65 Figure 34: Subcategory: Number of rules that discriminate and protect vested interests•••••••••••••••••••••••••••••••••••••••••• 65 Figure 35: OECD Product-Market Regulation Indicator, 2013••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 66 Figure 36: Barriers to Entrepreneurship, 2013••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 66 Figure 37: Cargo Transport Prices•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 67 Figure 38: Logistics Performance Indicator•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 67 Figure 39: The Five Pillars of Mexico’s Reformed Education Model•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 74 Figure 40: PISA Test Scores for Math and Reading among OECD Countries, 2015••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 75 Figure 41: Spending per Student at the Basic Education Level by State, 2013••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 76 Figure 42: Health Insurance and Health Spending in Mexico•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 82 Figure 43: OOP Spending and Insurance Coverage by Income Quintile••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 84 Figure 44: Financial Flows in the Mexican Health System•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 85 Figure 45: Disparities in the Use of Ambulatory Care and Cardiac Infarction Mortality•••••••••••••••••••••••••••••••••••••••••••••••• 85 Figure 46: Preventable Hospitalization and the Institutional Strength of Health Jurisdictions••••••••••••••••••••••••••••••••• 86 Figure 47: Expenditures on Social Assistance Programs in Mexico as a Percentage of GDP, 2003-2015•••••••••••••••••• 93 Figure 48: Expenditures on Social Assistance as a Percentage of GDP by Country, 2000-2015••••••••••••••••••••••••••••••••••94 Figure 49: Proyección del costo del sistema de pensiones a cargo del Gobierno Federal (% PIB), 2015-2080*••••••• 95 MEXICO POLICY NOTES Figure A1a: Distribution of beneficiaries by income quintiles, 2014••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 101 Figure A1b: Distribution of benefits by income quintiles, 2014••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 101 Figure A2: The ‘tax wedge’ in Latin America•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 101 7 CONTENTS Figure 51: Agriculture: Productivity Performance•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 108 Figure 52: Agriculture Land (thousands of hectares) and Contribution to Total Agricultural Output (%) by State•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 109 Figure 53: Water Use in Mexico•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 117 Figure 54: Transportation Breakdown by City Type••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 125 Figure 55: Solid Waste Production by State, 2013 (tons)••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••127 Figure 56: Final Energy Consumption, Baseline Scenario versus Transition to Energy-Efficient Technologies, 2010-2050 (petajoules)••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 128 Figure 57: Federal Public-Sector Revenue and Expenditure and Federal Public-Sector Debt, 2000-2017 (% of GDP)••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••136 Figure 58: General Government Fiscal Balances and Gross Government Public Debt, 2012-2017 (% of GDP)•••••••• 137 Figure 59: Tax Structures in Mexico, Latin America and the Caribbean, and the OECD, 2016••••••••••••••••••••••••••••••••••• 138 Figure 60: Vertical Fiscal Gaps in Selected Countries••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 147 Figure 61: Per Capita Own-Source Revenues by State•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 147 Figure 62: Property Taxes in Selected Countries (% of GDP)••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 148 Figure 63: Per Capita Total Revenues (after transfers) by State••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 149 Figure 64: State and Local Government Debt, 2005- 2016 (MXN billions)••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 150 Figure 65: The Aggregate SNG Debt Stock, 2005-2016 (% of GDP)•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 150 Figure 66: State Government Debt as a Share of Non- Earmarked Revenues, 2008-2016 (%)•••••••••••••••••••••••••••••••••••• 151 List of Tables Table A1. Principales parámetros de los diferentes regímenes de pensiones al nivel federal•••••••••••••••••••••••••••••••••••• 102 Table 1: National Wealth Estimates, Mexico and Upper-Middle-Income Country Average•••••••••••••••••••••••••••••••••••••••• 106 Table 2 : Public Expenditure, 2014-2017 (% of GDP)••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••139 List of Boxes Box 1. Initial results of pro-competition reforms between 2012 and 2017•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 62 MEXICO POLICY NOTES 8 1. Fostering Productivity Growth M exico’s economy has grown moderately over the last quarter century, with annual per capita GDP growth averaging just 1.2 percent between 1990 and 2017. The country’s weak economic performance reflected a decline in productivity, which fell by 8 percent during that period. Mexico’s productivity challenges are associated with large and widening re- gional disparities and the misallocation of resources between sectors and firms. Large productive firms integrated with Global Value Chains have not developed backward linkages to the rest of the economy and lagging regions. Meanwhile, widespread labor and firm informality contributes to the misallocation of productive resources. Reversing the decline in productivity will require an integrated strategy encompassing multiple policy areas and sectors. This broad strategy should include strategies discussed in other policy notes: (i) alleviating rigidities and distortions in labor markets and improving access to credit; (ii) alleviating existing rigidities and obstacles to compe- tition across sectors and sub-sectors while following through with the structural reforms enacted; (iii) designing and implementing effective interventions at the subnational levels to enhance both product and factor markets; (iv) dealing with the financing of social insurance schemes; and (iv) strengthening rule of law institutions at the federal and local levels. This note focuses on critical aspects of the diagnostic around the productivity dynamics in Mexico. It also links the aspects of the strategy above-highlighted to specific policy recommendations on other Policy Notes of this set given the cross-cutting nature of productivity growth. It also focuses on providing policy direc- tions on: (i) strengthening institutions and programs working directly on the productivity agenda; (ii) selected sub-national interventions to ease the regulatory burden; and (iii) proposing a broad and integrated strategy for fostering formalization. MEXICO POLICY NOTES 11 Chapter 1 and removing barriers to firm growth. Some of the reforms INTRODUCTION will need time to display their full impact, but their early results show that there is a significant possibility that they 1. For over two decades, Mexico’s economy has grown more would lead to a higher growth potential for the country moderately than those of other upper-middle-income coun- if they are fully implemented. Furthermore, these reforms tries. Between 1990 and 2016, its annual GDP per capita will require the complementary institutional and regulato- growth rate averaged just 1.2 percent, less than half the ry adjustment aimed at realigning incentives and reducing 3.5 percent average growth rate of upper-middle- income misallocation to achieve their full potential benefits. countries during the period.1 2. Decomposing the sources of growth reveals that a de- cline in productivity drove Mexico’s poor performance over KEY CHALLENGES the past several decades. Total factor productivity (TFP) 4. Mexico exhibits large regional disparities in productivity is the only component of growth that has consistently growth, and its regions appear to be diverging economi- diminished for the past 20 years, and falling TFP has par- cally over time. Stark differences in growth and poverty tially offset gains from labor and capital accumulation to indicators are evident between the southeastern, north- a lesser extent (Figure 1 and Figure 2). Historical and in- ern, and northern-central regions (Figure 3). Productivity, ternational comparisons highlight the key role of produc- measured by value added per worker, is five times higher tivity in economic growth. Mexico’s TFP growth rate was in Mexico City than it is in Chiapas (Figure 3a). Relative positive between 1950 and 1970. Had TFP continued to levels of urbanization, differences in connectivity with grow at the same pace during the past two decades, Mex- the U.S. market, infrastructure, education, among other ico’s GDP per capita would be 128 percent higher than its factors, are associated with these differences. Moreover, current level. Poland and Mexico had similar levels of GDP these disparities have widened over the last two decades per capita in 1990, and had Mexico’s productivity grown (Figure 3b). Labor productivity in the states of Aguascali- at the same rate as Poland’s—keeping factor endowments entes, Querétaro, Zacatecas, Nuevo León, and Chihuahua constant—its GDP per capita would be 56 percent higher increased at an average annual rate of over 1 percent be- than it currently is.2 tween 1993 and 2015, while labor productivity in Quintana Roo, Chiapas, Oaxaca, Tlaxcala, and Hidalgo declined.3 Eco- 3. The recent structural reforms have the potential to pro- nomic connectivity and market access contribute to these mote productivity growth but they will require a thorough disparities, and in some cases expanding market access by implementation. Given that the enactment of structural improving road infrastructure could boost employment reforms in 2013- 2015 coincided with a significant fall in by as much as 2.1 percent, output by 7 percent, and em- oil prices, it is hard to assess the recent impact of struc- ployment specialization by 3.4 percent.4 tural reform. The structural reforms in the energy, financial and telecom sectors are crucial to improving key product 5. Mexico’s high rate of informality, which also drags pro- and factor markets, while the education reform would im- ductivity growth, reflects its still burdensome labor and prove the quality of labor. Competition and labor reforms other regulations. States with higher informality rates also can play a key role in both improving resource allocation tend to have lower labor productivity. At micro level, for- Figure 1: Total Factor Productivity Index, Figure 2: Growth Decomposition, Mexico, Mexico, 1990-2016 (1990=100) 1991-2016 102 7.00 6.00 100 5.00 98 4.00 96 3.00 2.00 94 1.00 92 0.00 -1.00 90 1991-1995 1996-2000 2001-2005 2006-2010 2011-2016 -2.00 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 Averange Gross Product Growth Intermediate Inputs Growth Year Capital Growth Labor Growth TFP Growth Source: World Bank staff calculations with data from INEGI. Source: World Bank staff calculations with data from KLEMS, INEGI MEXICO POLICY NOTES 1 World Bank (2018) World Development Indicators (WDI). 12 2 World Bank staff calculations using data from Feenstra et al. (2015) (TFP) and INEGI (GDP per capita). 3 If labor productivity in the latter group of states were the same as in Aguascalientes, their current GDP per capita would be 81 percent higher. 4 Blankespoor et al., 2017. Figure 3: Labor Productivity at the State Level Chapter 1 (a) Labor productivity Index by State, 2016 (Mexico City =100) (b) productivity Divergence, 1993-2016 Ciudad de Mexico 2.0 Nuevo Leon Queretaro Coahuila Sonora Aguascalientes Baja California Sur (ln(VA/worker) 2016- ln(VA/worker) 1993)/T Quintana Roo Chihuahua 1.0 Baja California Jalisco Tamaulipas San Luis Potosi Colima Sinaloa Guanajuato 0 Durango Veracruz Zacatecas Morelos Yucatan Hidalgo Nayarit -1.0 Michoacan Mexico Puebla Tlaxcala Guerrero Oaxaca Chiapas 0 20 40 60 80 100 -2.0 12.2 12.4 12.6 12.8 13.0 13.2 VA/worker index (Mexico City=100) ln(VA/worker) 1993 Source: World Bank staff calculations based on data from INEGI mal firms are 84 percent more productive than the infor- even larger. In recent years, as labor productivity in the mal firms of the same size.5 Moreover, in the last twenty manufacturing sector declined, its share of total labor years, the number of informal firms and capital and labor fell. At the same time, productivity also decreased in the allocated to them increased significantly, being one of the services and commerce sectors, yet their share in total key factors contributing to misallocation.6 Informality also labor increased. However, reallocation between sectors slows the accumulation of human capital, compounding does not explain the sluggish productivity trend as main- the problem of inefficient input markets.7 In addition to taining sectoral labor shares constant at 1990 level would self-selection into the informal sector8 and despite recent have only marginally improved productivity growth. positive reforms, onerous labor regulations—including Instead, it is t misallocation within the three sectors the taxes, social security contributions, and firing costs— main culprit of low productivity growth. In fact, compar- are the root causes of informality as they generate, at ing the distribution of productivity across sectors (Fig- the same time, an implicit subsidy to informality and ure 4) with the distribution of productivity across firms an implicit tax on formality. 9 For example, firing costs, within each sector reveals that firm- level productivity is measured as severance pay for redundancy dismissal in significantly more dispersed and has a larger standard Mexico’s major cities, are double those in Brazil and five deviation, with most firms below the sectoral average and times those in United Kingdom and France.10 This high fir- a fat left tail of unproductive firms (Figure 5). Productiv- ing costs limit the ability of firms that hire formal workers ity dispersion is much higher in Mexico than it is in the to respond to shocks, therefore, generating incentives for United States or in any other Latin American country for informality.11 The regulatory roots of informality generate, which comparable data are available.12 These productivity in turn, a dynamic problem as it slows down formal firms’ differences persist even within narrowly defined sectors, growth, affecting the life cycle of firms and their invest- such as cut-and-sewn apparel manufacturing, where the ment decisions about innovation, technology, and labor most-productive firms are about 8 times more productive training. than the least-productive firms (Figure). Thus, the analysis suggests misallocation of factors between firms within 6. While productivity differences between sectors are large, sectors rather than misallocation between sectors as key productivity dispersion between firms within sectors are feature of the Mexican economy. 5 Busso et al., 2012. 6 Busso et al., 2012; Levy, 2018 MEXICO 7 Hanson, 2010; Arias et al., 2010. POLICY NOTES 8 Maloney, 2004. 9 Busso et al., 2012; Levy, 2018 10 For more information on the drivers of informality and possible solutions, see Policy Note 5 on Reforming Labor Policies to Encourage Formalization and Promote Productive 13 Inclusion. 11 Levy (2018) 12 Busso et al., 2013 Chapter 1 Figure 4: The Distribution of Sectoral Productivity vs. the Distribution of Firm-Level Productivity (deviation from the economic or sectoralmean) Within sectors (at 4 digits) the top 5% of productive .5 .5 The top 5% of productive sectors/activities can be up firms can be up to 1200% to 500% more productive more productive than the than the bottom 5% bottom 5% .4 .4 .3 .3 Density Density .2 .2 .1 .1 0 0 -2 0 2 4 -10 -5 0 5 10 Sectorial Median ln(VA/Worker) - National Median ln(VA/Worker) Firm ln(VA/worker)- Sectoral median ln(VA/worker) Source: World Bank staff calculations with data from ENAPROCE 2015, INEGI Figure 5: Total Factor Productivity in the Manufacturing Sector by Firm Size, Mexico and the United States 0.34 0.29 Proporción de establecimientos 0.24 0.19 0.14 0.09 0.04 -0.01 1/409 1/102 1/256 1/64 1/16 1/4 1 4 16 64 6 4 Productivity relative to manufacturing averange Firm size (number of empluyess) All firms [0-10] [11-100] [101-500] [+500] USA Distribution (1997) Source: World Bank staff calculations with data from INEGI and Pages (2010) 7. Micro, small, and medium enterprises (MSMEs) account and micro enterprises are unable to grow significantly. for 99.8 percent of all firms in Mexico and 76.4 percent of This unhealthy dynamic for the economy has also generat- employment, yet they only contribute 31.5 percent to total ed a “missing middle”15 effect in the manufacturing sector, value added.13 By contrast, MSMEs in countries such as which contains few highly productive small and medium Portugal and Spain account for a similar share of firms firms (Figure 5). and employment, but their contribution to value added is almost double that of MSMEs in Mexico.14 Due to compet- 8. The misallocation of resources among Mexican firms re- itive distortions, less-productive microenterprises in Mex- flects the country’s rigidities in product and factor markets MEXICO ico do not exit the market, while more-productive small and the disincentives to formalization due to regulatory POLICY NOTES 14 13 These figures are from Mexico’s 2014 Economic Census. 14 SME Finance Forum, MSME Country Indicators, 2011 15 Tybout, 2015 Figure 6: Example of the Dispersion of Labor Productivity between Firms within Narrowly Chapter 1 Defined Sectors: Cut-and-Sewn Apparel Manufacturing .4 .3 Density .2 .1 0 -4 -2 0 2 4 Firm Median ln(VA/Worker) - Sectorial Median ln(VA/Worker) kernel = epanechnikov, bandwidth = 0.3459Cut Source: World Bank staff calculations with data from ENAPROCE 2015 Figure 7: Indicators of Corruption and the Ease of Doing Business, Mexico and Comparators, 2016 85 80 75 70 500 1000 1500 2000 Prevalence of participation in corruption acts Source: World Bank staff calculations based on data from ENCRIGE 2016, INEGI, and the World Bank’s 2016 Doing Business report. distortions the social security obligations.16 Despite im- remain at the local level despite commendable reforms provements over the last years, limited access to credit at the national level (see below). If these distortions that markets contribute to the misallocation of resources and also lead to misallocation were eliminated, the realloca- the “missing middle” effect. Only 12 percent of firms have tion of resources from less-productive to more-productive access to financing, and microenterprises are especially firms couldsignificantly boost aggregate productivity in credit-constrained.17 There is still limited flow of credit to Mexico.19 In contrast, the corresponding estimates for the young, dynamic firms with high-growth potential.18 At the United States and China are just 43 and 86 percent, respec- same time, the still onerous costs of firing (including on the tively,20 underscoring the magnitude of the misallocation legal and regulatory side) and other labor rigidities deter of productive resources in Mexico. labor demand in the formal sector. Moreover, the payroll tax (rates) linked to social security of salaried workers (see 9. The service sector is especially vulnerable to imperfect Note 5 on Labor Markets) creates disincentives for em- competition, as most services are delivered at the local level ployers (taxes and other obligations) and employees (take where significant obstacles persist and trade cannot create home pay) for formalization. Additionally, many obstacles appropriate competitive pressure. In same subsectors a MEXICO POLICY NOTES 16 Busso, et al., 2012 17 ENAPROCE, 2015 15 18 Banerjee, 2014; Haltiwanger et al., 2013 19 By an order of magnitude up to 200 percent as suggested by Busso et al., 2012 20 Hsieh and Klenow, 2009 Chapter 1 Figure 8: Distribution of Management Practices, Mexico and the United States 0.15 U.S. Management distribution 0.13 Porcentaje de empresas manufactureras 0.11 0.09 0.07 0.05 0.03 0.01 0 0.25 0.5 0.75 1 -0.01 Management score Source World Bank staff calculations based on data from ENAPROCE 2015, INEGI, and Bloom et al., 2013. small group of firms in each local market enjoys large support investment in innovation and R&D, programmatic markups.21 High prices for inputs such as logistics, profes- duplication and fragmentation weaken their effectiveness. sional services and telecommunications services22 tend to For example, programs implemented by the National In- put smaller firms and startups at a disadvantage, especial- stitute of the Mexican Entrepreneur (Instituto Nacional del ly firms that rely heavily on such services, and given their Emprendedor Mexicano, INADEM) overlap with some Na- size cannot internalize them. Moreover, weaknesses in the tional Science and Technology Council (Consejo Nacional rule of law, excessive local regulations and corruption23 de Ciencia y Tecnología, CONACYT) programs. State- level exacerbate the misallocation of productive resources and interventions are also subject to duplication. Moreover, all encourage firms to remain small and “under the radar” to programs tend to focus on investment in the early stages limit their interactions with public authorities and benefit of the innovation process, without adequate follow-up at from incentives for smaller firms.24 higher levels of technological maturity. Innovation and R&D decisions and strategies also appear to be inconsis- 10. Beyond the misallocation of productive resources be- tent. For example, a tax credit for R&D was eliminated in tween firms, within-firm growth explains (year on year)25 2010 and then reinstated in 2017 without any apparent about two-thirds of total annual TFP growth in the man- assessment of its effectiveness. ufacturing sector,26 which also highlights the importance of building firm-level capabilities. Mexico is far from the 12. There is an important connection between high level of global technological frontier, and many manufacturing misallocation “between” firms and poor “within” firm pro- firms fall short of even the domestic frontier. Enhanced ductivity growth. It is important to stress that high levels management practices can boost firm-level productivi- of misallocation are likely to reduce incentive for firms ty, and recentstudies have found that they may account to accumulate firm-level capabilities and innovate as the for over 30 percent of cross-country differences in pro- returns from these investments are lower when misalloca- ductivity.27 Management practices in Mexico are gener- tion and distortions are high. ally poor, and the management scores of the country’s top firms are close to the median of the U.S. distribution (Figure 8).28 POLICY OPTIONS 11. Insufficient investment in innovation and research and 13. Reversing the decline in productivity will require an in- development (R&D) also affect productivity at the firm level. tegrated strategy encompassing multiple policy areas and Only about 14 percent of small, medium, and large Mex- sectors. This strategy should include: (i) alleviating rigid- ican firms invest in innovation, and less than 2 percent ities and distortions in factor markets (see Policy Notes report pursuing R&D activities. Although many programs 4, 5); (ii) alleviating rigidities and obstacles in product implemented at both the national and subnational levels markets (see Policy Notes 2, 3 and 6); (iii) dealing with the MEXICO 21 For more information on barriers to competition and reforms to improve market efficiency, see Policy Note 6 on Refining the Institutional and Policy Framework for Competition. POLICY NOTES 22 For more information on challenges in the telecommunications sector, see Policy Note 3 on Fostering Digital Development. 23 For more information on the drivers of corruption and specific proposal to address it, see Policy Note 14 on Fostering Transparency and Accountability to Curb Corruption. 24 A recent nationally representative firm-level survey (ENGRIGE 2016, INEGI) found that firms are most likely to engage in corruption to speed up procedures and avoid sanctions. 25 Year on year calculation tend to over-emphasize the importance of the “within” firm growth component. 16 26 Brown et al., 2015 27 Syverson, 2011; Bloom et al., 2016 28 ENAPROCE, 2015 financing of social insurance schemes (see Policy Notes 8 cymakers to track the evolution of productivity and its Chapter 1 and 9) and (iv) strengthening rule of law institutions at the determinants both at federal and state level.33 Fourth, it is federal and local levels. The rest of this note focuses only important that the design of the Special Program of Pro- on: sstrengthening institutions and programs working di- ductivity and Competitiveness, outlined in the Law as part rectly on the productivity agenda; selected sub-national of the National Development Plan (PND), ensures that the interventions to ease regulatory burden; and proposing UPE works on the basis of a well-defined annual workplan an ambitious and inter-sectoral strategy for fostering for- approved by the CNP rather than focusing on short-term malization. priorities and demands of the SHCP. 15. Strengthening the effectiveness of INADEM will be im- Strengthening Institutions and portant. INADEM plays a crucial role in supporting enter- Programs Working on Productivity prise development and entrepreneurship through its SME Issues Fund (Fondo PyMEs).34 INADEM operates numerous pro- 14. The role of National Productivity Committee (Comité grams, but only some of them are grounded in rigorous Nacional de Productividad, CNP) and Economic Productivi- evidence and effectiveness evaluation. INADEM provides ty Unit (Unidad de Productividad Económica, UPE) could be support to SMEs almost exclusively through “open calls” strengthened. Following the 2015 Productivity and Com- (convocatorias), yet a range of mechanisms (e.g., vouchers, petitiveness Law, which establishes the roles and attribu- innovative financing schemes, etc.) can be used to support tions of the CNP and UPE, it is necessary to ensure that the firm-level productivity, encourage the adoption of better implementation of this law complies with the following management practices, and improve access to finance for needs. First, UPE should become more independent and enterprises with high growth potential. INADEM could be be strengthened as a high-level institution, technically strengthen by: (i) launching an independent review of its strong, well-staffed and respected, tasked with propos- large program portfolio; (ii) strengthening the programs’ ing and assessing reforms to promote productivity and criteria to foster firm growth in terms of number of em- overseeing the implementation of the CNP’s recommen- ployees and other aspects; (iii) requiring that any public dations. This reform could follow the examples of the Pro- intervention be rigorously justified and that its impacts ductivity Commissions established by Chile and Australia, be regularly evaluated; and (iii) building INADEM’s orga- or Mexico’s own National Committee for the Evaluation of nizational capacity, especially in the area of monitoring Social Policy (Consejo Nacional de Evaluación de la Política and evaluation or outsourcing this to UPE with a clear de Desarrollo Social, CONEVAL). In Chile, the Productivity mandate. Commission29 is an independent and high-level institution tasked with proposing and assessing reforms to promote 16. A systematic assessment of enterprise development and productivity. The government is expected to either follow entrepreneurship support programs will be necessary to en- through on its recommendations, or provide a technical hance their impact on productivity. Under the institutional justification for not following them. Australia’s Productivity leadership of the Ministry of the Economy (Secretaría de Commission plays a similar role. In Mexico, CONEVAL pro- Economia), INADEM, CONACyT, and SHCP, the federal gov- vides a useful example of an independent advisory body ernment and many state governments invest substantial with substantial resources and institutional clout. Second, resources inprograms designed to promote productivity the CNP, while a high-level public-private body chaired by and innovation among entrepreneurs. However, dupli- the Minister of Finance30 with a strong mandate, needs to cation and fragmentation limit the effectiveness of these strengthen the enforcement of its recommendations,31 programs, and decisions to introduce or discontinue spe- and define clear implications and sanctions for lack of cific interventions are not based on rigorous empirical their implementation.32 Similarly, the CNP should maintain evidence. A thorough review of enterprise development its high-level profile and play a key role as the institutional and entrepreneurship programs could help to: (i) identify space responsible for coordination between key minis- opportunities to consolidate or harmonize programs, (ii) tries and non-governmental institutions (i.e. universities, assess the relative effectiveness of alternative program- private sector) to define strategic guidelines and recom- matic approaches, and (iii) evaluate the cost-efficiency of mendations to promote entrepreneurship, formalization different interventions. These evaluations should become and key multi-sectoral challenges. Third, the M&E chap- routine, and their results should inform the design of new ter of the law should be fully implemented and include, programs. Investment in enterprise development and en- among other evaluations, a periodic firm-level survey trepreneurship support must be accompanied by efforts encompassing a representative sample of manufacturing to promote competitive and well-functioning markets, as and services firms in each state which would allow poli- the incentives created by a competitive business environ- 29 See http://www.comisiondeproductividad.cl/ for Chile and https://www.pc.gov.au/ for Australia. MEXICO 30 While its “honorary” chair is the President of the Mexican Republic only twice the CNP meeting have been chaired by the President in the last 5 years. POLICY NOTES 31 These recommendations are “de jure” binding for Federal Government Officials but “de facto” there is no established sanction or consequence for the lack of their follow up. 32 For instance, in the case of the Chile’s productivity commission the Government is by law obliged to follow through its recommendations or provide a technical justification for not following them. 17 33 The National Productivity and Competitiveness Survey for MSMEs (Encuesta Nacional sobre Productividad y Competitividad de las Micro, Pequeñas y Medianas Empresas, ENAP- ROCE) would provide a useful example. See http://www.inegi.org.mx/est/contenidos/proyectos/encuestas/establecimientos/otras/enaproce/default_t.aspx 34 Fondo PyMEs’ budget has risen from MXP10 billion (US$750 million) in 2014 to MXP7.8 billion pesos (US$430 million). Chapter 1 ment are vital to the success of firm-level programs and Detailed recommendations for improving the regulatory industrial policies.35 framework and boosting competition at the local level are presented in Policy Note 6 on Refining the Institutional 17. Supporting the adoption of improved managerial prac- and Policy Framework for Competition. tices could enhance firm-level productivity. As most Mexi- can firms currently operate far from the technological frontier, improving firm productivity will require focusing Developing an Integrated Strategy on building firm capabilities and encouraging the adop- to Address Informality tion of good management practices before investing in 20. Informality is a complex, multi-policy, and multi-agen- innovation and R&D. Completing at least one managerial cy challenge, and addressing it will require a comprehen- improvement program should be a prerequisite for entre- sive strategy supported at the highest level of the admin- preneurs for participating in programs designed to pro- istration. Building on the positive 2017 labor reforms, mote innovation and R&D.36 which reduced firing costs and increased the flexibility of contracts, the strategy should support the further elimination of burdensome labor- market regulations by Implementing Interventions at the eliminating various contingent costs (including of the Sub-National Level firing process) that may have a good protection intent, 18. Improving connectivity to markets by investing in infra- but discourage formal hiring and thus formal protection structure and making logistic services more efficient could of more workers. It would also require a close look at the help lagging regions catch up to the rest of the country. Mex- taxation system that finances social insurance programs ico’s declining productivity is partly driven by its economi- as well as other taxes and subsidies at the firm level. And cally disadvantaged regions, and promoting regional con- would also need a review of other regulatory issues such vergence will require targeted geographical interventions. as contract enforcement. Moreover, there seems to be In addition, a demand- driven suppliers’ development an endogenous relationship between productivity and program could connect large investors to local SMEs while informality, and thus other policies that impact produc- consolidating existing programs that support produc- tivity may need to come into play. Tackling this thorny tivity and promote the development of firm capabilities. issue is difficult in a piece meal fashion. It requires the A site-specific skills development program, agreed and coordination of all the issues above discussed, and cost-shared with private firms, could attract investors and some other. This requires that the highest government better equip workers to meet the demands of employers. authority sets a structure, leadership, and discipline for Reducing the mismatch between workers’ abilities and the the work and reforms needed. skills demanded by firms will require both investment in workforce training and the creation of systems to reduce 21. A first step is a thorough revision of the many positive information asymmetry in the labor market. measures already implemented and of all barriers to formal- ity (the overall “formality tax”) that are still pending to be 19. Regulatory streamlining is needed to improve the busi- tackled. Then it is critical that a multi-disciplinary team be ness climate at the subnational levels (and federal level as working on the task, and can make fit all the reform pro- well). Competitive markets (with fair market mark-ups) posals together to ensure internal consistency of policies require simple and consistent regulations both at federal and incentives, and fiscal viability. This process should also and local level. Excessive regulatory compliance costs dis- bring all stakeholders to the table. Other Policy Notes, in- tort market competition, encourage informality, and invite cluding Policy Note 4, 6, 9 and 13 give some ideas that could corruption. Local governments could greatly simplify busi- be part of the discussion, including in the areas of taxation, ness regulations by establishing one-stop shops for licens- social insurance programs, firm and competition regulatory es and permits and enabling investors to interface with issues, technology and business services, access to finance, the public administration in a more transparent manner. and access to public procurement, among other. MEXICO POLICY NOTES 18 35 Aghion et al. 2015, Iacovone et al. 2016a, 2016b. 36 This approach was developed by CORFO in Chile. Other experiences that could provide relevant lessons include programs to improve managerial capacity in Japan, Korea, and Singa- pore (Cirera and Maloney, 2017, 2018). Chapter 1 References Aghion, Philippe, Jing Cai, Mathias Dewatripont, Luosha Du, Ann Iacovone, L., Pereira-López, M., & Schiffbauer, M. (2016a). ICT Use, Harrison, and Patrick Legros. (2015). “Industrial Policy and Competitive Pressures, and Firm Performance in Mexico. The Competition.” American Economic Journal: Macroeconomics, World Bank Economic Review, 30, S109-S118. 7(4): 1-32. Iacovone, L. & Pereira- Lopez, M. & Schiffbauer, M. (2016b). “Com- Arias, Javier, Oliver Azuara, Pedro Bernal, James J. Heckman, and petition makes IT better : evidence on when firms use IT more Cajeme Villarreal.(2010). “Policies to Promote Growth and Eco- effectively,” Policy Research Working Paper Series 7638, The nomic Efficiency in Mexico.” Institute for the Study of Labor World Bank. Discussion Paper 4740 Iacovone, L., Jiménez, F., Sánchez-Bella, P., Yuki, K. (2017). Building Banerjee, Ryan (2014). SMEs, Financial Constraints and Growth, shock absorbers: The role of better local business regulations BIS Working Paper N. 475. in response to a large external shock, Mimeo, Washington Blankespoor, B., Bougna, T., Garduno Rivera, R., & Selod, H. (2017). D.C.; Banco Mundial. Roads and the geography of economic activities in Mexico. INEGI (2018), Banco de Información Económica (BIE), Aguascalien- Bloom, Brynjolfsson, Foster, Jarmin, Saporta-Eksten and Van Re- tes, Mexico, INEGI. Retrieved from: http://www.inegi.org.mx/ enen, J. (2013). Working papers, sistemas/bie/ U.S. Census Bureau, Center for Economic Studies. INEGI (2016). ENAPROCE 2015, Aguascalientes, Mexico, INEGI. INE- Bloom, N., Sadun, R., & Van Reenen, J. (2016). Management as a Tech- GI (2017). ENCRIGE 2016, Aguascalientes, Mexico, INEGI. nology? (No. w22327). National Bureau of Economic Research. INEGI (2017), Sistema de Cuentas Nacionales de Mexico: Produc- Brown, J. D., Crespi, G. A., Iacovone, L., & Marcolin, L. (2015). Pro- tividad Total de los Factores, 1991-2016, Aguascalientes, ductivity Convergence at the Firm Level: New Evidence from Mexico, INEGI. Retrieved from: http://www.inegi.org.mx/est/ the Americas. Growth, 2, 1-5. contenidos/proyectos/cn/ptf/default.aspx Busso, M., Fazio, M. V., & Levy Algazi, S. (2012). (In) formal and (un) International Monetary Fund (2017).Country Report No. 17/147 productive: the productivity costs of excessive informality in Levy Algazi, S. (2018). Under-Rewarded Efforts: The Elusive Quest Mexico. IADB Working Paper. for Prosperity in Mexico. Inter-American Development Bank. Busso, M., Madrigal, L. & Pagés, C. (2013). Productivity and re- Maloney, W. F. (2004). Informality revisited. World development, source misallocation in Latin America. The BE Journal of Mac- 32(7), 1159-1178. roeconomics, 13(1), 903-932. Ministerio de Economía y Finanzas de Perú (2017). Estrategia de Cirera, X., & Maloney, W. F. (2017). The Innovation Paradox: Develop- Formalización de la Economía Peruana. ing-Country Capabilities and the Unrealized Promise of Techno- Pagés, C. (2010). The Age of Productivity: Transforming Economies logical Catch-Up. World Bank Publications. from the Bottom Up (No. 31278). Inter-American Develop- Cirera, X., & Maloney, W. F., Sarrias, M. (2018) Management ment Bank. Quality as an Input for Innovation in Developing Coun- Restuccia, D., & Rogerson, R. (2017). The causes and costs of misal- tries. Mimeo. location. Journal of Economic Perspectives, 31(3), 151-74. Feenstra, Robert C., Robert Inklaar and Marcel P. Timmer (2015), SME Finance Forum (2014). MSME Country Indicators. Re- “The Next Generation of the Penn World Table” American Eco- trieved from: https://www.smefinanceforum.org/data-sites/ nomic Review, 105(10), 3150-3182, available for download at msme-country-indicators. www.ggdc.net/pwt Syverson, C. (2011). What determines productivity?. Journal of Haltiwanger, J., Jarmin, R. S., & Miranda, J. (2013). Who creates Economic literature, 49(2), 326- 365. jobs? Small versus large versus young. Review of Economics Tybout, James (2015). The missing middle revisited. American Eco- and Statistics, 95(2), 347-361. nomic Review Papers and Proceedings, May 2015. Hanson, G. H. (2010). Why isn’t Mexico rich?. Journal of Economic World Bank (2018). World Development Indicators (WDI), Wash- Literature, 48(4), 987-1004. ington, D.C. Retrieved from: http://databank.worldbank.org/ Hsieh, C. T., & Klenow, P. J. (2009). Misallocation and manufactur- data/reports.aspx?source=world-development-indicators ing TFP in China and India. The Quarterly journal of econom- ics, 124(4), 1403-1448. MEXICO POLICY NOTES 19 2. Implementing Energy Sector Reform T he comprehensive energy reforms passed in 2013-14 are creating new market structures and stimulating investment in the energy sector, with significant potential for future eco- nomic growth. However, further implementation efforts will be necessary to realize the full benefits of the reforms and complete the transition to a competitive, well-regulated energy mar- ket that effectively leverages cleaner energy sources and takes full advantage of modern technolo- gies. Moreover, inclusive growth requires that firms and households have access to affordable and reliable power sources, across the country. Expanding the share of renewable and lower-carbon alternatives in the generation mix can also help ensure that economic growth is environmentally sustainable. Going forward, the following priorities for the energy sector should be considered: (i) sustained nationwide investment in infrastructure, particularly attracting private investment and financing, and improving investment prioritization (ii) stronger institutional capacity while providing policy certainty, (iii) improved price signals, (iv) policies and regulations that support the adaptation of the system to emerging technology innovation, and (iv) the accelerated adoption of digital and data-driven solutions by market operators and incumbents. INTRODUCTION AND prices, improving transparency, and promoting environ- REFORM PROGRESS mental sustainability. The reforms created a new policy, legal, and regulatory framework for the energy sector. 22. In December 2013, the Government of Mexico amend- They converted two major state-owned enterprises— ed the Constitution to introduce a set of sweeping ener- Mexican Petroleum (Petroleos Mexicanos, PEMEX) and the MEXICO gy-sector reforms covering the oil, natural gas, and electric- Federal Electricity Commission (Comision Federal de Elec- POLICY NOTES ity subsectors. These ambitious reforms were designed tricidad, CFE) into “state productive enterprises,” launched 21 to attract private investment and increase economic a series of bidding processes for private participation in efficiency across the supply chain while lowering energy the oil, gas, and electricity subsectors, and created new Chapter 2 Figure 9: The Composition of Wholesale Electricity Costs in Mexico 90 -7% 80 -11% 70 Dollars per MWh ( 2015) 60 50 40 30 20 10 0 2012 2013 2014 2015 Fuel Investment Operation and Maintenance CO2 Source: IEA, 2017 Note 1. Mexico introduced a tax on fossil fuel sales in January 2014 (excludes natural gas) as carbon emissions reduction measure. market platforms to promote competition and attract million bbls/d by 2040.38 Reform of the corporate structure new players.37 and governance of PEMEX, an overhaul of the company’s pension system introducing more flexibility in its labor 23. These reforms have an especially strong potential to relations as well as significant cost cutting measures in boost FDI and medium-term growth. Expected FDI expect- the areas of procurement and investment should, if con- ed through the reforms is between US$150-170bn over tinued, increase cost efficiency and allow the company to the next years, which could boost the FDI current struc- concentrate on profitable exploration and production op- ture and can have a significant impact on GDP growth portunities, in partnership with other companies. Despite potential. Ending PEMEX’s state monopoly in the oil sec- the relevance of oil in the energy sector modernization, tor is expected to reverse the decline in oil production energy security and as a means boost to overall output and exports observed during the past decades. Reforms growth, this note focuses on the challenges faced in elec- to the power and natural gas subsectors can have a great tricity, natural gas and sustainable energy as the Bank has impact on firms and households. If implementation takes developed expertise through policy dialogue and project place as expected, they would reduce the cost of a key implementation activities in these areas in the case of firm input and a major household expenditure category. Mexico. This in turn would generate important economy-wide improvements in productivity and household disposable 25. Significant investment commitments in the electricity income. These subsectors are also closely linked, as nat- and hydrocarbon sectors as well as the entrance of new firms ural gas has replaced oil as the primary fuel for power have already marked the beginning of the transformation. generation, helping to reduce the marginal cost and car- Lower prices for imported natural gas and oil, and the sub- bon emissions. stitution of natural gas for fuel oil in power generation, have contributed to an 18 percent decline in the average cost 24. Hydrocarbon sector reform focuses on increasing the of electricity supply since 2013. This has already reduced productivity of PEMEX while attracting foreign investment the cost of what is both an important item of household to the industry. Private companies can now participate in expenditure and a key input to commercial and industrial the exploration and extraction of hydrocarbons as well as businesses. However, the impact of the reform on electric- in refining, gas processing, transportation, storage distri- ity price through increased investment and competition bution and petroleum market activities. Several rounds to in generation is expected to materialize over the medium tender contracts for upstream investments awarded 107 term. Late-2017 saw the announcement of the results of hydrocarbon exploration and extraction licenses between the third long-term electricity auction to take place since 2015 and 2018 that, if successful, could generate an esti- the energy reform, relating to the installation of a further mated investment of US$161bn. The low oil price during 2.6 GW in mostly-renewable power generation capacity the early stages of reform implementation may have pre- (on top of the 4.9 GW in capacity arising from the two vented and delayed investments, though the contracts auctions in 2016). Together, contracts awarded through awarded as well as additional tenders could increase oil the three auctions are expected to generate $8.6bn in MEXICO POLICY NOTES production from less than 2 million bbls/d in 2018 to 3-3.5 investment by 2021. The average prices of winning bids 22 37 38 For a detailed description of the reform process, see Chanona, 2016; Vitor et al, 2017; and Comision Reguladora de Energia, 2018. IEA (2016) Mexico Energy Outlook have fallen in successive auctions, resulting in some of the Chapter 2 lowest prices globally for such contracts ($20.57 / MWh in KEY CHALLENGES the third auction, down from $33.47 / MWh in the second auction in 2016). These results hold out the promise of Market Competition, Equity, and even more competitive pricing benefiting end-consumers Efficiency in the future. In addition, new regulations on distributed generation (net metering, net billing and total sale) allow Electricity consumers to generate, consume, store, and sell electric- 27. The power sector reform established a complex market ity to the grid with “behind-the-meter’ small-scale power architecture, one that will take time to fully implement. generation in households and businesses, opening new Although the new system has replicated most of the opportunities for higher quality of service and lower prices standard features of power –sector reforms around the (e.g.; rooftop solar photovoltaic with batteries). Distribut- world—notably the Pennsylvania-Jersey-Maryland (PJM) ed generation in Mexico is growing at exponential rates model—it is also unique in two critical aspects. First, since 2013.39 In the natural gas industry, the extension of whereas most electricity markets allow for competition at pipeline infrastructure south from the U.S. border, along least in the distribution segment, and in many Latin Amer- with increases in their capacity, have also allowed more ican and Caribbean (LAC) markets transmission expansion Mexicans to benefit from imports of low-cost shale gas. is also competitive, Mexico’s transmission and distribution Lower gas and electricity prices havealready benefitted system remains under the control of the incumbent utility, the dynamic manufacturing sector in northern and cen- CFE. Second, Mexico’s electricity sector is highly complex tral Mexico, although sustained investment in natural gas in terms of the number of markets, products, and types of domestic production and pipeline infrastructure, will be contracts, as well as the regulations and procedures for necessary to expand access to natural gas throughout the overseeing and administering transactions.40 Mexico’sen- country. ergy-sector agencies now face the challenge of rapidly im- plementing complex institutional and regulatory frame- 26. Mexico’s reformed energy-sector institutions have works, which may strain administrative capacity. achieved significant progress in establishing new market structures, but key challenges remain. Further efforts will 28. Residential energy subsidies are regressive and costly. be necessary to foster competitive, equitable and eco- Falling energy prices have enabled the government to nomically efficient markets, to decarbonize the energy start adjusting electricity subsidies, and the average tar- mix, adopt cutting-edge technologies to boost efficiency, iff was reduced by about 10 percent between 2010 and and create a pipeline of strategic projects with the needed 2015.41 The structure of the residential subsidy, however, financing. remains highly regressive, and further efforts to phase it out could improve both expenditure efficiency and equity while generating new fiscal space.42 The subsidy to res- Figure 10: PPP Procurement Efficiency, Mexico and Comparators Preparation of PPPs 90 Comparators 1 60 OECD Frontier 2 PPP Contract Procurement USA Frontier Management 30 of PPPs EAP Tiger Frontier 3 Mexico+A9 Unsolicited Proposals 1. Comparators: China, Malaysia, Turkey, Romania and Thailand 2. OECD: Australia, Canada, Chile, France (PPP), Italy, Korea, Rep., Poland, Portugal, Turkey, United Kingdom 3. EAP Tigers: China, Korea and Singapore MEXICO POLICY NOTES 39 Reaching 434 MW in 2017 and expected to grow to 6.7 GW in 2023 (CREG, 2018). 40 For example, the market allows multiple approaches for structuring bilateral contracts: (i) bilateral power- purchasing agreements (PPAs), (ii) equity investments with PPAs, (iii) equity 23 investments without PPAs, (iv) design-build-finance-operate contracts (DBFOs), and (v) PPAs via competitive procurement. 41 IEA, 2017. The reforms began the process of transferring the subsidy burden from CFE to the Ministry of Finance (SHCP) by introducing it as an explicit item in the national budget. 42 SENER, 2017. The total cost of Mexico’s electricity subsidies reached approximately US$5.5-5.75 billion in 2016. Chapter 2 idential electricity cost an estimated $6.2bn, or 0.54% of to reduce the risk that price fluctuations resulting from GDP, in 2017.43 transmission congestion pose to market participants.45 SENER and the Energy Regulatory Commission (Comisión 29. Further cost reductions will require increased efficiency Reguladora de Energía, CRE) must design the nodal pricing across the value chain, including through the accelerated system, define initial FTR allocations, and subsequently adoption of advanced technologies and reduced distribu- recalculate subsidies in a more progressive way to better tion losses. Compared to many global and regional peers, target low-income households. Key challenges include: (i) Mexico fares relatively well on core performance indica- allocating FTRs appropriately, tors such as electricity tariff levels and network losses, yet its system is still far from international best practice fron- 32. (ii) ensuring that price signals effectively attract invest- tiers. For example, electricity tariffs for residential and in- ment in transmission expansion, and (iii) enhancing insti- dustrial users in Mexico are now closer to OECD levels, but tutional efficiency.46 In addition, using legacy FTRs as the they remain well above those in the United States despite basis for introducing progressive subsidies presents an low natural gas prices, and commercial users in Mexico important institutional challenge, since energy subsidies continue to face high electricity tariffs. The recent decline in Mexico are not handled by the CRE, the National Ener- in electricity tariffs largely reflects external factors, and gy Control Center (Centro Nacional de Control de Energía, considerable scope remains—especially for the CFE—to CENACE), or SENER, but rather by the Ministry of Finance further reduce costs across the value chain by harnessing and Public Credit (Secretaría de Hacienda y Crédito Público, clean-energy resources and adopting new technologies SHCP). Thus, decision-making around energy subsidies is such as smart grids and digital solutions. Transmission controlled by an entity whose aims and distributional con- losses are mainly caused by technical inefficiencies, and cerns extend beyond the electricity sector. Recent analysis distribution losses caused by theft, metering inaccuracies, suggests, moreover, that the current “cost-plus” regime and data-handling errors.44 Reducing these losses will re- may not be sufficient to regulate network investments, quire expanding and modernizing the T&D infrastructure. and that the CRE will need to consider other incentives in its future transmission-tariff methodology, implying a shift 30. The mobilization of private sector investment could play from cost-of-service to incentive-price regulation.47 a larger role in financing investment in T&D infrastructure, if the project preparation and structuring were improved. The 33. The process of restructuring the CFE into a corporatized CFE still holds a large share of the generation market, at public-sector utility is underway, but it remains incomplete. about 62 percent of total installed capacity, and it main- The authorities began unbundling the vertically integrat- tains its T&D monopoly. However, the new legal framework ed utility in January 2016 to promote competition, reduce allows it to form strategic alliances or use PPPs to develop barriers to entry, increase operational efficiency and trans- new T&D lines. Greater private participation could help form the CFE into a financially sustainable entity. The CFE address some of the industry’s most pressing challeng- has been restructured into a holding company with sub- es, such as reducing energy losses, deploying smart-grid sidiaries separated for functional and accounting purpos- technology, digitalizing the grid, managing congestion, es. As part of a new emphasis on transparency, the CFE is and integrating renewable energy into the grid. However, obliged to disclose corporate and technical information, although in general Mexico has a relatively strong capacity and it is now subject to performance evaluations. Howev- at the federal level to develop PPPs in infrastructure (Fig- er, the process of modernizing the CFE has been gradual, ure 10), the capacity of CFE to prepare and structure PPPs and there is a risk that its newly formed subsidiaries may in T&D is evolving and demonstrate particular scope for continue to communicate among themselves, and even improvement (notably on identifying risks and assessing with CENACE, whose personnel previously performed sim- the market conditions and bankability). ilar functions under the vertically integrated CFE. 31. The new regulatory framework mandates a shift from 34. As a large incumbent electric utility, the CFE’s power and a subsidized, zone-based pricing system to locational mar- potential strategic behavior represents a key challenge to ginal pricing, which is expected to improve the operational market competition and a significant source of uncertainty efficiency of the electricity network, but accomplishing this for potential market participants. The CFE’s transformation shift will require allocating financial transmission rights will be gradual, especially the shift in its organizational (FTRs) transparently and creating effective incentives to in- culture from that of a state-owned enterprise to that of vest in transmission expansion. Mexico’s transmission grid a corporate entity. The CFE’s ongoing transformation, needs significant upgrades and expansion. The energy re- and the progressive delinking of its decision-making pro- form introduced a new system of nodal prices to refine and cesses from government control, represents a key barrier improve price signals for transmission expansion and FTRs to entry for new market participants and to the efficient MEXICO POLICY NOTES 43 World Bank based on SENER, 2017. 44 In 2012, the CFE reported that annual distribution losses accounted for 16 percent of transported volumes, representing US$3 billion in lost revenue. 45 An FTR is a financial instrument that entitles the holder to receive compensation for costs that arise when the transmission grid is congested. FTRs allow market participants to hedge 24 against potential price effects. 46 Zenon and Rosellon, 2017; Kunz et al., 2017. 47 Zenon and Rosellon, 2017. implementation of pro-competition reforms. For exam- Multiple factors have contributed to the decline in natural Chapter 2 ple, the CFE can directly influence price formation in the gas production, but the most important is PEMEX’s limited wholesale market, including the energy-capacity and capacity to finance investments (including in new technol- certificates markets, as well as the determination of nodal ogies) to lead the exploration and production process for prices and the allocation of FTRs. Establishing truly com- both conventional and unconventional gas. U.S. pipeline petitive wholesale markets will hinge on the capacity of export capacity to the Mexican border has increased from energy-sector institutions, and particularly the CRE and 2.9 Bcf/d in 2015 to 7 Bcf/d in 2017, but a lack of down- the Federal Economic Competition Commission (Comisión stream infrastructure capacity has caused export volumes Federal de Competencia Económica, COFECE), to monitor to fall in recent years. The total cost of investments in the CFE’s behavior and enforce prohibitions on collusion improving the connectivity of the Mexican gas grid and and conflicts of interest. Because potential investors and deploying the natural-gas-based generation capacity private participants will continue to observe market dy- planned for 2020 is estimated at US$9.7 billion. namics and the CFE’s strategic behavior, the success of ef- forts to control the CFE’s power and reducing information 37. The country will need to evaluate different options and asymmetries will be vital to attract new investments and design a long-term strategy that delivers the most secure increase private-sector participation. and cost-effective solutions for consumers. For example, ex- panding its import pipeline infrastructure would help reap the benefits of the recent boom in U.S. shale production, Natural Gas while developing domestic gas fields would reduce im- 35. Mexico’s consumption of natural gas has increased sig- port dependence. The International Energy Agency (IEA) nificantly in recent years, driven both by demand from the forecasts a significant increase in Mexico’s demand for nat- power sector and from industry, and Mexico has become ural gas through 2040—at which point it will constitute 70 increasingly dependent on natural gas imports, especially percent of all primary energy—but only a modest increase from the United States. Demand for natural gas in Mexico in domestic gas production (Figure 11). In the baseline has grown at an average annual rate of 3.5 percent over scenario, Mexico will continue to rely on U.S. pipeline the past decade, due to rising demand from the electric- imports to satisfy at least half of its gas demand until the ity sector and from the metals and chemicals industries, mid- 2030s. which use it as a direct input. Industrial demand for natural gas is concentrated in northeastern and central states, and 38. Mexico needs to step up investment in its domestic a stronger U.S. economy and a shift toward energy-price gas-production capacity, as well as physical and institutional parity on both sides of the border will cause Mexican in- infrastructure. A necessary first step would be to improve dustrial demand to rise by another 0.2 billion cubic feet the technical performance of the gas sector across the val- per day (Bcf/d) by 2020.48 ue chain, optimizing production and reducing gas flaring and losses in operational units, storage, transportation 36. While Mexico’s demand for natural gas has risen, gross and delivery.49 The IEA predicts that Mexico will be able production has gradually fallen since 2009, due to declining to increase its domestic production after 2025, primarily non-associated gas production from mature legacy basins. by tapping the country’s significant unconventional gas Figure 11: The Recent and Projected Evolution of Natural Gas Production and Imports, 2005-2040 120 60% bcm Imports: Pipeline (United States) 100 50% LNG 80 40% Production: 60 30% Unconventional gas Conventional gas 40 20% Imports as share of demand (right axis) 20 10% 0 0% 2005 2010 2015 2020 2025 2030 2035 2040 MEXICO Source: IEA, 2016 POLICY NOTES 48 49 Goldman Sachs Commodities Research, March 2016. In 2016, PEMEX flared close to 500 million cubic feet a day (mcf/d) of natural gas. 25 Chapter 2 Figure 12: Competitive Gas Markets, Mexico and a Hypothetical Best-Practices Frontier, 2017 Inst. Org. and Reg. Fmk 5 4 3 Transparent flexible Supply diversification transactions 2 Best Practice Frontier 1 MEX 0 MEX-reform Regulated Tariffs/TPA Market opening /rules Rules Unbundling +Level Playing Field Source: Elizondo-Azuela, Barbalho et. Al. (2018) reserves,50 though this will require significant investment in the country’s energy mix, with oil increasingly being and expertise as well as efforts to improve the invest- displaced by cheaper, cleaner natural gas, which met just ment climate to attract new companies into exploration over a third of primary energy demand in 2015 (Figure 13). and production. A solid regulatory environment, includ- Oil still accounted nearly half of energy demand, with coal ing further efforts to transition toward market-based accounting for a further 7%. Although renewable energy price formation, will be essential to attract the required -including hydropower- represents only 8% of total prima- investment.51 Expansion of the gas pipeline network has ry energy demand and 17% of electricity generation, the failed to keep pace with the boom in gas demand, and targets introduced in the 2016 Energy Transition Strategy the pipeline system has reached its maximum capacity are ambitious: share of renewable energy in electricity to on multiple occasions.52 In addition, the gas distribution increase to 35% by 2024, 37.7% by 2030, and 50% by 2050. grid remains limited and gas pipeline networks are barely present in the southernmost parts of the country. Com- 40. In addition to supply-side measures to increase domestic pared to both other competitive gas markets in the region production and improve the functioning of the natural gas and a hypothetical best-practice frontier, Mexico still has market, continued efforts to diversify the energy mix can re- considerable scope to improve the functioning of its en- duce Mexico’s dependence on imported gas and accelerate ergy market. Such institutional improvements, allied to the decarbonization of the economy. These objectives can the full transformation of PEMEX into a state productive be achieved by increasing the use of renewable energy, enterprise and the full implementation of the new pricing enhancing energy efficiency, and improving demand-side methodologies and wholesale price index promulgated management—as natural gas already represents 60 per- by the CRE in late-2017, can help remove remaining bot- cent of total power generation. Emerging technologies are tlenecks to the needed investment in infrastructure across expected to play a key role in enhancing the overall tech- the natural gas supply chain. nical and economic efficiency of clean energy solutions. Mexico’s renewable energy potential is considerable. While hydro power generation has long been part of the energy Energy Mix mix, the scope for its expansion is limited by the country’s 39. Despite the shift from oil to natural gas in recent de- topography. Moreover, reduced reliance by households cades, and significant investment in renewable capacity, on traditional biomass to meet their energy needs in re- fossil fuels still account for 90% of Mexico’s primary energy cent decades has been one factor behind the fall of the supply. The first decade and a half of the 21st century saw renewables share in the energy mix. Mexico is already legal a 25% increase in Mexico’s primary energy demand, to framework for further development of the sector. However, reach 187 million tons of oil equivalent (Mtoe) by 2015. it is in solar and wind energy where perhaps the greatest This is roughly in line with economic growth over the scope lies for boosting the share of renewables in the en- same period. Recent years have seen a significant switch ergy mix, as evidenced by the globally competitive pricing MEXICO POLICY NOTES 50 Options for tapping unconventional gas reserves include: (i) the development of non-associated gas in deep-water fields in the Coatzacalcos area, (ii) the further development of the Chicontepec basin, and (iii) the development of shale gas in the Burgos and Sabinas basins. Because these measures will be costly, and their implementation will be slow, they will have a limited effect on import dependence and supply vulnerabilities in the short term. However, over the longer term domestic production could expand significantly, given substantial private capital, the adoption of best practices, improved managerial skills, strong environmental stewardship in ecologically fragile areas, and appropriate regulatory and 26 fiscal regimes. 51 In June 2015, the CRE eliminated the previous price-setting methodologies to incentivize natural gas production. 52 Vitor et al., 2017; IEA, 2017 Figure 13: Primary Energy Demand, by fuel, 2015 Chapter 2 World Mexico Nuclear 2% Renewable Coal Renewable 13% Nuclear 7% 8% 5% Coal 28% Natural Gas Natural Gas 22% Oil Products 35% 48% Oil Products 32% Source: Based on data from the International Energy Agency of winning bids in the country’s recent series of long-term trative efficiency to ensure the energy sector’s social and electricity auctions. Although Mexico’s solar power poten- environmental sustainability.55 For example, the slow and tial derived from average daily irradiation of around 5.5 uncoordinated development of appropriate regulations, kilowatthours per square metre (kWh/m2) (SENER, 2012) is including licensing rules and the issuance of permits roughly double the levels seen in Germany, for example, in- to minimize social and environment impacts, as well as stalled solar generation capacity in Mexico is less than 1% weaknesses in oversight and enforcement reduce the ef- of that in Germany. While supportive policies in Germany fectiveness of sustainable energy policies. Administrative have led to installed capacity reaching (IEA, 2016). There is processes are complex and involve multiple institutions as significant scope to boost both solar and wind generation well as subnational governments. Some functions, such as capacity across the northern and southern parts of the the approval of SENER’s social impact assessments, are still country, with significant investment in both expected to subject to acute bottlenecks, and there are still gaps in the come onstream in the coming years on foot of contracts regulatory framework, especially in terms of the minimum awarded in the long-term electricity auctions. requirements for social and community consultations.56 41. Further reducing the share of coal and oil in electricity 43. Emerging technological innovations could generate sub- generation and final consumption will require removing con- stantial efficiency gains and carbon emissions reductions, but straints on the production and use of natural gas. The devel- adopting new technologies will require market participants opment of domestic non-associated gas will require strong to modernize their assets. The emergence of cost competi- environmental regulations and oversight. The energy tive, clean energy and more efficient digital technologies— sector reforms introduced fiscal incentives to promote the including new sources of renewable energy, improved production of non-associated gas.53 Building institutional batteries, smart grids, intelligent sensors, wireless device capacity to develop and enforce regulations that protect connectivity, electric vehicles, etc.—is revolutionizing communities and the environment will be crucial to en- energy supply chains and service-delivery models across sure the sustainable development of shale or other non- the world. The authorities are taking steps topromote the conventional gas sources.54 Ensuring sound regulation, adoption of new technologies. However, among the most oversight, and enforcement will be challenging, given the important challenges facing Mexico will be to modern- complexity of shale gas development and the limited ex- ize PEMEX and the CFE, build their capacity for strategic perience of Mexico’s environmental enforcement agencies. planning and technological adoption, and transform their business models to accommodate digitalization, data ana- 42. Mexico has set ambitious targets for climate-change lytics, interfacing with consumers, improved harnessing of mitigation and clean energy generation, and the govern- demand-side resources, and other innovations. ment has introduced important policy and procurement measures to deliver on these commitments. However, 44. Access to finance remains a significant challenge for Mexico still lacks the institutional capacity and adminis- clean energy projects. The authorities have developed a MEXICO 53 These projects are exempted from royalties when gas prices are equal to or below US$5 per MBTU POLICY NOTES 54 See Castro-Alvarez et al. (2018) for a thorough discussion of the U.S. experience with horizontal drilling and hydraulic fracturing, and lessons as well as recommendations for Mexico are elaborated and discussed in 27 55 Government of Mexico, 2016; SENER, 2016; and IEA, 2017. 56 On June 1st, 2018, SENER introduced a resolution that presents the general provisions for the preparation of Social Impact Assessments in energy infrastructure projects, and associ- ated administrative procedures for their review and approval. Chapter 2 new, market-listed investment vehicle to attract private SENER, and CENACE. An effective monitoring framework capital for the financing of energy infrastructure: the In- should be complemented by the introduction of remedial vestment Trust for Investment in Infrastructure (Fideicom- and enforcement measures to discourage the exercise of iso de Inversion en Energia e Infrastructura, Fibra E)57 Their market power, collusion among CFE subsidiaries, or other asset base, however, must be sufficiently mature to deliver forms of anticompetitivebehavior. The CRE and COFECE stable and predictable revenue flows to bondholders. Ex- should work together to strengthen the capacity of regu- isting gas pipelines, along with transmission and distribu- latory agencies to gather information, redress complaints tion assets, are considered ideal for monetization through through a routine process, and oversee the vertical and Fibra E, given the low-risk nature of their operations and horizontal unbundling of the electricity and gas industries. their use of long-term contracts. However, a preference among institutional investors for stable, proven assets 48. The CRE and CENACE should counterbalance the in- creates a barrier to investment in renewable energy and centives of incumbents to abuse market power. Welcome demand-side projects in areas such as energy efficiency or measures would include nodal markets, day-ahead and distributed generation, which are less mature and which real-time markets, contracts markets, FTRs, clean energy have less-stable revenue flows. certificates, and ancillary markets. The CRE and COFECE should be afforded sufficient administrative indepen- dence and adequate financial and human resources to POLICY OPTIONS fulfill their mandates. It is important also to ensure coordi- nation among different enforcing and regulating entities. Creating an energy regulators’ group composed of rep- Creating Competitive, Equitable resentatives from COFECE, the CRE, and the Office of the and Economically Efficient Federal Prosecutor for the Consumer (Procuraduría Federal Markets del Consumidor, PROFECO) would help align the medium- 45. The CFE and the CRE could reduce electricity produc- and long-term strategic objectives and actions of sectoral tion costs by improving the operational efficiency of power regulators. plants and by reducing technical and nontechnical losses in the distribution network. The deployment of storage tech- 49. The government should take steps to reduce natural gas nology can also introduce higher operational efficiency as losses across the value chain and to increase gas production. the share of variable renewable energy increases, however Key priorities include reducing gas flaring, increasing gas this will depend on amendments by CREG to ancillary ser- production, and building storage capacity. Existing fiscal vices regulation. Both SENER and the CRE could promote measures to promote shale gas production could be com- the uptake of smart grid and emerging digital technolo- plemented by: (i) the establishment of a sound regulatory gies to boost energy efficiency and support the develop- environment by the CRE that uses market-based price for- ment of renewable energy resources. mation and calibration, and (ii) the elaboration of a solid legal framework by SEMARNAT and the Agency for Safety, 46. Residential electricity subsidies could improve their Energy and Environment (Agencia de Seguridad, Energía y progressivity while reducing fiscal costs. Coordinating the Ambiente, ASEA), backed by adequate monitoring and en- allocation of FTRs and designing progressive electricity forcement capacity to prevent and address negative social subsidies will involve the CRE, CENACE, SENER, and the and environmental impacts. SHCP. Higher progressive of subsidies could also be ac- companied by a reduction in the fiscal costs (which cur- 50. The government should expand the urban gas pipeline rently run above 0.5% of GDP). SENER could complement network. This will require the CRE and CENEGAS to develop these measures by introducing a program to support the a functional gas market by: (i) monitoring the efficiency of installation of rooftop solar panels in the residential sec- the regulatory framework to ensure that pricing promotes tor, as distributed generation for low-income households investment in the distribution pipeline network, (ii) level- has the potential to reduce the volume of subsidies for ing the playing field for distribution companies, (iii) ensur- grid-connected electricity services. ing open access to the pipeline system, (iv) maintaining close cooperation and regular information flows between 47. Competition could be stimulated. Policymakers should the CRE and CENEGAS, and (v) monitoring and controlling accelerate the functional and accounting separation of the power of incumbents. CFE’s subsidiaries and strengthen its corporate gover- nance by insulating its decision-making processes from Decarbonizing the Energy Mix and government influence. Establishing a consolidated system Developing Sustainable Energy for collecting, processing, and analyzing relevant market data along entire energy value chain will support effective 51. The government should streamline the authorization regulation and oversight. Creating an integrated informa- of clean energy infrastructure projects by enhancing the MEXICO POLICY NOTES tion system will require the active involvement of the CRE, efficiency of the ESIA approval and licensing process. SEN- 28 57 The Mexican Securities Industry Association estimates that Fibra E could help raise US$70 billion in additional investment. ER, SEMARNAT, and PROFEPA should work together to market pricing at distributional nodes, and from time-of- Chapter 2 improve institutional coordination and administrative use tariffs to intraday and real-time spot pricing. The SHCP efficiency, simplify the approval processes by creating a should update Fibra E guidelines and redesign the vehicle dedicated window, and evaluate the potential adoption to level the playing field among all energy projects while of complementary instruments such as sectoral and giving appropriate consideration to the characteristics of technical standards with strong monitoring and strin- clean energy projects. gent sanctions. SENER and the CRE should monitor the performance of renewable energy auctions to identify 54. The government should further strengthen existing bottlenecks or delays in the construction of plants. Fi- externality-pricing mechanisms. SENER, SEMARNAT, and nally, SENER and the CFE should promote investment in the SHCP should use the proceeds of the carbon tax to transmission and distribution infrastructure to support support clean energy financing. The carbon tax should be the integration of renewable energy, and they should expanded to include natural gas, and its rates should be introduce clear rules for implementing T&D expansion aligned with the relative carbon intensity of different fuels. projects via PPPs. 52. To help consumers save energy and reduce carbon emis- Promoting the Adoption of New sions, the CRE and CFE should digitalize the grid, share infor- Technologies mation, and modernize billing. With the support of SENER 55. SENER should work with other ministries to plan ener- and the CRE, CENACE should establish a data platform or gy sector infrastructure from a multi-sectoral perspective. hub to enable the secure storage of metered data on cus- Infrastructure plans should focus on economic efficiency tomer usage, telemetry data on network operations, and and low- carbon development, particularly in cities, which other information, as participants in a dynamic retail mar- offer opportunities to expand electric transportation and kets require timely and nondiscriminatory access to data digital infrastructure. The government should build the relevant to competition. capacity of CENACE, the CRE, and the CFE to leverage dig- ital solutions. Policymakers should also provide proactive 53. The authorities should improve the accuracy of market regulatory support to innovative business models such price signals. The CRE should shift from zonal or nodal as local energy services, bundled services energy-service locational marginal pricing to real and reactive locational companies, and peer-to-peer energy. MEXICO POLICY NOTES 29 Chapter 2 References Alcaraz, Carlo, Villalvazo, Sergio. 2016. The effect of natural gas Secretaria de Energia, 2016. Estrategia de Transicion para Promover el shortages on the Mexican Economy, Energy Economics 66 Uso deTecnologias y Combustibles mas Limpios, https://www.gob. (2017) 147-153. mx/cms/uploads/attachment/file/182202/20161110_1300h_Es- International Energy Agency, 2017. Energy Policies Beyond IEA trategia_CCTE- 1.pdf Countries: Mexico 2017. OECD/IEA, Paris, France. Secretaria de Energia, 2012. Prospectiva de Energías Renovables International Energy Agency, 2016. Mexico Energy Outlook, OECD/ 2012 – 2026, (Prospects for Renewable Energy 2012-2026), IEA, Paris, France. www.gob.mx/cms/uploads/attachment/ file/62954/Prospec- Vitor, Richard H.K., Sheldahl-Thomason H.; January 2017. Mexico’s tiva_de_Energ_as_Renovables_2012-2026.pdf Energy Reform, Harvard Business School, Paper 717-027. Castro-Alvarez, Fernando, Marsters P. Ponce de Leon Barido D., Comision Reguladora de Energia, 2018 (a). Informe de Labores Kammen, D. 2018. 2016-2017, https://www.gob.mx/cms/uploads/attachment/ Sustainability Lessons from Shale Development in the United States file/276001/Informe_de_Labores_2016_- for Mexico and other _2017_de_la_CRE.pdf Emerging Unconventional Oil and Gas Developers, Renewable and Comision Reguladora de Energia, 2018 (b). Distributed Genera- Sustainable Energy Reviews, 82 (2018), 1320-1332. tion in Mexico and Thoughts on Regulatory Innovation, pre- Zenon, Eric, Rosellon, Juan. 2017. Optimal Transmission Planning sentation at the World Bank workshop “Distributed Energy Under the Mexican New Electricity Market, Energy Policy Journal Markets: International Experience and Impact of Disruptive 104 (2017) 349-360. Technologies”, May 15, Washington DC Kunz, Friedrich, Rosellon J., Kemfert C. 2017. Introduction of Nodal Chanona Robles, Alejandro. 2016. Tracking the Progress of Mexico’s Pricing into the New Mexico Electricity Market Through FTR Allo- Power Sector Reform, Wilson Center, Mexico Institute, https:// cations, The Energy Journal, Vol. 38, SI1. www.wilsoncenter.org/sites/default/files/tracking_progress_ OECD, 2017. Driving the Performance of Mexico’s Energy Regulators, of_mexicos_power_sector_re form.pdf OECD Publishing, Paris. Oxford Institute for Energy Studies, June 2017. Mexican Energy SENER. 2017. Solving Mexico’s Electricity Subsidy and Enregy Pov- Reforms, Issue 109, University of Oxford. erty Granting Solar Bonuses for PV Solar Rooftops, draft paper Government of Mexico, 2016. Mexico’s Climate Change shared by Subsecretaria de Transicion Energetica. Mid-Century Strategy, https://unfccc.int/files/focus/long- World Bank, PPIAF; 2017. Benchmarking Public-Private Partner- term_strategies/application/pdf/mexico_mcs_final_cop- ship Procurement: Assessing Government Capability to Pre- 22nov16_red.pdf pare, Procure, and Manage PPPs, World Bank, Washington DC. MEXICO POLICY NOTES 30 3. Fostering Digital Development I n 2013, the launch of the National Digital Strategy, coupled with a sweeping telecommunica- tions reform, led to great advances in Mexico’s information and communications technology (ICT) sector and in its broader digital economy. Prices for telecommunication services have fallen significantly and access for the population has expanded following the reform. The new “red compartida” it’s a critically important reform and investment initiative that needs to continue implementation. Looking ahead, there still significant challenges and opportunities, including re- ducing access gaps and regional disparities in services. More competition across ICT markets will require further efforts on the regulatory side. A growing digital environment needs the strength- ening of the cybersecurity regulatory framework to protect critical infrastructure and data. Crit- ically, to encourage productive inclusion in an increasingly technologically advanced workplace, the government may want to consider a stronger role of digital technology in the education cur- ricula at all levels. Existing programs to promote technological uptake, including for girls at school and women, would also need to be scaled up. Mainstreaming the current interoperability of sys- tems, digital-identity initiatives, and open data platforms will also be essential. These initiatives could have a large impact in terms of fiscal savings of public resources, improving government transparency and supporting anti-corruption efforts. CONTEXT AND REFORM ment highlighted the ICT reform as part of the Pact for PROGRESS Mexico, an agreement signed by all major political parties before the general elections of December 2012. The tele- MEXICO 56. Information and communications technology plays a vi- communications reforms included the approval of Consti- POLICY NOTES tal role in supporting productivity growth, competitiveness, tutional amendments in June 2013 and the passage of the 33 and productive inclusion. Recognizing the pivotal impor- Federal Telecommunications and Broadcasting Law in July tance of a robust digital ecosystem, the Mexican govern- 2014. The reforms encompassed commendable measures Chapter 3 designed to: (i) increase competition in the telecom sec- 59. The government has also implemented a series of legal tor; (ii) develop fixed and mobile telecommunications in- and regulatory reforms aimed at increasing access to afford- frastructure, including a national backbone for broadband able quality telecom services. Mexico’s revamped regula- networks; and (iii) ensure equal access to telecommunica- tory regime for ICT is built on two pillars, an autonomous tions. While these reforms created a framework for greater and capable regulatory agency with real sanctioning ICT connectivity, a thriving digital economy requires a power and new operating rules for the sector. The new strong regulatory and institutional framework. Policies agency, the Federal Telecommunications Institute (Institu- and regulations need to be fully implemented, updated to Federal de Telecomunicaciones, IFT), has various powers and new ones introduced to keep up with rapidly chang- to restrain “dominant market operators,” companies whose ing technologies and business models. Finally, account- market share exceeds 50 percent. These include: (i) man- able institutions are vital to ensure that new advances in datory interconnection and infrastructure-sharing offers; digital technology empower citizens.58 Mexico’s National (ii) approval over final and wholesale prices that can be Digital Strategy (Estrategia Digital Nacional, EDN) seeks to matched by other competitors; (iii) explicit prohibitions address priority challenges in the ICT sector, focused in on commercial plans differentiating between on-net and the areas of connectivity, public institutions, the economy, off-net prices; and (iv) measures to ensure accounting education, health, and citizen participation. separation, and non-discriminatory network access. These rules have opened telecommunications markets to new 57. In recent years, Mexico has made considerable prog- domestic firms and foreign competitors, partly reducing ress in expanding access to telecommunications services, the dominance of Mexico’s telecom incumbent operator digitization technologies, and the delivery and creation of over the mobile broadband market. The IFT has also es- public services through digital solutions. The 2013 reforms tablished a databank for telecommunications services, established that telecommunications are public services improving transparency, data collection, and analytics. of general interest, and therefore the government must guarantee the universal coverage of high-quality telecom- 60. The authorities have consolidated and strengthened munications services provided through open competition the legal framework for implementing the EDN. The EDN supported by a strong and independent regulatory au- strives to expand digital inclusion to enable all social sec- thority. To achieve this goal, the Secretariat of Communi- tors to benefit from ICTs. The EDN includes two initiatives cations and Transport (SCT) launched the Transportation focused on digital inclusion and digital skills: (i) Mexican and Communications Infrastructure Investment Program Connection Points (Puntos México Conectado), a network (2013-2018), which enabled the creation of a nationwide to increase the standard of living and take advantage of in- Shared Network (Red Compartida)—a wholesale broad- come-generating opportunities provided by ICTs; and (ii) band wireless network that offers highspeed data capac- Code X (Codigo X), a project designed to increase ICT par- ity to other network and virtual operators. When fully ticipation among women and girls.60 Recent modifications implemented, the Red Compartida will extend broadband to the regulations for completing government processes service coverage to 92.5 percent of the population. The and measures to increase data openness have created the operator, Altan Networks, has already reached its first statutory basis for implementing EDN initiatives to pro- milestone for urban service, but the bulk of the total in- mote the growth of the digital economy. In addition, the vestment requirements estimated at US $7 billion is still to government published guides, manuals, and procedures be deployed over the next few years. in the Official Gazette. The procedures are mandatory for all government agencies, while the guides and manuals 58. In February 2017, the SCT announced the Digital Con- provide clarity on compliance with the new procedures. nectivity Program, which will complement the Red Compar- tida by implementing a proposed 10 projects59 focused on the nationwide provision of fixed and mobile telecommu- KEY CHALLENGES nications infrastructure. The Digital Connectivity Program includes (i) a passive infrastructure project that will allow 10,500 public buildings to be rented for telecommunica- Connectivity tions purposes; (ii) the network backbone (red troncal), 61. Prices for telecommunications services have fallen and currently in its design phase, which will create a public-pri- access has expanded, but significant access gaps and dispar- vate partnership to leverage the existing fiber-optic infra- ity remain compounded by competition and regulatory chal- structure of the Federal Electricity Commission (Comision lenges. At the end of 2017, while the National Consumer Federal de Electricidad, CFE) to create a robust national tele- Price Index of Mexico cumulative increased 17.31 percent communications network; and (iii) the Connected Mexico with respect to the end of 2013, the Communications Price (Mexico Conectado) project, which has created 101,322 Wi- Index decreased 23.76 percent in the same period (Fig- Fi hotspots in schools and other public buildings in rural ure 14). Recent reforms and policy actions have boosted and remote areas. the number of households with internet access to 47 per- MEXICO POLICY NOTES 34 58 The World Development Report 2016, “Digital Dividends”: http://documents.worldbank.org/curated/en/896971468194972881/pdf/102725-PUB-Replacement-PUBLIC.pdf 59 Banda Ancha para todos, Programa de Conectividad Digital, SCT, https://www.gob.mx/cms/uploads/attachment/file/207900/PCD_FINAL .pdf 60 Education characteristics of Mexican population. INEGI, 2014 Figure 14: Drop in Prices in telecommunications services following the reforms Chapter 3 Mexico price Index, Communications price Index, Fixed Internet price Index, and Mobile Telephony price Index by year 150 100 50 0 2010 2011 2012 2013 2014 2015 2016 2017 Mexico Price Index Communications Price Index Fixed Internet Price Index Mobile Telephony Price Index Source: OECD, Internet access. Total, % of all households, 2005 – 2017 Figure 15: Household Internet Access Rates Figure 16: Fixed and Mobile Broadband among OECD Countries, 2010 and 2016 Penetration in Mexico by State, 2016 Colombia Colombia Chiapas Chiapas Mexico Mexico Oaxaca Oaxaca Tabasco Tabasco Brazil Brazil Campeche Campeche Costa Rica Costa Rica Hidalgo Hidalgo Chile Chile Guerrero Guerrero Lithuania Lithuania Veracruz de Veracruz de Ignacio de la Ignacio de Llave la Llave Zacatecas Zacatecas United States United States Tlaxcala Tlaxcala Portugal Portugal Quintana Roo Quintana Roo Turkey Turkey Durango Durango Latvia Latvia Michoacan de Michoacan Ocampo de Ocampo San Luis San Potosi Luis Potosi Slovenia Slovenia Yucatan Yucatan Italy Italy Chihuahua Chihuahua Hungary Hungary Nayarit Nayarit Poland Poland Puebla Puebla Guanajuato Guanajuato Czech Republic Czech Republic Aguascalientes Aguascalientes Spain Spain Sinaloa Sinaloa Belgium Belgium Tamaulipas Tamaulipas Austria Austria Baja California Baja Sur California Sur Sonora Sonora France France Morelos Morelos Estonia Estonia Jalisco Jalisco Ireland Ireland Colima Colima Germany Germany Coahuila de Coahuila Zaragoza de Zaragoza Baja California Baja California United Kingdom United Kingdom Queretaro Queretaro Netherlands Netherlands Nuevo Leon Nuevo Leon Norway Norway Distrito Federal Distrito Federal Korea Korea Mexico Mexico 0 0 20 20 40 40 60 60 80 80 100 100 120 120 0 0 20 20 40 40 60 60 80 80 100 100 120 120 Teledensidad de Teledensidad Internet móvil de Internet móvil en en líneas líneas por cada 100 por cada hogares 100 hogares 2016 2016 2010 2010 Accesos de Accesos de BAF residencial por BAF residencial cada 100 por cada hogares 100 hogares Source: IFT, Banco Datos Telecomunicaciones. Price Index Base Year 2013 Source: IFT, Banco de Informacion de Telecomunicaciones, 2016 cent in 2016. However, Mexico’s access rate remains well communications sector and competitors seeking access to below the OECD average of 82 percent in 2015, as well as unbundled local loops.61 This continued asymmetry partly the averages for comparable countries (Figure 15). Within reflects delays by the market dominantoperator in imple- Mexico, access rates vary substantially by state (Figure 16). menting the Electronic Management System (EMS). These Despite Mexico’s successful regulatory reforms, limited delays stem from a slow and burdensome IFT processes competition continues to constrain access to high-quality and the failure of the preponderant operator to meet IFT broadband services. deadlines. When complete, the EMS will provide conces- sionaires and other firms with complete and updated 62. The IFT has acknowledged that information asymme- information on the preponderant operator’s network and MEXICO tries persist between the incumbent operator in the tele- those facilities subject to shared access or co-location. POLICY NOTES 61 4OECD Telecommunication and Broadcasting Review of Mexico 2017: http://www.oecd- ilibrary.org/docserver/download/9317021e.pdf?expires=1516744894&id=id&accname=gue st&checksum=FB2AFDBD0F82198 E03F3BB3BB7E92B19 35 Chapter 3 Figure 17: Market Share by type of Market, 2017 1% 2% 2% 5% 14% 15% 14% 55% 70% 22% America Movil Telefonica America Movil Grupo Televisa Megacable-MCM AT&T Other Totalplay Axtel Other Source: IFT, Banco de Informacion de Telecomunicaciones, 2016 63. Expanding access to digital services intensify risks to cy- tion between government institutions. In 2011, the govern- bersecurity, and Mexico will have to take the necessary steps ment signed the Interoperability and Open Data Scheme to protect its critical infrastructure. Mexico experiences the for the Federal Public Administration Agreement (Esquema second- most cyberattacks in Latin America, and the num- de Interoperabilidad y de Datos Abiertos de la Administración ber of attacks increased by 40 percent between 2013 and Pública Federal, EIDA), which defined the regulatory and 2014. Cyberattacks affected approximately 10 million peo- policy framework for integrating processes and sharing ple in 2014, and each year they impose an annual cost of platforms among federal government institutions. The between US$3 and US$5 billion dollars.62 Mexico’s Nation- EIDA designated six “trusted sources” of personal identifica- al Cybersecurity Strategy (Estrategia Nacional de Cibersegu- tion that can be used to complete government procedures ridad, ENC) calls for the Scientific Unit of the Federal Police online. These trusted sources include the Unique Popula- and the National Response Center for Cyber Incidents to tion Registration Code (Clave Única de Registro de Población, launch public awareness campaigns focusing on cyberse- CURP), birth certificates, and e-signatures. Trusted sources curity issues. The ENC’s regulatory framework has yet to allow agencies to share information through standardized be elaborated, and several important challenges inhibit fields established as part of the government’s digital proce- its implementation: limited capability from public and pri- dures. The CURP can now be used to complete 44 govern- vate sectors and the lack of a regulatory framework, the ment procedures. Three procedures do not need a physical critical infrastructure, and a harmonization between fed- birth certificate, and 478 government procedures can be eral and local policies. completed with an e-signature. 66. Once the basis for interoperability has been established, Digital Inclusion and Digital Skills the challenge is to incorporate an increasing number of pro- 64. Mexico’s key challenge in the areas of digital inclusion cedures and expand access to digital public services. Trusted and digital skills will be to scale up current initiatives, ensure sources such as the CURP or e-signature greatly expedite their sustainability, and replicate their success at the subna- government processes, yielding substantial savings for tional level. Two years after their establishment, the Puntos both citizens and the public administration. México Conectado centers are approaching their maximum capacity of 400,000 registered members. Meanwhile, the Codigo X program has thus far reached only about 1,000 The Legal Framework and Open women and girls. Data 67. The new legal and institutional framework for digital government requires continued work to keep up with grow- Interoperability and Digital ing digital economy. The legal and institutional framework Identity increased the volume of digital procedures and services 65. Interoperability is necessary to create a unique digital and created rules, guidelines, standards, and mechanisms MEXICO POLICY NOTES identification system and facilitate the exchange of informa- to solve technological problems in each agency. Continu- 36 62 OAS study Security Trends in Latin American and the Caribbean. ity and compliance will be required to further elaborate 70. Adopt new administrative procedures to increase com- Chapter 3 the legal framework and continue the process of digitizing petition and reinforce the innovative regulatory framework. procedures and services. A long-term digital government The level of competition in wholesale markets directly policy will need to be continuously updated, and mecha- influences the retail price of internet services. As in other nisms to ensure compliance among subnational govern- ICT subsectors, information asymmetry and market dom- ments will be necessary to consolidate the use of digital inance by the incumbent can diminish competition. The procedures and services by the government, the private government should analyze the causes of information sector, and civil society. Maintaining a strong regulatory asymmetry in the internet-service subsector, the obstacles framework is vital to the development of the digital econ- to establishing an EMS, and the methodology for estab- omy and to promoting inclusion, connectivity, civic inno- lishing rates for wholesale services. The authorities should vation, open contracting, and other objectives. then consider imposing additional conditions to the dom- inant market operator designed to decrease tariffs and 68. The quality and quantity of open data must be signifi- expand access to essential network facilities. cantly improved and needs to be utilized more effectively to enable the growth of the digital economy and enhance digi- 71. Obligate infrastructure sharing across public and private tal government services.63 The General Directorate of Open passive infrastructure to lower costs and increase access to Data (Dirección General de Datos Abiertos, DGDA) was cre- broadband services. Major cities in Italy, Japan, Korea, ated to implement Mexico’s open data policy, which was Spain, and Sweden successfully employed public-private established in 2013 as part of the EDN. The open data pol- partnerships to leverage the full value of public infrastruc- icy both mandates availability of data in open standards, ture. Maximizing the functionality of existing ICT networks and facilitates the link between the demand and supply and related infrastructure can reduce spending on public of open data. However, the quality and quantity of open works and substantially reduce retail costs. data is lower than in OECD countries leading this Agen- da, and Mexico’s private sector and civil society should be 72. Strengthen cybersecurity regulatory framework, in- strengthened to use open data effectively. Certain data- crease public awareness and prioritize implementation. The bases are not consistently available in open data standards government may address this challenge by: (i) launching to the public, and technical constraints inhibit theuse of a national cybersecurity awareness campaign designed open data at the state level. Moreover, public institutions, to encourage stakeholders to take steps to protect infor- civil-society organizations, journalists, the private sector, mation collected and shared through servers, networks, and academia have relatively little experience using open transactions, and other means; promoting stakeholder data and are therefore they are not able to leverage its full investment in designing a national cybersecurity training potential. Only nine of Mexico’s 32 states are members of program; and establishing appropriate data-privacy regu- the Open Data Network, and the sophistication of data lations. openness varies widely across local governments. Digital Inclusion and Digital Skills POLICY OPTIONS 73. Promote technological uptake of the Puntos Mexico Conectado program by increasing the availability of internet services in public spaces and providing training in digital Connectivity skills. To increase the value of Mexico’s existing ICT infra- 69. Revise the Universal Service Strategy to reduce coverage structure, the government should strengthen the digital disparities between rural and urban areas. The SCT and the inclusion strategy and work with state governments to IFT should jointly analyze alternative options for providing ensure its sustainability and increase its impact. To en- coverage in rural areas and incorporate them into a new courage productive inclusion in an increasingly techno- Universal Service Strategy. Universal service policies aim logically advanced workplace, the government may want to facilitate telecommunications access in areas where to consider a stronger role of digital technology in the ed- the costs of providing service may exceed the commercial ucation curricula at levels, starting from early childhood. returns. Universal service frameworks can also advance It could redefine the strategy for encouraging women to specific policy objectives by providing services to target pursue careers in science, technology, engineering, and groups, such as low-income households or people with mathematics. disabilities. The decision-making process should clearly define the type of services required, establish whether Interoperability and Digital market mechanisms will be able to provide those services Identity at affordable rates, assess the merits of imposing a univer- sal service obligation relative to alternative policies, and 74. Scale up the current interoperability and digital-identity evaluate those alternatives in light of broader policy ob- initiatives to include more government procedures and expand MEXICO jectives and other ongoing programs.64 the availability of e-government services. Subnational admin- POLICY NOTES 63 64 OECD, Open Government Data Review of Mexico, 2016 UNIVERSAL SERVICE POLICIES IN THE CONTEXT OF NATIONAL BROADBAND PLANS, OECD, https://one.oecd.org/document/DSTI/ICCP/CISP(2011)10/FINAL/en/pdf 37 Chapter 3 istrations are responsible for collecting information crucial corruption. The data generated by Mexico’s subnational to the delivery of public services. To promote the further administrations have great value to citizens, NGOs, the pri- digitization of public services, subnational administrations vate sector, and other public administrations. A national should be integrated into the existing Federal Government open data program should include: (i) the development interoperability framework. The government should analyze of a guide for integrating the “open data enabler” into the most valuable public services used by citizens and the state and municipal development plans and creating the private sector and harmonize its digitization mechanisms at necessary legal framework; (ii) measures to promote the the national level. A clear case for this, which would reduce development of local open data regulations; (iii) an analy- overlapping of benefits and improve public spending effi- sis of existing government funds that subnational admin- ciency, is using ICT to support a unique national identifier istrations can use to implement open data policies; (iv) the to the beneficiaries of social assistance and social insurance sharing of lessons learned by states and municipalities programs (see Note on Social Protection). that have implemented open data initiatives; and (v) col- laboration between subnational governments and inter- national experts in open data and e- government services. The Legal Framework and Open Data 76. Increase quantity and quality of open data and stim- 75. Design and implement a national open data program ulate demand for its effective use. Open data programs in collaboration with subnational administrations. Overall, are most successful when they adopt an “ecosystem” ap- these type of data platforms at the federal and sub-na- proach, under which governments invest not only in sup- tional levels would be key to support the government plying data but also in building the capacity of the public transparency agenda, and could be used effectively by the sector, priorities and data demands of different groups; (iii) newly established anti-corruption agency technical sec- build the technological capacity of the public and private retariat in its monitoring and risk assessment functions. sectors; (iv) support existing open data communities; and Moreover, open data, in which Mexico has already made a (v) increase data literacy among potential users, including lot of progress, can help streamline government services, journalists, nonprofit organizations, academics and soft- encourage innovation, improve public safety, and reduce ware developers. MEXICO POLICY NOTES 38 Chapter 3 Annex 2. Broadband Subscriptions Figure 18 Fixed Broadband subscriptions, OECD countries (June 2017) Fixed Broadband subscriptions, OECD countries (June 2017) 1.2.1. OECD Fixed broadband subscriptions per 100 inhabitants, by technology, June 2017 50 45 40 35 30 25 20 15 10 5 0 Mexico Denmark Korea United Kingdom Czech Republic Slovenia France Netherlands Iceland Belgium Sweden Canada Luxembourg Greece United States New Zealand Japan OECD Hungary Estonia Ireland Austria Finland Lithuania Israel Latvia Slovak Republic Poland Colombia Switzerland Norway Germany Portugal Australia Spain Italy Chile Turkey DSL Cable Fibre Satelite Fixed wireles Other Figure 19 Mobile Broadband subscription, OECD countries (June 2017) Mobile Broadband subscription, OECD countries (June 2017) 1.2.2. OECD Mobile broadband subscriptions per 100 inhabitants, by technology, June 2017 180 160 140 120 100 80 60 40 20 0 Mexico Greece Czech Republic Canada Japan United Kingdom Finland Estonia United States Denmark Sweden Latvia Iceland Korea OECD Ireland New Zealand Norway Spain Poland Switzerland Netherlands Austria Lithuania Italy Luxembourg Chile France Slovak Republic Germany Belgium Turkey Israel Slovenia Portugal Hungary Colombia Australia Data and voice subscriptions Data-only subscriptions Total (where breakdown not available) MEXICO POLICY NOTES 39 4. Developing a Deeper, More Inclusive Financial Sector R eforms over the last 4 years in the sector have been welcome and highly positive in deep- ening of the financial sector. Credit (as a share of GDP) to the private sector increased and people with bank account increased by more than 10 percentage points due to the reforms. Yet there is a way ahead to implement the reforms and pass further reforms. Credit to the private non-financial sector is just 41.6 percent of GDP, well below the 72.5 percent average for the other LAC-5 countries and far below the 143.2 percent average for emerging markets world- wide. Moreover, a large share of the population still lacks access to financial services, perpetuat- ing economic inequality (37 percent of adults had an account in a formal financial institution in 2017). MSMEs provide 71 percent of the employment, but only 20 percent of them use bank credit due to affordability and access issues. The country’s financial sector remains highly concentrated, and stronger competition may lead to more efficient intermediation. To fill the gaps the 13 DFIs owned by the federal government have grown rapidly and now support one-third of total credit to the private sector. Despite progress the authorities would need to take further measures to maximize the financial sector’s contribution to inclusive growth, which include: (i)strengthening and consolidating NBFIs, including cooperatives; (ii) developing financial infrastructure includ- ing the development of asset-based lending (ABL) to expand access to finance for MSMEs; (iii) incentivizing banks to pilot ABL programs and products, for example with DFIs offering a second- loss partial credit guarantee; (iv) shifting from quantitative lending targets in DFIs to a system based on multiple development outcome indicators to sharpen efficiency in achieving intended objectives; (v) placing greater emphasis in crowding-in private financing by adjusting the focus of DFI instruments which can also be used to bring in the capital markets. Additionally, there is a great source for growth and inclusion that needs to be harnessed. Mexico has one of the largest MEXICO POLICY NOTES financial-technology subsectors in Latin America. To foster financial inclusion and access through this new and vibrant segment, the issuance of secondary regulations to the Fintech law and the adoption of adequate oversight mechanisms should be given a high priority. 41 Chapter 4 Figure 20 Private Non-Financial Sector Credit to GDP 150 125 100 75 50 25 0 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 Mexico Argentina Brazil Colombia Emerging markets Chile Turkey Russia South Africa Source: Bank of International Settlements 79. Recognizing the role of finance for development, re- CONTEXT AND REFORM forms have aimed to expand financial services. In particular, PROGRESS the 2014 Financial Reform Law67 increased the role of fed- eral Development Finance Institutions (DFIs). Moreover, 77. A deeper Mexican financial sector would help increase the 2016 National Financial Inclusion Policy and the 2017 productivity and reduce inequality. A well-functioning fi- National Financial Education Strategy included reforms nancial sector transforms savings into productive invest- designed to: (i) strengthen the institutional capacity of ment, manages a range of economic risks, and provides the National Banking and Securities Commission (Comis- the necessary infrastructure for financial transactions. ión Nacional Bancaria y de Valores, CNBV); and (ii) disburse government payments and transfers from the largest so- 78. Mexico’s financial sector remains small relative to the cial- protection programs through dedicated accounts. country’s level of development. It allocates only a moderate Complementary efforts to address gaps in the financial in- amount of capital and is dominated by a banking sector frastructure, accelerate the rollout of new products, facili- that is too small for the size of the economy. Credit to the tate the spread of access points, increase the transparency private non-financial sector amounts to just 42.9 percent of financial services, and enable large-volume payments of GDP (Figure 20), of which bank lending represents would enhance the impact of recent reforms. 19.4 percentage points.65 This level is well below the 72.5 percent average for the other LAC-5 countries and far 80. Recently, on March 2018, Mexico adopted an umbrel- below the 143.2 percent average for emerging markets66 la law on Fintech (Ley para Regular las Instituciones de worldwide. A large share of the population lacks access to Tecnología Financiera). The new law is the first one of its financial services, perpetuating economic inequality. Only kind worldwide and provides a regulatory framework for 29 percent of adults in the bottom four income deciles the authorization, operation and supervision of Fintech have an account at a financial institution, below the LAC institutions (Instituciones de Tecnología Financiera, ITFs) fo- average of 41 percent and the upper middle-income av- cusing on two particular types: crowdfunding institutions erage of 63 percent. Although Mexico’s pension system and electronic payment funds institutions. The Law also covers 73 percent of the population, contributory pension introduces the legal underpinnings for a regulatory sand- schemes cover just 26 percent, limiting the size of pension box environment for innovative companies, the concept savings. Similarly, the size and depth of capital markets are of open data for non-confidential aggregate data and for well behind levels in peer countries. Whereas responsible transactional data with consumers’ consent through the macroeconomic management and a resilient financial Application Programming Interfaces (APIs) and a provi- system have helped maintain financial stability in Mexico, sion to recognize virtual assets and regulate their usage oversight authorities lack independence, which could im- and operation in Mexico. The Fintech Law provides an pede their ability to manage stress opportunity to expand access to financial services while enabling innovation in the industry. MEXICO 65 According to the Bank of International Statistics (BIS), as of Q4 2017. Credit is provided by domestic banks, all other sectors of the economy and nonresidents. The private non-finan- POLICY NOTES cial sector includes non-financial corporations (both private-owned and public-owned), households and non-profit institutions serving households as defined in the System of National Accounts 2008. In terms of financial instruments, credit covers loans and debt securities 66 Emerging markets in this note comprise Argentina, Brazil, Chile, China, Colombia, the Czech Republic, Hong Kong SAR, Hungary, India, Indonesia, Israel, Korea, Malaysia, Mexico, Poland, Russia, Saudi Arabia, Singapore, South Africa, Thailand and Turkey 42 67 The 2014 financial reform was based on four main pillars: (a) increase competition in the financial sector by inhibiting anti-competitive practices; (b) encourage credit through devel- opment banks by strengthening their legal framework, mandate and operating capacity; (c) explore ways that financial authorities can encourage private financial institution to channel more credit to productive activities and; (d) maintain a stable and solid financial system. Figure 21 Share of Adults with an Account at a Formal Financial Institution, 2017 (%) Chapter 4 90 80.2 80 74.3 75.8 70.0 69.2 68.6 70 60 54.4 48.7 50 45.8 40 36.9 30 20 10 0 Argentina Brazil Chile China Colombia Mexico Russia South Turkey Regional Africa Median Source: Global Findex database 2017 nance and financial inclusion. Altogether, pension funds, KEY CHALLENGES mutual funds and insurance account for 37 percent of fi- nancial sector assets. 81. Despite significant reform progress, challenges remain ahead. The key challenges facing the Mexican financial 84. Banking infrastructure in Mexico is heavily concentrat- sector fall under five broad categories. They are: (i) ex- ed in urban areas. Only 29 percent of the rural population panding financial inclusion; (ii) providing credit and fi- has an account at a formal financial institution, far below nancial services to micro, small and medium enterprises the LAC average of 41 percent. As in other LAC countries, (MSMEs); (iii) strengthening financial-sector oversight;68 Mexican banks have used agent banking to reach areas (iv) enhancing the efficiency of DFIs; and (v) boosting the where a traditional branch would not be viable. However, supply of long-term savings and investment. Mexico’s agent-banking model remains focused on urban areas70 and relies heavily on large retail chains rather than Financial Inclusion small shops. Agent- banking services also tend to be lim- ited to taking deposits and accepting loan repayments. 82. Expanding financial access among lower-income house- There is also a large gap in coverage of agents between holds remains a top priority. Mexico’s unbanked popula- northern Mexico and the center and south of the country. tion is far larger than the country’s level of economic and Within the five states with the largest number of commis- financial-sector developmentwould predict. The share sion agents per adult, four are located in the north of the of adults with an account at a formal financial institution country. In contrast, the five states with the smallest num- slightly decreased from 39 percent in 2014 to 37 percent ber of agents per adult are located in the south and center in 2017, far below the level of most comparator countries of the country. (Figure 21). Among households in the bottom 40 percent of the income distribution, 74 percent of adults remained 85. Mexico’s financial sector is concentrated, and stronger unbanked in 2017.69 Developing these initiatives, includ- competition may lead to more efficient intermediation. The ing in urban microfinance, among others, requires adopt- financial sector includes a diverse range of financial inter- ing responsible finance policies and a framework for finan- mediaries and many small institutions, but assets remain cial education and consumer protection. concentrated in large banks and financial conglomerates, and intermediation margins have been rising. The seven 83. Banks dominate Mexico’s financial sector, but in recent largest banks account for about 80 percent of total bank years the growth of the nonbank sector has outpaced the assets, and the sector is concentrated relative to Latin growth of banks. Commercial banks account for over 41 American peers. With economies of scale, more concen- percent of the financial sector’s assets, and development tration does not necessarily lead to less efficient out- banks represent another 10 percent (Figure 22). Regulated comes, and various measures of competition estimated and unregulated non-deposit-taking financial companies over 2012-15 illustrate the competitive conduct of banks. (sociedades financieras de objeto múltiple, SOFOMES), sav- The H-statistic, a commonly used measure of output price ings and credit institutions, and deposit warehouses are elasticity with respect to input prices, is higher than in small, representing a combined 6 percent of the financial other major Latin American markets suggesting compet- sector’s assets, yet they play an important role in microfi- itive conduct is strong. Meanwhile, a Lerner index higher MEXICO POLICY NOTES 68 Including in emerging areas such as Fintech, as well as in financial consumer protection, among several others. 43 69 According to Mexico’s National Financial Inclusion Survey (Encuesta Nacional de Inclusión Financiera), between 2012 and 2015 the share of adults with a transaction account (“cuen- ta de depósito” or “cuenta de ahorro”) at a regulated financial-services provider increased from 35.5 to 44.1 percent. 70 Agent banking only covers 19 percent of rural areas. Chapter 4 Figure 22: Financial Sector Structure Insurance companies Development banks 6.2% 10.2% Deposit warehouses 0.1% Economic development funds 1.1% Pension funds Commercial banks 18.7% 41.8% Brokerage firms 3.0% Sofomes 4.9% Mutual funds Popular savings and credit entities 13.0% 1.0% Source: World Bank staff calculations based on data from Banxico. than in other major Latin American countries suggests have not yet recovered after the global financial crisis that mark ups are high. The Boone indicator, a measure and have been unable to access private credit. Mexico’s of profit elasticity with respect to marginal costs, is not nonbank financial institutions (NBFIs) have struggled to as strong as in the peer countries suggesting significant overcome efficiency issues and have not taken advantage room for competition improvements. The diverging con- of the agent-based models. Popular financial companies clusions from measuring different aspects of competition (sociedades financieras populares, SOFIPOs) face high reg- illustrate its complex nature and impact on efficient out- ulation costs and have generally been unable to increase comes. The country has liberalized entry both by allowing profitability and would likely benefit from greater consol- foreign banks into the market, allowing small non-bank idation, while many savings and loan cooperatives (so- intermediaries to operate, and by introducing niche bank ciedades cooperativas de ahorro y préstamo, SOCAPs) face licenses that allow for the entry of less complex banks with competitiveness challenges and would benefit from more commensurately lighter regulatory burdens. professional management. 86. Mexico has one of the largest financial-technology sub- sector in Latin America, with 238 registered startups in 2017. Financial Services for MSMEs Startups in Mexico are concentrated in lending (23% of 88. MSMEs remain underserved by the formal financial sec- startups), payments and remittances (22%), enterprise and tor. Mexican MSMEs provide 71 percent of employment personal financial management (15 and 10%, respective- and earn 51 percent of business income. However, 73 ly), crowdfunding (9%), and insurance (6%). The latter is percent of MSMEs are informal, and only 20 percent use on the rise, leading the highest growth segments in 2017. bank credit. MSMEs receive just 11 percent of all commer- According to the Finnovista report, particular features of cial bank credit, a significant share of which is backed by the Mexican market that make it a fertile environment for DFI guarantees. MSMEs often lack access to appropriate fintech include: (i) a high Internet and smart mobile devic- types of credit, as high levels of credit risk limit the range es penetration; (ii) a strong ecosystem of entrepreneurship of products available to MSMEs. and e-commerce; (iii) a low banking penetration, and; (iv) an undeveloped consumer lending offer. Most fintech 89. The 2016 Financial Sector Assessment Program found startups are currently concentrated around Mexico City, that although progress has been made in easing financial Monterrey and Guadalajara. Before the Fintech Law, non- constraints among MSMEs, challenges remain in terms of banks were not allowed to issue mobile money accounts. affordability and access. MSMEs with access to finance re- With the necessary legal framework and physical infra- ported facing relatively favorable credit conditions, espe- structure in place, both the e-money issuers created by cially for credit backed by DFI guarantees, with relatively the Fintech Law (i.e. the IFPEs) as well as banks will be in a low interest rates and average maturities of 24-36 months. better position to develop products and business models However, MSMEs that were unable to obtain credit cited that can reach the unserved in remote areas, most likely high interest rates and cumbersome procedures as bind- through mobile money eco-systems. ing constraints. MSMEs with relatively few employees and MEXICO low debt levels, and those in the initial stages of develop- POLICY NOTES 87. Small nonbank financial institutions have the poten- ment, reported facing significantly less-favorable condi- tial to expand financial-service provision to lower-income tions than larger, more well-established MSMEs. Pervasive 44 groups, but their small size, limited funding capacity, and informality further hinders access to finance in the MSME low profitability remain key challenges. Many SOFOMES sector. Figure 23: DFI-Backed Credit Chapter 4 10 9 8 7.7 7.4 7.9 7 7.5 7 7.2 6.8 6.2 6.5 6 6.2 5 2013 2014 2015 2016 2017 2018 Original target path Actual Source: SHCP and Diario Oficial de la Federación, December 16, 2013, third section. extent to which DFIs (i) reach the segments of the financial Financial-Sector Oversight market they aim to support, (ii) effectively balance costs 90. Responsible macroeconomic management and a and benefits, (iii) promptly phase out obsolete programs, resilient financial system have helped maintain financial (iv) crowd out private finance, or (v) contribute to the ex- stability in Mexico. The 1994 crisis underscored the impor- cessive accumulation of risk. Moreover, the number of fed- tance of a stable financial system to Mexico’s economic eral DFIs is high relative to emerging market peers. They performance. Five of the country’s seven largest banks have been established at different points in time to serve are subsidiaries of large international banking groups, objectives considered important at the time. However, the which helps mitigate domestic risks, but also increases large number of DFIs reduces the flexibility to reorient the exposure to external shocks, as the 2008 global financial DFI activities to current priorities. crisis demonstrated. These institutions form the core of systemic risks in the country. No major financial institution has failed in recent years, but a large number of housing- Other Issues finance providers struggled in the aftermath of the global 92. An inadequate supply of long-term savings and financing financial crisis, and many smaller institutions remain finan- instruments constrains investment options for pension-fund cially weak despite generally favorable economic condi- managers and limits the availability of financing for large tions. Both institutional capacity and independence are infrastructure projects. Capital markets in Mexico are com- critical to oversight authorities’ ability to handle threats paratively shallow vis-à-vis peer countries and their further to financial stability. Whereas the Central Bank enjoys sub- deepening would benefit areas of the economy and the stantial independence, the CNBV has more limitations on financial system. As well, Mexico’s pension funds have too this front. few contributors, contribution amounts are insufficient, and asset returns will not provide an adequate income level for future retirees. Under the defined-contribution Development Finance Institutions schemes, pension payments are expected to replace just (DFIs) Efficiency 28-34 percent of labor income. Pension assets are exces- 91. The 13 DFIs owned by the federal government have sively invested in government instruments, which entail grown rapidly and now support one- third of total credit to little risk but offer low returns. Meanwhile, infrastructure the private sector. Following the 1994 crisis, the govern- projects that could yield significantly larger returns have ment liberalized financial markets and eased restrictions difficulty obtaining funding. Some progress has been on the entry of foreign banks, yet these measures failed to made in developing corporate equity and debt markets, adequately deepen the financial sector, and the govern- but equity markets and specialized funds for financing in- ment responded by enhancing the role of DFIs. New leg- frastructure remain small, expensive, and illiquid. Despite islation in 2013 facilitated a rapid expansion in DFI credit its potential to help close Mexico’s large infrastructure gap, and set ambitious quantitative financing targets (Fig- the market for infrastructure-related bonds is not fully de- MEXICO ure  23). These targets, however, created incentives to in- veloped. Infrastructure bonds can consolidate public and POLICY NOTES crease the volume of lending without adequate emphasis private financing into a single instrument, and projects 45 on development impact and on associated risks including that generate a direct financial return (e.g., via user fees the risk of crowding out private finance. It is unclear the or other tariffs) are especially well suited to bond issues. Chapter 4 ly-warning indicators and improving coordination between POLICY OPTIONS the CNBV and auxiliary supervisory bodies. Financial Inclusion Financial Services for MSMEs 93. The issuance of secondary regulations to the Fintech law 96. Development of financial infrastructure through as- and the adoption of adequate oversight mechanisms should set-based lending (ABL) could expand financial access among have a high priority. The Fintech Law provides an opportunity MSMEs. The government’s financial inclusion policy calls for to enhance the enabling environment for financial inclusion developing private credit products aimed at underserved through the adoption of innovative business models that market segments, particularly MSMEs. ABL can advance this have broad reach. The authorities have outlined a roadmap objective by providing MSMEs with revolving lines of credit for secondary regulations to follow the implementation of the to cover startup and operational costs on an ongoing basis. Law within 6, 12 and 24 months. Two cross-cutting issues for However, ABL requires a modern financial infrastructure, secondary regulation that need to be taken into consideration including: (i) a clear legal framework that allows for flexible are: (i) monitor and protect against regulatory arbitrage and transactions while maintaining legal certainty between the maintain a level playing field especially between Fintech and parties involved; (ii) an efficient system for registering inter- incumbent entities; and (ii) ensure consistency across second- ests in collateral and that provides priority to the first party ary regulations associated with various financial sector laws. to register as against subsequent third-parties; and (iii) en- Given that Fintech firms may operate within the mandates of multiple agencies, strong coordination mechanisms on a forcement mechanisms that enable creditors to recover the bilateral basis between agencies as well as in relation to two outstanding amount of the loan by realizing the collateral in new bodies created by the law (the Fintech Innovation Group an extrajudicial manner after a debtor’s default.71 Emerging and the Interinstitutional Committee) would be key. Moreover, areas to include, among others, include the use of electronic strong focus is needed to develop internal capacity, processes invoices by MSMEs and Fintech firms. and procedures for licensing, monitoring and supervising fintech services and providers under the new legal and reg- 97. The government could incentivize banks to pilot ABL ulatory framework. In addition, recent developments across programs, for example by having DFIs offer a second-loss the world and in Mexico, show the need for a comprehensive partial credit guarantee (SLPCG). This guarantee would al- cyber security framework for the financial sector industry. low banks to develop and introduce ABL type products, putting Mexico’s legal, registry and enforcement reforms 94. To facilitate financial access, development of remote ac- to the test, while providing a safety net for lenders in case cess channels, in particular those supporting the widespread elements of the system do not perform as designed. The operation of mobile banking, should be prioritized, while at SLPCG would become eligible after the bank exhausts its the same time seeking enhanced coverage of bank, non-bank rights and remedies under the secured transactions sys- and third-party physical access points. An appropriate com- tem, standardizing the value of movable collateral and bination of physical access points and remote channels, hedging against both enforcement risks and risks that the and a balance between in-house services and third-party collateral’s liquidation value will be less than its assessed providers, can decrease both time and travel costs on the value. These measures should be accompanied by efforts demand side and transaction and overhead costs on the to raise awareness within the financial sector regarding supply side, enabling financial firms to sustainably expand ABL process and requirements, and the authorities could operations in remote and underserved areas of the coun- offer advisory assistance to financial institutions interest- try. Streamlining the regulatory and policy framework and ed in piloting ABL credit programs for SMEs. facilitating the adoption of enhanced business models could mitigate distortions and ease constraints on the ex- Financial-Sector Stability pansion of financial access points and remote channels. 98. Increasing the institutional, supervisory, and budget- 95. The authorities should continue to strengthen and consol- ary independence for some financial-sector regulators is idate NBFIs, including cooperatives. Increasing the aggregate important. Whereas the Central Bank rightly enjoys sub- scale of the SOFOMES, SOFIPOs, and SOCAPs sector would stantial independence, which should be upheld, other lower costs and expand the range of financial products and key financial-sector agencies like CNBV, the National services. Priority measures include: (i) swiftly implementing Commission for the Retirement Savings System (Comisión corrective actions through the merger, acquisition, or liqui- Nacional del Sistema de Ahorro para el Retiro, CONSAR), the dation of NFBIs when improprieties and regulatory breaches National Insurance and Bond Commission (Comisión Na- are detected; (ii) requiring NBFIs to develop multiyear strate- cional de Seguros y Fianzas, CNSF), and the Institute for the gic plans; and (iii) tightening supervision by developing ear- Protection of Bank Savings (Instituto para la Protección al MEXICO POLICY NOTES 71 Mexico is advancing reforms in all three areas: (1) reforms to the secured transactions law took place between 2002 and 2014; (2) creation of the collateral registry tool place in 2010; and (3) development of summary and extrajudicial remedies took place in 2014. To determine whether these collective reforms have reached a level for the introduction of ABL the IFC consulted three top banking law firms in 2017 regarding the legal feasibility of ABL. All three broadly agreed that ABL is now possi- 46 ble in Mexico, though additional legal and regulatory reforms could facilitate its introduction. In addition, the IFC also commissioned an operational and technological diagnostic of the movable-collateral registry (“Registro Unico de Garantias”) to determine if its structure and features are also presently suited for the introduction of ABL. The National Law Center consultancy, which performed the diagnostic, concluded that the system as it stands supports the introduction of ABL, although further technological updated are necessary. Ahorro Bancario, IPAB) have limited autonomy. Although it 103. DFIs could reform their products to foster a more di- Chapter 4 is funded by supervisory fees, the CNBV’s budget is subject verse and effective infrastructure- finance market. These to SHCP approval. The longstanding salary freeze may im- include state banks such as the National Bank of Public pede the CNBV’s ability to retain high-quality staff which Works and Services (Banco Nacional de Obras y Servicios Pú- is in high demand in this complex sector. Moreover, CNBV blicos, BANOBRAS) and others. Consolidating loan-guaran- staff receive inadequate legal protections. tee products into basic guarantee streams could help in- vestors better understand the products which can be sep- arated by creditor financing risks versus project sponsor DFI Efficiency contractual risks. Creating refinancing facilities could help 99. Consolidating DFIs could help align their activities with as a mode of providing contingent loan guarantees for fol- current development finance priorities and increase their low up funding. The use of guarantee instruments could efficiency. For example, when Infonacot was established to help promote continued investment financing through lend to low-income workers, it filled a gap in the market, capital market instruments after short-to-medium-term but payroll deduction loans are now widely available from bank credits expire. Finally, expanding the use of first-loss commercial banks a merger with BANSEFI could be consid- guarantees to creditors could incentivize bond investors ered. Similarly, a merger between NAFIN and Bancomext to invest in public-private partnerships. These instruments has been considered in the past and could be re-evaluated. would promote better use of commercial long-term credit rather than DFI credit competing with the private sector. 100. Shifting from quantitative lending targets to a system based on multiple outcome indicators could sharpen effi- ciency incentives among DFIs. Refining the overarching Other Issues strategy and objectives for DFIs could increase their effi- 104. Focusing more on higher-yield investments while reducing ciency and enhance their impact. Ongoing evaluations, administrative costs would help in the aim that future retirees pro-transparency efforts, and improved coordination have better replacement rates. Given low anticipated in- could provide the basis for a revised DFI strategy. The come-replacement rates in the defined-contribution pension results framework for DFIs should be expanded beyond systems, retiring contributors will likely experience a sharp lending volume to include financial-inclusion indicators, drop in income. This could present a serious fiscal liability if the number of new borrowers, and other outcomes di- political pressures lead to the adoption of supplementary rectly relevant to development objectives. Having over pension programs. A package of policies on the revenue, achieved towards their quantitative goals, DFIs could shift expenditure, investment, and administration sides of the pen- focus to target these objectives. sion system would need to be envisaged. 101. Policymakers could place greater emphasis on crowd- 105. Establishing a long-term lifecycle benchmark for pen- ing-in private financing by changing the focus of DFI in- sion institutions would better align the incentives of asset struments. DFIs use both guarantees and second-tier managers with those of contributors. Compensating asset lending to promote the growth of the private credit mar- managers based on investment returns encourages them ket. However, much of DFI lending could potentially be to seek short-term gains. The pension supervisor, CONSAR, crowding out the development of private sector credit has already strengthened incentives for asset managers on a long-term basis. Further shifting DFI’s funding (e.g., to make longer-term investments, but adopting a single via more strategic and broader use of guarantees), risk, long-term benchmark designed by an independent com- and client-interaction activities for the private sector mittee would be a clear improvement over the current could more effectively leverage private capital and liquid- system of multiple undisclosed benchmarks. ity, encouraging more sustainable development of the financial sector for long term finance. Thus, greater use 106. In order to increase the availability of investable securi- of guarantees instead of loans or by providing second ties capital market development should be incentivized via a tier finance rather than direct finance, and exiting from multi-pronged approach. Corporate governance improve- markets where private financial institutions can serve the ments and market regulations could be streamlined for market will help maximize DFIs’ impact for the level of risk mid-sized firms, and measures taken to promote issuance of assumed. private sector bonds. As discussed above, DFIs can support bond issuances through targeted risk guarantees in specific 102. DFI corporate governance could be improved. A recent areas and alternative assets such as PPP projects. Mexico’s evaluation of DFI governance recommended revising the membership in the Pacific Alliance and MILA73 should be composition of board members and reforming mecha- further exploited to augment the diversity and liquidity of nisms for electing CEOs.72 Better governance could en- available securities on the market, including, inter alia, more hance DFI efficiency. issuances of private sector fixed income securities. MEXICO POLICY NOTES 72 To facilitate financial access, development of remote access channels, in particular those supporting the widespread operation of mobile banking, should be prioritized, while at the same time seeking enhanced coverage of bank, non-bank and third-party physical access points. An appropriate combination of physical access points and remote channels, and 47 a balance between in-house services and third-party providers, can decrease both time and travel costs on the demand side and transaction and overhead costs on the supply side, enabling financial firms to sustainably expand operations in remote and underserved areas of the country. Streamlining the 73 Mercado Integrado de Latinoamericano, an initiative integrating the stock markets and listing and trading practices, among Mexico, Chile, Colombia and Peru. 5. Reforming Labor Policies to Encourage Formalization and Promote Productive Inclusion M exico has low unemployment rates by the standards of its regional peers, its labor force is increasingly better educated, and the share of workers employed in the informal sec- tor fell from 60 percent in 2012 to 56.5 percent in 2017. These are important achieve- ments that were brought about by policy reforms over the last years. Yet labor-force participa- tion rates among women remain stubbornly low, dropout rates in upper secondary school are high, young workers face limited economic opportunities, and a large share of informal workers (among the largest in Latin America) inhibits the growth of productivity and wages. Indicators of allocative efficiency in the Mexican labor market are low, and wages are barely rising. Multiple barriers constrain women and youth from accessing job opportunities. Across the entire labor force, skills mismatches and misallocation hinder the expansion of firms. Successfully fostering productive inclusion will require a multifaceted formalization strategy that promotes formal em- ployment addressing the incentives faced by firms and employees. Such a strategy must integrate a multiplicity of actors and institutions to coordinate a wide array of policies. The main policies under the proposed formalization strategy should include: tax policy, social protection reform, ad- MEXICO POLICY NOTES ministrative simplification, reduction of burdensome regulation (particularly at the sub-national level), enhanced enforcement, cutting the costs for hiring and firing, and reducing the length of legal procedures in labor courts, among other. Fostering participation in the formal labor force of 49 Chapter 5 women and youth is also critical. Policies geared toward this objective include: facilitating access to affordable, high-quality childcare; promoting publicly funded afterschool programs; facilitating school-to-work transitions, including through wage subsidies to formal firms to take on young workers and train them to perform a job; and enhancing access to job-search and training support as part of a broader set of active labor market programs. In this context, strengthening the rela- tionship between the education system and the private sector to equip young workers and wom- en with the skills demanded by employers will also be necessary. In terms of wage legislation, Mex- ico is the country in LAC with most room to increase its low minimum wage without significantly affecting compliance or the wage distribution. However, it is crucial to assess potential impacts on productivity, human capital investment and wage inequality, to understand the advantages and drawbacks of different levels of increase in the wage floor under the specific dynamics of the Mexican labor market. CONTEXT AND REFORM the self-employed; and, those hired in formally registered PROGRESS firms that do not comply with the obligation of enrolling them in contributory social insurance. The discussion of 107. Mexico’s unemployment rate is very low by the stan- policy action requires the distinction of formalization of dards of both Latin American and OECD countries, yet the informal firms, entailing regulation enforcement for regis- composition of the workforce reveals serious inclusion tering and taxing them; and formalization of employment, challenges. The labor-force participation rate (LFPR) varies entailing assurance of social protection, registration of substantially by gender, discouraged workers make up a labor contracts and law enforcement for the protection significant share of the economically inactive population, of labor rights. Moreover, taxes, social programs and labor and informality is pervasive. Of the 82 million Mexicans be- regulations are parts of a single incentive structure that tween the ages of 15 and 65, just 63.2 percent (52 million) has large implications for productivity.76 participate in the labor market, leaving an economically inactive population of 30 million. The LFPR for women is 35 109. Widespread informality suppresses wage growth. In percentage points lower than it is for men. Of the roughly 2017, 56.5 percent of employed workers were engaged in 50 million employed workers, 35 million are salaried em- the informal sector, and more than 60 percent of wage em- ployees and 10 million are self-employed. The remaining 5 ployees earned less than three times the minimum wage. million (or just over 9 percent of the employed population) Informality is especially high in poorer areas of Mexico, are divided equally between employers and unpaid family which contributes to regional and rural-urban disparities. workers. While Mexico’s unemployment rate is low at just Rates of formal employment are highest in the states of 3.6 percent (1.8 million people), an estimated 16.1 percent Nuevo Leon, Coahuila, and Chihuahua and lowest in Chi- of the economically inactive population (5 million people) apas and Oaxaca.77 Gender disparities are also substantial, are willing and able to work but are discouraged or believe as the rate of formal employment among men significant- there no suitable available jobs ly exceeds the rate among women. Although the sector that has the greatest proportion of formal workers is man- 108. The informal economy is complex and heterogeneous. ufacturing (62 and 55 percent of men and women respec- Recognizing the diversity of actors, motivations and caus- tively), informality remains high across sectors showing es behind informality is essential for policy action. The that the phenomenon is not sector-specific. Asymmetric definition of informality used in this note includes all those regulation of the relations between workers and firms for workers engaged in economic activities in a context in salaried and non-salaried workers compounds misalloca- which they cannot invoke the legal or institutional frame- tion of resources towards the informal sector and contrib- work to govern their economic inclusion, independently of utes to low productivity. the circumstance or reason for it.74 Informal employment is therefore comprised of workers with no salaried contrac- 110. The reform process over the last decade brought at- tual relations or subordination of a worker to a firm or boss tempts to increase flexibility to the labor market and facil- in exchange for a wage, such as small family firms, farmers itate the formalization of firms and jobs. In 2012, Mexico’s in sharecropping agreements, workers remunerated on Congress approved a set of reforms aimed at increasing the basis of units sold, among others;75 it also includes labor market flexibility especially for young workers. The MEXICO POLICY NOTES 74 INEGI (2014) “La informalidad laboral, marco conceptual y metodológico” 75 According to Mexico’s labor and social security laws, there is no obligation that non-salaried workers be enrolled in contributory social insurance. The distinction between salaried and non-salaried contracts in Mexico is significant, as many institutions and policies have been designed separately providing different benefits for workers depending on the type of 50 contract they hold (Anton et al, 2013; Levy 2018). 76 Levy (2018) 77 INEGI, ENOE 2017 reforms also increased enforcement by adding to the re- 37 percent, while the share of the labor force with some Chapter 5 sources available to inspection of employers’ fulfilment of complete secondary school education increased from 31 the regulation, and increasing the penalties for noncom- to 35 percent. However, while educational attainment is pliance. Since 2014, measures designed to reduce the rising among younger workers, their informality rates are cost of formalization, such as simplified tax procedures, the same as those of earlier, less educated cohorts.79 have played a role in encouraging firms to join the formal sector. As part of the fiscal strategy “Crezcamos juntos”, 112. Strict labor regulations along with lengthy and costly the efforts to increase the tax base have been successful dismissal procedures diminish incentives for formality.80 in formalizing firms and their employees as well as in- Burdensome labor legislation implies high costs in hiring creasing revenue. On the social security side, IMSS’ strat- and firing, place an implicit tax on formal employment, in- egy against informality implied a significant reduction of centivizing firms away from formal contracts. Restrictions the administrative burden for enrollment, reaching 2.9 on the hours of work and non-standard work, increases million additional formal workers enrolled between 2012 the hiring costs of formal hiring.81 The severance pay re- and 2017.78 quirements are very high, placing the burden on individu- al firms and discouraging formal hiring. The average sever- D  espite progress driven by positive reforms, challenges ance pay for redundancy dismissal in Mexico is more than remain ahead. double compared to that of in Brazil, and more than qua- druple of those in United Kingdom and France.82 Uncer- tain, complex, and lengthy legal procedures for dismissal CHALLENGES further diminish incentives for formality. Most dismissals end up being contested in labor courts, where processes may take 3 or more years, and the dismissal cannot be de- Lowering the Cost of Formalization clared effective until the court reaches a verdict.83 for Firms and Workers 111. The supply of educated workers is increasing relatively 113. Informality is linked to low productivity and low wages. quickly, but firms and employment are both growing faster in Informality is associated to firm size, being more prevalent the informal sector than in the formal sector, across sectors. among small firms. Such firms employ the vast majority of Between 2005 and 2017, the share of the labor force with Mexican workers, especially in the south. The 2012 labor at least some high school education increased from 24 to reform introduced more-flexible hiring practices, but fell Figure 24: The Share of Remunerated Workers without Access to Social Security Benefits by State (%) 62.53 Michoacan 61.90 60.03 Puebla 59.48 57.75 Guerrero 56.29 56.00 Chiapas 55.58 52.28 Zacatecas 47.01 46.18 Nayarit 45.64 44.06 Campeche 42.34 41.56 Colima 37.73 35.96 Sinaloa 35.06 34.45 Jalisco 33.64 32.22 Querétaro 31.89 31.84 Ciudad de Mexico 30.75 29.45 Baja California Sur 28.89 27.46 Tamaulipas 25.78 24.32 Chihuahua 23.69 21.87 Nuevo Leon 21.10 0 10 20 30 40 50 60 70 % Source: INEGI. ENOE Note: Data are for 3Q 2017. 78 IMSS 2017, Estadisticas de asegurados. 79 Levy and Szekely (2016) 80 There are several other factors that may contribute to the high levels of informality, i.e. social protection system, tax system, and product market regulations (barriers to entry, licens- MEXICO ing etc.), which are covered in Policy Notes for Social Protection, Fiscal, and Productivity respectively. POLICY NOTES 81 The maximum number of hours for daytime work is designated as 8 hour and 7 hours for night shifts s (Art. 61 Federal Labor Code), with a 100% premium for overtime work (Art. 66 Federal Labor Code) 82 The severance pay for redundancy dismissal for a worker (average of severance pay with 1, 5 and 10 years of tenure) in Mexico is 22 weeks of salary, compared to 9 weeks in Brazil, 4 51 weeks in United Kingdom, and almost 5 weeks in France. This difference is more drastic for workers with short tenures: 14 weeks in Mexico, compared to 1.7 weeks in Brazil, 0 weeks in United Kingdom, and 0.9 weeks in France (Doing Business, 2018). http://www.doingbusiness.org/data/exploretopics/labor-market-regulation 83 See Policy Note on Rule of Law for further details on the functioning of Labor Courts and the linkages with informality. Chapter 5 Figure 25: The Evolution of Real Labor Income, 2005-2017 (MX) $2,000 $1,900 $1,800 real wage in MX pesos $1,700 $1,600 $1,500 $1,400 $1,300 I III I III I III I III I III I III I III I III I III I III I III I III I 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 Source: ENOE 2005-2017 Figure 26: The Evolution of the Female Labor-Force Participation Rate, 1992-2016 65 60 55 % of female population ages 15+ 50 45 40 35 30 25 20 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 Mex Bra Chl Col OCDE LAC Source: World Development Indicators 2017 short of creating unemployment insurance, increasing the workers are informal.86 Increasing the minimum wage is share of temporary workers.84 Administrative barriers and argued to reduce income inequality and raise the living high employer and employee social contributions levied standards for impoverished workers. However, the mea- at low and very low-income levels continue to encourage sure could be inadequate if it does not go hand in hand the allocation of resources to small, low-productivity firms, with other actions to increase productivity and ultimately which employ less-educated workers, depressing wage reduce informality. rates85 . Integrating Women into the Paid 114. More than 90 percent of the workers making up to Labor Force the minimum wage are informal. Currently, the minimum wage is under the poverty line (minimum wellbeing line) 115. While Mexico’s female LFPR is improving, it still lags the defined by CONEVAL. During the first trimester of 2018, rates of most OECD countries and regional peers. Only 45.5 there were 8.6 million people whose labor income was percent of working-age Mexican women participate in the up to the minimum wage, representing 16 percent of labor force, well below the averages of 53 percent for LAC MEXICO the employed population. The great majority of these countries and 51 percent for the OECD. Mexico’s female POLICY NOTES 52 84 Approval of the unemployment insurance element by the legislature is still pending. 85 Levy and Lopez-Calva (2018) 86 INEGI, Tabulados ENOE I-2018. Figure 27: The Labor-Force Participation Rate by Education Level, Gender, and Age Chapter 5 Incomplete primary Complete primary 20.0 40.0 60.0 80.0 1 0 0 .0 20.0 40.0 60.0 80.0 100.0 Participation rate (%) Participation rate (%) 15-24 25-34 35-54 55-64 65+ 15-24 25-34 35-54 55-64 65+ Age group Age group Male Female Male Female Complete Secondary Mid & high Education 20.0 40.0 60.0 80.0 1 0 0 .0 20.0 40.0 60.0 80.0 100.0 Participation rate (%) Participation rate (%) 15-24 25-34 35-54 55-64 65+ 15-24 25-34 35-54 55-64 65+ Age group Age group Male Female Male Female Source: ENOE 2017. Note: Data are for 3Q 2017. LFPR is below the rates of regional peers such as Brazil (56 Fostering Youth Inclusion percent), Colombia (58 percent) and Chile (52 percent), the latter two had female LFPRs in the 1990s that were close 117. Young workers in Mexico face elevated rates of unem- to or below the rate for Mexico (Figure 26). The gender dif- ployment and informality, as well as precarious employment ferential in LFPR is more pronounced among young adults conditions.90 Although youth unemployment rate in Mex- and those with lower levels of educational attainment ico (6.9 percent) is low compared to most of its regional (Figure 27). A recent study estimated that Mexico’s gender peers (for example 18.4 and 30.5 percent in Colombia and gap in LFPR costs the country a staggering 22 percent of Brazil respectively), it is two to three times higher than the per capita GDP.87 rates for other age groups. An estimated 67 percent of employed youth work in the informal sector, 10 percent- 116. Multiple barriers hinder women’s entry into the labor age points higher than the average for the working-age force, and many women are forced to balance the demands population. In addition to the high turnover and exclusion of unpaid work against paid work and education. In Mexico, from social security programs that characterize work in women shoulder nearly 77 percent of all unpaid house- the informal sector, informal jobs are less likely to impart work. The average woman spends six hours each day the skills valued in the formal labor market, reinforcing doing unpaid housework, compared to an average of two the economic exclusion of informal workers. Moreover, 29 hours for men. A large household labor burden presents a percent of workers aged 15 to 19 earn less than the mini- serious challenge for women attempting to attend school mum wage, and 46 percent earn just one to two times the or work a full- or even part-time job. Moreover, almost minimum wage.91 30 percent of employees in Mexico work very long hours (more than 40 hours in a usual week),88 far above the OECD 118. A large share of 92Mexico’s youth is not in employment, average of 13 percent, which compounds the challenge education or training (NEET). In 2016, a quarter of Mexi- of balancing multiple obligations and acts as a barrier to cans aged 20-24 were in this group, far above the OECD women’s entry in the labor market.89 Social norms also are average of 16 percent (Figure 26). Being out of school or a determinant factor of labor supply: Mothers, and espe- training during adolescence has a negative long-term cially younger mothers, are far less likely to work or attend impact on productivity, lowering lifetime wages and school than are non-mothers. employment prospects and slowing overall economic MEXICO 87 Cuberes and Teigner (2017) POLICY NOTES 88 According to OECD statistics, in Mexico the average usual weekly hours worked is 45.6, one of the highest among OCDE and LAC countries. 89 OECD, Better Life Index, 2016. 53 90 Unless otherwise noted, youth are considered those in the 15 to 29 age range. 91 Ibid. 92 INEGI, 2017a. Chapter 5 Figure 28: Youth Not in Education, Employment or Training (NEET), Mexico and OECD Countries 45 40 35 30 25 20 15 10 5 0 Iceland Czech Republic Netherlands Japan Denmark New Zealand Sweden Mexico Greece Luxembourg Germany Switzerland Australia Estonia Norway Russia Austria Slovenia Hungary Israel OECD average Lithuania Slovakia Poland United States Costa Rica France Colombia Belgium Chile Canada Latvia Turkey Brazil Ireland Spain Italy Finland Portugal 15-29 year-old-men 15-29 year-old-women Source: OECD (2018), Youth not in employment, education or training (NEET) growth.93 For any given age cohort, a 1 percentage point graduates who pursued general studies are less likely increase in the share of youth NEET is associated with a 7 to find a job than those who graduated from vocational percent reduction in earnings for that cohort 20 years lat- tracks. er.94 The prevalence of population NEET also has negative implications for poverty reduction and shared prosperity, 120. NEET status has important intergenerational implica- as almost 60 percent of Mexico’s population NEET come tions. A sizeable share of Mexico’s youth is not accumu- from households in the bottom 40 percent of the income lating the human capital necessary to fully leverage eco- distribution.95 Mexico’s low female LFPR contributes to the nomic opportunities. Meanwhile, the higher incidence of large share of population out of work, training or work. NEET status among youth from households in the lowest Young women in Mexico are nearly four times as likely as quintiles of the income distribution perpetuates poverty young men to be NEET, and 35 percent of young women and inequality. Moreover, disengagement from education are currently in this status— the second-highest rate in and the labor market is associated with other social prob- the OECD. Motherhood is a key driver of NEET status, and lems that increase vulnerability to poverty traps, including Mexican mothers in all age groups are far less likely to be incomplete education, teen childbearing, youth unem- engaged in paid employment than are mothers in most ployment, crime, and violence. OECD countries. Among men, dropping out of high school is the most reliable predictor of NEET status, followed by previous unemployment.96 Roughly four out of five youth Addressing Skills Mismatches NEET were previously employed, and only one in five be- 121. Roughly one-third of Mexican firms cite difficulty came a NEET immediately after leaving school. finding qualified workers as an obstacle to productivity and expansion. More than 40 percent of Mexican employers 119. Unemployment and informality among young workers struggle to fill vacancies—particularly in semi-skilled can have lasting negative effects on employability and wag- trades—compared to an average of 31 percent in compa- es in adult life. In any six-month period, only 17.1 percent rable countries.99 The lack of technical specialists is most of high school graduates that have been employed in the acute in the mechanics (22 percent), automotive (18 per- informal sector successfully transition to a formal job. By cent), aerospace (16 percent), electrician services (12 per- contrast, 85 percent of high school graduates that obtain cent), and electronics (10 percent) subsectors, inter alia.100 formal employment stay in the formal sector.97 In addition, The positions that are hardest to fill are for machinery op- less-skilled workers experience greater wage losses and erators, salespeople, administrative staff, technicians, and larger persistence effects.98 In the short term, high school certified workers. MEXICO 93 Szekely and Karver (2015) POLICY NOTES 94 de Hoyos et al. (2016) 95 de Hoyos et al. (2016) 96 de Hoyos, Rogers and Székely (2016) 97 Calderon (2015) 54 98 See, e.g.: Heckman and Borjas (1980); Arulampalam, Gregg, and Gregory (2001); Cruces, Fields, Jaume and Viollaz (2015). 99 Hays (2016); Baz et al. (2014) 100 Manpower Group (2015) 122. This mismatch between labor supply and demand re- extent to which workers value additional income more Chapter 5 flects a combination of factors, including inadequate skills, than the benefits provided by social security contributions ineffective signaling, and high search costs. Enrollment and payroll taxes. Research shows that low-income work- rates in upper secondary and tertiary education have im- ers are especially likely to favor immediate income over proved, but progress has been uneven among regions and long-term benefits.103 The benefits of formality for workers population groups. The share of Mexican adults with up- are mainly associated with services, for example health. per secondary and tertiary education remains below the Ensuring good quality of these services and aligning the OECD average. 48.9 percent of employers attempting to services provided with their perceived cost can also help fill vacancies in semi-skilled trades report seeking candi- encouraging formality. Considering alternative ways of dates who have completed high school (educación media financing social insurance premia (i.e., increasing indirect superior), while 45.6 percent report seeking candidate with taxation, as discussed in Social Protection Policy Note) to a college education (educación superior), and only 20.9 alleviate the tax- burden associated with formalization. percent report hiring individuals with a middle-school Moreover, the success of these reforms will also depend on education or less (educación básica).101 Firms hiring semi- complementary implementation of other measures aimed skilled technical staff report encountering a range of chal- at reducing labor legislation rigidities, such as reduction lenges, including a lack of technical talent (33.5 percent), in severance pay requirements, that inhibit labor demand. long learning curves (26.9 percent), low performance (14.3 percent), high training costs (13.7 percent), and insuffi- 125. Promote careful design and rigorous evaluation of cient education and technical skills (11.5 percent). A full pilots to evaluate the potential effects on formalization 72 percent of firms report that they are willing to pay more of targeted reductions in labor taxation. The experiences for candidates with hands-on experience, training, and of advanced economies suggest that in-work tax credits practical knowledge appropriate to the position. Firms of and benefits targeted to low-wage and part-time earners all sizes state that they would pay an extra 22 percent, on could increase the demand for low-skilled formal labor.104 average, to workers with the right experience and skills.102 The evidence points, however, to higher probability of expected effects on formal employment under certain conditions,105 namely (i) Minimum wages present a bind- POLICY OPTIONS ing constraint for employment creation in the country; (ii) services financed with payroll taxes are perceived to be of low value; (iii) few constraints to the expansion of Lowering the Cost of Formalization labor demand; and of especial importance, (iv) workers for Firms and Workers have the skills to work in the industries that are expect- 123. A multifaceted “formalization strategy” that promotes ed to expand as a result of a payroll tax cut. For example, formal employment and addresses the multiple incentives in the case of Colombia, binding minimum wages could to firms and employees to become or remain informal is explain the relatively larger employment effects than in needed. Such a strategy must integrate a multiplicity of ac- other countries.106 In Turkey, the accompanying measures tors and institutions to coordinate a wide array of policies. of income tax credits and subsidies on land and electricity The main policies under this strategy to generate the right consumption could have been key to the relatively large incentives to formalization should include: tax policy ad- employment effect.107 justments, social protection reform, administrative simpli- fication, reduction of burdensome regulation (especially 126. Existing evidence also shows that benefits of employ- at the sub-national level), enhanced enforcement, cutting ment creation need to be weighed against a potentially the costs for hiring and firing, and reducing the length of large reduction in the amount of taxes collected and the legal procedures in labor courts. Achieving large gains in need for alternative sources of financing social insurance.108 reducing informality will also require strong enforcement in the labor justice system and the overall rule of law. 127. Provide direct incentives to firms for formalization through targeted programs to informal firms.109 The ben- 124. Consider reducing payroll taxes (compensating with efits of formality for firms are associated with access to increased indirect taxation) to promote formalization of market opportunities and productivity-enhancing public employment. Eliminating some of the ancillary benefits goods, for example access to finance, public procurement, currently financed by social security contributions and technology adoption and other types of relevant busi- payroll taxes, such as housing support, could reduce the ness services. Ensuring provision of such benefits would tax burden associated with formalization. The impact of improve the incentives to formal establishment of firms’ this change on formal employment will depend on the businesses. 101 Manpower Group (2015) 102 Mourshed et al. (2012) MEXICO 103 Cunningham and Maloney (2001); Levy (2008); and, Anton, Hernandez and Levy (2013) POLICY NOTES 104 OECD (2015, 2016) 105 Pagés (2017) 106 Kugler et al. (2017) 55 107 Betcherman et al. (2010) 108 See Social Protection Policy Note for the exploration of alternative financing for social insurance. 109 See Policy Note on productivity for a more exhaustive discussion. Chapter 5 increasingly prominent role in reducing information asym- Facilitating the Entry of Female metries in the labor market, improving employability, and Workers into the Paid Labor Force enhancing workforce adaptability. However, job-interme- 128. Facilitate access to affordable, high-quality childcare diation services and more intensive interventions such as care to boost female participation in the labor force. Efforts career counseling and remedial education must be adapt- to expand coverage and enforce mandatory preschool via ed to address the unique challenges faced by younger the Children’s Program for Working Mothers (Programa workers. Linking employment services with comple- Estancias Infantiles para Madres Trabajadoras) have had mentary interventions and with private-sector firms and significant positive effects in maternal employment.110 industry associations can facilitate the school-to- work Further efforts to increase affordability and improve the transition. These can include wage subsidies paid to for- quality of childcare programs, along with more-flexible mal firms to take on young workers and provide them the operating hours and scheduling options, could signifi- training they need to perform a job. These wage subsidies cantly increase female LFPR. should have a sunset clause for each worker and each firm. 129. Promote publicly funded afterschool programs to fur- 132. Enhance access to job-search support and workforce ther reduce the opportunity cost of employment among training to increase the likelihood of obtaining formal em- mothers. The gap between work and school schedules can ployment.112 While multiple programs attempt to foster make it difficult for both parents to hold paid jobs. Studies youth inclusion, most are fragmented, duplicative, and/ have shown that afterschool programs can increase the or underfunded. Evidence suggests that a combination likelihood of work, the number of hours worked, and the of programs yields better results than individual interven- quality of employment among working mothers.111 tions.113 A comprehensive approach to promoting youth employment that incorporates intermediation, skills train- 130. Expand access to flexible work arrangements for both ing, wage subsidies, and self-employment support, inter men and women to ease gender disparities in paid employ- alia, has proven to have a greater impact than the sum ment and household labor. Both in the public and private of the individual programs.114 Other key elements of suc- sectors, policies that encourage flexible work arrange- cessful active labor market policies include private-sector ments are critical to women’s prospects for employment engagement, adequate screening and targeting of vul- and career advancement. Options include parental leave, nerable groups, and providing sustained support early temporary part-time employment, telecommuting, and in the job-search process.115 Like other forms of support, compressed workweeks. In 2014, paternity leave was rec- job-search assistance is more effective when combined ognized as a right in Mexico, and employers—not the so- with intermediation services that match employers with cial security system—are now obliged to offer five days of jobseekers.116 paid paternity leave. While paternity leave is an important step toward greater paternal engagement in childrearing, the legislative and policy framework for ensuring adequate Addressing Skills Mismatches and work-life balance among workers with children is still very Skill Misallocation limited. Evaluating the impact of existing programs aimed 133. Strengthen the relationship between the education at promoting work-life balance, such as the Family-Re- system and the private sector to equip workers with the sponsible Company (Empresa Familiarmente Responsable) skills demanded by employers. If schools are unable to program, could strengthen the analytical underpinnings communicate to firms what is being taught in the class- for flexible work policies. There is also need of exploring room, education attainment will have limited signaling the prevailing attitudes related to parental involvement in value for employers, and if firms are unable to communi- household activities and childcare, and measuring if and cate the type of skills they need, the school system may how these attitudes can be influenced by behaviorally not adequately equip students to succeed in the labor informed interventions to empower parents (particularly market. While employers continue to demand cognitive fathers) to play an active role in their children’s develop- and technical skills, a recent survey found that 40 percent ment. of Mexican firms identified socioemotional skills—in- cluding communications, customer relations, teamwork, and leadership—as the most difficult skillset to find.117 Fostering Youth Inclusion Recent education reforms have highlighted a growing 131. Facilitate school-to-work transitions and tackle school awareness of the importance of socioemotional skills, but dropouts to improve employment prospects of younger these skills remain insufficiently valued by the education workers. In Mexico, public employment services play an system. MEXICO 110 Mateo and Rodriguez-Chamussy (2016). POLICY NOTES 111 Contreras et al. (2010); Berthelon and Kruger (2011) 112 CONEVAL (2005) 113 Blundell et al. (2013), Graversen and van Ours (2008), Gueron and Hamilton (2012) 114 Kluve et al. (2016) 56 115 IEG, 2012, Kluve et al. (2016) 116 See policy note on productivity. 117 Baz et al. (2014) 134. Promote demand-driven training and internship pro- the National Commission on Minimum Wages (CONSAMI). Chapter 5 grams to bridge the gap between the skills imparted by Reviewing the existing international evidence, it is clear the education system and those demanded by employers. that the effects on productivity, human capital invest- Mexican firms cite lack of work experience as the main ment and employment are context- specific. Although reason for not filling vacancies, followed by inadequate the research of the impacts of changes in minimum wag- education, lack of technical skills, and lack of socioemo- es on the labor market outcomes, such as employment tional skills.118 The international experience suggests that and informality, and on human capital accumulation. has well-targeted training and internship programs that ad- been extensive, up to date it has not led to a consensus equately reflect the need of employers can enhance the in either the theoretical or the empirical literature.120 The- efficiency of the labor market.119 oretically, predictions of the effects on employment and informality point in different directions, depending on 135. Expand the availability of training programs to enable the assumptions about characteristics of the market and workers to adapt to the evolving needs for more sophisti- the size of the change in the minimum wage. Empirical cated technical skills. Producing graduates equipped to evidence for Mexico suggests that small increases in the work in high- tech sectors is vital to Mexico’s continued minimum wage are likely to trigger minimal or no chang- growth. While recent labor reforms have enhanced skills es in the labor market (employment and informality) and certification, options for training and professional certifi- school attendance.121 In particular, the increase in the cations that can be obtained while studying for a degree minimum wage resulting from the alignment of different are needed. minimum wage zones in 2012 seem to have had no neg- ative effects on employment and positive -but small- ef- fects on the probability of being an formal worker among Reviewing minimum wage those affected by the wage rise policy.122 Nonetheless, legislation the studied increases in the minimum wage have been 136. Enhance empirical evidence on the advantages and relatively small and departed from very low initial levels. drawbacks of increasing the wage floor - under the specific More research is needed to estimate potential thresholds dynamics of the Mexican labor market. Understanding po- at which further increases may start producing effects tential impacts of wage legislation is crucial to guide the and to increase robust evidence as base for CONSAMI to institutional wage-setting scope and mechanisms under fulfill its mandate. MEXICO 118 Manpower Group (2017a) POLICY NOTES 119 European Commission (2015) 120 Cunningham et al (2016) 57 121 Perova and Trujillo (2016) 122 Campos-Vazquez et al (2017) Chapter 5 References Anton, A., F. Hernandez and S. Levy (2013) “The end of informal- swers from A Continuous Time Model of Heterogeneity and ity in Mexico?: fiscal reform for universal social insurance”. In- State Dependence” Economica, 47 (187), 247-83. Hendra, R., ter-American Development Bank, Washington, D.C. Greenberg, D., Hamilton, G., Oppenheim, A., Pennington, A., Argente, D. and J. L. Garcia. (2015) “The price of fringe benefits Schaberg, K., and Tessler, B. (2016). “Encouraging Evidence on when formal and informal labor markets coexist”. IZA Journal a Sector-Focused Advancement Strategy: Two-Year Impacts of Labor Economics 4:1 from the Work Advance Demonstration.” Washington, DC: Arulampalam W., P. Gregg and M. Gregory (2001) “Unemployment MDRC. Scarring” The Economic Journal, Vol. 111, Issue 475, 577-584. Independent Evaluation Group (2012) “Youth Employment Pro- Baz, V., Martinez, J., Garcia, E., Cañedo, A.P., and Meza, M. (2014). gramas: An Evaluation of World Bank and International Fi- Encuesta de Competencias Profesionales 2014. Ciudad de Mex- nance Corporation Support”. IEG 2012. ico: Centro de Investigacion para el Desarrollo, A.C. (CIDAC). ILO (2017) “Labour Overview Latin America and the Caribbean”, Berthelon, M. E. and D. I. Kruger (2011) “Risky behavior among ILO Regional Office for Latin America and the Caribbean. Lima: youth: Incapacitation effects of school on adolescent moth- 2017. erhood and crime in Chile” Journal of Public Economics, 2011, ILO (2018) “Notes about Public Employment Services in Selected vol. 95, issue 1, 41-53. Latin American and Caribbean countries. Mexico” Betcherman, G., D. Meltem, and C. Pagés. (2010). “Do employment INEGI. 2005 to 2017. “Encuesta Nacional de Ocupación y Empleo. subsidies work? Evidence from regionally targeted subsidies (ENOE)” Instituto Nacional de Estadística y Geografía. Mexico. in Turkey.” Labour Economics 17:4 (2010): 710−722. INEGI. (2017a) . “Encuesta Nacional de Ocupación y Empleo. Blundell R., M. Costa Dias, C. Meghir, and J. Van Reenen. 2004. Eval- (ENOE)” Instituto Nacional de Estadística y Geografía. Base de uating the Employment Impact of a Mandatory Job Search datos, tercer trimestre 2017. Mexico. Program. Journal of the European Economic Association vol. 2, Kluve, J., Puerto, S., Robalino, D., Romero, J.M., Rother, F., Stöterau, issue 4, 569-606. J., Weidenkaff, F., and Witte, M. (2016). Interventions to Improve Busso, M., M. V. Fazio, and S. Levy. (2012) (In)formal and (un) Labor Market Outcomes of Youth: A Systematic Review of Train- productive: The productivity costs of excessive informality in ing, Entrepreneurship Promotion, Employment Services, and Mexico, Inter-American Development Bank working paper Subsidized Employment Interventions. IZA discussion paper number IDB-WP-341, August 2012 Series. DP10263. Campos-Vazquez, R. M. and N. Lustig. (2017). Labour income in- Kugler A. D., M. D. Kugler and L. O. Herrera-Prada, (2017). “Do Pay- equality in Mexico. Puzzles solved and unsolved. WIDER Work- roll Tax Breaks Stimulate Formality? Evidence from Colombia’s ing Paper 2017/186. November 2017. Reform” Economia Journal, The Latin American and Caribbean Campos-Vazquez, R. M., G. Esquivel and A. S. Santillan. (2017) “The Economic Association -LACEA vol. 0 (Fall 2017), 3-40. impact of the mínimum wage on income and employment in Levy, S. (2008). Good Intentions, Bad Outcomes. Social Policy, In- Mexico”, CEPAL Review No. 122, Santiago, Chile. formality and Economic Growth in Mexico. Brookings Institu- CONEVAL. (2015). Informe de la Evaluación Específica de Desem- tion Press, Washington, D.C. peño 2014-2015, Programa de Apoyo al Empleo. Mexico City: Levy, S. and M. Szekely. (2016) “¿Más escolaridad, menos informal- CONEVAL.Contreras, D., P. Sepulveda and S. Cabrera (2010) idad? Un análisis de cohortes para México y América Latina”, El “The effects of lengthening the school day on female labor Trimestre Economico, Vol. 83, Num. 332 supply: Evidence from a quasi-experiment in Chie” Working Levy, S. (2018) “Under-Rewarded Efforts. The Elusive Quest for Papers 323, University of Chile, Department of Economics. Prosperity in Mexico” Inter-American Development Bank, Cruces, G., G. Fields, D. Jaume and M. Viollaz (2015) “The Washington, D.C. growth-employment-poverty nexus is Latin America in the Levy, S. and L. F. Lopez-Calva. (2018) “Persistent Misallocation and 2000s. Mexico country study” WIDER Working Paper 2015/079 the Returns to Education in Mexico,” Cunningham, W.V., X. del Carpio, L. Iacovone, J. M. Moreno, L. Pa- The World Bank Economic Review, Forthcoming. bon and E. Perova. 2016. “El Salario Minimo y la Productividad Mateo Diaz, M. and L. Rodriguez-Chamussy (2016). “Cashing-in on Empresarial, Laboral y General con un Enfoque en el Caso de Education”, World Bank Publications, Latin American Develop- Mexico”. Mimeo ment Forum Series, The World Bank. Cunningham, W.V. and W. F. Maloney (2001) “Heterogeneity Manpower Group. (2015). Puestos Tecnicos 2015. Retrieved from among Mexico’s Microenterprises: An Application of Factor https://www.manpowergroup.com.mx/uploads/estudios/ and Cluster Analysis.” Development and Cultural Change 50 (1): Puestos_Tecnicos_2015.pdf 131-156. Manpower Group Mexico. (2016). Encuesta de Expectativas de Em- de Hoyos, R., A. Popova, R. F. Halsey (2016) “Out of school and out pleo ManpowerGroup Mexico: Q1 2017. Mexico. of work: a diagnostic of ninis in Latin America,” Policy Research Manpower Group. (2017a). Encuesta de Escasez de Talento Working Paper Series 7548, The World Bank. 2016/2017. Resultados de Mexico. Retrieved from https://www. de Hoyos, R., H. Rogers, and M. Székely (2016) “Out of School and manpowergroup.com.mx/uploads/estudios/Escasez-de-Tal- Out of Work: Risk and Opportunities for Latin America’s Ninis.” ento-MX.pdf World Bank, Washington, DC.European Commission. (2015). Manpower Group (2017b). Encuesta de Expectativas de Empleo European Employment Policy Observatory Review. Upskilling Manpower Group Mexico Q4 2017. Unemployed Adults: The Organization, Profiling and Targeting Retrieved from https://www.manpowergroup.com.mx/uploads/ of Training Provision. encuesta_de_expectativas/MX_0417.pdf Graversen, B. K. and J. van Ours (2008) “How to help unemployed Messina, J. and J. Silva. (2018) Wage Inequality in Latin America: Un- find jobs quickly: Experimental evidence from a mandatory derstanding the Past to Prepare for the Future. Latin American activation program” Journal of Public Economics, 2008, vol. Development Forum. Washington, DC: World Bank. 92, issue 10- 11, 2020-2035. Mourshed, M., Farrell, D., and Barton, D. (2012). Education to em- Gueron, J. M., and G. Hamilton. 2002.“The Role of Education and ployment: Designing a system that works. McKinsey&Company. Training in Welfare Reform.” Welfare Reform and Beyond Policy Retrieved from http://www.compromisorse.com/upload/es- Brief. Washington, DC: The Brookings Institution.http://www. tudios/000/222/Education-to-Employment_FINAL.pdf MEXICO POLICY NOTES brookings.edu/papers/2002/04poverty_gueron.aspx.Hays Observatorio de Salarios (2017). Informe del Observatorio de Sal- Recruiting Experts Worldwide. (2016). Reporte Laboral Mexico arios 2017. Puebla: Universidad Iberoamericana Puebla 2016 OECD (2015) OECD Employment Outlook 2015, OECD, Paris. OECD 58 Heckman, J. and G. Borjas (1980) “Does Unemployment Cause (2016) Better Life Index, Country Reports. OECD, Paris. OECD Future Unemployment? Definitions, Questions and An- (2016a) OECD Employment Outlook 2016, OECD, Paris. Padilla-Perez, R. and F. G. Villarreal. (2017). Structural Change and The World Bank Group (2013b). Mexico Reform Agenda for Inclusive Chapter 5 Productivity growth in Mexico 1990- 2014. Structural Change and Sustainable Growth. and Economics Dynamics 41 (2017) 53-63. Washington, D.C: World Bank Group. Pagés, C. (2017) “Do payroll tax cuts boost formal jobs in devel- The World Bank Group. (2017). World Development Indicators [Data oping countries?” IZA World of Labor 2017: 345 doi: 10.15185/ file]. Retrieved from http://databank.worldbank.org/data/re- izawol.345 ports.aspx?source=2&series=NY.GDP.MKTP.KD&country=M EX Perova, E. and J.D. Trujillo. (2016) “The Effects of a Change in the Székely, M. and J. Karver. (2015) “Youth Out of School and Out of Minimum Wage on Schooling, Employment and Informality: Work in Latin America: A Cohort Approach” Policy Research The Case of Mexico”, Mimeo Working Paper No. 7421. World Bank, Washington D.C. The World Bank Group. (2013a). Country Partnership Strategy for the United Mexican States for the period FY2014-2019. Washing- ton, D.C: World Bank Group. MEXICO POLICY NOTES 59 6. Implementing and Expanding Competition Reforms123 M exico has laid out a significantly improved base for a comprehensive national competi- tion policy. During the last five years, it has substantially revised its competition policy at the constitutional and legislative levels, and has taken steps toward establishing a new institutional framework to foster and regulate competition. Operationalizing the revamped com- petition framework, and implementing and enforcing sector-specific competition reforms, would materialize in accelerated productivity growth over the medium term. But further measures will be necessary to unleash the full potential of the new competition policy in Mexico. Key priorities include embedding competition principles at the subnational level, where ineffective regulations continue to: inhibit interstate trade, create local monopolies and oligopolies, and stifle local entre- preneurs and farmers. Moreover, the country could further improve regulation for access to, and interconnection of, networks and transport services to keep input costs competitive. The authori- ties can also pass sector-specific reforms and strengthen regulatory oversight to further enhance contestability in key markets such as telecommunications and agriculture. On the latter, it could eliminate regulatory barriers to domestic trade in agricultural goods. To face intermediary market power, it could make available information on prices to small agriculture producers across local markets, enhance the storage infrastructure for smaller farmers, and tackle anti-competitive be- havior. To improve the checks and balances in the system, the authorities could grant COFECE as well as other bodies (e.g. Federal Telecommunications Institute, IFT) the power to contest national and subnational laws that (appear to) violate the constitutional principle of competition. MEXICO POLICY NOTES 123 This note was prepared by Tanja Goodwin (Senior Economist, MTI), Martha Licetti (Lead Economist, Markets and Competition Policy Global Lead, MTI), and Soulange Gramegna (Consultant, MTI), with inputs from Seidu Dauda (Economist, MTI). 61 Chapter 6 CONTEXT AND REFORM Box 1. Initial results of pro-competi- PROGRESS tion reforms between 2012 and 2017 Energy market reform (2013): 137. Continued progress on Mexico’s competition policy and institutional framework could substantially accelerate • Expected investment between 175 and 200 US$bn. growth and boost shared prosperity. Barriers to competition • Around 70 new energy firms. slow the reallocation of productive factors, contributing Telecommunication reforms (2013) low productivity. A decade ago, the Central Bank of Mexi- co estimated that constraints on competition reduced the • Prices for mobile broadband services have fallen GDP growth rate by 1 percentage point per year.124 Recent between 61 and 75 percent. studies by the OECD and the World Bank (2017) corrobo- • Growth of mobile broadband subscriptions by 380 rate this estimate. percent. Payment systems reforms (2014) 138. Limited competition has a particularly negative impact on lower-income households, especially in rural areas.125 The • Concentration in point of sales among acquirers has exercise of market power in the food, beverage, and phar- dropped by at least 30 percent. maceutical markets costs Mexican households in the low- Reform of competition legal framework (2013) est income decile about 20 percent more than households in the highest decile.126 In rural areas, the cost to house- • Double in fines for anti-competitive practices holds in the lowest decile is 23 percent higher than the Sources: Government of Mexico, 2018; Prices for mobile broadband baskets cost to households in the highest decile. The entry of for- (high, medium and low-usage), in USD PPP, and mobile broadband subscrip- tions per 100 inhabitants (OECD 2017a); HHI of number of POS (Banxico, 2016); eign supermarkets into Mexico boosted the welfare of the SHCP (2018). average household by an estimated 6.2 percent of initial household income, underscoring the scope of the gains the awarding of the wholesale broadband shared-network to increased competition. However, wealthier households concession to an operator that competes with the domi- were most likely to use supermarkets and therefore most nant telecom provider, the awarding of contracts to private able to benefit from the entry of new competitors.127 firms in the petroleum sector, and the initiation of COFECE’s first investigations and corrective measures in the airport 139. Enhancing competition in the services sector could ac- and local trucking sectors. A complementary presidential celerate GDP growth. Regulations in Mexico’s services sec- initiative, Everyday Justice (Justicia Cotidiana), launched in tor are not conducive to competition. Because firms across 2015, placed special emphasis on improving the regulatory economic sectors rely on transportation, telecommunica- environment.128 With support from the World Bank Group, tions, energy, and professional services, pro-competition the National Commission for Regulatory Improvement reforms in the services sector can lower economy-wide in- (Comisión National de Mejora Regulatoria, CONAMER, for- put costs. A simulated reduction in the regulatory restric- merly COFEMER129) systematically reviewed existing rules tiveness of Mexico’s services sector suggests that targeted and regulations in three priority sectors across all 32 states reforms could yield an additional 0.4 to 0.5 percentage to identify and remove anticompetitive barriers.130 These points of GDP growth each year, ceteris paribus. achievements reflect a broad understanding among public officials and opinion leaders that a robust competition poli- 140. Spurred by a growing national consensus regarding cy not only requires effective law enforcement, but also the the economic damage caused by anticompetitive behav- adoption of pro-competition regulations. iors and policies, Mexico has made significant progress in elaborating and implementing the legal framework for 141. The structural reforms supported by COFECE have start- competition over the past five years. The Pacto por México ed to show results in opening key sectors to competition at the set in motion constitutional changes that established the national level (see Box 1). As of February 2018, the Secretar- Federal Commission for Economic Competition (Comisión iat of Energy (Secretaría de Energia, SdE) reported that con- Federal de Competencia Económica, COFECE), overhauled cessions totaling US$175 billion had been awarded to 69 the regulatory framework for the telecommunications sec- private-sector firms.131 Reforms in the telecom sector have tor, and opened the energy sector to private investment, caused mobile broadband prices to fall to among the lowest among other key reforms. Recent achievements include levels in the OECD.132 In the financial sector, the introduction 124 The New York Times, 2006. 125 World Bank & OECD, 2017. 126 Urzúa, 2013. 127 Atkin, et al., 2015. MEXICO 128 Not only was “better regulation” established as a constitutional principle, but a new General Law on Regulatory Improvement obliged subnational governments to enhance their POLICY NOTES regulatory policies and procedures. 129 According to the General Law of Better Regulation, approved in May 2018, and its 10th transitory provision, the former Comisión Federal de Mejora Regulatoria will now be denominat- ed Comisión Nacional de Mejora Regulatoria (CONAMER), and all previous mentions to COFEMER shall now refer to CONAMER. 130 The World Bank Group has been supporting Mexico’s efforts to strengthen competition policy since 2012. It developed the Markets and Competition Policy Assessment Tool (MCPAT), 62 a groundbreaking methodology to identify, prioritize, and address the most harmful subnational regulations. 131 Government of Mexico, 2018. 132 OECD, 2017a. Figure 29: The Perceived Effectiveness of Antimonopoly Policy Chapter 6 6 5 4 3 2 1 0 Mexico Argentina Paraguay Hungary Colombia Israel Slovak Republic Slovenia Costa Rica Czech Republic Poland Peru Estonia South Korea Brazil Uruguay Chile 2012 2014 2016 Source: WEF, Global Competitiveness Report. (Higher value = effectiveness) of “aggregators” for payments and transfers, open technolo- vent the abuse of market dominance, regulate mergers, gy for different card networks, and interoperability among and pursue advocacy cases, often changing the dynamics clearing houses for card payments have reduced concentra- of entire markets. New tools, such as an effective leniency tion in the points of sale (POS) segment.133 program and criminal sanctions, appear to have increased both the detection and deterrence of cartel agreements, 142. As COFECE itself has become increasingly effective as but COFECE will need to continuously expand its range of an antitrust authority, Mexico has been recognized as one investigative techniques to uncover more sophisticated of the leaders in pro-competition reform. In 2014, Global types of collusion. Mexico now has an opportunity to as- Competition Review, a leading antitrust journal, awarded sume a lead role in Latin America’s efforts to foster open COFECE 3 out of 5 stars, placing it on par with the com- and competitive markets and pioneer new tools, such petition agencies of Chile, Israel, and New Zealand.134 In- as private enforcement of anti-competitive practices dicators published by the World Economic Forum (WEF) (where firms initiate legal actions against their suppliers suggest that the perceived effectiveness of antimonop- for forming cartels, for example, and as opposed to public oly policy has increased, contrary to the trend among enforcement). This has been underdeveloped in LAC in Mexico’s comparators. COFECE has increasingly applied general, and Mexico has a mature system where this can its policy and enforcement tools to disrupt cartels, pre- work. Figure 30: The Frequency with which COFECE Applies Competition Policy Tools* 120 100 80 60 40 20 0 2011 2012 2013 2014 2015 2016 2017 Cartels Abuse Merger Advocacy * The figure shows the total number of incidences in which COFECE took an administrative action designed to alter market dynamics. The “cartels” and “abuse of dominance” categories include fines imposed to punish anticompetitive behavior; “mergers” includes conditioning or blocking mergers; and “advocacy” includes changing an existing anticompetitive government MEXICO policy or preventing one from being established, investigating barriers to competition and issuing binding recommendations, and conducting studies on market dynamics that led directly to reforms. During this period, COFECE also issued 60 binding opinions regarding individual bids or auctions for concessions and licenses. POLICY NOTES Source: Author’s own elaboration, based on data from Global Competition Review, COFECE’s annual reports and the WBG Anti-Cartel Enforcement Database. 133 Banxico, 2016. 134 Global Competition Review, 2017. 63 Chapter 6 Governments can intervene in markets as: (i) a buyer, KEY CHALLENGES via public procurement, (ii) an legislative and executive power, via licensing and permits, (iii) a referee of mar- 143. Despite the significant reform progress, competition kets, via controls on the abuse of market power, (iv) a constraints still impact key services markets, with negative regulator, via laws on network industries and others spillover effects on large segments of the Mexican econ- with natural monopolies, (v) a financial supervisor, via omy. Anticompetitive behavior, trade protection mea- rules for banking, financial, and insurance firms, and sures, and incomplete or weakly enforced regulatory (vi) signatory of international treaties, via trade agree- mechanisms distort important input markets and enable ments and import-export laws.141 While the government the exercise of significant market power. For instance, has strengthened its role as a referee and a regulator in between 2009 and 2012 seven shipping lines formed a some sectors, many government interventions are not cartel and divided up the entire market for roll-on, roll- conducive to competition, especially in the areas of off cargo, which includes automobiles and agricultural licenses and permits, network regulation, and interna- equipment.135 Eliminating the exercise of market power tional trade. by global shipping cartels could increase trade across Latin America by an estimated 15 percent.136 Tariff and 146. Many legislative and executive actions by subna- nontariff barriers in the steel industry also appear to be tional governments do not mirror reform progress at the reinforcing market dominance of major firms, adversely federal level and unintentionally damage competition affecting large parts of Mexico’s automotive and con- in local markets. States and municipalities maintain struction industries. Lack of effective regulation poses restrictive regulations that the federal government challenges in the railway sector. Mexico’s railways move abandoned decades ago. State governments have about 80 million metric tons of cargo each year and are particularly strong regulatory discretion in the agri- critical to the production of cereal grains and other bulk culture, construction, transportation, and profession- commodities. Railways also play a key role in automotive al-services sectors, while municipalities wield consid- value chains linked to the US. However, a lack of effective erable authority over local construction, retail, and interconnection regulation may reduce their efficiency real estate.142 While systems have been established to and raise freight rates.137 align the federal regulatory framework with pro-com- petition principles, no corresponding mechanisms are 144. Over the past five years, regulators have discovered in place at the local level. For example, the federal gov- cartel agreements and other anticompetitive practices af- ernment opened the trucking industry to competition fecting transactions with a total value of up to 2 percent in the 1990s, but several states continue to limit the of annual GDP. Between 2003 and 2006, a conspiracy number of licensed operators and even grant monop- among six pharmaceutical companies to rig public pro- olies to trucking unions in specific municipalities. Sim- curement bids cost the Mexican Institute for Social Se- ilarly, while national price controls have been largely curity (Instituto Mexicano del Seguro Social, IMSS) more abolished in product markets where competition is than US$30 million in overcharges.138 In 2014, a single viable, some municipalities continue to regulate prices cartelization agreement among pension funds affected for certain consumer goods.143 a volume of sales equal to 2.3 percent of GDP.139 Be- tween 2010 and 2014, explicit anticompetitive agree- 147. New data reveals that numerous barriers to com- ments between firms that were revealed and sanc- petition distort the level playing field in key sectors for tioned by the authorities affected a volume of annual local economies, such as agriculture, transport, public sales equivalent to 1.1 percent of GDP (see Annex 1). As procurement and professional services. A series of joint massive as these figures are, anticompetitive behavior World Bank-CONAMER studies revealed that in at least likely affects a much larger share of the economy, as 2417 incidences across 32 states, subnational regula- only an estimated 10 to 30 percent of cartels are ever tions limit entry, reinforce dominance, facilitate collu- detected.140 sive outcomes, or distort the level playing field.144 Data collected between 2017 and 2018 reveals that in one 145. The government’s main challenge will now be to state alone, up to 165 of such barriers were identified in align its market interventions with pro- competition prin- three priority sectors (Figure 31). As part of a 2016 ex- ciples while continuing to police anticompetitive behavior. ercise designed to identify the “most absurd regulatory 135 COFECE, Segundo Informe Trimestral, 2017, p. 29-30, available at http://www.senado.gob.mx/sgsp/gaceta/63/2/2017-07-19- 1/assets/documentos/2IT2017.pdf 136 Hummels et al., 2009. 137 COFECE, 2017. 138 “Estimación de los beneficios obtenidos por la sanción de un cártel en licitaciones públicas del IMSS en México”, p. 4, COFECE, 2015. Available here: http://www.cofece.mx/attach- ments/article/37/IMSS_Evaluacion_ex-post.pdf 139 COFECE, 2015. MEXICO 140 Combe et al., 2008; Miller, 2009. POLICY NOTES 141 Tirole, 2016. 142 Substantial economic regulation is defined as those norm-giving instruments that regulate the economic activity of a specific sector, including entry, operations/service provision, prices or tariffs or the role of associations. See Licetti et al., 2016b. 143 For meat from a different Mexican state to be sold in Sonora, the vendor requires a prior favorable opinion by the Regional Cattle-Raising Union. The state law on cargo transportation 64 explicitly establishes a preference for granting concessions to firms that already operate in the market and especially to older firms. Sonoran state law also explicitly allows agricultur- al unions to establish minimum prices for their produce. 144 Licetti et al., 2016a; Licetti et al., 2016b. Figure 31: Number of regulatory restrictions Figure 32: Subcategory: Number of rules that Chapter 6 identified in key sectors (overall) reinforce dominance or limit entry Figure 33: Subcategory: Number of rules that Figure 34: Subcategory: Number of rules that are conducive to collusive outcomes or discriminate and protect vested interests increase costs to compete in the market Source: CONAMER, MCPAT application under Justicia Cotidiana. Note: The prevalence of regulations of different nature and effects varies: While Baja California features more rules that discriminate and protect vested interests, Tamaulipas features more rules that are conducive to collusive outcomes or increase costs to compete in the market, and Queretaro features 26 rules that reinforce dominance and limit entry. obstacle,” COFECE already found 220 “regulations that ments have been associated with a more than 6 percent limit the capacity of individuals or firms to compete in increase in annual retail sales.147 In Mexico State, more the market, or restrict the options of goods and services stringent barriers to market entry are associated with for consumers, without achieving the social or eco- fewer competing supermarket chains.148 Restrictive reg- nomic benefit they intended.”145 In 7 states, regulations ulations can also facilitate anticompetitive practices. For discriminate against agricultural producers from other example, many municipalities either allow incumbent states. Local suppliers are given an explicit advantage tortilla vendors to bar the entry of new competitors, es- in public procurement in 11 states and in the transport tablish minimum distances between tortilla vendors, or sector in 13 states. Professional associations act as li- maintain rules that allow for price coordination.149 Since censing authority and can propose fee schedules for 2005, COFECE has uncovered at least six cartel agree- their services in another 13 states. ments among tortilla producers.150 148. These barriers affect firm competitiveness, artificially 149. The regulatory framework for network industries inflate consumer prices, and even the efficiency of public and other services sectors is still not fully-adequate to spending. In Tabasco, for example, monopoly transpor- curb the abuse of market power or incentivize efficiency. tation unions overcharge for transport of construction In 2013151, the aggregate score for Mexico’s product mar- materials, increasing the cost of road works by an esti- ket regulation still lagged behind those of other recent mated 15 percent.146 In Oaxaca, reforms that ease regu- OECD members, with barriers to entrepreneurship, trade lations on the operating hours of commercial establish- and investment imposing especially serious constraints 145 COFECE, 2017c. 146 Licetti, et al., 2016b; Teravaninthorn & Raballand, 2008; Osborne, Pachón, & Araya, 2014. MEXICO 147 Licetti & Dauda (forthcoming) POLICY NOTES 148 Licetti et al., 2016a. 149 In Michoacán, 28 municipalities establish minimum distances between tortilla stores, and 9 municipalities require consent from competitors for new tortilla shops to enter in the market. The minimum tortilla prices set collectively by local producers in Lazaro Cardenas municipality are 30% higher than the average prices for the occidental region. (García & 65 García Soto, 2010) 150 Government of Mexico, 2016. 151 Data on Product Market Regulation is collected by OECD (and WBG in case of several non-OECD countries) every 5 years. The 2018 data was not yet available as of July 2018. Chapter 6 Figure 35: OECD Product-Market Regulation Indicator, 2013 3.5 30% 3.0 29% 31% 2.5 21% 38% 16% 2.0 24% 23% 27% 5% 24% 16% 9% 32% 1.5 6% 10% 37% 14% 18% 39% 30% 33% 38% 35% 34% 33% 29% 1.0 42% 43% 45% 38% 30% 41% 40% 39% 0.5 45% 43% 39% 62% 33% 33% 56% 47% 49% 42% 44% 35% 51% 46% 41% 35% 0.0 Estonia Hungary Czech Republic Slovenia South Korea Mexico Costa Rica Slovak Republic Chile Poland Peru Colombia Israel Paraguay Brazil Uruguay Argentina State control Barriers to entrepreneurship Barriers to trade and investment Source: OECD PMR Database; WBG-OECD PMR Database, 2013, as of March 2018. Note: Some figures may not sum up to 100% due to rounding issues. Figure 36: Barriers to Entrepreneurship, 2013 3.5 3.0 18% 2.5 21% 33% 26% 2.0 24% 22% 36% 20% 21% 35% 16% 21% 41% 21% 31% 1.5 20% 29% 18% 25% 29% 17% 31% 37% 46% 36% 39% 38% 1.0 26% 39% 33% 53% 53% 37% 62% 44% 46% 0.5 52% 45% 61% 41% 42% 36% 44% 43% 46% 36% 32% 28% 18% 21% 0.0 13% Costa Rica Slovenia Czech Republic South Korea Mexico Colombia Paraguay Israel Brazil Peru Argentina Slovak Republic Estonia Poland Hungary Chile Uruguay Complexity of regulatory procedures Administrative burdens on startups Regulatory protection of incumbents Note: Some figures may not sum up to 100% due to rounding issues. Source: OECD PMR Database; WBG-OECD PMR Database, 2013, as of March 2018. on competition (Figure 35). Mexico’s regulatory protec- tion of incumbent firms posed a greater barrier to en- Low levels of competition trepreneurship than in any other recent OECD member continue to distort incentives (Figure 36). While major reforms have opened some in several transport and sectors to competition through recent reforms some logistics sectors. federal product market regulations still constrain com- petition in some sectors and add to the recently uncov- 150. Anticompetitive railway mergers are likely responsible ered subnational barriers to competition. Such federal for the rapidly increasing cost of railway transport. The for- restrictions include, inter alia, a lack of third-party access mer antitrust authority opposed railway mergers in 2002 in electricity, restrictions on the number of competitors and 2006, but a higher court approved the mergers in in electricity generation, and the involvement of profes- 2011. Since then, rates for railway freight transport have ris- MEXICO sional associations in establishing and enforcing entry en substantially, outpacing price increases in other trans- POLICY NOTES regulations in road transportation. portation subsectors, and currently, average per-kilometer fares on rail sections operated by an interconnecting con- 66 cessionaire are more than eight times higher than on the Figure 37: Cargo Transport Prices Chapter 6 390 Merger between Ferromex and Ferrosur (Nov. 2005) 340 290 240 190 140 90 Dic 2003 May 2004 Oct 2004 Mar 2005 Ago 2005 Ene 2006 Jun 2006 Nov 2006 Abr 2007 Sep 2007 Feb 2008 Jul 2008 Dic 2008 May 2009 Oct 2009 Mar 2010 Ago 2010 Ene 2011 Jun 2011 Nov 2011 Abr 2012 Sep 2012 Feb 2013 Jul 2013 Dic 2013 May 2014 Oct 2014 Mar 2015 Ago 2015 Ene 2016 Jun 2016 Nov 2016 Abr 2017 Sep 2017 Rail Sea Road Producer Price Index (Transport, mail and telecom) Note: Nominal non-seasonally adjusted producer price index (December 2003 = 100) until June 2012, projected until January 2018 based on corresponding nominal non-seasonally adjusted producer price index (June 2012=100) for each cargo transport service. Source: INEGI Figure 38: Logistics Performance Indicator 4 3 2 1 0 Mexico Czech Republic South Korea Costa Rica Brazil Slovenia Chile Slovak Republic Estonia Poland Hungary Israel Paraguay Colombia Peru Argentina Uruguay 2012 2016 Source: WBG, Logistics Performance Index (LPI) 2012 - 2016. Note: LPI – 5=best performance; sample=155 countries in 2012, 160 countries in 2016. Israel’s score in 2012 corresponds to 2014, since it was not available in the 2012 LPI database. rest of the rail route (Figure 37).152 Although COFECE has lenge, especially given Mexico’s substantial infrastructure identified potential lack of competition in interconnection gap. Landing slots in the Mexico City International Airport services between rail networks, no corresponding regula- (Aeropuerto Internacional de la Ciudad de México, AICM) tory adjustment has yet been implemented.153 High rail- are concentrated among a small number of airlines. In freight costs likely have negative spillover effects on other 2014, two airlines controlled 65 percent of the slots and transportation and logistics subsectors, given the impor- accounted for 67 percent of all flights. About 30 percent tance of the railway network to competition in broader of daily flights were assigned a slot shortly before takeoff container shipping and intermodal transportation and its or landing. Competitor airlines have difficulty obtaining function as a substitute for road transportation.154 terminal and runway access, and the resulting lack of com- petition on routes in and out of Mexico City accounts for 151. The allocation of user rights for congested transpor- 40 to 80 percent of the difference in airfare between AICM tation infrastructure is a critical competition-policy chal- and other Mexican airports.155 Following an investigation, MEXICO 152 COFECE, 2017. POLICY NOTES 153 In March 2017, COFECE’s Investigation Authority published a report concluding preliminarily that there was lack of competition in interconnection services in several key railway lines. In 2018, the Plenary of the Commission judged that there was not yet sufficient information to confirm the lack of competition in the entire network, but also indicated that the 67 existence of effective competition could not be confirmed either. 154 OECD, 2017b. 155 Ros, 2010. Chapter 6 COFECE declared landing and take-off slots, and use and options. Authorities could promote more local whole- control of airport infrastructure at AICM ‘essential facili- sale markets, farmers markets and collection centers that ties’, without which other airlines will not be able to com- offer auction-based price-determination mechanisms. pete. COFECE proposed corrective measures, including Better storage facilities for grains, and a proper regulatory increased transparency, the elimination of “grandfather framework for warehouse storage quality, receipts and clauses,” and the authorization of secondary trading of ter- electronic auctioning platforms could moderate seasonal minal and runway access, but these measures have not yet price hikes. Stalls in traditional markets can often only be been fully implemented.156 reassigned according to discretionary rules involving ne- gotiations with existing stall holders. Secondary tradition 152. Additional efficiency-enhancing solutions, such as of stall rights or auctions could facilitate entry. As pointed the adoption of optimal appointment and vehicle-booking out earlier, restrictions to the trade of agricultural goods systems, could be applied to congested maritime ports or among states can further reduce contestability. multimodal transportation infrastructure. Mexico’s logistics performance is relatively strong by the standards of Latin 155. Redesigning government interventions in the agri- America, but it has not improved as much as other recent cultural sector could further level the playing field and en- OECD member countries in Eastern Europe (Figure 38). hance contestability. As in many other countries, Mexico’s Mexico currently ranks 5th from the bottom among recent agribusiness input markets are highly concentrated39. In OECD members in terms of “timeliness of logistics.” 2009, 95 percent of planted hybrid seeds were produced only by two corporations.160 COFECE has recommended 153. The pace of implementation of the new regulatory legalizing parallel imports to address the risk of further framework for the energy and telecommunications sector, harm to competition. For machinery and equipment, sub- both at the national and subnational levels, could weak- sidy programs should not be conditional on certifications, en its benefits to consumers. While recent reforms have which only 8 companies have been able to achieve. More helped foster competition, Mexico’s telecommunications broadly, subsidy programs can be highly overlapping. and digital-services markets remain concentrated by the Corn, chili and tomato producers can be eligible for up to standards of comparable countries. As of end-2016, the 14 different federal programs.161 Some subsidy programs dominant operator, Telcel, continued to hold 71 percent of are available only to a limited number of producers (such all subscriptions in the mobile broadband market. In the as concessionaires for water wells with high usage). Water energy sector, while new private competitors to the na- rights should be efficiently assigned – especially in scarci- tional oil and gas company, Mexican Petroleum (Petróleos ty situations162. Barriers to foreign investment in the cargo Mexicanos, PEMEX), may reduce gasoline prices, subna- transport sector could be lifted to increase contestability tional regulations often establish minimum distances of in cold chain transport which is underdeveloped in Mexi- more than 1 kilometer between gas stations, reducing co. In general, associations and market participants should the benefit to consumers by granting significant pricing not take part in regulatory activities. 163 power at the point of sale. These issues are discussed in greater detail in the policy notes on energy and telecom- 156. Pro-competition measures are essential to encourage munications. a highly concentrated banking sector to operate efficient- ly. Measured by bank assets, the Mexican banking sector 154. Information asymmetries and other market failures was more concentrated than in Argentina or Colombia, could be exacerbating market power along agricultural val- between 2007 and 2011. As of 2013, 74 percent of all ue chains. Between 2005 and 2014, food prices in Mexico credits and 80 percent of all ATMs were held by five banks. grew by 23.3 percent above non-food items, more than in While this concentration is in part explained by network other OECD countries.157 While this price increase benefit- effects and economies of scale, it is also associated with ted producers more than consumers in most cases, there is higher profit margins than in Brazil, Uruguay and Chile164. evidence that price increases are passed on to consumers COFECE’s investigations have revealed initial evidence for faster than price reductions.158 Around half of agricultural explicit anti-competitive practices in the banking sector. production in many products is sold to through interme- Allegedly, various intermediaries agreed to jointly limit diaries.159 While intermediaries can offer efficiencies in ag- competition in auctions of the country’s public debt mar- gregating produce from small individual producers, they ket165. The competition authority has started to successful- may be able to exercise significant market power when ly advocate for pro-competition measures in several mar- primary producers lack alternative commercialization ket segments, including payment services. Most recently, 156 COFECE, 2017b. Presentation by COFECE to International Chamber of Commerce in Mexico (April, 2016). 157 (COFECE, 2015) 158 In most of the cases, the price increase for consumers was below the price increase for producers (except for beef, eggs, soft drinks and beer), but price transmission is often asym- metric (see COFECE, 2015). MEXICO 159 At least one study suggests that intermediation increases food prices by 400 percent (IPD, 2014) POLICY NOTES 160 Namely, Monsanto and Pioneer. See Luna et al. (2012). 161 (COFECE, 2015) 162 For additional recommendations on how to improve economic regulation of the water sector that enables private sector participation and promotes efficiency in operation and service delivery, incl. water tariff reform, more PPP involvement, competitively tendered concessions and management contracts, refer to Policy Note 1. 68 163 COFECE, 2015) 164 https://www.cofece.mx/cofece/images/Estudios/COFECE_trabajo_investigacion_prot.pdf#pdf 165 Filing number IO-006-2016, available here: https://www.cofece.mx/investiga-cofece-la-intermediacion-de-valores-de-deuda- emitidos-por-el-gobierno-mexicano/ as per the Fintech Law, approved in March 1st, 2018, fi- Chapter 6 nancial entities will be prohibited from charging different POLICY OPTIONS rates to FinTech Institutions and other customers. Further 159. The government could align its market interventions remaining challenges are discussed in greater detail in the more closely with pro-competition principles, especially in policy notes on financial services. its roles as a regulator of network industries, a gateway to international trade, an issuer of licenses and permits, and 157. Some trade-policy measures may restrict or distort a referee of competition. COFECE can coordinate actions foreign competition. While Mexico’s economy is relative- across the public sector and provide the analytical inputs ly open, its tariffs are still high compared to those of its necessary to continuously refine and enhance the gov- major trading partners. Mexico imports rice, beans, beef, ernment’s understanding of market conditions and com- and chicken from countries with which it holds free-trade petition dynamics. Meanwhile, CONAMER can support agreements, while imports of these goods from poten- reforms involving subnational governments. To facilitate tially more competitive countries are diverted due to re- interinstitutional collaboration, COFECE, CONAMER, and maining import tariffs.166 In some sectors, trade measures other agencies that regulate or intervene in markets could may be facilitating the exercise of significant market develop an information-sharing system to identify and power (and lead to prices above the competitive level). address barriers to competition, and even allow for a com- For example, Mexico recently renewed an ostensibly plaint and redress mechanisms for firms and citizens.170 temporary six-month unilateral increase in steel tariffs of 15 percent for the fourth time. Steel also has the highest 160. The government can ensure that state and local au- number of tariff lines subject to anti-dumping measures. thorities have the tools and support by CONAMER,, and the In addition, all but one of the 114 tariff lines that require incentives to remove barriers to competition in priority sec- an import approval (aviso automático),167 referred to a tors. This will entail: (i) fully implementing the three-pri- steel product. Concomitantly to these various protection- ority-sector reform plan initiated under Justicia Cotidiana, ist measures, steel imports fell by 23.5 percent between led by CONAMER in partnership with state and local gov- 2015 and 2016. The steel sector is traditionally prone to ernments; (ii) helping subnational governments replicate high concentration, oligopolistic market structures, and successful structural reforms adopted under Justicia Cotid- collusive agreements. Barriers to imports may accentuate iana at the local level across all states; (iii) implementing domestic market power: Within the steel-transforma- the Programa de Reforma a Sectores Prioritarios under tion subsector numerous countervailing duties apply to CONAMER and based on the World Bank’s Markets and specific products for which there are few producers, or Competition Policy Assessment Tool;171 and (iv) setting up only one producer, in Mexico. In February 2018, COFECE a mechanism to reward subnational governments for im- opened an investigation into alleged cartel agreements plementing pro-competition reforms.172 The government in the steel market168. can also take advantage of new rules obliging subnation- al authorities to conduct regulatory impact assessments 158. Overall, operationalizing the new institutional frame- (manifestaciónes de impacto regulatorio) by embedding work for competition will inevitably encounter resistance a simplified competition checklist and COFECE’s review from entrenched interests. The ineffective application process into the assessment methodology to prevent the of new regulatory powers or state- and municipal-level establishment of new obstacles to competition, such as rules could still undermine the implementation of these price controls, restrictions on interstate trade, or the grant- reforms. New investigative and analytical tools developed ing of local monopoly rights in sectors where competition by COFECE and CONAMER have shed light on remaining is viable. gaps in the policy and institutional framework, especially at the subnational level. Advancing the reform agenda will 161. The government can pass sector-specific reforms and require broad ownership among stakeholders, including strengthen regulatory oversight to enhance contestability the general public. Outreach and awareness-raising efforts in key markets such as transport, telecommunications and will be a key challenge to spread knowledge of the bene- agriculture. Priority actions include: (i) facilitating open fits of competition169 and build an understanding of good access to infrastructure by, inter alia, revising the rules for regulatory principles and best practices among state and allocating slots at AICM, consider regulating interconnec- local governments, and awareness of business on compe- tion services between railway networks, creating transpar- tition law – especially at the local level. ent booking systems for port services, and analyzing other 166 COFECE, 2017a. 167 The “aviso automático” is an automatic authorization to import or export merchandise. 168 Filing number DE-020-2017, available here: https://www.cofece.mx/investiga-cofece-posibles-practicas-monopolicas- absolutas-en-el-mercado-del-acero-adquirido-por-autori- dades-publicas/ 169 In 2015, 26% of Mexicans associated competition with “the state offering more and better products and services,” and 7% with “better firms growing until there’s only one left.” See: Ipsos Public Affairs & USAID, 2015. MEXICO 170 Models for such institutional mechanisms that can act upon complaints about government-imposed barriers on competition include the Australian Government Competitive Neutrality POLICY NOTES Complaints Office (AGCNCO) that provides independent advice to the Government following private sector complaints about unfair competition from the public sector, and the Commission for Elimination of Bureaucratic Barriers in Peru that can eliminate illegal or unjustified public sector actions that hinder market entry or permanence. 171 The Ley General de Mejora Regulatoria stipulates in the transition clauses that one of the tasks for CONAMER will be to publish guidelines within one year on the Programa de Refor- 69 ma a Sectores Prioritarios (Art. 11.VII). 172 For example, Australia implemented – as part of its National Competition Policy program - monetary compensation mechanisms (competition payments) to states and territories that implemented committed competition reforms. Payments were a share of additional revenue raised by the Australian Government due to reforms. Chapter 6 intermodal transportation bottlenecks; (ii) continuing to existing anti- dumping measures and countervailing promote open competition in telecommunications and duties; (ii) refrain from repeatedly renewing temporary digital services by using regulatory oversight to prevent tariff increases to create effectively permanent tariff the abuse of market power by incumbent firms; and (iii) barriers; (iii) replace the aviso automático system with eliminating regulatory barriers to domestic trade in agri- an electronic notification mechanism; and (iv) assess cultural goods, making available information on prices to options for ending the outdated Sectoral Promotion agriculture producers across markets, and strengthening Programs (Programas de Promoción Sectorial, PROSEC) the storage infrastructure for smaller farmers. and extending special tariffs to entire industries in a nondiscriminatory manner; and (v) subject all normative 162. The government can strengthen federal checks and instruments issued by government to the regulatory im- balances mechanisms to remove and prevent anti-competi- provement process to prevent the adoption of anticom- tive policies and actions by public bodies. This will require: (i) petitive policies. undertaking a comprehensive review to identify and elim- inate federal-level barriers to competition in key sectors; 164. Finally, the government can increase the elimination (ii) expanding mandatory regulatory impact assessments and deterrence of anti-competitive practices by further pri- to cover draft legislation and additional policy areas; and oritizing COFECE’s resources, and incentivizing compliance. (iii) granting COFECE as well as other bodies (e.g. Federal This would entail: (i) improving the system for differenti- Telecommunications Institute, IFT) the power to contest ating complex and fast-track mergers to free up resources national and subnational laws that appear to violate the for cartel enforcement, advocacy, and other functions; (ii) constitutional principle of competition. developing a class-action and private-damages frame- work to deter anticompetitive agreements; and (iii) en- 163. The government can remove trade policy measures couraging a culture of compliance by launching aware- that unduly restrict contestability in domestic markets. To ness programs, especially among firms operating at the accomplish this, the authorities will need to: (i) review state and local levels. MEXICO POLICY NOTES 70 Chapter 6 References Agencia EFE, 2017. Agencia EFE. [En línea] Griffith, R., Harrison, R. & Macartney, G., 2007. Product market Available at: https://www.efe.com/efe/america/mexico/mexico- reforms, labour market institutions and unemployment. The reporta-80-000-millones-de-dolares-en- inversion-por-refor- Economic Journal, 519(117). ma-energetica/50000545-3354911 Hummels, D., Lugovskyy, V. & Skiba, A., 2009. The trade reducing [Último acceso: 29 January 2018]. effects of market power in international shipping. Journal of Aghion, P., Braun, M. & Fedderke, J., 2008. Competition and pro- Development Economics, 1(89), pp. 84-97. ductivity growth in South Africa. Licetti, M. & Dauda, S., forthcoming. The impact of Shopping Hour Economics of Transition, 4(16), pp. 741-768. Regulations on Retail Sector Performance: Evidence from Mexico, Atkin, D., Faber, B. & Gonzalez-Navarro, M., 2015. Retail Globaliza- Washington, D.C.: World Bank Group. tion and Household Welfare: Evidence from Mexico.. National Licetti, M., Goodwin, T. & Villarán, L., 2016a. Competition Reform Bureau of Economic Research. in Tabasco State: Combatiendo regulaciones que restringen la Chiquiar, D. & Ramos-Francia, M., 2009. Competitiveness and competencia a nivel sub-nacional: Estado de México, México, Growth of the Mexican economy. Banco de Mexico, Working Washington, D.C.: World Bank Group. Papers, Volumen 11. Licetti, M., Goodwin, T. & Villarán, L., 2016b. Combatiendo regula- COFECE, 2014. Control de Concentraciones, s.l.: s.n. ciones que restringen la competencia a nivel sub-nacional: Esta- COFECE, 2015. Public version of the final ruling with file number I0- do de Tabasco, México, Washington, D.C.: World Bank Group. 003-2015, s.l.: s.n. Licetti, M. y otros, 2018. Strengthening Argentina’s integration in COFECE, 2015. Reporte sobre condiciones de competencia en el sec- the global economy : policy proposals for trade, investment, and tor agroalimentario. s.l.:s.n. COFECE, 2017a. Política comercial competition. Washington, D.C.: World Bank Group. con visión de competencia, Mexico City: COFECE. Milenio, 2017. Cofece lleva a suprema corte caso de slots en AICM. COFECE, 2017b. COFECE notifies Mexico City’s International Airport [En línea] (AICM) of measures to regulate slot allocation. [En línea] Available at: http://www.milenio.com/negocios/slots-aicm-aero- Available at: https://www.cofece.mx/cofece-notifica-medidas-pa- puerto-ciudad_mexico-cdmx-cofece- suprema_corte-sc- ra-regular-la-asignacion-de-horarios- de-despegue-y-ater- jn-horarios_0_1071493079.html rizaje-en-el-aicm/ [Último acceso: 30 January 2018]. [Último acceso: 29 January 2018]. Miller, N., 2009. Strategic leniency and cartel enforcement.. Amer- COFECE, 2017c. Resultados: Premio para identificar el obstáculo ican Economic Review, 99(3), pp. 750- 68. más absurdo para competir y emprender, 2016-2017, Mexico NYTimes, 2006. www.nytimes.com. [En línea] City: COFECE. Available at: http://www.nytimes.com/2006/06/03/world/amer- COFECE, 2017. Final Report in Investigation DC-002-2016. [En línea] icas/03slim.html?pagewanted=2&_r=1 [Último acceso: 29 Available at: https://www.cofece.mx/cofece/images/AI/Dict- January 2018]. amen_DC-002-2016.pdf OECD, 2017a. OECD Telecommunication and Broadcasting Review COFECE, 2017. Segundo Informe Trimestral, s.l.: s.n. of Mexico, Paris: OECD Publishing. OECD, 2017b. Review of the Combe, E., Monnier, C. & Legal, R., 2008. Cartels: The probability of Regulation of Freight Transport in Mexico. Paris: OECD Publish- getting caught in the European Union. s.l.:s.n. ing. El Financiero, 2015. Suprema Corte avala caso contra farmacéuti- Productivity Commission, 2005. Review of National Competition cas; pide rebajar multas. [En línea] Available at: http://www. Policy Reforms. Report No. 33 ed. Canberra: s.n. elfinanciero.com.mx/economia/suprema-corte-pide-reba- Ros, A., 2010. A Competition Policy Assessment of the Domestic Air- jar-multas-por- fraude-de-farmaceuticas-al-imss.html line Sector in Mexico and Recommendations to Improve Compe- [Último acceso: 30 January 2018]. tition, s.l.: Documentos Cofeco/OCDE. Expansión, 2017. El regulador multa a 7 navieras por colusión en Sekkat, K., 2009. Does competition improve productivity in devel- transporte marítimo de autos. [En línea] oping countries?. Journal of Economic Policy Reform, 2(12), pp. Available at: https://expansion.mx/empresas/2017/06/09/el-reg- 145-162. ulador-multa-a-7-navieras-por- colusion-en-transporte-mari- Symeonidis, G., 2008. The effect of competition on wages and timo-de-autos productivity: evidence from the United Kingdom. The Review [Último acceso: 30 January 2018]. of Economics and Statistics, 90(1), pp. 134-146. Forni, L., Gerali, A. & Pisani, M., 2010. Macroeconomic Effects Of Syverson, C., 2011. What determines productivity?. Journal of Eco- Greater Competition In The Service Sector: The Case Of Italy.. nomic literature, 2(49), pp. 326-65. Tirole, J., 2016. Economics for Macroeconomic Dynamics, 05(14), pp. 667-708. the Common Good. s.l.:Princeton University Press. García, F. & García Soto, C. E., 2010. Brujula de Compra - Inhiben Urzúa, C. M., 2013. Distributive and regional effects of monopoly Ayuntamientos Competencia en Tortillas, Mexico City: Profeco. power. Economía Mexicana. Nueva Época , Volumen 22.2. Global Competition Review, 2017. Rating Enforcement 2017, Lon- World Bank; Organisation for Economic Co-operation and De- don: Law Business Research Limited. Gobierno de México, velopment, 2017. A Step Ahead : Competition Policy for Shared 2016. Evolución de Precios. [En línea] Prosperity and Inclusive Growth. Washington D.C.: World Bank. Available at: https://www.gob.mx/cms/uploads/attachment/ World Bank, 2014. The unfinished revolution : bringing opportunity, file/234822/XII_Evolucion_de_los_precios_- good jobs and greater wealth to all Tunisians. Washington, D.C.: _febrero_2016.pdf [Último acceso: 29 January 2018]. World Bank Group. MEXICO POLICY NOTES 71 7. Improving the Education System’s Equity M exico’s 2013 education reforms introduced key changes to the sector’s financial and personnel management strategy. The reforms established a professional system for hiring, evaluating, training, and promoting teachers and provided full autonomy to the National Institute for the Evaluation of Education. Further action will be necessary to consolidate these gains and address remaining challenges. For example, the evaluation of incumbent teachers continues to be politically contentious, and a comprehensive approach to improving the equity of the education system would need to be defined and implemented. Moreover, important compo- nents of the education sector are not yet aligned with the objectives of the reform program. The country’s teacher-training colleges (normales) do not adequately prepare teachers for the new education curriculum, and the results of teacher evaluations do not inform training and staffing decisions. Going forward, the authorities could focus on (i) improving the design and implemen- tation process of the professional teaching service (servicio profesional docente), (ii) aligning all elements of the new education model toward learning, (iii) creating a national early child develop- ment program, (iv) adjusting the intergovernmental transfers for education, making them more based on needs (per student) and with a stronger equalization component for lagging states, and (iv) reorienting the sector’s financial and personnel career policies and incentives to improve the equity of the education system. MEXICO POLICY NOTES 73 Chapter 7 enrolled in the Mexican education system, only half ac- INTRODUCTION quire the necessary foundational skills to fully leverage their productivity in the labor market or responsibly ex- 165. Human capital is the most important determinant of ercise their rights as citizens. Educational outcome indi- productivity and the engine of long-term economic growth. However, years of schooling only partially reflect the con- cators are even lower among children from marginalized tribution of human capital to productivity, growth, and households, exacerbating the country’s already vast eco- improved social and economic wellbeing, whereas the nomic and social inequalities. Despite its relatively large quality of education— measured in terms of acquired budget, which equals 6 percent of GDP and accounts for skills—drives the relationship between education and over 20 of total public spending, Mexico’s education sec- development.173 tor remains characterized by poor overall performance and stark disparities between students from households 166. Poor educational outcomes have important long-term at different income levels.174 welfare effects, both at the household level and at the na- tional level. Children from poor households often enter 168. In 2013, the government launched an ambitious ed- school with limited cognitive and socioemotional skills. ucation reform program designed to improve educational Despite significant improvements, Mexico’s elementary outcomes, particularly among students from poor house- schools are still ill- equipped to remediate those deficits, holds and marginalized communities. The reform program diminishing the chances that students from poor house- established a professional teaching service (servicio pro- holds will graduate from the upper secondary level. fesional docente, SPD) with standard systems for hiring, Most students who drop out of upper secondary school evaluating, training, and promoting teachers. It asserted leave the education system without the minimum skills the independence of the National Institute for the Eval- to secure a job in the formal sector. Instead, many take uation of Education (Instituto Nacional para la Evaluación low- paid, low-productivity jobs in the informal sector, in- de la Educación, INEE) to oversee all evaluation functions creasingly the likelihood that they will remain poor. This throughout the education system, including measuring process is one of the mechanisms through which poverty and monitoring the performance of students, teachers, and inequality are transmitted from one generation to the school directors, supervisors, and schools. And it used next. At the national level, this process slows the growth of human capital, increases the burden on the social pro- school grants and technical assistance to strengthen tection system, and contributes to Mexico’s low overall school autonomy and promote school-based manage- productivity. ment (SBM). Building on these reforms, the government recently adopted a new competency-based curriculum, 167. As is the case in other Latin American countries, Mex- introduced equity-enhancing policies, and formulated a ico’s educational indicators are well below what its level of strategy to improve the education system’s governance. per capita GDP would predict, jeopardizing the country’s These changes have given rise to a new education model, long-term development prospects. The results of the 2015 based on five pillars, which will be implemented over the PISA show that while a large majority of 15-year-olds are next several years (Figure 39). Figure 39: The Five Pillars of Mexico’s Reformed Education Model Compentency Based Curriculum Teacher Governance Professional Development Equity School Based Management MEXICO POLICY NOTES 74 173 World Bank, 2018. World Development Report. Washington, DC: The World Bank 174 World Bank. 2016. Mexico: Public Expenditure Review. Washington DC: The World Bank Figure 40: PISA Test Scores for Math and Reading among OECD Countries, 2015 Chapter 7 a. Reading b. Mathematics Percentile Percentile 25th 50th 75th 25th 50th 75th OECD average OECD average Singapore Singapore Finland Hong Kong SAR, China Hong Kong SAR, China Taiwan, China Korea, Rep. Macao SAR, China Macao SAR, China China (B-S-J-G) Poland Korea, Rep. Taiwan, China Finland China (B-S-J-G) Poland Russian Federation Russian Federation Latvia Vietnam Croatia Malta Vietnam Latvia Lithuania Lithuania Chile Croatia Malta Kazakhstan Cyprus Malaysia Bulgaria Romania Uruguay Bulgaria Malaysia Cyprus Romania United Arab Emirates United Arab Emirates Chile Turkey Moldova Trinidad and Tobago Turkey Argentina Montenegro Montenegro Uruguay Costa Rica Trinidad and Tobago Colombia Albania Kazakhstan Thailand Mexico Mexico Moldova Argentina Jordan Georgia Thailand Costa Rica Albania Qatar Brazil Lebanonl Georgia Colombia Qatar Peru Peru Jordan Indonesia Indonesia Tunisia Brazil Dominican Republic Macedonia, FYR Macedonia, FYR Tunisia Kosovo Kosovo Algeria Algeria Lebanonl Dominican Republic 300 400 500 600 300 400 500 600 Score Score Non-OECD OECD OECD interquartile range imum competency levels, and a small fraction reach the KEY CHALLENGES OECD average (Figure 40). 169. During the last 30 years, Mexico’s primary and lower 170. Education outcomes vary substantially across regions. secondary enrollment rates have significantly improved, For example, the populations of Mexico City and Nuevo but retention rates at the upper secondary level and over- León have an average of 10.5 and 9.8 years of education all educational attainment indicators continue to lag those respectively, while the average for Chiapas is just 6.3. of Mexico’s regional and OECD comparators. For instance, Children in southern Mexico not only spend fewer years only 6 out of 10 students who enroll in upper secondary in school, they also learn less than their peers in other ultimately graduate, and graduation rates are even lower areas of the country. In the 2017 PLANEA test, roughly among students from poor and vulnerable households. three-quarters of students finishing lower secondary Mexican students score poorly on national and interna- (grade 9) in Guerrero and Tabasco were not adequately MEXICO tional standardized tests, with wide gaps between stu- proficient in math, compared to about half in Puebla and POLICY NOTES dents from poor and non-poor households. Mexico’s PISA Mexico City. These regional disparities account for much 75 scores in mathematics, reading and science are relatively of the differences in education trajectories between stu- low: only half of all Mexican students achieve the min- dents from poor and non-poor households; only 7 per- Chapter 7 cent of students form marginalized households finish 172. Many factors contribute to the gap in educational upper secondary on time and with minimum learning outcomes between students from higher- and lower-in- outcomes versus close to 30 percent among non-margin- come households, including differences in early childhood alized households.175 development and in the quality of education services. Poor- er households lack critical information about early child- 171. Poor educational outcomes reflect both a lack of stu- hood development. They also have fewer resources to in- dent preparation—especially among students from margin- vest in children’s health and education and relatively low alized households—and the inability of the system to meet expectations regarding the returns to such investment. the needs of inadequately prepared students. Household Consequently, parents in lower-income communities are factors play a crucial role in educational attainment, ex- less likely to engage in activities that can contribute to plaining up to half the variation in outcome indicators ob- their children’s cognitive and non-cognitive development. served at any given point in time. Because the education A lack of early stimulation has especially significant and system must provide foundational skills to all students, lasting effects on the lifetime accumulation of cognitive regardless of household factors or other initial conditions, and socioemotional skills. In extreme cases, such as when teacher quality is the most important determinant of suc- pregnant women are undernourished, children from the cess. Before the 2013 reforms, teachers in Mexico were poorest households are at a disadvantage from birth. selected on a discretionary basis; once hired, they were When these children enter the formal education system, neither evaluated nor motivated to improve their perfor- they already face a significant skills deficit compared with mance. The resulting process of adverse selection attract- children from non-poor households. Moreover, children ed relatively poor candidates to the teaching profession.176 from the poorest households often attend the worst-per- Before 2013, in-service teacher training was driven by sup- forming schools, which suffer from deficient infrastructure, ply and based on unclear criteria. Although the reforms inadequate learning materials, and unqualified teachers. created the possibility of developing a teacher-training Rather than leveling the playing field for students at differ- system based on evaluation results, such a system has yet ent income levels, the Mexican education system tends to to be established. exacerbate social and economic inequalities. Figure 41: Spending per Student at the Basic Education Level by State, 2013 50,000 60% 45,000 50% 40,000 35,000 40% % of State Budget 30,000 MXN 25,000 30% 20,000 20% 15,000 10,000 10% 5,000 0 0% Estado de Mexico Michoacan Guerrero Tlaxcala Hidalgo Quintana Roo Morelos Tabasco San Luis Potosi Oaxaca Jalisco Guanajuato Puebla Queretaro B.C. B.C.S. Durango Campeche Zacatecas Sinaloa Sonora Coahuila Yucatan Aguascalientes Veracruz Chiapas CDMX Colima Tamaulipas Nuevo Leon Nayarit Chihuahua Spending per Student Educación Básica State National Average Spending per Student Educación Básica Federal Share of State Budget Source: World Bank (2016) MEXICO POLICY NOTES 76 175 De Hoyos, Estrada and Vargas (2018) “Predicting individual wellbeing through test scores: evidence from a national assessment in Mexico”, Policy Research working paper; no. WPS 8459. Washington, D.C 176 Estrada (2015) Rules rather than discretion, teacher hiring and rent extraction. http://cadmus.eui.eu/bitstream/handle/1814/36535/MWP_2015_14.pdf 173. Mexico spends just 0.6 percent of its GDP on devel- teacher training: (i) providing complementary learning Chapter 7 opmental programs for children under the age of five, and materials, (ii) focusing on a specific subject, and (iii) con- spending is heavily concentrated among children in the ducting follow-up visits. The SPD could better leverage three-to-five age group.177 Mexico’s universal preschool the information generated by in-service teacher evalua- policy for children between the ages of three and five is tions to increase the effectiveness of its training strategy. responsible for the overwhelming majority of spending Until recently, teacher training in Mexico was entirely on children under five. Mexico spends just 0.02 percent supply-driven, and all teachers received the same instruc- of GDP on early stimulation for children under three, well tional content. Teacher evaluations implemented as part below the levels of other middle-income countries in the of the SPD provide the Secretariat of Public Education region. By contrast, Colombia and Chile spend 0.19 and (Secretaría de Educación Pública, SEP) and the National In- 0.25 percent of GDP, respectively. stitute for the Evaluation of Education (Instituto Nacional para la Evaluación de la Educación, INEE) generates all the 174. A regressive distribution of financial resources contrib- necessary information to align the provision of teacher utes to the positive correlation between school quality and training with the demands of Mexican teachers. Training household income level. In relatively affluent areas such as materials, focus areas, and follow-up schedules could be Mexico City and Nuevo León, investment per student at customized to reflect the needs of teachers, increasing the basic education level is above the national average of the relevance of training and enhancing its impact on roughly MX$20,000 per year, whereas in Guerrero, Chiapas, student learning. and Oaxaca, the poorest states in Mexico, investment per student is significantly below the national average (Fig- 178. In-service teacher-training strategies should be de- ure 41). The unbalanced distribution of resources reflects signed at the school level and incorporated into school-im- both federal transfers with limited equalization power (for provement plans in line with the five pillars of the SPD and needs and revenue capacity) and differences in education the government’s SBM strategy. The SBM strategy intro- investment by state. duces new managerial practices focused on improving student learning. It provides school directors with tools to monitor learning outcomes in math and reading compre- POLICY OPTIONS hension and enables administrators to observe teachers in the classroom via a technique known as Stallings class- room observations. The SBM strategy also enables school Leveraging Performance directors to accurately appraise teacher performance and Incentives to Improve Teacher provide timely feedback on areas for improvement. Incor- Quality porating the results of SPD teacher evaluations into the 175. Align all the elements of the education system into a SBM strategy could increase the amount of performance coherent whole. Harmonizing the different components information available to school directors, allowing them to of the education system would enhance the effectiveness design more effective teacher-training plans. It would also of the reforms and create the foundation for a long-term strengthen the coherence of education policy and more national education strategy. The adoption of the new ed- fully align the education system with the objective of im- ucation model is a strong first step in aligning the system, proved student learning. but some parts of the SPD remain inconsistent with the overall approach to education quality. Adopting a National Early Childhood Development Policy 176. Strengthen the effectiveness of Mexico’s teacher-train- ing colleges (normales). Without improving the quality of 179. A significant share of the observed developmental the institutions responsible for training teachers, the SPD gaps between children from households at different income will have a limited impact on student learning outcomes. levels emerge before age three, and early childhood devel- The normales could improve their effectiveness by apply- opment programs should primarily target disadvantaged ing the SPD’s rules and evaluation criteria to their own in- children. In Mexico, however, spending on children under structors, and over the medium term, the authorities could three is both low overall and disproportionally focused mandate that the normales adopt merit-based systems for on children in urban areas. The center-based services selecting and promoting instructors. The normales could offered by the Secretariat of Social Development (Sec- also draw on the results of the SPD evaluations to inform retaría de Desarrollo Social, SEDESOL) and the Mexican their institutional improvement strategies. Institute of Social Security (Instituto Mexicano del Seguro Social, IMSS) have the greatest coverage. IMSS centers are 177. In-service teacher training could enhance student relatively well-funded and have low children-to-teacher learning, but the system must incorporate some additional ratios. In rural areas, the National Council for Educational features into its design. The international evidence has Development (Consejo Nacional de Fomento Educativo, MEXICO identified three best practices for effective in-service CONAFE) provides community-based early stimulation POLICY NOTES 177 IADB, 2016. 77 Chapter 7 programs run by local non-professional teachers for chil- 183. Quality assurance. National measures and standards dren under age four. However, the quality of CONAFE pro- to supplement program-specific standards are required. grams is far below international standards, and extremely While Mexico has made significant progress in collecting high overhead costs appear to indicate low expenditure anthropometric measures for children age 0-3, there is efficiency.178 very little information about child cognitive and socio- emotional development, as well as on the quality of child- 180. The successful implementation of an early childhood care provision. The availability of data is also key for the development policy depends crucially on the quality of the evaluation of single programs. individual programs and the infrastructure that supports those programs. According to the literature there are four 184. Human resources. While human resource develop- elements of the system’s infrastructure that are key for ment may be included as an individual program feature, the delivery of quality, equitable and sustainable services: the actual availability of a pool of well-trained potential governance, finance, quality assurance, and human capac- personnel depends on elements that transcend the indi- ity. In each of these four areas Mexico has scope for signif- vidual programs. Mexico might take advantage of the re- icant improvements. form of the Escuelas Normales in order to create teachers with ECD specific profiles. 181. Governance. One possible solution to improve horizon- tal (across sectors) and vertical (across government levels) coordination could be the creation of a boundary spanning Realign the National Education entity (BSE) – an institution with an explicit mandate to co- Budget to Provide an Equitable ordinate efforts among the myriad of relevant institutions. Distribution of Resources Similar to the approach followed by De Cero a Siempre 185. To equalize opportunities between students at (DCAS) in Colombia, the central coordinating agency might different income levels, investment in education should not be part of any particular line ministry, but rather of the be allocated progressively, and it should reflect student president’s office, signaling a strong political mandate for need. Revising the resource allocation criteria used in coordination. The new agency might potentially rely on the joint federal-state system for education financing Sedesol as operational structure on the ground. could produce a more equitable and efficient distribu- tion of resources. For example, funds could be allocated 182. Financing. Proper financing mechanisms ensure suf- on a per student basis, adjusted according to the relative ficient resources initially and secure funding for the life of marginalization of each school’s local community, and the program. In order to maximize both efficiency and eq- complemented by additional results-based financing. uity, ECD policies should target the most disadvantaged The results indicators used to allocate financing could with high quality provisions. Mexico spends too little in include final outcomes, such as improvements in stan- policies for children age 0-3 belonging to the most dis- dardized test score, as well as instrumental goals such advantaged households. Using as benchmarking the cost as the implementation of the SPD, the SBM approach, or per child of a cost-effective ECD program implemented other key policies. in Peru (Cuna Mas), we have simulated the cost of cov- ering all children who belong to the “población objetivo” 186. Based on the entry examinations, the authorities could of Prospera and Conafe. This would require increasing create a system of financial and professional/career progres- the spending per capita from 0.02 to 0.20 percent of the sion incentives designed to encourage high-performing GDP, that could be partly funded through an increase in teachers to work in schools in marginalized communities. income tax and the reduction in inefficient public spend- Despite its limitations, the entry examination into the SPD ing. Nuñez (2017) finds that in 2015 19.7 percent of the is a sound proxy for future teacher performance, and it al- public expenditure was used for items different from lows the authorities to identify the most effective teachers those originally budgeted for. entering the education sector. MEXICO POLICY NOTES 78 178 Brookings, 2017. 8. Strengthening the Healthcare System R ecent reforms have expanded access to health care significantly and improved health outcomes. National statistics indicate that the share of the population without access to healthcare has been halved in less than a decade, falling from 38.4 percent in 2008 to 15.5 percent by 2016. Seguro Popular covered 52.6 million Mexicans who previously lacked health insur- ance. Despite increased spending, a combination of allocative and technical inefficiencies weak- ens the impact of the health budget. Out-of-pocket spending remains high, especially among lower-income households. Large disparities in the quality of health services across states com- pound larger regional socioeconomic inequalities. Nationwide, an aging population and changing lifestyle factors are driving an increase in chronic noncommunicable diseases and degenerative conditions, which are intensifying spending pressures on the healthcare system, including for long term care. Overall, the health system is fragmented, contributing to cost and service inefficiencies. The role of primary care providers has been weak, and local entities responsible for managing primary care vary substantially in their capabilities to handled it. Dealing with growing needs and medium-term spending pressures while maintaining fiscal sustainability for the system will re- quire further reforms. They include, among other, (i) strengthening pharmaceutical regulation, procurement, and supply-chain to help reduce expenditure across systems; (ii) better coordinating between healthcare schemes (e.g., through reciprocity agreements for exchange of services) to pursue efficiency gains and a better use of facilities; (iii) kickstart the process of harmonization of benefits packages across insurance systems; (iv) start working on a more substantial administra- tive integration of the insurance systems, including patient data; (v) integrating some financing and investment planning functions within each health insurance scheme; and (vi) reorganizing MEXICO health jurisdictions around a primary-care-based model. POLICY NOTES 81 Chapter 8 were about 43 percent of the OECD average and below CONTEXT AND REFORM the Latin American average. Mexico public spending on PROGRESS health is also lower than most individual regional com- parators, with the notable exception of Argentina. Addi- 187. During the first half of the 2000s, Mexico implemented tionally, important savings have been achieved through major reforms that greatly expanded insurance coverage the consolidated purchasing of pharmaceuticals and the and substantially improved health outcomes. Aside from role of the General Health Council (Consejo General de the private sector Social Protection System for Health Salud), established to coordinate the adoption of new (IMSS) and that for public sector employees (ISSSTE), a pharmaceuticals and medical devices for six insurers.180 noncontributory health care system for vulnerable house- Its role is determining, through a cost-benefit analysis, holds, better known as Seguro Popular, was established whether those insurers should only purchase products in 2003. By 2012, Seguro Popular covered 52.6 million deemed safe and effective by the Federal Commission for Mexicans who previously lacked health insurance. Since the Protection Against Health Risk (Comisión Federal para then, access to health services has continued to expand. la Protección contra Riesgos Sanitarios). National statistics indicate that the share of the popula- tion without access to healthcare has been halved in less 189. Despite these advances and growing public spending, than a decade, falling from 38.4 percent in 2008 to 15.5 Mexico continues to face important challenges in ensuring percent in 2016.179 Since 2000, the maternal mortality rate equitable access to quality health services. Despite an in- has fallen by 50 percent, and infant mortality has dropped crease in total health spending from 4.9 percent of GDP by more than 40 percent. in 2000 to 6.3 percent in 2014, Mexico’s health spending remains relatively below comparators. In 2014, Mexico’s 188. The establishment of Seguro Popular significantly total health expenditure was about half the OECD aver- increased the amount of public resources invested in the age and below the Latin American average of 7.2 percent. health sector. Between 2004 and 2014, public health Public sector health spending, including both govern- spending as a share of GDP rose by approximately 43 per- ment schemes and compulsory health insurance, also rep- cent. Importantly, Seguro Popular significantly reduced resents a modest share of total health spending: in 2015, the incidence of catastrophic and impoverishing health public spending comprised 52 percent of total health spending among lower-income households. Between spending, versus an OECD average of 70 percent and low- 2004 and 2014, catastrophic health spending by house- er than most individual regional comparators, with the holds the poorest income quintile declined by 11 per- notable exception of Argentina. cent, and impoverishing health spending fell by 54 per- cent (González Block & Martinez, 2015). Mexico’s health 190. The health system is fragmented, contributing to a spending remains relatively low by global and regional number of costs and service inefficiencies. Several discon- standards. In 2014, Mexico’s total health expenditures nected sub-systems provide healthcare, and employ- Figure 42: Health Insurance and Health Spending in Mexico a. Insurance Coverage by Income Quintile, 2016 b. Health Spending by Type, 2003-2014 4 60 43.7 20.8 55 50.4 3.5 61.2 50 76.4 % of GDP % of THE 87.3 3 45 75.5 54.5 40 2.5 47.6 35 37.4 23.0 2 30 11.4 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 Q1 Q2 Q3 Q4 Q5 Total Social Security Seguro Popular Health expenditure, Out-of-pocket health expenditure Private Insurance Other public (% of GDP) (% of total expenditure on health) MEXICO POLICY NOTES 179 It refers to people with access barriers, even though they may have an insurance affiliation. Source: National Council for the Evaluation of Social Development Policy (CONEVAL), https://www.coneval.org.mx/Medicion/Paginas/Evolucion-de-las-dimensiones-de-pobreza.aspx, https://www.coneval.org.mx/Medicion/EDP/Paginas/Datos-del-Modulo-de-Condi- 82 ciones-Socioeconomicas.aspx. 180 In 2014, the consolidated procurement of pharmaceuticals yielded an estimated savings of 0.03% of GDP. However, even consolidated procurement still involves several discreet processes, and the procurement system remains largely confined to pharmaceuticals. ment status determines insurance coverage and access ment budget, resulting in a very low rate of gross fixed Chapter 8 to provider networks.181 The Mexican Social Security In- capital formation, which reached just 0.1 of GDP percent stitute (Instituto Mexicano del Seguro Social, IMSS) covers in 2015. The lack of coherent prioritization criteria within private-sector workers and their families. The Institute for and across subsystems is also one of the causes of ineffi- Social Security and Services for State Workers (Instituto ciencies and inequities. de Seguridad y Servicios Sociales de los Trabajadores del Estado, ISSSTE) covers federal government employees. 193. Despite the expansion of insurance coverage, out-of- The poor and unemployed are enrolled in Seguro Popular. pocket (OOP) spending remains high and disproportion- Self-employed, non- salaried, and informal workers are ately impacts the poorest households. OOP spending fell covered by one of several federal programs managed by from 55 percent of total spending in 2003 to 44 percent the Secretariat of Health (Secretaría de Salud), including in 2014, but remains a key source of inefficiency and a Seguro Popular and the conditional cash- transfer pro- major barrier to health access among the poorest house- gram IMSS-Prospera. Despite widespread coverage, in- holds. In Mexico, OOP health expenditures are largest for come levels are closely correlated with health outcomes: ambulatory care, and pharmaceutical costs are especially infant mortality rates, for example, are 20 times higher in high. OOP spending is also highly regressive: 45 percent the most marginalized (poorer) municipalities than in the of total ambulatory OOP health costs are borne by house- least marginalized. holds in the first three income quintiles, and 28 percent are borne by non-insured households across all income 191. An aging population and an ongoing epidemiological quintiles, especially non-insured households in quintile transition are intensifying pressure on Mexico’s healthcare four.182 Also, the poorer quartile of the population spends system. The dependency ratio—i.e., the number of people a significantly higher (37 percent) higher proportion of over age 64 relative to the number between the ages of their income on health than the richest quartile. The data 15 and 64—rose from 7.4 percent in 1985 to 10 percent do not indicate why high levels of OOP spending persist, in 2016. The aging of the population, compounded by but they may reflect dissatisfaction with the quality or ac- behavioral factors such as diet and exercise, is shifting cessibility of services provided by institutions covered by the disease burden from acute communicable diseases health insurance, prompting patients to seek care from to chronic degenerative diseases. By 2015, all five lead- private health providers as shown in Figure 43a. OECD, ing causes of death in Mexico were noncommunicable 2016.183 diseases, and the country’s obesity rate was among the highest in the world. The 2016 National Health and Nutri- 194. Health access has expanded, but insurance coverage tion Survey (Encuesta Nacional de Salud y Nutrición, ENSA- of services for high costs treatments is unequal. Seguro NUT) found that 72.5 percent of adults were overweight Popular now covers 50.4 percent of the insured popu- or obese, along with 36.3 percent of adolescents, and 33.2 lation, while the various social security programs cover percent of children age 5 to 11. Managing chronic diseases 47.6 percent.184 The coverage packages provided by Se- associated with old age and lifestyle factors tends to be guro Popular and the social security programs have con- far more complex and expensive than treating acute com- verged over time and are now very similar, particularly municable diseases, and addressing the challenges facing coverage for primary care. In addition, the fact that per- the Mexican health sector will require changing the way sons from higher level income quintiles are also affiliated healthcare is organized, funded, and managed at both the to Seguro Popular indicate that many households view national and regional levels. SPSS enrollment as a complement to social security en- rollment rather than a substitute.185 However, coverage for high- cost treatments still differ significantly. About KEY CHALLENGES 38 percent of IMSS affiliates lose their medical benefits each year and become eligible for Seguro Popular.186 At any given time, an estimated 13.4 percent of the popu- Improving the Efficiency, Scope, lation lacks a comprehensive coverage of high-cost con- and Equity of Health Services ditions, though Seguro Popular still covers adverse health 192. Health spending suffers from both technical and al- events.187 While Seguro Popular reduces the incidence of locative inefficiencies. Administrative costs accounted for catastrophic health expenditures,188 these effects vary almost 10 percent of total health spending in 2014, the across population groups and health conditions. More- highest level in the OECD. These high administrative costs, over, the estimated reduction in catastrophic health ex- which chiefly stem from the fragmentation of the system, penditures is only a few percentage points, implying that squeeze the fiscal envelope for the health sector’s invest- the risk is still substantial. 181 OECD, 2016. MEXICO 182 González Block, 2015. POLICY NOTES 183 OECD, 2016. 184 The remaining 2% are covered by specific social security arrangements for public institutions such as the armed forces or the national oil company. 185 See, for instance, Kurowski and Villar, 2012. 83 186 Guerra et al., forthcoming. 187 ENSANUT, 2016. 188 See: Ávila-Burgos, et al., 2013; King, et al., 2009; Grogger et al., 2015; Knox, 2016. Chapter 8 Figure 43: OOP Spending and Insurance Coverage by Income Quintile a. OOP Spending by Income Quintile and Healthcare Provider, 2012 b. Distribution of out-of-pocket ambulatory health expenditures according to income quintiles and financial protection status 8.1 29.1 9.0 % % 5.3 12.0 15.9 8.0 2.6 3.3 24.3 8.0 9.4 0.2 14.0 12.4 0.0 0.3 10.4 10.7 6.4 0.3 1.6 5.6 5.2 3.2 2.8 1 2 3 4 5 1 2 3 4 5 Income quintile Income quintile Public Own institution Private No protected Protected Source: González Block MA 2015 based on ENSANUT 2012 Source: Gonzalez Block 2017 based on ENSANUT 2016 varies across states and is especially low in the poorest Consolidating the Healthcare regions, contributing to the overall inefficiency of the System sector.189 Chronic disease and homicide rates contribute 195. The fragmentation of Mexico’s healthcare system con- to disparities in life expectancy at the state level that tributes to inefficient spending and use of assets and ser- also reflect the socioeconomic differences that extend vices. The current structure prevents the materialization beyond the health sector.190 States with high marginaliza- of economies of scale, results in redundancy, unequal ac- tion rates exhibit the largest health risks in terms of sexu- cess to services, and the suboptimal utilization of human al protection, tobacco use, and alcohol consumption, po- and physical capital. Moreover, the size and effectiveness tentially indicating differences in the scope and efficacy of cross-subsidies among public health institutions are of public health outreach. The service quality provided by difficult to measure due to the lack of a unique roster each insurance scheme also varies, with particularly large of beneficiaries. The fragmentation of the health system gaps in the quality of care for myocardial infarction mor- into separate, vertically integrated insurance programs tality.191 Mortality rates within 30 days of hospitalization also limits patient choice and weakens efficiency incen- for cardiac infarction vary from 3 percent among high-in- tives. It also inhibits access to care, since patients can only come, mostly privately insured patients with access to use healthcare providers within their insurer’s network, private care to 6.5 percent among IMSS-insured patients regardless of geographic location, service availability or cared for at IMSS facilities and reach a staggering 59 per- medical condition, and is the primary determinant of the cent among patients cared for elsewhere. This disparity already noted high administrative and insurance costs is due largely to the fact that Seguro Popular only covers (almost ten percent of total health expenditures, com- cardiac infarction cases in patients under age 60. Hospi- pared to three percent in the OECD). The absence of sys- tals are often unwilling to devote substantial resources tem-wide planning produces an inefficient allocation of to treating uninsured patients, as they are unlikely to be facilities and resources, leading to excess supply in some reimbursed. areas and shortages in others. Each insurer maintains its own pharmaceutical and medical-device supply chains 197. Efforts to harmonize the activities of insurance provid- and develops its own standards of care, which can vary ers have been only partially effective. The effectiveness of substantially. the General Health Council (Consejo General de Salud) has been limited because, in practice, insurers still conduct in- 196. The problems of fragmentation are compounded by ternal reviews of medical products and make independent differences in resource allocation and socioeconomic condi- decisions regarding the definition of their prioritization tions across states. The geographic distribution of public criteria, benefit packages and use of medical devices and health resources does not reflect the needs of the pop- pharmaceuticals—decisions that are often based on bud- ulation, as transfer to states, despite improvements, fall getary constraints. Consequently, individuals covered by short in equalizing local resource availability and spend- different insurers have access to very different medicines MEXICO ing needs. In addition, financial management capacity and services. POLICY NOTES 84 189 Knaul et al., 2012. 190 Gomez-Dantés, 2016. 191 World Bank, 2013. Figure 44: Financial Flows in the Mexican Health System Chapter 8 Financing sources FIRMS HOUSEHOLDS Para-statal income Tributary Employer Employee Affilation Non-tributary income income contributions contributions fees FASSA (Ramo 33) Federal State Government Governments Ramo 12 Ramo 12, Ramo 19 Ramo 12 Financing Social security Pre paid Schemes Seguro IMSS- Household MoH/SESA IMSS/ISSSTE/SEDENA*/ private popular prospera out-of pocket payments SECMAR*/PEMEX* insurance Hospitals, health Hospitals, health clinics, and Hospitals, health clinics, and clinics, and Private providers healthcare physicians of MoH/SESA physicians of social security physicians of providers institutions IMSS-Prospera Beneficiaries Population without social security Population with social security Self-employees, workers in the informal sector, Workers in the formal sector, Population with the capacity to pay unemployed and government employees Source: Adapted from Avila-Burgos et al. (2016), Gonzalez-Block et al. (forthcoming) and Frenk and Gómez-Dantes (2015) Figure 45: Disparities in the Use of Ambulatory Care and Cardiac Infarction Mortality a. Relative Gap in the Probability of Using Ambulatory Care between b. Disparities in Mortality Rates due to Acute Myocardial Patients in the Highest and Lowest Income Quintiles, 2012 Infarction, 2015 58.70 Ambulatory care provider None 1.7 AMI Mortality rate Any provider 0.9 Private 0.5 6.50% ambulatory care 2.5 for not using Supply barriers Reasons 3.90% Demand barriers 1.4 Unisured (50.4% of Social security (40.9% Private care (0.7% total population) of total population) of total population) Source: Adapted from Bautista Arredondo et al. (2013) Source: Azpiri-López (2016); ENSANUT-MC (2016). try. The focus of health care provision is often curative, Developing a High-Quality, with relatively few preventive interventions. In the IMSS, Equitable, Outcome-Driven primary care is the purview of a small number of family Healthcare System physicians, who are trained in-house, while the Ministry 198. The role of primary care providers has been weak, and of Health uses general practitioners, noncertified family local entities responsible for managing primary care lack physicians, or social-service interns.192 The rate of avoid- clearly defined roles and their capabilities vary substantially. able hospitalizations in hospitals run by the Ministry of Despite the introduction of a new comprehensive health- Health increased from 13.4 per 10,000 insured people in care model (Modelo de Atención Integral de Salud, MAI), 2001 to 19.7 in 2011, even though the expansion of Seguro in practice there is no single model of care in the coun- Popular increased access to primary care.193 Lower-income MEXICO POLICY NOTES 85 192 Comisión Nacional de Protección Social en Salud, 2014. 193 Lugo Palacios et al., 2015. Avoidable hospitalizations may reflect a low quality of care by primary-care providers that fail to prevent or treat a condition that later requires hospitalization. Chapter 8 Figure 46: Preventable Hospitalization and the Institutional Strength of Health Jurisdictions a. Diabetes Hospital Admission in Adults, 2006 and 2011 600 500 400 300 200 100 0 Iceland UK Canada Israel Swede Denmark New Zeland Czech R Italy Switzerland Norway Spain Portugal Australia Ireland Finland OECD Belgium USA Latvia Germany Singapore Poland Chile Korea Mexico Austria Hungary 2006 2011 Source: OECD Health Statistics 2014 b. Institutional Strength of Jurisdictions: National average Acciones o actividades 100 80 60 40 Encargado con Reglamentos entrenamiento 20 0 Encargado Fondos específicos Source: The World Bank – HNSP Mexico 2017 households with low levels of educational attainment and 199. The use of parallel, inconsistent data-collection and households living in highly marginalized areas are less like- information systems complicates sectoral monitoring, over- ly to use ambulatory care providers than wealthier house- sight, and policymaking. States, social security institutions, holds with greater educational attainment in less- margin- healthcare providers, and health jurisdictions have cre- alized areas. Health jurisdictions (jurisdicciones sanitarias) ated multiple data-collection systems that do not follow are responsible for directing, managing, and operating the same parameters and are not interoperable, i.e. they primary care facilities, along with the State Health Services are not able to exchange and/or make use of the same (Servicios Estatales de Salud, SESAs). Health jurisdictions are information. The inadequate scope and quality of the tasked with designing and implementing health programs available data inhibit the adoption of a patient-centered that reflect local needs, whereas SESAs are responsible model of care. Multiple initiatives with data-integration for providing human, material, and financial resources to components are underway, including the rollout of the medical units. However, the health jurisdictions and SESAs National Health Information System (Sistema Nacional de have not been adapted to suit the post-Seguro Popular Información Básica en Salud, SINBA), the Comprehensive health system. The institutional capacity of the health Health Quality System (Sistema Integral de Calidad en Sa- jurisdictions varies, but their capabilities are generally lud), Seguro Popular’s Health Payroll System (Sistema Nómi- limited. Capacity differences reflect an uneven allocation na en Salud), and IMSS Digital, among others. However, the of resources, an outdated geographical distribution of institutional complexity of the health sector, legal restric- facilities, and a lack of normative instruments to regulate tions, political cycles, and insufficient budgetary resources MEXICO their responsibilities under the MAI. Consequently, health to finance the large- scale deployment of new technology POLICY NOTES jurisdictions lack the capacity and the resources necessary have hindered these efforts. Integrated information sys- to address the new challenges and responsibilities intro- tems will be critical to the success of various interventions 86 duced by the MAI, which emphasizes primary care as the currently being discussed by policymakers, such as the basis for the overall model of care. strategic procurement of supplies, the unification of ben- efit plans, the adoption of risk- based budgeting, and the la Negociación de Precios de Medicamentos y Otros Insumos Chapter 8 use of uniform evaluations for medical technology.1194 para la Salud) and improving both the distribution of med- icines across states and the management of drug stocks in individual health units; and (iii) harmonizing prioritization POLICY OPTIONS criteria across the different subsystems. Improving the Efficiency, Scope, Consolidating the Healthcare and Equity of Health Services System 200. Adopt performance incentives and other innovative 202. In the short term, better coordination between health- instruments to improve results and resource allocation. The care schemes could yield major efficiency gains. In the short government could explore innovative mechanisms to term, reciprocity agreements enabling the exchange of allocate the budget transparently across states based on services between different healthcare schemes could projected expenses, local health needs, and demographic maximize the value of healthcare facilities and infrastruc- characteristics. Performance-based payment systems and ture— the emergency obstetric care agreement should other financial incentives could encourage greater invest- serve as a model for cross-scheme collaboration. ment in prevention, increasing the system’s ability to re- solve health problems (resolutive capacity), and integrate 203. Kickstart the process of harmonization of benefits patient management across administrative levels and packages across insurance systems. The authorities have healthcare providers. Adopting performance benchmarks made progress in aligning primary-care interventions, and could enhance hospital and provider- network autonomy. criteria have been established to update benefits based on Developing internal markets, i.e. markets for health prod- cost-effectiveness, affordability, financial protection, the ucts and services, especially those related to primary care opinion of the scientific community, supply and demand services, could ensure more equitable financing for prima- factors, and social acceptance. The Health Secretariat ry care and essential hospital interventions. The govern- could lead efforts to define an explicit reference package, ment could also consider introducing a risk-based formula strengthen the process for evaluating medical technolo- for resource allocation that accounts for differences across gy, update benefits based on the established criteria, and regions—based on more rigorous and comprehensive launch a round of technical negotiations with the social data—and increasing the contribution of private resourc- security institutions to discuss parameters and a timeline es, such as copayments and complementary private insur- for harmonizing benefits. This process can help to reduce ance, to fund the care of non-poor patients. At a provider disparities in the quality and comprehensiveness of care. level, options such as Public Private Partnerships (PPP) have shown potential for efficiency gains, contributing to 204. Eliminating beneficiary duplication can render fiscal fill the infrastructure gaps and introducing international savings. Eliminating duplications in Seguro Popular cov- standards of quality, with little or none fiscal impact. erage alone could have saved 0.1 percent of GDP in 2014 alone. There are, in addition, further potential savings 201. Strengthen pharmaceutical regulation and sup- from eliminating IMSS and ISSSTE duplications. There is an ply-chain to help reduce expenditure across systems. The urgent need to consolidate beneficiaries in a single verifi- international experience has shown that controlling phar- able roster. Efforts are underway to create a unique roster maceutical prices through market mechanisms can be of beneficiaries across public health institutions, and the very effective in reducing healthcare costs. Policymakers government is coordinating the process of information can stimulate competition among suppliers of generic exchange and cross-checking. However, this project is still drugs while controlling prices for expensive patent-based in its early stages and has not yet been used to eliminate name-brand-name drugs. The prices of drugs covered by redundant payments. benefit plans should be frequently revised to reflect evolv- ing market dynamics, technological innovations, and new 205. Start working on a more substantial integration of ad- information about therapeutic effectiveness. This can be ministrative systems. This would entail automatically regis- can be complemented by the mandatory substitution of tering former IMSS beneficiaries with Seguro Popular and generic alternatives and centralized procurement. Imple- establishing mechanisms to coordinate enrollment and menting these measures will require: (i) building the ca- the management of clinical data. In addition, a transparent pacity of the National Center for Technological Excellence process for purchasing services across all insurance sys- in Health (Centro Nacional de Excelencia Tecnólogica en tems (i.e., service-exchange agreements) could optimize Salud) to conduct robust, independent assessments of the utilization of health infrastructure in areas covered health technology; (ii) strengthening supply-chain man- by both social security institutions and Seguro Popular, agement for pharmaceuticals by supporting the Coordi- while also expanding overall coverage and improving the nating Commission for Negotiating Prices for Medicines financial outlook of both healthcare providers and public MEXICO and Other Health Supplies (Comisión Coordinadora para insurers. Leveraging new technologies for managing ap- POLICY NOTES 194 OECD, 2016. 87 Chapter 8 pointments based on capacity could further boost the effi- payment and budgeting. The government’s strategy for ciency of health-sector resources by taking full advantage integrating electronic health records should be informed of underutilized infrastructure. by two promising initiatives: Seguro Popular’s Health Pay- roll System and IMSS Digital. Establishing an integrated 206. There are also potential inefficiencies linked to the in- national health information system will require dedicated tegration of financing and investment planning. This current budgetary resources and well-designed rules for manag- system hampers any opportunity for strategic purchasing ing health data that reflect international best practices and thus getting the best value for the money. In addition, and enjoy broad political support. provider payment mechanisms are not linked to results; they remain mostly based on historical budgets with little or no incentive for cost- containment, or improving ser- Developing a High-Quality, vice delivery to patients. Greater investment coordination Equitable, Outcome-Driven between healthcare schemes would also increase the im- Healthcare System pact of future capital spending. 208. Reorganize health jurisdictions around a prima- ry-care-based model. As the frontline units of the health 207. Consolidate data-collection systems and adopt mech- system, health jurisdictions should be central to the gov- anisms to improve the quality and comprehensiveness of ernment’s strategy for preventing and managing non- data. The authorities could strengthen, integrate, and ex- communicable diseases and chronic health conditions. pand the various databases used by different elements of Building the capacity of health jurisdictions will require: the health sector, with the goal of building a national elec- (i) updating their operational guidelines, (ii) publicizing tronic health records system encompassing both patients effective strategies for preventing and managing chronic and providers. This system would allow policymakers to diseases, (iii) allocating the funds necessary to implement track data on patients and illnesses across levels of care, those strategies, (iv) adapting personnel training to reflect geographic locations, insurance systems, socioeconomic the evolving disease profile, and (v) increasing the autono- groups, and other domains. More precise and comprehen- my and capacity of jurisdictions to address local challeng- sive administrative data will be vital to accurately assess es. Finally, the Secretariat of Health should create forums provider performance, build a strong primary-care-ori- for representatives of health jurisdictions to share their ented health system, and introduce new mechanisms for experiences and learn from one another. MEXICO POLICY NOTES 88 Chapter 8 References AMIS (2016), Diseño de productos innovadores de aseguramien- Gutiérrez JP (2013), “Gaps in effective coverage by socioeconomic to en colaboración Government- privada. Análisis de necesi- status and poverty condition”. Salud Publica Mex 2013;55 sup- dades de salud y de protección financier. Presentación al pl 2:S106-S111. Comité Ejecutivo de AMIS. AMIS, noviembre. Gutiérrez JP, García-Saisó S, Espinosa-De La Peña R, and Balandrán Azpiri-López, JR (2016), “Observations in relation to myocardial DA (2016a), “Desigualdad en indicadores de enfermedades infarction: Are there three different Mexicos?” Medicina Uni- crónicas y su atención en adultos en México: análisis de tres versitaria 18, no. 72: 181-84. doi:10.1016/j.rmu.2016.06.003. encuestas de salud.” Salud Pública de México 58, no. 6: 666. Avila-Burgos L, Cahuana-Hurtado L, Montañez-Hernandez J, Ser- doi:10.21149/spm.v58i6.7923. van-Mori E, Aracena-Genao B, and Del Río-Zolezzi A (2016). “Financ- Gutiérrez JP, García-Saisó S, Espinosa-De La Peña R, and Balandrán ing Maternal Health and Family Planning: Are We on the Right Track? DA (2016b), “Desigualdad en indicadores de comportamien- Evidence from the Reproductive Health Subaccounts in Mexico, tos de riesgo en adolescentes en México: análisis de dos en- 2003–2012.” Plos One 11, no. 1. doi:10.1371/journal.pone.0147923. cuestas de salud.” Salud Pública de México 58, no. 6 (2016b): Barraza-Lloréns M, Panopoulou G, and Yadira Díaz B (2013). “In- 657. doi:10.21149/spm.v58i6.7922. come-related inequalities and inequities in health and health care Gutiérrez JP, García-Saisó S, Espinosa-De La Peña R, and Balandrán DA utilization in Mexico, 2000 - 2006.” Revista Panamericana de Salud (2016c), “Desigualdad en indicadores de resultados y acceso a la Pública 33, no. 2: 122-30. doi:10.1590/s1020-49892013000200007. salud para niños en México: análisis de tres encuestas de salud.” Sa- Bautista-Arredondo S, Serván-Mori S, Arantxa Colchero M, lud Pública de México 58, no. 6: 648. doi:10.21149/spm.v58i6.7921. Ramírez-Rodríguez B, and Sosa-Rubí SG (2014), “Análisis del Gutiérrez JP, García-Saisó S, Espinosa-De La Peña R, and Baland- uso de servicios ambulatorios curativos en el contexto de la rán DA (2016d), “Monitoreo de la desigualdad en protección reforma para la protección universal en salud en México.” Salud financiera y atención a la salud en México: análisis de las en- Pública de México 56, no. 1: 18. doi:10.21149/spm.v56i1.7319. cuestas de salud 2000, 2006 y 2012.” Salud Pública de México Cahuana-Hurtado L, Chilian-Herrera O, Bautista-Arredondo S, Con- 58, no. 6: 639. doi:10.21149/spm.v58i6.7920 treras-Loya D, Franco-Marina F, González-Vilchis JJ, and Lozano Human Reproduction Program (2015), WHO Statement on Cae- R. (2015), “Corrección de la mala clasificación de las muertes por sarean Section Rates. Geneva, WHO. sida en México. Análisis retrospectivo de 1983 a 2012.” Salud Institute of Medicine (1994), America’s Health in Transition: Pro- Pública de México 57: 142. doi:10.21149/spm.v57s2.7602. tecting and Improving Quality. Washington, DC: The National Castro Arantxu VS and Kaufman H. (2015), “Assessing equitable Academies Press. care for Indigenous and Afrodescendant women in Latin Instituto Nacional de Salud Pública Prospera, Unicef (2016), De- America”. Rev Panam Salud Publica 38, no 2 :96–109 sarrollo infantil temprano de niñas y niños de tres a cuatro Comisión Nacional de Protección Social en Salud (2016), Catálogo años en hogares beneficiarios de PROSPERA: identificación de Universal de Servicios de Salud 2016. México, SSa. factores de riesgo y de protección. Mexico City, INSP. Consejo Nacional de Evaluación de la Política de Desarrollo Social Kurowski, C., and Manuela Villar-Uribe. 2012. “Mexico’s Social (CONEVAL), https://www.coneval.org.mx/Medicion/Paginas/ Protection in Health and the Transformation of State Health Evolucion-de-las-dimensiones-de-pobreza.aspx, https://www. Systems.” Washington, DC: World Bank. coneval.org.mx/Medicion/EDP/Paginas/Datos-del-Modu- Leyva-Flores R, Infante-Xibille C, Gutiérrez-Reyes JP, Quinti- lo-de-Condiciones- Socioeconomicas.aspx no-Pérez F (2013), “Inequidad persistente en salud y acceso Danese-Dl, Santos LG., Sosa-Rubí SG and Valencia-Mendoza A (2011), a los servicios para los pueblos indígenas de México, 2006- “Analysis of changes in the association of income and the utiliza- 2012” Salud Publica Mex;55 supl 2:S123-S128 tion of curative health services in Mexico between 2000 and 2006.” Linares-Pérez N, López-Arellano O (2012), “Inequidades en salud BMC Public Health 11, no. 1. doi:10.1186/1471-2458-11-771. en México”. Gaceta Médica de México; 148:591-7 El Financiero (2015), ¿Cómo se distribuye el ingreso en México? Lugo-Palacios, D G, & Cairns J (2015), “Using ambulatory care sen- Ciudad de México, 16th July. sitive hospitalisations to analyse the effectiveness of primary Guerra Guerra G, Ridríguez E, Doubova S, Saldago de Snyder N, care services in Mexico.” Social Science & Medicine 144 (2015): Borja VH & González Block MA (Pending publication), Annu- 59-68. doi:10.1016/j.socscimed.09.010. al loss of job-related right to healthcare due to employment Secretaría de Salud (2102), Observatorio del Desempeño Hospita- turnover affects over a third of private sector workers in Mexi- lario 2011. Dirección General de Evaluación del Desempeño, co and affects primarily the elderly). Demography. Secretaría de Salud, México. Gómez-Dantes H, Fullman Nancy, Lamadrid-Figueroa H, Cahua- Secretaría de Salud, Seguro Popular, Prospera (2016), Encuesta Nacio- na-Hurtado L, Darney B, Avila- Burgos L, Correa-Rotter R (...) n& nal de Salud y Nutrición, ENSANUT de Medio Camino 2016. Re- Lozano R. (2016), Dissonant health transition in the states of sultados del Módulo PROSPERA. Mexico City, SSA. (Presentation). Mexico, 1990–2013: a systematic analysis for the Global Bur- Shamah Levy T, Cuevas Nasu L, Rivera Dommarco J, Hernández Ávi- den of Disease Study 2013”. The Lancet 388:2386-2402 la M (2016), Encuesta Nacional de Salud y Nutrición de Medio González Block MA (Coord.) (2015), Diagnóstico de los retos al Siste- Camino 2016. Informe final de resultados. Mexico City, INSP. ma Naciona de Salud Universal. México, Cofemedir-Funsalud. Shengelia B, Tandon A, Adams OB, Murray C (2005), Access, utili- González Block MA & Martínez González G, (Coords) (2015) Hacia zation, quality, and effective coverage; an integrated concep- la cobertura universal de la protección financiera de la salud tual framework and measurement strategy. Social Science and en Mexico. Ciudad de México, AMIS. Medicine.;61: 97-10. González Block MA and Scott J. (2010), “Equidad y salud. Retos, Strouse, C., R. Perez-Cuevas, M. Lahiff, J. Walsh, and S. Guendelman avances y perspectivas”. In Cortes F & Oliveira O (Coords), Los (2016). “Mexicos Seguro Popular Appears To Have Helped Reduce grandes problemas de México. V. Desigualdad social. pp 211-234 The Risk Of Preterm Delivery Among Women With Low Education.” Gonzalez Perez GJ, Vega Lopez MG, Cabrera Pivaral CE, Vega Lo- Health Affairs 35, no. 1: 80-87. doi:10.1377/hlthaff.2015.0594. pez A, Muñoz de la Torre A. (2010), “Marginación Social y Mor- Villalobos A, Allen-Leigh B, Salazar-Alberto J, De Castro F, Barrien- talidad Evitable en México”. En: Gonzalez Perez GJ, Vega Lopez tos-Gutierrez T, Leyva- Lopez A, et al. (2017), Quality of repro- MG, Cabrera Pivaral CE (Coordinadores). Desigualdad social y ductive healthcare for adolescents: A nationally representa- equidad en salud: perspectivas internacionales. Universidad tive survey of providers in Mexico. PLoS ONE 12(3): e0173342. de Guadalajara. Guadalajara, Jalisco. doi:10.1371/journal.pone.0173342 Guendelman S, Gemmill A, Thornton D, Walker D, Harvey M, Wagstaff A, Dmytraczenko T, Almeida G, Buisman L, Hoang-Vu Eo- MEXICO Walsh J and R Perez-Cuevas (2017), “Prevalence, Disparities, zenou P, Bredenkamp C, Cercone JA, Diaz Y, Maceira D, Molina S, POLICY NOTES And Determinants Of Primary Cesarean Births Among First- Paraje G, Ruiz F, Sarti F, Scott J, Valdivia M, and Werneck H (2015), Time Mothers In Mexico” Health Affairs 36, no.4:714-722 doi: “Assessing Latin America’s Progress Toward Achieving Universal 10.1377/hlthaff.2016.10842017 Health Coverage”. Health Affairs 34, No. 10:1704-1712. 89 9. Enhancing Social Protection A fter a prolonged period of stagnation, poverty rates have more recently fallen. This reduc- tion can partly be attributed to social protection programs, which have expanded over the past decade and continue to play an increasingly important role in strengthening the economic resilience of the poor and vulnerable. However, reaching the most marginalized and poorest remains a challenge. The proliferation of a myriad of social assistance programs with overlapping purposes and beneficiaries (at least 5,491 social programs operate today at the feder- al, state and local levels) reduces efficiency and the effectiveness of the social assistance system. Sustainably raising the incomes of the poor and expanding their access to opportunities continue to be challenging especially considering the limited spending and orientation of productive and labor inclusion programs. At the same time, multiple social insurance schemes, with disparities in benefits and coverage, exist in parallel, and along with lack of clarity in roles and integration across SA and SI, limit labor mobility, and potentially create unintended disincentives to work and save. The ageing population of Mexico and the need to provide a fiscally sustainable and equitable pension system will only magnify these challenges. Successfully addressing them will require a concerted effort on multiple fronts. The SA system will require consolidation, eliminating overlap- ping and poor performing programs and better funding those that reach the poor meeting ade- quate objectives. This will also require a more sophisticated SP infrastructure, specially a unique digital ID for all beneficiaries of all non-contributory and contributory SP programs and linking administrative databases across all programs. Designing adequately funded productive programs that are better adapted to improve social mobility and using a social intermediation approach for the high-priority communities, will help upward mobility of the poor. Providing a fiscally sustain- able, equitable, and an integrated pension system will require: consolidation and integration of schemes; consideration of alternative ways of financing social insurance premia (e.g. relying less MEXICO POLICY NOTES on payroll taxes and shifting the burden to indirect taxation); and parametric reforms of the exist- ing schemes (e.g. gradual equalization of retirement ages, increase in contribution rates, raising level of non-contributory with an affluence test). 91 Chapter 9 212. Mexico’s social protection system includes numerous CONTEXT AND REFORM programs with diverse objectives and varying degrees of PROGRESS coverage and effectiveness. The country’s contributory social insurance schemes offer pensions and health in- 209. After remaining broadly unchanged for several years, surance to formal sector workers, helping protect them poverty rates have recently fallen. Between 2010 and 2014, against income shocks and smoothing consumption the official multidimensional poverty headcount index over the long term. In addition, non-contributory health stagnated at 46 percent, mostly explained by the lack of insurance (Seguro Popular) and non-contributory (so- improvement in the income poverty rate, which in turn cial) pension provide coverage for those not include in reflects the limited opportunities for the poor to earn an contributory schemes. A wide range of social assistance income. Recent data from the National Council for the (SA) programs—including Mexico’s flagship condition- Evaluation of the Social Development Policy (Consejo al cash- transfer scheme (Programa de Inclusión Social, Nacional de Evaluacion de la Politica de Desarrollo Social, PROSPERA199)—constitute an important part of the social CONEVAL) indicate that the multidimensional poverty protection system. SA programs are designed to directly headcount rate, which is now measured according to a or indirectly reduce poverty and develop human capital. new methodology, fell to 43.6 percent in 2016, while the Labor market programs, which promote employability extreme poverty rate declined from 11.3 percent in 2010 and link workers to job opportunities, constitute a small to 9.5 percent in 2014 and reached 7.6 percent in 2016. but important part of the broader SP system. Similarly, the monetary poverty rate, measured according to the new methodology, fell from 53.2 percent in 2014 to 213. Mexico’s flagship poverty-reduction programs had a 50.6 percent in 2016 as labor income rose.195 demonstrably positive impact on human capital develop- ment and mitigated the negative impact of shocks. The cov- 210. However, disparities in poverty rates and access to ser- erage of programs, particularly PROSPERA and the Elderly vices between regions and demographic groups continue to Adults Program (Programa Adultos Mayores, PAM), has persist. In 2016, more than 40 percent of poor households expanded substantially over the past decade, boosting were located in just five states: Chiapas, State of Mexico, consumption, improving nutrition, health and education Oaxaca, Puebla, and Veracruz. These states have some of outcomes, increasing household resilience, and directly the country’s lowest average per capita income levels,196 reducing poverty.200 Overall, flagship SA programs201 are highest inequality indicators, largest indigenous popula- progressive. Between 2012 and 2016, more than 60 per- tions, and widest gaps in educational attainment. These cent of SA beneficiaries were in the bottom two per capita states are also home to numerous remote communities consumption quintiles, and more than 60 percent of ben- with limited connectivity to the national economy or the efits (78 percent for PROSPERA) went to households in the public administration; their labor markets are character- bottom two quintiles (See Annex Figure 47.a and 47.b). ized by low-productivity jobs and high rates of informality; Moreover, during the global financial crisis, such programs and they are among the most populous states in Mexico. reduced economic losses among the poor, containing the Nationwide, poverty and extreme poverty rates among increase in poverty rates to just 1 percentage point be- indigenous groups are far above the national average.197 tween 2006 and 2008.202 Had the coverage of PROSPERA’s predecessor, OPORTUNIDADES, not been expanded, the 211. Social protection plays a key role in reducing poverty income of households in the poorest quintile could have and promoting equity and resilience. Effective social pro- fallen by 25 percent or more.203 Social protection programs tection systems reinforce the economic resilience of low- have also enabled households to cope more effective- er-income households by shielding them against adverse ly with recent natural disasters, such as the September shocks faced throughout the life-cycle, and improve equi- 2017 earthquake, by rapidly mobilizing and channeling ty by reducing (extreme) poverty through redistribution. resources to the affected areas. They also expand economic opportunity by preserving and building human capital and by providing access to 214. Recent efforts led to increased coverage of the pen- better employment opportunities, which can help indi- sion system, though it remains far more limited than origi- viduals and households escape poverty. Well-functioning nally envisioned. Between 2012 and 2017, the number of social protection policies and programs provide a founda- employees covered by social security (IMSS) increased tion for inclusive growth, social stability, and sound mac- by more than 3 million compared to an increase of just roeconomic management.198 over one million in the 2006-2011 period. During the 195 World Bank, 2018. 196 Per capita incomes in State of Mexico, Oaxaca, and Veracruz stagnated from 2010 to 2014. 197 As of 2016, three in four indigenous people were poor according to multi-dimensional poverty index, while only about 40 percent of the non-indigenous population lived in conditions of poverty. ENIGH. MEXICO 198 World Bank, 2012. POLICY NOTES 199 PROSPERA succeeded two earlier programs, OPORTUNIDADES and PROGRESA, and reflect a continuous policy intervention. 200 Hoddinott et al., 2000; Gertler et al., 2006; Angelucci and De Giorgi, 2009; Gertler, 2004; Skoufias and Parker, 2001. 201 This analysis relies on information from the Household Income and Expenditure Survey (Encuesta Nacional de Ingresos y Gastos de los Hogares, ENIGH) and is therefore limited to the programs covered in the survey, which include PROSPERA, the Elderly Adults Program (Programa Adultos Mayores, PAM), the Temporary Employment Program (Program Empleo 92 Temporal, PET), the Direct Rural Support Program (Programa de Apoyos Directos al Campo, PROCAMPO) and various scholarships classified as “other social programs.” 202 Grosh, Bussolo, and Freije, 2014 203 Ibid. same period, the percentage of the labor force consid- reflects a broader disparity in development indicators and Chapter 9 ered to be working in the informal sector204 was reduced social service provision between northern and southern by three percentage points from 59.5 to 56.5 percent.205 Mexico.210 The disparities between indigenous and non-in- This progress may have been due to increased labor digenous peoples are also extensive; in 2016, extreme inspections as well as attempts to incorporate small poverty for indigenous populations was six times higher enterprises through favorable tax treatment.206 Efforts than for non-indigenous populations.211 Even with PROS- to increase voluntary pension contributions were only PERA’s extensive and largely successful efforts to reach the marginally successful. indigenous population, more than half of the indigenous extreme poor are not covered by the any of the three large scale programs.212 CHALLENGES 216. Despite the significant progress made over the past decades, multiple interventions by institutions at different Efficiency of Social Assistance levels of government, and by different institutions within System each level of government, reduce the overall efficiency of the 215. While Mexico’s flagship social assistance programs system. Mexico operates at least 5,491 social protection perform well overall, reaching the marginalized and the and social development programs at the federal, state, poorest remains a challenge. Flagship SA programs have and local levels.213 While steps were taken to consolidate significantly improved access to healthcare and education social programs recently, especially at the federal level,214 among poor households at a modest fiscal cost. Indeed, the proliferation of many small-scale programs persists. social programs were responsible for an estimated 63.3 Inadequate coordination results in social spending inef- percent of the net reduction in multidimensional poverty ficiencies, duplicative coverage and coverage gaps, a less observed over the past decade.207 At about 1.2 percent of equitable distribution of benefits, and weak transparency GDP, Mexico’s SA spending is similar to that of other coun- and accountability. The most recent assessment by CON- tries in the region (Figure 47 and Figure 48), but relatively EVAL identified 20 programs that were 100% similar with low both in absolute terms and compared to the OECD av- up to three other programs.215 The fragmentation of social erage of 2.2 percent of GDP.208 Despite the effectiveness of programs also creates unintended disincentives to work SA programs, the territorial differences persist in multiple and save.216 The government is currently developing an dimensions, hampering the economic and human capital Integrated Social Information System (Sistema Integrado development of marginalized regions. In 2016, about 40 de Información Social, SISI),217 which will consolidate data percent of poor households (and 60 percent of extremely on potential and actual beneficiaries of social programs at poor households) were located in just five states209 which all three levels of government. Figure 47: Expenditures on Social Assistance Programs in Mexico as a Percentage of GDP, 2003-2015 1.4 1.2 1.0 0.8 0.6 0.4 0.2 0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 CCT Source: World Bank UCT Atlas of Social Social Protection pensions Indicators Public of Resilience and Equity Works In kind Fee waiver Other Social Assistance 204 Please see Labor and Productivity Policy Notes for more details on informality. 205 Among Latin American countries, only Mexico and Ecuador did not reduce the share of workers in the informal sector between 1993 and 2013. See: OECD, 2016. 206 Kaplan, 2017. 207 World Bank, 2018 208 World Bank, 2016 209 Chiapas, State of Mexico, Oaxaca, Puebla, and Veracruz. 210 In 2016, the share of households lacking access to basic services was four times higher in the southeastern states (Chiapas, Guerrero, and Oaxaca) than in the northern border states (Baja California, Sonora, Chihuahua, Coahuila, Nuevo Leon, and Tamaulipas). Similarly, educational outcomes and access to social security are far lower in southern states than in northern states, while the incidence of food security was 1.5 times higher in the south. 211 As of 2016, with the exception of lack of access to health care services indicator, the difference between the indicators of social deprivations for indigenous peoples was at least 12 MEXICO percentage points higher than for the non-indigenous group and, in some cases, much larger, such as the lack of basic household services. ENIGH (2016). POLICY NOTES 212 ENIGH (2016). The three large scale-programs that are covered in the survey are: PROSPERA, PAM, and PROCAMPO. 213 CONEVAL, 2017. 214 As of 2016, there are 152 social programs at federal level, a significant reduction from 2010 when 273 federal social programs were present. 93 215 https://www.coneval.org.mx/Evaluacion/IEPSM/Documents/Consideraciones_para_el_proceso_presupuestario_2018.pdf 216 World Bank, 2012. 217 The SISI will contain information on both potential and actual beneficiaries of social programs at all three levels of government. Chapter 9 Figure 48: Expenditures on Social Assistance as a Percentage of GDP by Country, 2000-2015 3.0 2.5 2.0 1.5 1.0 0.5 0 Argentina Brazil Chile Colombia Costa Rica Ecuador Mexico Peru Uruguay 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 Source: World Bank Atlas of Social Protection Indicators of Resilience and Equity and are neither targeted nor tailored to the needs of poor Expanding Access to Opportunities households; thus, are underutilized by the poor. 217. Early indications suggest that social assistance pro- grams support a sustained increase in the human capital of 219. Increasing labor participation and productive inclusion the poor, nevertheless sustainably raising their income con- among poor households requires addressing household-lev- tinues to be challenging. There are early indications that el constraints. For instance, the likelihood that vulnerable these investments are paying off; children with exposure mothers will get a job increases if adequate childcare ser- to PROGRESA- OPORTUNIDADES-PROSPERA exhibited vices are provided, and improving the employment pros- better educational attainment, geographic mobility, labor- pects of mothers can have positive developmental impli- market outcomes, and household economic indicators in cations for their children. In 2016, over half of the econom- early adulthood.218 However, despite steady expansion ically active population worked in the informal sector and in coverage and evidence of protection against negative lacked access to the childcare services offered to formal impact of shocks, most of the poor are still not able to get workers. While SEDESOL provides subsidized childcare out of poverty. International experience indicates that to some working mothers, this program component has social programs that provide complementary programs just 300,000 potential beneficiaries and does not explicit- and services (i.e. Cash Plus) have a greater impact than ly target poor mothers.220 Similarly, a large portion of the cash transfers alone. These include promoting support for vulnerable youth is neither employed nor in education or employment and income-generating opportunities along training. However, most existing programs are fragment- with greater access to social services and measures to ed and underfunded limiting their impact. Improving promote overall wellbeing. In recent years a series of steps labor participation and productive inclusion among poor were taken towards this (e.g. establishment of programs households will require a more comprehensive strategy aimed at improving beneficiaries’ employment prospects with adequate funding. and access to social services, linking beneficiaries with income generating activities and paid work etc.) but im- plementation so far has proven challenging. Social Security 220. Lack of clarity in roles and lack of integration across SA 218. Active labor market programs are limited. Between and SI potentially contributes to incentive compatibility is- 2003 and 2013, Mexico spent less than 0.2 percent of its sues. The lack of integration of contributory pensions and GDP on active labor-market programs, significantly less health insurance with the non-contributory programs has than Argentina, Brazil, Chile, Colombia, and Uruguay.219 led to misallocation of scarce resources and poor target- Most of the resources in the broader definition of active ing. For example, participating in a contributory pension labor market programs goes to income-generation and program may disqualify an individual from participating productive programs. A significant share of these pro- in noncontributory pension programs. Receiving even a grams focuses on agricultural production, where they modest contributory pension benefit bars beneficiaries tend to benefit large agricultural producers with limited from obtaining a noncontributory pension—effectively impact on poverty. There are some signs of increased ac- imposing a 100 percent marginal tax on contributory pen- cess of the poor to these programs: for instance, the share sion income, which along with the eligibility conditions of PROSPERA beneficiaries that participate in productivity for Seguro Popular can encourage individuals to remain and labor-inclusion programs is on the rise following its re- in the informal sector. Meanwhile, the lack of connectiv- MEXICO design in 2014. However, these programs remain limited, ity between beneficiary registries enable households to POLICY NOTES 94 218 Parker and Vogl 2017; Kugler and Rojas, 2018 219 World Bank Atlas of Social Protection Indicators of resilience and Equity (ASPIRE) 220 The labor market policy note provides a more detailed analysis of employment constraints and related interventions. manipulate the system by having one household member that covers formal salaried employees combining an in- Chapter 9 enrolled in the contributory scheme, who obtains benefits dividual account with a solidarity fund was prepared but for the entire household, while other household members has not been approved by the Congress.226 Lack of an un- collect benefits from the noncontributory schemes. The employment insurance scheme combined with lengthy inability to identify overlapping benefits also applies to and costly dismissal procedures and high severance pay health insurance, reflecting a lack of coordination between requirements diminishes incentives for formality.227 Tem- state and federal social programs.221 While the perverse porary payments to unemployed workers would buffer incentives created by social programs may discourage the impact of economic shocks at the household level and formal employment,222 the available evidence does not help stabilize aggregate spending during downturns in suggest that they are the primary cause of informality.223 the business cycle. 221. Disparities in the benefits packages offered by dif- 223. The aging of Mexico’s population over the next two de- ferent SI schemes inhibit labor mobility. Social insurance cades will intensify pressure on pension systems and health programs for formal workers are designed to address risks spending. As fertility rates fall and life expectancy rises, the related to old age, ill health, disability, and survivorship.224 number of people over age 60 as a share of those ages 20- Different schemes for public and private sector workers are 59 is projected to double from 18 percent to 37 percent offered by the national, state, and municipal governments, between 2015 and 2040. Mexico’s aging population will and there are separate pension schemes for employees in substantially increase spending on pensions and health- the oil, banking and utility sectors. The benefits offered care. The 1997 pension reform introduced a privately by these schemes vary, and the prospect of shifting from managed, defined-contribution system to replace the de- a more- generous to a less-generous benefits package fined- benefits scheme that covered workers in the formal limits labor mobility.225 Moreover, the lack of integration private sector. While the reforms made the system more between contributory pension and health insurance pro- financially sustainable, liabilities inherited from the main grams and non-contributory programs contributes to the defined-benefit system will not be cleared until the middle misallocation of scarce resources and undermines benefi- of the century. ciary targeting. 224. In the meantime, federal pension spending is projected 222. Mexico’s numerous social protection programs do not to double as a share of GDP by 2030, and spending on state include unemployment insurance, and recent attempts to government pensions is also expected to increase rapidly in introduce unemployment insurance have been unsuccess- the coming decades. The state and local government pen- ful. A draft law for an unemployment insurance scheme sion schemes are of particular concern since most have Figure 49: Projections of federal pension spending (% of GDP), 2015-2080 4.0% 3.5% 3.0% 2.5% (% PIB) 2.0% 1.5% 1.0% 0.5% 0.0% 2015 2018 2021 2024 2027 2030 2033 2036 2039 2042 2045 2048 2051 2054 2057 2060 2063 2066 2069 2072 2075 2078 IMSS ISSSTE 65 y Más (II) Total (65 y Mas, % PIBpc) Gobierno Federal (65 y Mas, % PIBpc) 221 These issues are discussed in detail in the health sector policy note. 222 Levy, S., 2008; Aterido et al., 2011; Azuara and Marinescu, 2013. 223 Please see Labor and Productivity Policy Notes for more details on informality and potential underlying drivers. 224 Unemployment insurance is not covered in Mexico. 225 This may contribute to the misallocation of labor across sectors limiting productivity. See Productivity Note for an exhaustive discussion on the low total factor productivity and MEXICO underlying challenges. POLICY NOTES 226 The funding of the scheme was one of the main reasons for the lack of its approval, with the 3% contribution rate proposed to be drawn from the current 5% contribution that goes into Infonavit, a housing savings scheme. 227 The dismissal costs present a significant problem in the labor market, since virtually all dismissals end up being contested in labor courts, where processes may take 3 or more years. 95 The parties to the dismissal argue on the payment of severance pay, forgone earnings and other compensations, while the dismissal cannot be declared effective until the court reaches a verdict. Even when the court reaches a positive verdict, a significant share of the workers does not receive any compensation. See Policy Notes on Rule of Law and Labor Market Policies. Chapter 9 not reformed and there are no estimates of current spend- can facilitate the implementation of more coherent and ing much less projections of future spending. Spending on sophisticated policies. the non-contributory pension program will increase as the population ages. In fact, if the non-contributory pension 228. Most middle-income countries, including all except scheme maintains its value relative to the average income Brazil in Latin America, have introduced a national, digital level and covers the entire elderly population that lacks a identification system, while Mexico still lacks one. These contributory scheme, pension spending would more than systems generate unique numbers by using biometrics double in the next two decades.228 to ‘deduplicate’ the database, not allowing the same in- dividual to be registered more than once. In a growing 225. Even as spending increases due to the legacy IMSS number of countries, the agency managing the identifi- scheme, cohorts that entered the system after the 1997 cation system offers authentication services to govern- reforms will see sharply lower pensions than those grandfa- ment and even private sector entities. By checking the thered in the reform. The replacement rates for workers that number against biometric or demographic information, have contributed to a pension scheme during most or all of fraudulent transactions such as ghost beneficiaries can their working- age years are projected to drop by as much as be significantly reduced. On the other hand, Mexico has half compared to those that began working before 1997.229 planned to introduce a unique identification number This trend reflects the relatively low contribution rate of the for over two decades, but this policy has yet to be im- defined-contribution scheme – 6.5 percent compared to plemented. The Unique Population Registry Code (Clave at least 10 percent in similar schemes in the region. Such Única de Registro de Población, CURP), is not effective, a drastic decline in benefits would likely prove socially and and attempts to replace it with a biometrically verified politically difficult and may prompt policymakers to extend identification number have proven unsuccessful. Despite the guarantees of the defined-benefit scheme. Alternative- improvements in the CURP registration system, there ly, there could be pressure to increase the non-contributory are more unique identification numbers than there are pension and allow workers with contributory pensions to people in Mexico, indicating that many individuals hold receive them (as is the case today in Chile). multiple “unique” numbers.232 In the absence of a reliable national identification system, major social programs 226. An additional adequacy problem stems from low con- spend tens of millions of dollars on their own discrete, tribution densities, as many workers spent years outside the noncompatible systems. formal sector. Most Mexican workers have, at one time or another, established an individual account with a pen- 229. A robust identification system that is used by all of sion-fund management company known as AFOREs.230 the major government programs help also identify ineligi- However, because the share of their working lives spent in ble households by allowing public agencies to cross-check the formal sector is directly related to their income levels, multiple administrative databases (e.g., property registries, many low- income workers will reach retirement with bal- tax records, etc.) to target beneficiaries and scale benefits ances that are too low to provide adequate pensions. At progressively.233 Multiple programs can use the same tar- any given point in time, just over half of the labor force is geting system, which reduces program costs. Connected not contributing to a pension scheme, and some workers databases can also be used to migrate beneficiaries from have never contributed. one program to another: for example, the ability to con- nect databases allowed Chile to phase out its noncontrib- utory pension program much more smoothly than Mexi- Infrastructure to Support a More co. Chile’s experience also illustrates how linked databases Efficient Social Protection System using unique identification numbers can improve trans- 227. Key elements of the infrastructure required for an in- parency: by inputting their identification numbers into a tegrated social protection system are not yet in place. Pro- government website, Chilean households can view all the grammatic fragmentation and disconnected information benefits that they receive, while public agencies can view systems hinder efforts to integrate and rationalize SA and a mirror image showing which programs are serving the SI programs both within and between levels of govern- same households. Strict rules regarding data protection ment. Without unique identification numbers, the author- and data sharing ensure that households’ personal infor- ities cannot cross-check beneficiary databases to correct mation is protected. Availability of high quality adminis- errors and identify fraud. For instance, in the health sector, trative data was key in the success of the Chilean system, double enrollment the IMSS and Seguro Popular health while the quality of existing administrative data remains insurance systems results in an estimated US$300 million a challenge in Mexico, with issues of incomplete and in- dollars per year in fraudulent outlays.231 In addition to re- consistent data and lack of regular quality and internal ducing waste, the use of unique identification numbers consistency checks. MEXICO POLICY NOTES 228 World Bank, 2015. 229 Inter-American Development Bank, forthcoming. 230 Administradoras de Fondos para el Retiro. Please see Financial Sector Policy Not for the challenges faced by pension fund management companies. 231 World Bank, 2017. 96 232 There are an estimated 180 million CURPs issued for a population of around 130 million people. 233 40 For instance, in Turkey, the unique ID allows the government to use 28 databases (e.g. property, taxes, etc.) to differentiate between those that need a full subsidy and those that can afford to pay part of the health insurance premium. the social assistance programs. While flagship social assis- Chapter 9 POLICY OPTIONS tance programs are progressive in Mexico, the difficulty in reaching the poorest and most marginalized could be al- Improving the Efficiency in Social leviated by moving towards more dynamic targeting sys- Assistance System and Reaching tems. In Mexico, demand based open applications do not the Marginalized and Poorest exist (i.e. applications to programs are not accepted con- tinuously) and recertification happens very infrequently 230. Operationalize SISI, which will be an important tool for (e.g. 8 years for PROSPERA). Thus, a poor household who identifying inefficiencies, with a proper legal framework to was not present during the new application/recertifica- ensure its use and sustainability. Improving the well-being tion process will not be protected and will have to wait of poor households require integrated support provided until the next application period many years in the future. by multiple agencies at different levels of government. Introducing more frequent recertification, and moving Adopting modern management tools would enhance towards an open demand-based application, through planning and facilitate coordination and reduce inefficien- changes in the operational rules/regulations would allow cies. To this end, operationalizing the SISI will be crucial to for improving the inclusion of the poor, and exclusion of identify duplications and reducing fragmentation, and it the no-longer poor in a more dynamic manner. In fact, the will allow for a deeper analysis of the overall distribution social assistance programs in most of Europe, as well as of spending on social programs at various levels. To avoid most OECD countries, have open demand-based applica- risks that could jeopardize the SISI’s successful implemen- tation and consolidation, its governing legal framework tions, with yearly or biyearly recertification. needs to be reformed, establishing SISI as the information system for all social programs and mandate its use in man- 233. Other instruments, such as better targeted FAIS234 agement of such programs. could help narrow the gap in access to social services across regions and groups, particularly access to basic services. 231. Identify duplications, coverage gaps, and opportuni- While FAIS is not an SP tool per se, it has the potential to ties for consolidation/harmonization to inform the redesign increase territorial equity in access to basic services, espe- for a more efficient and coherent SA system. An in-depth cially given the oversight role SEDESOL has on its planning assessment of social assistance system from different and targeting of FAIS. FAIS is a US$3.5 billion decentral- perspectives, including identifying duplications, gaps, ized fund allocated to subnational governments. It aims opportunities for consolidation/harmonization (e.g. use to close the social infrastructure gap and improve social of a harmonized targeting system across programs, com- development indicators. FAIS support is particularly im- bination of small programs with similar objectives and/or portant to small and poor municipalities, and it represents target population etc.), as well as the institutional coordi- an average of 47 percent of the budget of municipalities nation aspects, will be a key first step towards improving classified as “very highly marginalized.” An improved allo- efficiency and enhancing the coherence of the SA system. cation criteria and enhanced performance management The availability of critical information from SISI – such as system for municipalities and institutional capacity-build- social development indicators, employment status, cover- ing at different levels of government could enhance the age of existing programs, and regional, demographic and impact on this mechanism. household-level characteristics for actual and potential beneficiaries, will facilitate such analysis. While the evi- Strengthening and Expanding dence is an important element, redesigning the SA system Access to Social and Economic may prove politically complex, given the federal structure Opportunities for the Poor in Mexico, and the limited coordination and articulation of social programs across different levels of Government. 234. Consolidate programs, eliminating poor performing The recent reform of the welfare system in the UK, consol- ones and designing programs that are better adapted to idating and simplifying almost all programs into one sin- poor with adequate funding. First and foremost, the de- gle scheme (Universal Credit), indicates that such reform/ sign of productive inclusion programs needs to consider redesign is feasible, if strong political will exists, and the the poor and vulnerable. For instance, targeted outreach reforms are carefully planned, communicated in advance, efforts could raise awareness of the available programs and implemented. and provide information on the application process and any potential implications for participating in other social 232. Explore ways to expand reach of social assistance sys- programs. Preliminary indications from linking PROSPERA tem to poorest and most marginalized, including through beneficiaries to productive-inclusion programs and social more dynamic targeting. Social assistance programs can be services suggest the lack of adequate capacity/funding a critical entry point for other services. The monetary trans- was a binding constraint. An evaluation of the implemen- fers may remove the income barrier, while co-responsibili- tation and impact of this experience could yield important MEXICO ties or some programs further facilitate/encourage access lessons. In addition, the programs need to address the POLICY NOTES to other services. A robust targeting mechanism, is key for main constraints faced by the poor. For instance, poor 234 Fondo de Aportaciones para la Infraestructura Social. 97 Chapter 9 youth are more likely to drop out without finishing high of technology. A combined life-cycle approach to social school. This has important scarring effects for their labor intermediation services could ensure that all members of market outcomes and lifetime earnings. This high share of the poor households are linked to essential services, most drop outs, as well as high rates of inactivity in this group of which would be provided by local governments. For indicate that greater effort is required to ensure school example, parents of young children need access to appro- completion and to facilitate the school- to-work transition priate health, nutrition, and parenting information—some to improve employment outcomes and boost income of which is provided by PROSPERA—as well as information levels. The international evidence indicates that programs and access to early childhood development programs. need to be designed in close collaboration with the private Access to appropriate child development/stimulation pro- sector to ensure that they provide access to income-gen- grams within the first 1,000 days of life is especially vital for erating activities or wage employment by developing the children from poor and vulnerable households, as by age specific skills demanded by firms. In addition, making na- 3 the effects of socioeconomic disparities are already evi- tional employment services more responsive to the needs dent. In Mexico, spending on early childhood development of poor households—for example, by training front-office programs is low and not necessarily targeted to the areas workers and counselors on techniques for working with that are most in need. Moreover, only half of the 1.5 million poor and vulnerable applicants—would enhance the im- children that comprise CONEVAL’s target population are en- pact of productive-inclusion programs. Indeed, evidence rolled in early childhood development programs. Similarly from European Union suggests that well-targeted intern- working age adults need to be better informed about and ship and training programs which reflect the need of em- linked to employment support as well as and livelihoods ployers have positive employment effects,235 especially if programs. Innovative technological approaches can help combined with effective referrals/intermediation.236 Given address supply-side gaps, especially in remote areas. For the very limited resources devoted to these programs, en- example, Mexico’s high degree of cellphone penetration suring adequate funding will be critical to expand access would enable the authorities to deliver program informa- to productive opportunities. tion directly to beneficiaries via text message. 235. Explore a social-intermediation approach for high priority areas and marginalized communities. Social-inter- Providing a Sustainable, Equitable mediation services are a form of case management: they and an Integrated Social provide information and psychosocial support based on Insurance System for a Rapidly a set of results/goals and mutual responsibilities defined Aging Population for each household according to its specific circumstanc- 237. Gradually integrate the main contributory pension es. The international experience indicates that combining schemes. Many OECD countries are consolidating their cash transfers with social services (e.g., healthcare, coun- pension schemes, and most now include civil servants and seling, and productive inclusion) tailored to the specific private sector workers in the same scheme. This major pol- needs of households have a greater impact than cash icy shift reflects both the need to increase labor mobility alone. Multiple countries in Latin America and the Carib- and the lack of justification for disparities between pen- bean have implemented social-intermediation services, sion systems. Consolidation may also lower administrative notably Chile and Colombia, as have various European costs by leveraging economies of scale. countries. In Mexico, the States of San Luis Potosí and Guanajuato have already implemented a social- interme- 238. Ensure portability with minimal transaction costs diation approach with the robust support of their state across pension schemes. Even when schemes are not ful- governments with mixed results. In addition, PROSPERA ly consolidated, portable pension benefits can enable designed and implemented a pilot program (Acom- workers to move between the public and private sectors pañamiento Familiar) in Guerrero State, with support from without losing pension wealth in a way that discourag- the World Bank. These experiences underscore the impor- es efficient labor mobility. This applies not only to IMSS, tance of political ownership in implementing and sustain- ISSSTE, as well as schemes covering workers in public sec- ing this approach and highlight potential challenges relat- tor banks, PEMEX and even state governments. ed to cross-sectoral management as well as gaps in supply of services. While implementing the social-intermediation 239. Integrate contributory and non-contributory systems approach nationwide at scale will neither be fiscally fea- and applying better targeting. Currently, the receipt of a con- sible nor necessary, the targeted deployment of this ap- tributory pension essentially disqualifies an individual from proach in high-priority areas, to address the challenges receiving the non-contributory pension. This effectively of most marginalized communities, could maximize the places a tax on contributions, encouraging people (along value of limited resources. with other benefits such as Seguro Popular) to remain in the informal sector. The incentive problems may become more 236. Address important gaps, such as early childhood devel- significant to the extent that the non-contributory benefit MEXICO POLICY NOTES opment, potentially through social intermediation and use is increased in the future. It is also not clear that everyone 98 235 European Commission (2015) 236 Blundell et al. (2013), Graversen and van Ours (2008) that is not receiving a contributory pension should qualify ing for the legacy IMSS scheme. A gradual convergence Chapter 9 to receive the non-contributory pension as it is meant as an of new entrants into the reformed system, perhaps with anti-poverty pension. A better approach would apply an af- supplementary occupational schemes would facilitate fluence test based on household income and then apply a portability and take advantage of economies of scale in partial offset for contributory pension income as is done in administration. While there is little information available Chile. This potentially reduces fiscal costs while improving regarding state and local plans, it is likely that they would incentives for formalization. require a number of parametric reforms ranging from lon- ger career averaging periods for the calculation of the ini- 240. Consider alternative ways of financing social insur- tial DB pension to automatic inflation indexation. Finally, ance premia, especially for workers with low wages. The tax the tax treatment of pension income should be reviewed wedge in Mexico is among the largest in Latin America and special exemptions that benefit high income individ- (see Annex figure 2) contributing to the various incentives uals restricted. in the Mexican labor market that encourage informality. As pointed out by Levy (2018) the combination of the in- 243. Consider raising contribution rates for the defined-con- troduction of non-contributory pensions and health along tribution scheme to the levels originally planned. It is well with the increased tax wedge (from both income taxes known that the defined contribution scheme will not and social insurance taxes) add to other distortions that provide adequate pensions to most workers, even those encourage informality. The shift away from payroll tax fi- with fairly good histories of consistent contributions nancing has already taken place in some countries (e.g., However, increasing contribution rates from the current Uruguay, Colombia) with a positive impact on formal em- level of 6.5 percent to about 10 percent which is close to ployment.237 both the original target level and the levels of comparable countries combined with the aforementioned increase in 241. Over the medium term, and if the fiscal envelope al- retirement age, could mitigate the impending decline for lows it, introduce unemployment insurance that combine a future cohorts. solidarity fund with individual accounts to foster risk-pool- ing while limiting moral hazard. The design will need to be 244. Continue the implementation of ongoing initiatives to complemented by search requirements as well as comple- reduce informality. Evidence from other countries such as mentary services to facilitate quick re-employment (e.g. Brazil, Colombia, and Costa Rica suggest that a combina- reporting on search efforts periodically, participating in tion of incentives and enforcement can increase the cover- the designated training or placement programs etc.). The age of self-employed workers. financing of the scheme can partly be obtained through unbundling of some services that are covered under the 245. If benefits of non-contributory pension are increased, current payroll tax (e.g. housing), as proposed previously, they should include applying an affluence test to better but will need to be carefully communicated to ensure buy- target resources in fiscally sustainable way. Non-contribu- in. Having a well-designed unemployment benefit system tory pension benefits in Mexico are indeed low relative to will reduce reliance on severance pay as income support those of non-contributory schemes in comparable coun- (which, unlike unemployment benefits puts financial the tries. This will create pressure to supplement low pen- burden on individual firms), and will enable a reform of sions. At the same time, limited fiscal resources should be severance pay at the same time. Furthermore, having a focused on the elderly poor. Projections that assume con- carefully designed unemployment benefit system in place stant benefit levels relative to per capita income indicate will allow income-smoothing for unemployment risk at that spending will already triple from 0.2 to 0.7 percent the household level and will also serve as an automatic of GDP over the next 20 years. Raising the benefit levels stabilizer for aggregate economic shocks. The financial without parallel measures is not sustainable. Costs could cost of the scheme will need to be considered within the be reduced by applying an affluence test (as in the Chilean overall fiscal framework and the priorities, while factoring case which limits benefits to households in the lowest 60 in its multitude benefits. percent of the population. Raising the benefit level will likely prove necessary to prevent a large increase in pover- 242. Reform the existing defined-benefit schemes, in- ty rates among the growing elderly population. cluding the legacy IMSS scheme to improve both the fiscal sustainability and equity of the pension system. Currently, Mexico’s various pension schemes provide highly dispa- Building the Infrastructure to rate benefits to different workers, particularly those in the Support a More Sophisticated public or quasi-public sector.238 Retirement ages should be Social Protection System gradually equalized across these schemes and incentives 246. Establish a unique, digital identification number for for early retirement eliminated. Retirement ages should be all Mexican citizens. This will be essential to upgrade the linked to changes in life expectancy going forward includ- social protection system. A digital identification system MEXICO POLICY NOTES 99 237 See Anton (2014). 238 World Bank (2016): Análisis del Sistema de Pensiones: Parte 2: Portabilidad e Integración de Sistemas de Pensiones Contributivos y no Contributivos: Recomendaciones de Política, Diseño e Implementación Chapter 9 would allow for accurate and timely verification of bene- social assistance services at the local level. It reduced the ficiaries. It would also allow the government to compare processing of a large number of paper documents and individual and household data from different databases. the time spent on the applications by guaranteeing the Augmented by an interoperability framework and secure availability of information from other institutions. Also, data-sharing protocols, a national digital identification through the use of this system 10% of social assistance system could enhance targeting, reduce fraud and dupli- benefits were identified to be duplicates. cation, and enable the implementation of more sophisti- cated social policies. 248. Invest in improving the quality of existing adminis- trative data. Consistent efforts to improve data quality 247. Link administrative databases to enhance the efficiency contributed to the success of the linkages of administra- and impact of social protection programs. Many compara- tive databases in many countries. Ensuring consistency ble countries have built integrated identification systems in the use of identification classifiers and use of metadata that link dozens of major databases (e.g., tax, property, standards, would alleviate the issues of incompleteness auto registration, financial assets, etc.) In Chile and Turkey, and missing classifiers. In addition, introducing quality creating this type of system required multiple years and controls, which imply cleaning of existing but messy data; a strong political commitment to breaking down barriers and undertaking internal quality checks when the existing between institutions. The social information system in data is not consistent internally or/or with other databases Turkey was key to reducing the time and cost of providing will further improve the quality of administrative data. MEXICO POLICY NOTES 100 Chapter 9 ANNEX- FIGURES Figure A1a: Distribution of beneficiaries by income quintiles, 2014 Other social assistance programs Subsidized health insurance Public works program Conditional cash transfer program Cash transfer program Other non-contributory social pension Federal non-contributory social pension Contributory pensions 0% 20% 40% 60% 80% 100% Q1 Q2 Q3 Q4 Q5 Source: World Bank staff calculations based on ENIGH 2014 and using ADePT Figure A1b: Distribution of benefits by income quintiles, 2014 Public works program Conditional cash transfer program Cash transfer program Other non-contributory social pension Federal non-contributory social pension Contributory pensions 0% 20% 40% 60% 80% 100% Q1 Q2 Q3 Q4 Q5 Source: World Bank staff calculations based on ENIGH 2014 and using ADePT Figure A2: The ‘tax wedge’ in Latin America Argentina Brazil Uruguay Colombia Costa Rica Mexico Panama Chile Bolivia Paraguay El Salvador Nicaragua Dominican Republic Ecuador Peru Venezuela Jamaica Guatemala Trinidad and Tobago Honduras LAC OECD 0% 5% 10% 15% 20% 25% 30% 35% 40% Employer SSC Employee SSC Income Tax Source: OECD MEXICO POLICY NOTES 101 Chapter 9 Table A1. Main parameters of the different federal pension schemes. Federal Social IMSS ISSSTE PEMEX Pension Program Year of the reform Implemented at the 1997 2007 No reform national level since 2013 Coverage Private sector workers Federal government workers PEMEX workers Elderly without and some workers of the a contributory pension state governments Retirement age 65 60 (in 2018) 55 65 Required weeks of 1250 weeks 15 years of service 25 years of service None contribution to qualify for pension Contribution rate 5.15% (Employer), 5.175% (Employer- No contribution No contribution 1.125% (Employee), Government Agency), 0.225% (Government) 1.125% (Employee), 0.225% (social contribution) Replacement rate From 75% of the average of From 50% with 15 years of From 80% of the average $580 pesos per month the base salary of the latest work to 95% with 29 years base salary over the last 5 years with 60 years of age of work. additional year of work, plus to 100% of the average of 4% for each additional year the base salary of the latest of work after 25 years of 5 years contributed with 65 work, until reaching 100% as years of age. maximum. Minimum pension 24 years of service 15 years of service Not applicable Not applicable guranteed *State-level Information not available. Social contribution of the federal government (IMSS) to each worker’s pension fund, based on the range of minimum wages earned by the worker up to 15 minimum wages: 7 minimum wage; 1.01 to 4 minimum wages; 4.01 to 7 minimum wages; 7.91 to 10 minimum wages; 10.01 to 15 minimum wages. MEXICO POLICY NOTES 102 Chapter 9 References Antón, A., F. Hernández and S. Levy (2012), “The End of Informality Grosh, M., del Ninno, C., Tesliuc, E., Ouerghi, A. (2008). “For Protec- in Mexico? Fiscal Reform for Universal Social Insurance”, In- tion and Promotion: The Design and Implementation of Effec- ter-American Development Bank. tive Safety Nets.” Washington, DC: World Bank. Antón, A., F. Hernández (2011), “Financing Universal Social Insur- INEGI. (2017). “Resultados de la Encuesta Nacional de Ocupacion ance in Mexico.” mimeo. y Empleo, Cuarto Trimestre de 2016.” Boletin de Prensa No. Avila-Parra, C., Escamilla-Guerrero, D. 2017. “What are the Effects 66/17. Aguascalientes, Mexico. of Expanding a Social Pension Program on Extreme Poverty Inter-American Development Bank (forthcoming), “Pension and Labor Supply? Evidence from Mexico’s Pension Program Trends in LAC and Path for Reform”. Kaplan, D. (2017), “La evo- for the Elderly.” Wprld Bank Policy Research Papers No: 8229. lucion de empleo registrado con IMSS”, presentation at the World Bank, Washington, DC Inter-American Development Bank. Blundell R., M. Costa Dias, C. Meghir, and J. Van Reenen. 2004. Eval- Levy (2018). Under-Rewarded efforts: The elusive quest for pros- uating the Employment Impact of a Mandatory Job Search perity in Mexico, Inter-American Development Bank. Program. Journal of the European Economic Association vol. 2, Levy, S., (2008), Good Intentions, Bad Outcomes: Social Policy, issue 4, 569- 606. Informality and Economic Growth in Mexico, Brookings Insti- CONEVAL. (2017). “Informe de Evaluacion de la Politica de Desar- tution Press rollo Social 2016.” Federal District: Government of Mexico OECD (2016), “Taxing Wages in Latin America and the Caribbean.” CONEVAL. (2014). “Indicadores de acceso y uso efectivo de los ser- Rodriguez-Castelan, C. 2017. The Effects of Mexico’s Social Infra- vicios de salud de afiliados al Seguro Popular.” Federal District: structure Fund (FAIS) on Regional Development.” Mimeo. Government of Mexico Schulz, P. (2004). “School Subsidies for the Poor: Evaluating the Edmonds, E. V. (2006). “Child Labor and Schooling Responses to Mexican Progresa Poverty Program.” Journal of Development Anticipated Income in South Africa,” Economics, Volume 74, Issue 1, pp. 199-250. Journal of Development Economics 81, No. 2, pp. 386-414. Silva-Mendez, J. 2017. “Mexico’s Social Infrastructure Fund: Char- European Commission (EC). (2015). European Employment Policy acteristics and Relevant International Experience.” Mimeo. Observatory Review. Upskilling Unemployed Adults: The Or- World Bank (2012). “The World Bank 2012-2022 Social Protection ganization, Profiling and Targeting of Training Provision. Strategy: Resilience, Equity, and Opportunity.” Fernald L., Gertler P., Neufeld L. (2009). “The 10-year Effect of Opor- World Bank (2015). “Analysis Cuantitativo del Sistema de Pen- tunidades, Mexico’s Conditional Cash Transfer Programme, on siones” informe del Banco Mundial para el Ministerio de Ha- Child Growth, Cognition, Language, and Behaviour: a Longitu- cienda, 2015. dinal Follow-Up Study.” World Bank (2016a). “Mexico Public Expenditure Review.” Wash- Gertler, P. (2004). Do Conditional Cash Transfers Improve Child ington, DC: The World Bank. https://openknowledge.world- Health? “Evidence from PROGRESA’s Control Randomized Ex- bank.org/handle/10986/25062 periment.” American Economic Review, 94(2): 336-341. World Bank (2016b). “Mexico: Análisis del Sistema de Pensiones, Graversen, B. K. and J. van Ours (2008) “How to help unemployed Parte 1 – Análisis Cuantitativo.” Washington, DC. World Bank. find jobs quickly: Experimental evidence from a mandatory World Bank (2017). “Identity Management Systems Analysis: Mex- activation program” Journal of Public Economics, 2008, vol. ico.” Washington, DC: The World Bank 92, issue 10-11, 2020-2035. World Bank (2018). “Mexico Systematic Country Diagnostic” Wash- ington, DC: The World Bank MEXICO POLICY NOTES 103 10. Fostering Resilient, Productive and Low-Carbon Rural Territories N atural resources are an important contributor to Mexico’s output and a major compo- nent of its national wealth. However, unsustainable exploitation and increasingly severe climatic events are rapidly depleting the country’s natural capital. Mexico faces the chal- lenge of conserving and sustainably managing its forests while also meeting a growing demand for timber products. Public programs and investments in multiple sectors could be aligned to reduce deforestation, forest degradation, and other natural resource deterioration, while at the same time generating economic opportunities, including for the rural poor. Mexico’s agricultural sector has performed below potential, particularly in some areas of the country in which land frag- mentation and subsistence agriculture prevails. Addressing specific logistics and facilitation infra- structure (including storage) gaps, reducing market (price) information asymmetries, improving intermediaries’ competition in captured markets, and reducing other market access constraints to smallholder farmers and small-scale forestry enterprises could bolster productivity and incomes in lagging regions. MEXICO POLICY NOTES 105 Chapter 10 ecological resilience to the impact of climate change while CONTEXT AND REFORM also sustainably increasing the productivity of natural cap- PROGRESS ital and enhancing its contribution to income generation and employment. 249. Mexico’s natural resources—which include agricultural land, forests, fisheries, water, and coastal areas—are an im- 252. Mexico has become a leader in the global climate portant contributor to GDP and a major source of employ- agenda. The country has taken important steps to support ment. In 2014, natural capital represented 13.2 percent a resilient and low-carbon growth path, via policy and of the country’s national wealth.239 Key resources include institutional reforms. Cross- sectoral instruments include cropland, pastureland, and subsoil assets, which together the National Development Plan, the National Climate comprise nearly 80 percent of Mexico’s natural capital (Ta- Change Strategy, the National REDD+ Strategy, and Mex- ble 1). Mexico’s diverse geography, abundant natural re- ico’s National Strategy on Biodiversity. Mexico introduced sources, generally favorable climate, and geographic prox- a Climate Change Policy in 2012 to support the transition imity to large and dynamic markets drive the growth of its to a competitive, sustainable and low carbon economy, agricultural sector. Crop and livestock production employs reduce vulnerability of the population and ecosystems, about 13.5 percent of Mexico’s labor force. Mexico also has and assign competencies within government. Interna- 88 million hectares of forest, which cover almost 45 per- tionally, the country committed to reducing greenhouse cent of its territory. Forested areas are home to about 12 gas (GHG) emissions by up to 22 percent by 2030 in its Na- million people, many of whom depend directly on natural tional Determined Contribution (NDC). In fact, Mexico was resources for their livelihoods. the first developing country to pledge its NDC. Under the 250. In addition to their direct economic value, natural re- Bonn challenge, the country pledged to restore over eight sources provide critical ecosystem services. Healthy ecosys- million hectares of degraded land by 2020. Despite these tems regulate the hydrological regime and improve water important steps, much remains to be done to realize the quality, control coastal erosion, and provide habitats for a country’s ambitions to move toward a productive, resilient wide range of species. Mexico is among the world’s most and low carbon development path, for which the agricul- biodiverse countries and contains an estimated 10 percent ture, forest and other land uses sector (AFOLU) is expected of global biodiversity. to be an important contributor. 251. However, a combination of unsustainable exploitation 253. One of the main achievements in terms of natural re- and the intensifying effects of climate change are straining source protection, but particularly in building social capi- Mexico’s natural resources. Unchecked agricultural ex- tal, comes from community incentive programs to improve pansion, overgrazing, water pollution, poorly managed forest management, established by the government in the coastal development, and illegal logging contribute to 1990s. These programs cover a comprehensive range of an ongoing process of deforestation240 and land degra- activities on forest lands related to social organization, dation241 that is diminishing Mexico’s natural capital. Soil capacity building, land-use planning, sustainable forest erosion affects almost half of the national territory, and 38 management and protection, as well as the extraction, percent of Mexico’s rivers are considered highly polluted. processing, and marketing of forest goods and services. Integrated management and planning of rural territories A key example is the Payment for Environmental Ser- can conserve natural resources and build economic and vices (PES) program, the largest such program in Latin Table 1: National Wealth Estimates, Mexico and Upper-Middle-Income Country Average Average for Upper- Average for Upper- Per Capital Per Capital Total Mexico Middle- Income Mexico Middle- Income Total Wealth Natural Capital Countries Countries (2014 US$) 110,471 112,798 (2014 US Dollars) 14,629 18,960 Share by subsector % % Share by subsector % % Produced Capital 36.1 25.3 Forests 14.8 9.1 Natural Capital 13.2 16.8 Protected Areas 6.9 8.6 Human Capital 53.7 58.3 Crop and Pastureland 38.7 47.4 Net Foreign Assets -3.1 -0.4 Subsoil Assets 39.6 34.9 Source: The Changing Wealth of Nations 2018; World Bank. MEXICO POLICY NOTES 239 National wealth includes produced capital, natural capital, human capital, and net foreign assets. It complements GDP and other macroeconomic indicators by measuring changes in a country’s underlying asset base. 106 240 The General Law on Sustainable Forest Management defines deforestation as the complete or partial removal of vegetation from forest lands for non-forestry purposes. 241 The General Law on Climate Change defines degradation as the reduction in the carbon content of natural vegetation, ecosystems, and soils caused by human activity. Forest degra- dation includes both the reduced capacity of forest ecosystems to provide environmental services and damage to their productive capacity. America, which, since 2003, has spearheaded the appli- industries. As part of its NDC under the Paris Climate Ac- Chapter 10 cation of economic instruments for forest conservation cord, Mexico committed to reducing emissions from the and the promotion of sustainable forest management agriculture and livestock sector by 22 to 36 percent rela- practices. tive to baseline projections between 2020 and 2030. 254. The development of community forestry in Mexico 257. Block pricing for water and low water tariffs promote can be attributed to the system of collective land ownership the misuse of limited water resources. While this problem and sound community governance structures: this system is not unique to agriculture, given that approximately 78 is unique in terms of its scope and impact on local com- percent of Mexico’s water use is for agriculture, ensuring munities, in which an estimated 61 percent of Mexico’s 66 efficient water use by the sector is critical. The expansion million hectares of temperate and tropical forests belong to of agriculture and agribusiness activities in the north of agrarian communities. Many of these forests have greater the country has created especially severe environmental potential for timber production than the national average. pressures; concentrated economic activity and increased Despite the productive potential of communally-owned population density have led to overexploitation of land forests and the successful experiences in community for- and water resources and the degradation of local ecosys- est management to date, most forest dwellers continue tems. Nationwide, approximately 80 percent of Mexico’s to live in conditions of great marginalization and poverty, agricultural land exhibits some level of degradation due and few agrarian forest communities invest in their forests to overgrazing, excessive pesticide use, and/or improper or engage in sustainable and profitable forest manage- water management. ment over the long term. While forest management can become a significant source of income and employment 258. The agricultural sector not only contributes to climate for forest dwellers, public policies on forest management change, it is also highly vulnerable to its effects. Projections must be maintained and strengthened, and efforts must to year 2050 indicate that climate change could reduce be made to promote access to finance and involve the Mexico’s agricultural output significantly. The scenarios private sector. suggest reductions in potential maize production of be- tween 12 and 27 percent. In northern Mexico, farmers are exposed to a range of extreme weather events, such as CHALLENGES drought and frost.243 In southern states such as Guerrero, Chiapas, and Oaxaca, farmers face climate-related risks such as floods and pest infestations. Nationwide, approxi- Natural resource degradation and mately 76 percent of Mexico’s total cultivated area is rain- vulnerability to climate change in fed, which leaves production highly sensitive to changes Mexico’s rural territories in rainfall and temperature. Addressing the short- and 255. Although deforestation and forest-degradation rates long-term risks posed by climate change will require sub- have declined in recent years, the clearing of forested land stantial investments in both adaptation and mitigation. for agriculture and cattle ranching continues to be a chal- lenge. A 2015 Forest Resource Assessment found that for- The Unexploited Potential of the est cover has declined by 5.4 percent since 1990, reflecting Forestry Sector more than 3.7 million hectares lost during this 15-year pe- riod.242 Deforestation rates vary, but are highest in tropical 259. Mexico faces the challenge of conserving and sus- dry forests and tropical rainforests. High poverty rates and tainably managing its forests while also meeting a growing limited economic alternatives in forested areas incentivize demand for timber products, which currently exceeds pro- activities that generate short-term returns, leading to the duction by a factor of three. In 2006, 80 percent of the vol- unsustainable exploitation of forest resources. The low ume of timber production came from communally owned marginal productivity of many agricultural and forest ac- forests.244 Over the last two decades, community-based tivities encourages farmers and ranchers to expand crop forest enterprises (CFEs) have flourished in Mexico. How- and pastureland rather than intensifying production. Un- ever, while Mexico has the capacity to more than double sustainable logging, fuel-wood collection, forest fires, and its current production of standing timber, the rising num- pests and diseases further damage and degrade Mexico’s ber of CFEs has not yet been matched by a commensu- forests, and deforestation is a significant cause of GHG rate increase in timber production. Instead, legal timber emissions. production fell from 9.4 million cubic meters in 2000 to 5.9 million in 2012, even as domestic demand for timber 256. Agriculture and cattle ranching are among Mexico’s increased. This trend is due in part to the failure of sectoral largest sources of GHG emissions. Agriculture and cattle incentives to balance conservation and economic objec- ranching are also the third-fastest-growing sector in terms tives. In 2015, almost 1,500 harvesting permits245 were of emissions generation, after the automotive and mining granted, but harvesting only occurred in 46 percent of the MEXICO POLICY NOTES 242 Mexico has between 66 and 88 million hectares of forest, depending on the definition used. 107 243 A drought in 2011 cost corn farmers an estimated US$645 million and bean farmers US$430 million. 244 Merino Pérez et al., 2014. 245 Based on an analysis of the SEMARNAT series on the extent of authorized production and the volume of real production. Chapter 10 Figure 51: Agriculture: Productivity Performance 0.040 0.035 0.030 0.025 0.020 0.015 0.010 0.005 0.0 Mexico Colombia Brazil Argentina Chile United -0.005 States -0.010 -0.015 Output growth TFP growth Input growth Figure 1 (b) Labor productivity, Output per worker (2004-2006 USD$/worker) 12000 10000 80000 60000 40000 20000 0 1992 1994 1996 1998 2000 2002 2004 2008 2010 2012 2014 Mexico Colombia Brazil Argentina Chile United States Source: Calculations based on data from USDA ERS - International Agricultural Productivity. approved area. A recent study246 revealed that the most strong capital injection. Currently there is also little pri- accessible forested areas could produce 60 million cubic vate sector investment in timber product value chains. meters of round-wood, while the priority areas identified Other CFE capacity constraints include challenges in by the National Forestry Commission (Comisión Nacional managing forests and in processing to add value; lack of Forestal, CONAFOR)—including major watersheds and the technical, administrative, management, and innovative states of Campeche and Quintana Roo—could produce capacities; limited market access; lack of access to basic up to 16 million cubic meters. Even with lower projections, communications for the purposes of marketing; and lim- the benefits of improving forest production are significant. ited access to key information on market prices for tim- For example, if the 2018 national forest production goal ber and real timber production costs. CFEs need to have of 11 million cubic meters of round- wood per year is met better access to finance and a closer partnership with the and annual forest productivity is increased from 1 to 3 private sector if they are to better access Mexico’s timber percent, it is expected that the number of formal jobs gen- production market. erated in the forestry sector by 2025 could be on average 21,216 per year. Ensuring a Robust and Sustained 260. The challenge of meeting timber demand is further Agriculture Growth Pattern hampered by CFE’s limited capacity. Access to finance is 261. Mexico’s agricultural sector has performed below ex- one of the main challenges facing CFEs, in which their pectations; improving the sector’s productivity can increase MEXICO POLICY NOTES long-term competitiveness can only be assured through its contributions to inclusive economic development. Total 108 246 IMCO and World Bank (forthcoming). Figure 52: Agriculture Land (thousands of hectares) and Contribution to Total Agricultural Chapter 10 Output (%) by State 2500 14 12 2000 10 1500 8 6 1000 4 500 2 0 0 Michoacan Veracruz Oaxaca Jalisco Sinaloa Sonora Guanajuato Puebla Mexico City Durango Chiapas Baja California Chilhuahua Tamaulipas Coahuila Agriculture Area ('000 ha) Agriculture GDP (%) Source: SAGARPA/SIAP (2017) and INEGI (2016). Data for year 2015. factor productivity (TFP) has been a source of agricultural 263. Geographical and biophysical differences contribute growth in Mexico, but a decomposition of output growth to regional performance gaps, however, structural and suggests that the gains in productivity have been smaller technological factors play a major role in determining ag- compared to other countries (Figure 51(a)). Technological ricultural productivity and competitiveness across regions. change in countries like Chile and Brazil is happening fast- The northern and central states have the largest share of er, as reflected in more efficient use of inputs in the pro- high-value crops and high-yielding, irrigated commercial duction process. Furthermore, labor productivity is very production, while agricultural sectors in southern states low in Mexico, in which the output generated per worker tend to rely on less-productive, smaller-scale operations in Chile is 1.8 times higher than in Mexico, and the respec- using traditional cultivation techniques. Important dis- tive figures for Brazil and Argentina are 3.0 and 6.4 times parities in agricultural performance both between and higher than in Mexico (Figure 51(b)) within regions and across the production scale reduce average yields. Nationwide, the average corn yield is 262. Addressing disparities in agricultural performance about 3 metric tons per hectare, less than half the US between states and regions could benefit thousands of average of 8 metric tons. Studies have estimated that poor and vulnerable households. In 2015, five northern technological improvements implemented across less and central states—Jalisco, Michoacán, Sinaloa, Chihua- than half of Mexico’s corn- producing area could boost hua, and Sonora—contributed 50 percent to agricultural agricultural output by 8 percent and spur a 2 percent in- output, even though Mexico’s southern states are home crease in the growth of related industries such as animal to the largest share of agricultural producers and have feed. Irrigation is especially crucial, as the average corn an important share of Mexico’s agriculture land (e.g., Chi- yield on irrigated land is 8 tons per hectare, compared to apas and Oaxaca states) (Figure 52). Moreover, regional just 2.3 tons on rainfed land.247 disparities in agricultural production are widening over time: between 2004 and 2010, primary agricultural out- 264. Land fragmentation is a challenge to agricultural de- put grew by 2.5 percent in the north, 1.3 percent in the velopment. About 73 percent of plots are smaller than 5 center, and just 0.1 percent in the south. In 2015, Jalis- hectares, many of which are worked by semi-subsistence co’s agricultural sector comprised 6 percent of the state’s farming households employing traditional, rainfed pro- economic output and 8.5 percent of employment, and duction practices, especially in the central and southern contributed 12 percent to Mexico’s total agricultural pro- parts of the country. The small size of plots prevents the duction. By contrast, among southern states agriculture formation of economies of scale, except in cases where provides a far larger share of employment, but contrib- effective farmer organizations are in place. Low marginal utes much less to economic output. In Oaxaca, agricul- productivity also impedes financial access. Households ture accounts for 32 percent of employment but just 6 farming small plots in remote or isolated areas face ad- percent of economic output. In Chiapas, despite having ditional productivity and competitiveness challenges. A 1.9 million hectares of agriculture land, the sector con- recent study suggests that the land reforms of the ejido tributes just 3.7 percent to Mexico’s total agriculture system in 1992, which created opportunities for complete MEXICO POLICY NOTES economic output (Figure 52). property rights (Dominio Pleno) have limited impacts in 247 SAGARPA, 2016. 109 Chapter 10 terms of growth of agriculture added value, in the absence icies and programs designed to support sustainable ag- of comprehensive complementary agricultural and rural riculture and forestry development under the leadership development programs.248 of CONAFOR and the Secretariat of Agriculture, Livestock, Rural Development, Fisheries, and Food (Secretaría de Ag- 265. Logistical and value chain constraints result in ineffi- ricultura, Ganadería, Desarrollo Rural, Pesca y Alimentación, ciencies, and limit innovation and market inclusion. Mexico SAGARPA). In December 2016, SAGARPA, and the Secre- faces wide logistical constraints due to the country’s high tariat of Environment and Natural Resources (Secretaría de dependency on road transportation (which is costly, with Medio Ambiente y Recursos Naturales, SEMARNAT) signed poor road conditions and insecurity adding to costs), poor an agreement to collaboratively promote sustainable connectivity between production nodes and consumer rural development and address climate change. In the markets, lengthy cross border times, and limited quality short term, the government can strengthen cross-sec- storage infrastructure.249 Furthermore, although the agri- tor coordination by establishing a set of common policy culture and food system in Mexico has transformed rapid- orientations, such as mainstreaming zero-deforestation ly, it is still at a mixed stage of development. For example, approaches into agricultural support programs and sec- while this system has modern characteristics, consolidat- toral development planning.253 Preparing a joint workplan ed in segments such as supermarkets and other modern could help coordinate environmental and agricultural retail and wholesale markets, and characterized by grow- policies and maintain focus on priority areas. In the me- ing innovation and some levels of “dis-intermediation” dium-to-long term, launching joint programs with shared (supermarkets buying directly from processors, or urban goals and budgets could more effectively leverage public wholesalers buying directly from farmers), it is also largely resources, and establishing joint mechanisms to priori- traditional and fragmented, with high levels of intermedi- tize and design public investments at the national and ation, little value addition and high inefficiencies. These regional levels could maximize the impact of the invest- inefficiencies can be significant, in which food losses and ment budget. At the institutional level, further efforts will waste in Mexico have been estimated at 20.4 million tons be necessary to coordinate and leverage climate finance annually, with 70 percent of these losses occurring from through the creation of the Climate Change Fund (Fondo production to distribution.250 Achieving higher efficien- para el Cambio Climático, FCC).254 Performing sectoral envi- cies in the way off-farm components of the food system ronmental assessments of current programs and policies are carried out (processing, packaging, logistics, etc.) is could also help identify entry points to foster sustainable critical to improve outcomes for farmers and other value program and policy outcomes. chain actors, as well as for consumers.251 267. Enhance capacities for spatial territorial planning, management and monitoring. Mexico has recently taken POLICY OPTIONS important steps towards strengthening its policy and institutional framework to support territorial devel- opment, including the restructuring of the Ministry of Aligning Productive and Agrarian and Urban Development (Secretaría de Desarrol- Environmental Objectives to lo Agrario, Territorial y Urbano, SEDATU)255 and issuing of support Growth, Conservation, and the general law on territorial development,256which rec- Resilience252 ognizes the relevance of anchoring territorial planning 266. Align policies and programs around productive and and management strategies in cross-sectoral, as well as conservation objectives to address threats to Mexico’s nat- regional and local, priorities. The new law is expected to ural capital. Public programs and investments in multiple address regional unbalances, bring both urban and rural sectors could be aligned to reduce deforestation, forest perspectives into territorial planning, as well as climate degradation, and other natural resource deterioration, risks management approaches and environmental di- while at the same time generate improved livelihood op- mensions. A new National Strategy on Territorial Devel- portunities. An effective low-carbon rural development opment will be developed, with a long-term perspective. strategy will require an unprecedented degree of policy In the short-term, however, there are important areas and institutional coordination, particularly at the local of capacity strengthening that can enhance emerging level and between stakeholders in the agricultural and for- efforts to promote integrated territorial planning and estry sectors. The government has already introduced pol- management, as discussed below. 248 Property Rights Reform in Mexico: Impact on agricultural, rural, and structural transformations. Alain de Janvry, Elisabeth Sadoulet, Marco Gonzalez-Navarro, Kyle Emerick (TU), Eduardo Montoya, Matthew Pecenco, Daley Kutzman (NWU). Presentation at the Annual World Bank Conference on Land and Poverty, March 2018. 249 Mexico has low storage capacity (4 million m3) compared to Brazil (5.7 million m3) and Japan (34 million m3). Also, the available cold storage infrastructure is significantly more expensive (e.g., for tomato) in relation to its competitors. 250 Pérdidas y desperdicios de alimentos en México. Banco Mundial, 2017. 251 Some reports estimate that consumers in Mexico could pay up to 86 percent higher prices resulting from high levels of intermediation. MEXICO 252 The World Bank has recently approved two operations to support resilient, productive and low-carbon environments. POLICY NOTES 253 Cross-sector coordination should also entail the inclusion of environmental considerations and conditions into large agricultural support programs managed by SAGARPA, such as integrating eligibility criteria into the operating rules for agricultural activities to prevent the expansion of the agricultural frontier to the detriment of the forests. 254 The FCC was created by the General Climate Change Law, Chapter VII, Articles 80–86. 255 SEDATU’s mandate is the planning, coordination, administration, generation and execution of public policies for territorial ordinance, provide access to decent dwellings, urban and 110 rural development and granting legal certainty to agrarian populations, seeking to improve Mexicans’ quality of life, prevent human settlements in risk areas and assist in case of natural phenomena for immediate attention. https://www.gob.mx/sedatu/que-hacemos/ Last access: January 21st, 2018. 256 La Ley General de Asentamientos Humanos, Ordenamiento Territorial y Desarrollo Urbano, Nov 28, 2016. 268. Continue enhancing capacities for integrated environ- irrigation, incentives for renewable energy investment, Chapter 10 mental monitoring. Since the 1990s, Mexico has created efforts to promote the production of organic and min- numerous national parks and protected areas. While this eral-based fertilizers, and credit guarantees to support is an effective approach to conservation, national parks green investments. The impact of these efforts should be and protected areas require sustained budgetary outlays. assessed to identify and scale up effective interventions. Mexico has made significant progress in establishing an Further measures could support the development of integrated environmental monitoring system for protect- enhanced seed varieties and build an enabling environ- ed areas and compiling biodiversity and forest invento- ment for climate-smart investment. Adopting integrat- ries.257 In addition, the National Institute of Ecology and ed pest-management strategies, using price-correction Climate Change (Instituto Nacional de Ecología y Cambio mechanisms to discourage the misuse of water and land, Climático, INECC) and CONAFOR periodically conduct and creating contests, auctions, challenge funds and national assessments of GHG emissions. The government other programs to support innovative climate-smart could enhance its environmental oversight framework by solutions could enhance the efficiency of resource use. refining the carbon-monitoring methodology used by Plans to reduce emissions in specific sectors as part of the national monitoring, reporting, and verification sys- the NDC represents an opportunity to identify and pri- tem for land use, land-use change and forestry, and by oritize high-impact mitigation opportunities, sustainable further integrating monitoring systems at the territorial practices, technologies, and investments. With respect to level. the cattle ranching subsector, GHG emission reductions could be generated by adopting climate-smart systems 269. Enhance interinstitutional data-sharing mechanisms and practices, such as silvopastoral production systems, to enable authorities to more effectively coordinate policies fodder reserves/banks, supplemental feeding, and im- and programs. SEMARNAT’s environmental monitoring proved pasture management and animal husbandry, all systems could be linked with SAGARPA’s food and fisher- of which also contribute to improved productivity and ies information service, as well as the information systems enterprise resilience. The private sector can play a pivot- used by CONAFOR, SEDATU, state and municipal govern- al role in supporting the shift toward more sustainable ments, local authorities, and community leaders to sup- investment by adopting environmental-compensation port coordinated environmental planning and manage- approaches, signing zero-deforestation agreements, ment. These data-sharing mechanisms could be designed launching reforestation programs, and supporting the to support planning processes at the national, regional, development and testing of climate-smart technologies. and local levels, and they could integrate ecological anal- Market mechanisms can also be leveraged yses, including assessments of land-use suitability and climate risks. 272. to reduce water scarcity. The authorities could adopt a pricing and marketing system for wholesale water re- 270. Validate and scale-up approaches to improve local sources (aguas nacionales) that reflects regional and tem- territorial governance, planning and management. The poral changes in water levels, aligning the cost of water re- Sustainable Production Systems and Biodiversity Proj- sources with their relative scarcity. The government could ect implemented by the Commission for Knowledge also strengthen mechanisms for resolving water conflicts. and Use of Biodiversity (Comisión Nacional para el Con- Policymakers should discourage the incipient shift in ocimiento y Uso de la Biodiversidad, CONABIO) and the commercial agriculture toward the south of the country, Biodiversity in Productive Forests and Certified Mar- where water is still plentiful, by removing subsidies and kets Project implemented by CONAFOR attempt to link potentially imposing taxes. Investment incentives should productive investments with environmental outcomes. be used to encourage water preservation, collection, However, initiatives that would take these efforts to treatment and reuse schemes. scale and frame them within the context of integrated territorial management are relatively new in Mexico. 273. Engage vulnerable groups in efforts to halt defor- Two recently approved World Bank operations will help estation and forest degradation and implement integrated mainstream, to a territorial level, the approaches applied territorial management. Historically, rural households in earlier projects, however, close monitoring and evalu- without formal land titles have had very little input into ation will be necessary to continuously identify lessons decision-making processes regarding forest management and incorporate them during the implementation of and productive activities, even when these activities were these operations. critical to their livelihoods. Efforts to include vulnerable groups should draw on lessons learned from international 271. Promote climate-smart technologies and adopt prov- and national experience, including those fostering finan- en strategies for addressing climate challenges to bolster cial inclusion and supporting transparent mechanisms for resilience of producers in the agriculture and forestry sec- resource allocation. Special attention should be devoted tors. SAGARPA is advancing the country’s ambitious cli- to the specific obstacles that inhibit women’s participation MEXICO mate goals through, inter alia, a program to modernize in natural resource management. POLICY NOTES 257 The information system includes the Monitoring of Ecosystem Diversity in Natural Protected Areas of Mexico System (SAR- MOD), the Wide Coverage System for Monitoring Diversity (SAC-MOD); and the National Forest and Soil Inventory (INFyS), among others. 111 Chapter 10 and relatively large, well-off producers. A thorough analy- Supporting Productivity and sis of agricultural support programs could provide a sound Competitiveness of Agriculture/ basis for reform. International experience suggests that Forestry Sectors and Sustainable support to agriculture innovation and expanding access Employment to services and input markets by enhancing the roles of 274. Enhance the effectiveness of current support to ag- the public and private sectors, drives technological uptake riculture, forestry, and other productive resource-based and improve productivity and competitiveness. sectors, by focusing on building long-term productivity and competitiveness. The country has relied heavily on its com- 277. Address specific constraints to the competitiveness parative advantages to develop its agriculture and forest- of smallholder farmers and small-scale forestry enterprises ry sectors, particularly around its natural capital endow- to bolster productivity and enhance market integration. ments and geographical location; further gains will come SAGARPA’s and CONAFOR’s extension systems should from improving productivity and strengthening its com- be maintained and strengthened by collaborating with petitiveness position. On the supply side, accelerating the national and international institutions that have proven adoption of new technologies could sustain productivity effective in promoting rural development and building growth in the primary sector and reduce vulnerability to human capital among agricultural, forestry and related environmental hazards and economic shocks. service workers.260 In addition, further efforts to facilitate financial inclusion, including for green investments, and 275. Enhance the competitive position of Mexico’s agri-food mainstream innovative public-private partnership mod- and forest products via logistical improvements. Empirical els to facilitate access to markets and technical services, evidence highlights the relevance of investments in the such as the “productive alliance approach” could expand transport network on improved market access and local commercial opportunities for smallholder farmers, ejidos employment and specialization.258 In Mexico, despite im- and rural entrepreneurs. Farming and forestry workers portant investments in road infrastructure, bottlenecks with limited scope for commercialization may benefit remain particularly in hinterland connections to primary from an approach that integrates productivity support domestic markets and ports. Investments in port and road with health and nutrition, education, and social-protec- infrastructure need to be maintained and enhanced. There tion programs. For transitional producers with greater is also a need for continued investments in enhancing cold commercial potential, productivity improvements should chain and storage infrastructure. Furthermore, system be complemented by efforts to strengthen market link- investments supporting user-friendly price and climate ages, provide technical assistance, facilitate credit access, information can contribute to improving the ability of pro- and develop entrepreneurial and administrative skills. ducers, and private and public entities to make decisions Smallholder farmers and small forestry enterprises re- and can be an effective means to reduce market power of quire capital to grow, and they need access to a support- intermediaries and increase competition. Supporting the ive private sector to diversify and integrate into larger val- development of market access and linkages programs, as ue chains. Private-sector partnerships with entrepreneurs well as general programs aimed at enhancing producers and community- based forestry enterprises are especially and other key supply chain actors capacities in the post- vital to the sustainable exploitation, processing, and mar- farmgate segments, can contribute to improve supply keting of timber and non-timber forest products. While chain efficiencies, generate value addition and improve Mexico’s timber-processing sector is growing, there is competition. Strengthening phytosanitary and sanitary still considerable scope for improvements in quality and capacity could also further expand export opportunities, efficiency. For farmers and forestry producers that are al- while mainstreaming risk-based inspection approaches ready well integrated into markets, creating an enabling could help address key food-safety issues in the domestic business environment could help improve productivity market. and enhance competitiveness. 276. Increase effectiveness of public support to the agri- 278. Strengthen links between rural social and economic cultural sector. Rising concerns have emerged on the ef- programs to enhance opportunity for rural workers. In- fectiveness of public support for the agricultural sector, tegrating the information systems used by SAGARPA, which is seen as regressive and disproportionately bene- CONAFOR, SEDATU, and Mexico’s major health, education fiting large-scale producers, especially programs that pro- and social protection programs could enhance interin- vide direct payments based on cultivated land area or the stitutional collaboration. Establishing a comprehensive size of livestock herds. While direct payments to farmers system to monitor the delivery of support to rural com- have fallen in recent years,259 they continue to comprise munities and track its effectiveness could yield important about 80 percent of producer support. Large-scale “mar- information on beneficiary targeting and programmatic ket support” subsidies appear to privilege northern states design. MEXICO POLICY NOTES 258 Roads and the Geography of Economic Activities in Mexico. Policy Research Working Paper 8226. World Bank, October 2017. 259 Direct producer support fell from an average of 29 percent of agricultural spending between 1993 and 2004 to an average of 10 percent between 2014 and 2016, well below the OECD 112 average (OECD, 2017). 260 For example, the International Maize and Wheat Improvement Center (CIMMYT) implements the highly successful Sustainable Modernization of Traditional Agriculture program with the support of SAGARPA and MASAGRO. 11. Sustainably Managing Scarce Water Resources M exico’s water resources are diminishing, pollution levels are rising, competition between users is intensifying, and inefficiency in the use of water in the agricultural and oth- er sectors is significant. Rising global temperatures and shifting precipitation patterns are already affecting the country’s hydrological cycles, and the increasing strain on the country’s scarce water resources is leading to the overexploitation of groundwater and productivity losses. Water security and access is a critical priority. In this context, strengthening water management will require modernizing the current policy and institutional framework. A more efficient allo- cation and better valuation of water resources could also ease stress on hydrological systems, attenuate conflicts among water users, and reduce the risk of shortages. Improving coverage and quality of water and sanitation services will require addressing regional and rural-urban dispari- ties, improving infrastructure and finding efficiencies, including in the investment planning and execution process. Investment would also be needed for the construction and rehabilitation of key projects (e.g., dams, canals, pumping stations). Some options to achieve efficiencies and better use the fiscal space of the sector include: (i) introducing multi-year budget for CONAGUA (with a long-term planning that transcends political cycles) and other investments in the sector; (iii) establishing a pool/pipeline of priority projects at the national and subnational government lev- els; (ii) revisiting water pricing to improve service efficiency, manage sustainably water resources and reduce fiscal burdens from subsidies; (iv) formulating a water-infrastructure and resilience strategy within the budget process of the sector, and (v) adopting innovative financing strategies and instruments (including expanding the use of Public-Private Partnerships (PPPs) beyond wa- ter supply and wastewater) and developing performance-based management contracts for Water MEXICO Supply and Sanitation (WSS) services. Additionally, an assessment of governance and accountabil- POLICY NOTES ity measures in the sector is needed to identify gaps and overlapping competencies. 115 Chapter 11 to water or receive low-quality water services. More than CONTEXT AND REFORM 100 of the country’s 731 watersheds face severe shortag- PROGRESS OF THE SECTOR es, and the number of overexploited aquifers more than tripled from 32 of 653 in 1975 to 115 in 2016. Groundwater 279. Improving WSS quality is critical to achieving Mexico’s provides more than 65 percent of all water used by Mex- Sustainable Development Goals (SDGs). As of 2015, drink- ican cities. ing water and sanitation coverage reached 92.5 percent261 and 91.4 percent262 of Mexico’s population, respectively. 281. Water use in the agricultural sector265 is highly ineffi- However, almost 9 million people still lack access to drink- cient, and agricultural runoff increasingly pollutes freshwa- ing water, and 11 million lack access to sewerage, most of ter lakes and rivers. Mexico’s agricultural sector accounts whom are in poorer states such as Guerrero, Chiapas, and for 77 percent of total water use, but its efficiency rate is Oaxaca. Approximately only 57 percent of the collected just 40 percent266. In 2016, agriculture contributed with wastewater is treated, technical and nontechnical losses 4.1 percent to GDP. The government subsidizes water and consume 30-50 percent of the water in the distribution electricity for agricultural production, both of which con- networks, and very few WSS utilities achieve cost recovery. tribute to inefficient water use. The implicit electricity sub- Moreover, millions of households continue to receive WSS sidy provided to Mexican farmers for pumping groundwa- services that do not meet SDG standards for safety and ter has been estimated at about US$1 billion per year.267 service quality. In addition, approximately 50 percent of the wastewater discharged into the rivers comes from the agricultural 280. Rapid urbanization, robust economic and population sector, primarily nonpoint pollution sources268. Mexico has growth, and suboptimal resource management are putting approximately 6.5 million hectares of irrigated land, and increasing stress on Mexico’s water resources. Meanwhile, the federal government plans to expand irrigation to an- the inefficient use of water resources is compounding the other impact of overexploitation and water pollution. Mexico’s northern-central region encompasses 32 percent of the 282. 1.5 million hectares at a cost of almost US$5 billion over country’s water resources but is home to 77 percent of the next 10 years.269 Nevertheless, there are still positive ini- the population and contributes 79 percent to its GDP. By tiatives in terms of strengthening new dam-safety norms, contrast, the south-southeastern region encompasses 68 which have recently been formulated, and risk assess- percent of Mexico’s water resources, but is home to only ments of several dams are currently underway for better 33 percent of its population and contributes to 21 percent resilience and bulk distribution of water. In addition, the to GDP. The regional disparities between water availability government has recently introduced policies to promote and GDP contributions of northern and southern areas more efficient irrigation through irrigation-management of Mexico are exacerbated by copious water resources: transfer programs, but these programs remain limited to they cannot be considered as suitable as supplies of wa- a few locations. ter for the country’s rural population, mostly in the poor south, because of their levels of pollution263. Urban areas 283. Inefficient water management imposes large economic have taken precedence in higher investments due to their costs. The decline in water resources represents 28 percent political and economic power, whereas rural areas have of the total economic cost of natural-resource depletion, lower economies of scale due to scattered populations. As or twice the cost of deforestation.270 In 2015, the cost of Mexico’s population has grown, water availability per cap- declining water resources and the degradation of water ita has fallen sharply, dropping from 18,035 m3 per year quality represented half of a percentage point of GDP. in 1950 to 3,392 m3 per year in 2015. The National Water Moreover, while the pace of deforestation and fossil-fuel Commission (Comisión Nacional del Agua, CONAGUA) proj- extraction appear to be slowing, water depletion acceler- ects that water resources per capita will reach 3,250 m3 by ated by 0.7 percent per year between 2003 and 2015.271 2030.264 More than 35 million Mexicans have limited access 261 Drinking-water access rates range from 95.7 percent in urban areas to 81.6 percent in rural areas. 262 Sanitation access rates range from 96.6 percent in urban areas to 74.2 in rural areas. 263 Shretha, S., Anal, K., Salam, A, & van der Valk, M. (2015). Managing Water Resources under Climate Uncertainty: Examples from Asia, Europe, Latin America, and Australia. New York, NY: Springer 264 CONAGUA, 2016. “Numeragua Mexico.” Secretaría de Medio Ambiente y Recursos Naturales and Comisión Nacional del Agua. 265 The agricultural sector in the country consumes around 70 percent of available water in the country. 266 From the total amount consumed by the agricultural sector in the country, only 40 percent is actually used for irrigation. The remaining 60 percent is lost to evapotranspiration, surface runoff (whereby water ends up in the drain) and water losses of the water distribution network used for irrigated croplands. Water efficiency for irrigation is calculated through convey- ance efficiency (which represents the efficiency of water transport, canals, distribution networks), and field application efficiency (which represents the amount of water actually used in irrigated land. 267 In Mexico, electricity prices for the agricultural sector remain well below cost. See: OECD, 2017. “Mexico’s efforts to phase out and rationalize its fossil-fuel subsidies.” These subsidies also respond to political support to the agricultural sector and its organization committees that is creating faster depletion and unsustainable use of resources, as excess demand for irrigation is encouraged by the subsidy to electricity used for groundwater pumping, and where farmer groups pay less than 1/5 of the cost of generating and distributing electricity for pumping water. MEXICO 268 Of all non-municipal wastewater, 29.9 m3/s, or 15.8 percent of the total, are treated. 63.52 m3/s of non-municipal wastewater comes from sugar refineries and are used to irrigate POLICY NOTES sugarcane fields (See: Water Resources in Mexico: Scarcity, Degradation, Stress, Conflicts, Management, and Policy, second edition 2014, Springer). 269 Mexico ranks 19th worldwide in water-storage capacity, with 150 km³ of stored water and 667 big dams. Mexico ranks 6th in total irrigated area, with 6.9 million hectares under irri- gation. Cutting-edge irrigation technologies in the north and rain-fed solutions in the south can make viable expanding farming irrigated land, along with higher reuse and conversion from treated wastewater, given the limited replenishment and endowment of surface and underground water in the country. See: CONAGUA, 2016. 116 270 Instituto Nacional de Geografia e Informatica (INEGI) (2016) Cuentas Economicas y Ecologicas de Mexico. 271 In addition, Mexico’s 2016 water accounts show that the annual depletion of groundwater cost an estimated at 27.9 billion pesos in 2015, while the cost of untreated wastewater was 57.4 billion pesos. Figure 53: Water Use in Mexico Chapter 11 100% North, Center and Northwest 90% 1, 734m3/person/year 31 80% II 70% I 60% 77 80 Percent VI Southeast 50% III VII 13,097m3/ 40% IX person/year 69 30% VIII XII 20% XIII 23 10% 20 IV X 0% Natural Average Population GDP V XI Availability South North Source: CONAGUA, 2010 286. 19.3 percent, greater than the share with low levels of KEY CHALLENGES educational attainment (17.4 percent), very poor housing quality (12.0 percent), or inadequate access to basic health 284. Despite significant reforms, policy challenges remain services (15.5 percent). Other multidimensional poverty ahead. For instance, incomplete legal and institutional estimates274 indicate that water supply and sanitation ac- reforms weaken the efficiency of water-sector manage- count for 10 percent of nonmonetary poverty in rural ar- ment. Legislation in 1993 and 2004 embraced integrated eas, comparable to the impact of education (12.5 percent). water-resources management as critical to sustainable development. However, administrative and regulatory 287. Climate change is increasing the frequency and inten- challenges prevent the government from fully operation- sity of weather-related extreme events and changing precip- alizing this principle. CONAGUA is responsible for regulat- itation patterns, with a disproportionate impact on the most ing water resources, and for developing, operating, and vulnerable households. Weather-related disasters inflicted overseeing hydraulic infrastructure and other water-sec- US$44 billion in damages between 1980 and 2015, with tor investments. However, CONAGUA’s broad mandate floods and hurricanes together costing US$1.6 billion in reduces the efficiency of its regulatory functions and its 2015 alone.275 Deviations from historical rainfall trends can regulatory autonomy since it is responsible for the man- harm crop production, especially corn grown on rainfed agement of water resources, for service provision (bulk land,276 and the agricultural sector accounts for a large water) and the construction and operation of large water share of the economic losses caused by climate change. infrastructure. Moreover, legal instruments such as the Poor and vulnerable households are especially sensitive Financial System for Water (Sistema Financiero del Agua), to climate-related risks to agricultural production, food which was proposed in the 2004 Water Law272, have not prices, and food security277. Ongoing efforts to address been implemented, and most operators and water users’ water- security issues in the Valley of Mexico and among associations continue to depend on unpredictable, some- communities near its headwaters have yielded valuable times poorly targeted federal investment programs and lessons, and effective approaches can be scaled up at the subsidies, which represent 50 percent of total investment national level. However, policymakers continue to lack in the water sector. accurate and timely weather and climate information, and interinstitutional coordination is weak. The 2014-18 285. Limited access to water services is a major contributor National Development Plan, the National Infrastructure to nonmonetary poverty. In 2016, the monetary poverty Plan, and the National Water Program all call for greater rate was 36 percent (representing 44 million people living coordination between sectors, but the institutional frame- below the poverty line), and the extreme monetary pover- work does not support multisector investment in climate ty rate was 7.6 percent (representing 9.4 million people).273 change adaptation and mitigation projects. Under the government’s new multidimensional pover- ty-estimation methodology, the share of the population 288. Limited investment in key water management infra- with inadequate access to basic public services—includ- structure is exposing Mexico to higher climate and non-cli- ing water and sanitation—is 272 The 2004 Water Law also mandated the restructuring of CONAGUA’s key functions by transferring responsibilities from the central level to subnational Basin Agencies and Basin Councils. These Basin Agencies and Basin Councils were established, but many of their proposed functions—such as setting rates, collecting fees, and preparing financial plans—re- main under the purview of CONAGUA MEXICO 273 CONEVAL, 2016. POLICY NOTES 274 See, e.g., Oxford Poverty and Human Development Initiative (OPHI) Country Briefing: Mexico, 2017. 275 Maskrey, A. (2015) Evaluación global de reducción de riesgos por desastres 2015. United Nations- UNISDR. 276 The study found that a 10 percent increase in rainfall variations reduced output to be 0.8 percent for maize and other crops. See: Yunez and Meza (2015) The Effect of Rainfall Varia- 117 tion on Agricultural Households: Evidence from Mexico. https://ideas.repec.org/p/ags/iaae15/212457.html 277 The government has identified several priority actions to reduce the vulnerability of people and property to the impact of extreme weather events, including reforming national, state, and city water-planning processes. Chapter 11 matic related risks while increasing opportunity costs of 291. The water sector faces both cyclical and structural fi- underinvestment. Mexico ranks 19th worldwide in wa- nancial challenges, which weakens incentives to use public ter-storage capacity, with 150km3 stored in 667 large resources efficiently, and hampers government’s long-term dams. However, some of these dams were built more than planning of the sector. The sector is spending inefficient- 50 years ago, and increased investment in maintenance ly because of complex public investment management and rehabilitation will be vital to ensure their continued system and operating rules. The annual budget cycle dis- operation and to protect local populations from poten- courages multi-year investment projects. Most water and tial dam failures, and improve efficiency of hydropower sanitation funds are allocated annually with no guarantee generation. Noteworthy, that tightening fiscal constraints of future financing. An inefficient planning process results are still reducing investment for efficient management in frequent delays and creates incentives to misrepresent and distribution of water among competing users. The the status of ongoing projects. In some cases, local gov- pressures on federal, state and municipal budgets make ernments, particularly in disadvantaged areas, may lack the need for alternative sources of funding and a more the technical capacity to fulfill the complex requirements efficient public expenditure in the water sector more ur- of the planning process. Feasibility studies often ignore gent. However, ambitious capacity-building and changes the asset lifecycle, further favoring investment over op- in roles and responsibilities of planning, funding and erations and maintenance. The discretionary authority of implementing efficiency improvement programs will be CONAGUA for funding execution within and across mu- required to fully implement these norms and make spend- nicipalities creates delays and shifts in resource allocation ing more efficient. within a fiscal year. 289. Inefficient water use in the agricultural sector intensi- 292. Water is almost always priced below its true econom- fies pressure on scarce water resources. Because water for ic, cultural and social value in Mexico. Increasing tariffs to irrigation is not subject to abstraction charges, farmers levels that approach the actual value of water resources have little incentive to use water efficiently. This policy has would discourage waste, boost cost recovery and increase been the subject of intense policy discussion, as irrigated the financial envelope for infrastructure investment and agriculture accounts for most of non-hydropower water maintenance. However, raising water tariffs beyond the abstraction and consumption. The 2030 Water Resources level required to cover basic operation and maintenance Group has developed tools for public-private partnerships costs has thus far proven politically unfeasible. The feder- of large-scale irrigation projects through non-traditional al government recently reduced CONAGUA’s budget by PPP model for agricultural water use. The national infra- up to 70 percent, compounding an ongoing decline in structure fund (FONADIN) of the Secretary of Finance of the agency’s capacity to properly execute and supervise Mexico (SHCP) is exploring the viability of PPPs in irriga- public works, which has led to delays and cost overruns tion in water stressed by high-productive areas. in construction contracts. Nevertheless, the government is committed to realigning price incentives and restoring 290. Increasing urbanization, demographic growth, and financial balance to the water sector. Hence, the sector’s climate change are intensifying competition between rural budgetary process should aim for more flexibility accom- and urban water users. The Integrated Urban Management panied by greater transparency in decision making and (IUWM) framework can complement traditional engineer- reporting. ing approaches to resolving WSS challenges. IUWM is underpinned by three core concepts: (i) that cities both directly impact, and are fundamentally dependent on, POLICY OPTIONS the wider watershed; (ii) that traditional, supply driven, hard infrastructure approaches are not sufficient to close 293. Introduce multi-year budgetary planning for CONA- the water cycle and ensure sustainable services, and (iii) GUA to improve spending efficiency and revisit water that water management and WSS policies must incor- pricing to improve consumption efficiency. Strengthening porate all consequences of urban water use, including water security will require precisely valuing water re- its implications for ecological sustainability, and account sources, scaling up payments for environmental services, for them via instruments such as payments for environ- improving interinstitutional collaboration (all functions mental services. Under an IUWM approach, planning for of basin agencies and councils), and incorporating cli- the water sector is integrated with other urban sectors, mate resilience into planning processes at the federal, such as land use, housing, energy, industry, and transpor- state, and municipal levels. For rural areas, the govern- tation to overcome urban planning fragmentation with ment should further adapt water quality standards to the aim of improving system- wide performance. IUWM reflect basin-level objectives and expand the autonomy also incorporates the interests of other users in the river of Basin Agencies, Basin Councils, and municipal water basin— including other cities, regional industries, and the utilities over local WSS service delivery. For urban areas, local environment—which have unique water quantity many cities are vulnerable to an array of climate and MEXICO POLICY NOTES and quality needs that may evolve over time. An effec- non-climatic risks, and face the prospect of droughts tive IUWM approach usually requires coordinated action and/or floods. Existing planning and investment design across multiple jurisdictions, both within the municipality 118 frameworks are not sufficient to ensure water security in and in the larger river basin. urban areas. 294. Establish a pool/pipeline of priority projects at the the use of PPPs beyond water supply to include irrigation, Chapter 11 national and subnational government levels. This initiative flood management, water conservation, and wastewater also would need developing the feasibility studies for that treatment could enable the authorities to accomplish key pipeline of projects. This would also allow subnational sectoral objectives in a resource-constrained environment governments with more limited to take advantage of proj- and free fiscal resources for social projects. ects when resources become available. 298. Instrument private strategies within the programs 295. Formulate a water-infrastructure and resilience strat- of CONAGUA to advance sectoral policy objectives in an egy within the budget process of the sector278. To mitigate environment of tight budget constraints and foster com- risks related to water security, cities should mainstream cli- petition. First, by including Public Private Partnerships mate resilience into their urban planning and water-infra- (PPPs) beyond water supply and wastewater treatment. structure investment processes. Optimizing the use of ex- These options can help municipal governments through isting infrastructure, strengthening sectoral management, concessions or management contracts to improve water and promoting water reuse, recycling, and recovery could utility services locally. Second, by enabling CONAGUA to improve the systemic efficiency of urban water manage- tap into private-sector financing and investment in other ment, reduce the cost of WSS services, and help close the areas such as data collection (e.g., weather, climate, hydro- water cycle within cities. The federal government could logical), agricultural water management, and flood man- support these efforts by expanding and increasing pay- agement and prevention. Budget constraints provide an ments for environmental services in states with high levels opportunity to develop a financing strategy that improves of water overexploitation and scarcity. Across different lev- the efficiency of public expenditures, avoids cost overruns, els of government, promoting green-growth models that and incentivizes private-sector participation by expanding reflect scarcity, uncertainty, and stress on water resources the use of performance-based contracts and results-based could increase the structural efficiency of water use, and financing, impact investments, and climate finance. creating an interinstitutional framework for sharing data on water resources and environmental externalities could 299. Establish formal institutional incentives to integrate strengthen the analytical underpinnings of water policy. related sub-sectors into the overarching water management objectives and targets. Integrating water supply, drainage, 296. Increase WSS investments for the construction and sewerage, wastewater, and disposal services (i.e., on-site long-term operation of capital projects. A particular case of sanitation) into urban water supply and sanitation (WSS) increasing investments in the maintenance and rehabilita- infrastructure can boost efficiency by leveraging econo- tion of dams will help enhance climate-change resilience. mies of scales. Several municipalities in Mexico have be- Options to increase funding relate to scaling up the use gun exploring alternatives to traditional WSS planning and of public-private partnerships and performance-based investment processes under the rubric of integrated urban contracts to multiply the potential public funding into the water management (IUWM). IUWM can be more effective if sector279, while incentivizing an increase in the operational instrumented vis-à-vis with other important components efficiency of local operators. Moreover, federal transfers like setting tariffs with efficient water valuations, restruc- are the main source of investment capital, supported to turing of CONAGUA’s management and implementing varying degrees by local taxes and international aid in the functions, and introducing water competition to providers country. Due to this condition, the investment strategy of with PPP funding to align performance with cost recovery the sector should incorporate local sources of financing to thresholds. Also, by incentivizing the integration of related multiply the number of local projects, increase local rev- sub-sectors that enable better management of water and enues, and promote efficiency in project implementation natural resources, while treating effectively pollution and by introducing incentives to ensure that the health and waste the country can accelerate the pace to reach water productivity benefits from water supply and sanitation and environment SDGs. projects outweigh the public funds invested. 300. Design and implement economic instruments for bet- 297. Introduce and scale up PPPs in the water sector, includ- ter water resources valuation to encourage states and mu- ing for financing where the framework and tariffs can sup- nicipalities to adapt coordinated policies and programs to port improvements in operational and financial efficiency. build resilience against climate change and reinforce water This will require adopting strategies for long-term finan- security. The authorities should promote new economic in- cial sustainability of water supply and sanitation operators struments for equitably allocating water rights, controlling and infrastructure. Adopting innovative financing strat- pollution, valuing water resources, and balancing the in- egies and instruments, developing performance-based terests of competing water users. The government should management contracts for WSS services, and expanding also build the capacity of Basin Agencies, Basin Councils, MEXICO 278 Large funding gaps requires investments canalized though “brown field projects” bringing an additional level of complexity to the PPP schemes to be deployed but with potential of POLICY NOTES optimizing economic returns of those investments and enable resilience within cities and large municipalities. 279 It is worth mentioning that these arrangements are fostering competition rather than privatizing services. Competitive tendering for concession agreements, for instance, are being implemented by CONAGUA. The Programa de Indicadores de Gestión de Organismos Operadores (PIGOO) is benchmarking service and efficiency performance of operators reporting 119 to CONAGUA. These benchmarks are used to determine viability of contract expansions and modifications at the state, city and hydrographic regions. The indicators include the ratios between operation and maintenance costs and volumes produced and consumed per operator, as well as efficiency performance indicators of management and customer satisfaction. Chapter 11 and Irrigation Districts and Units to strengthen resilience and it is not of the same quality as in urban areas. There- in vulnerable areas. A long-term integrated strategy for fore, CONAGUA and some high- performing water utilities harmonizing irrigation programs and promoting water still face challenges in providing adequate services in the conservation (e.g., through volumetric pricing, water ex- poorest rural areas, where the financial framework and ca- traction charges), increasing productivity, and discourag- pabilities to manage water supplies locally are lacking280. ing waste and pollution should be designed as part of a comprehensive green growth agenda for water-sector in- 303. Consider harmonizing, modernizing and simplifying stitutions. Finally, integrating water value chains in cities, the water management framework over the medium- and increasing public awareness of water issues, and promot- long-term. This will imply conducting a deep review of ing policy dialogue on IUWM could improve the quality of subsidy programs281 that hinders efficient management urban water infrastructure. of water resources in agriculture. These subsidy programs for irrigation directly contribute to overuse and waste, and 301. Assess governance and accountability measures of a comprehensive water management framework strategy the sector to identify inefficiencies and overlapping com- should be improved gradually by phasing out these sub- petencies. Administrative reforms have stalled, budget sidies, promoting incentives for efficient water allocation constraints are reducing infrastructure investment, and among competing uses and establish stewardship princi- federal subsidies are decreasing. Revising the regulatory ples for a more efficient inter-sectoral water management. framework for water services could address operational Improving oversight of public investment in the water gaps and redundancies and clearly assign responsibilities sector, enforcing sector’s regulations, and strengthening at each level of government. An analysis of risks to eco- contract management could help prevent cost overruns nomic efficiency and financial sustainability in the water and maintain service quality over time. sector could inform a strategy for improving the distri- bution of public expenditures, including federal subsidy 304. Develop an integrated water information system with programs in the water sector. multi-sector information for effective decision-making pro- cesses. Institutional fragmentation and an unclear regu- 302. Complement the decentralization of CONAGUA’s func- latory framework weaken data quality, and inadequate tions with the design of transparent and efficient intergov- information undermines the effectiveness of sectoral ernmental financing mechanisms. These include the use of planning. More detailed data could enable the govern- intergovernmental capital matching grants to incentivize ment to pilot the use of extraction charges based on investments in the sector and incentives to improve reve- water valuations for large irrigation users and test their nue collection to improve the distribution of sector fund- effectiveness at scale. Better data could also improve pro- ing between the federal and subnational governments. gram targeting and reduce regional disparities in access In addition, strengthening the capacity of Basin Agencies to WSS services. Expanded monitoring of water quality in and Basin Councils could improve sectoral governance watersheds, tighter oversight of groundwater extraction and contribute to more efficient planning, resource allo- and quality, and expanded civil-society participation in cation, and investment at the local level. Moreover, the basin planning could support more effective water pol- decentralization of water and sanitation services has not icies. Finally, developing an integrated monitoring and closed the gap between urban and rural areas because evaluation system that includes data on financial flows, policies assign to each municipality the responsibility of investment projects, water resources, hydrologic condi- their own water management. Nevertheless, in most rural tions, and WSS services could greatly improve policymak- areas, water supply does not cover their entire population ing in the water sector. MEXICO POLICY NOTES 120 280 Silva-Rodriguez, J.A. (2016) Rural Water Supply in Mexico. Cuad. Desarro. Rural vol.13 no.78 SCielo. 281 The Tarifa 9 Program (electricity subsidy of water pumping for irrigation run by the Federal Commission of Electricity), the Program for Improving Water Efficiency in Agricultural Areas (a program run by SEMARNAT) and the Special Program for Energy Use in Agriculture (PEUA, Programa de Energía de Uso Agrícola, run by SAGARPA). 12. Fostering Efficient and Sustainable Urban Systems I nvestment in urban infrastructure and services has not kept pace with Mexico’s rapid urban- ization, leading to a range of social, economic, and environmental challenges in metropolitan areas. The most pressing problems are: (i) the difficulty to revert the effects of the expansion of single-use housing developments on the urban periphery of cities that was boosted in the 2000s; (ii) the traffic congestion, for which a very conservative estimated US$70 billion in urban public transportation investment would be required to reduce it to the average level in developed coun- tries; (iii) the environmental costs associated with deficient solid waste management systems, as only 78 percent of all solid waste produced in Mexican cities is sent to final disposal sites, and a mere 11.5 percent of disposal sites comply with environmental regulations; and (iv) the challenge of increasing economic productivity in cities while limiting climate change effects in the context of an inefficient use of energy. Policy action will be needed to cope with these challenges, nota- bly to: (i) renew government efforts to promote compact and inclusive urban development; (ii) strengthen urban planning institutions at the local level to attract investment and deliver services to a denser population; (iii) strengthen regulations on minibus services and renew the fleet with clean technology equipment; (iv) increase the funding and enlarge the scope of action of the Fed- eral Program to Support Mass Transit (PROTRAM); (v) focus increased attention to non-motorized transport infrastructure, travel demand management initiatives, use of cleaner technologies in transport, and transit oriented development interventions; (vi) improve the institutional architec- ture for the solid waste management sector, strengthen capacities at all levels of government and improve financing options; (vii) focus on the construction or rehabilitation of final disposal facili- ties, the closure of existing open dumps, the acquisition of state of the art collection equipment, and the installation of biogas-capture and waste-to-energy technologies; (viii) promote behavioral change regarding all the steps of solid waste management and recycling; (ix) raise awareness of MEXICO POLICY NOTES the benefits of energy-efficient investments; (x) balance the interests of multiple stakeholders in the municipal energy sector; and (xi) engage the private sector in expanding energy-efficient infrastructure. 123 Chapter 12 310. (iii) establishing an ambitious public bicycling shar- CONTEXT AND REFORM ing system in Mexico City; (iv) approving the National Law PROGRESS on Solid Waste Management and the National Program for Prevention and Integrated Solid Waste Management; and (v) 305. Over the past 30 years, urban development in Mex- launching studies to improve energy efficiency at municipal ico has been characterized by the rapid, uncoordinated, level in all the state capitals of the country. These objectives and dispersed growth of urban and peri-urban areas. The are mutually consistent and reflect Mexico’s commitments population of Mexican cities increased sevenfold, and under the Paris Climate Accord and other international the country’s 11 largest metropolitan areas are now nine agreements. The government has an opportunity to build times larger than they were in the 1980s.282 The creation on these efforts by extending the scope and quality of of vast public housing developments on relatively inex- urban services, including housing, transportation, solid pensive land far from city centers in the 1990s and early waste management, and water and sanitation, with a fo- 2000s drove the growth of urban sprawl in areas that cus on energy efficiency. lacked social services and were far from jobs, creating numerous social, economic, and environmental chal- lenges. MAIN CHALLENGES 306. Investments in urban housing and other urban services have not been able to keep pace with the rapid Addressing the Lasting Effects of urbanization. Nine million housing units were built be- Expansionary Housing Policies tween 2000 and 2014, yet the national housing deficit 311. The explosive growth of large, low-density, single-use remained broadly stable in absolute terms at 9.2 million housing developments over the past two decades has fu- units in 2016.283 Moreover, the construction of new hous- eled urban sprawl. Mexico began to radically transform ing was not accompanied by commensurate investment its housing sector in the early 2000s. The supply of low- in mass transit, and an estimated US$70 billion in public cost housing increased by about one million units each transportation investment would be required to reduce year between 2006 and 2011, but as housing developers traffic congestion in Mexican cities to the average level sought to produce more housing units while keeping land of developed economies.284 Just 78 percent of all solid costs low, they increasingly built on rural tracts far from waste produced in Mexican cities is sent to final disposal city centers. Limited attention was paid to the overall sites, and only functionality and accessibility of new developments on the outskirts of cities that lacked proximity and access to 307. 11.5 percent of disposal sites comply with environmen- basic public services such as education and health. Long tal regulations.285 In Mexico City alone, about 15 percent distances to jobs in urban centers also pose challenges of the 13,000 tons of solid waste produced daily ends up in terms of commuting times, costs, and air pollution. in the streets, clogging drainage systems and contributing Moreover, the resulting patchwork of dispersed housing to floods. developments substantially increased the marginal cost of providing electricity, water, and other services, exacerbat- 308. Urban service provision remains inadequate even ing social exclusion. as municipalities spend a large share of their budgets on services. The ongoing urbanization process and the ex- 312. In 2013, housing policies rightly shifted to promote pansion of the middle class are increasing pressure on more compact and sustainable cities. To discourage municipal budgets, limiting the fiscal space for invest- low-density urban expansion, the revised rules of the new ment. Inadequate service provision and underinvestment housing policy scale the amount of direct support by lo- in infrastructure have constrained the ability of cities to cation type, with larger subsidies offered to “well-located” boost economic growth, foster social inclusion, raise liv- housing units in dense urban areas, close to employment ing standards, leverage economies of agglomeration, and and with access to basic services. This was a very positive promote environmental sustainability. shift. Nonetheless, it brings the challenge of affordability to fore as “well-located” housing tends to be relatively 309. The authorities have adopted policies designed to mit- expensive, and sometimes higher-cost units are ineligible igate climate change at the city level. Reforms have covered for the program. In 2017, only 9.8 percent of direct sup- multiple sectors, including: (i) designing a program to port was allocated to units in core urban areas, and 19.3 include location parameters in the definition of housing percent went to areas that were fully serviced by infra- subsidies; (ii) developing a federal program to support structure networks.286 This suggests that the greater direct the design, investment and implementation of Bus Rapid support offered to core urban units was unlikely to com- Transit (BRT) options in the largest cities; pensate for higher land values in city centers, prompting MEXICO POLICY NOTES 282 La Expansión de las Ciudades 1980-2010, SEDESOL, 2012 283 CONAVI, 2016. 124 284 RAS engagement, WB 2012. 285 Informe de la Situación del Medio Ambiente en México, SEMARNAT 2015. 286 CONAVI, 2016. developers to produce high-end unsubsidized units rather bating congestion. Between 2009 and 2014, the size of the Chapter 12 than affordable subsidized units. Consequently, increasing national vehicle fleet increased by 23.1 percent, from 30.8 the stock of well-located housing that is within the finan- million to 38 million vehicles, and it is expected to reach 70 cial reach of lower-income households remains among million by 2030. Meanwhile, households in the lowest in- the most critical challenges in the housing sector. come decile spend approximately 13 percent of their total income on public transportation. In some large cities, the 313. Despite the efforts to increase the supply of low-in- cost of transportation exceeds one-third of the minimum come housing, a substantial housing deficit remains, wage. Moreover, it is not uncommon for commuters to and much of the current housing stock does not meet use multiple types of transportation in a single trip, which formal construction standards. Over 10 million housing significantly increases the time and cost of commuting.289 units were built between 2000 and 2018, but a similar Urban sprawl and congestion have also led to an increase increase in the number of households left the housing in pollution emissions. deficit largely unchanged in nominal terms at about 9.2 million units.287 Nationwide, 5.5 million units do not 315. Public transit remains the primary mode of transporta- meet formal construction standards, in most cases be- tion in urban Mexico, accounting for 60 percent of total trips, cause they lack access to basic services. This is known yet low-quality bus services continue to make up 90 percent as the qualitative deficit. Another 3.7 million units are of public transportation, with the remaining 10 percent cov- overcrowded, located in high-risk areas, or in such ered by Bus Rapid Transit (BRT) options. Buses are respon- poor condition that they must be fully replaced. This is sible for approximately 55 percent of the 30 million daily known as the quantitative deficit. Formal annual pro- trips in the Mexico City metro area—making them one of duction of new housing units would need to increase the world’s largest bus-transit systems. Meanwhile, rapid by 70 percent to both meet new demand and eliminate bus lines, subways, and light rail systems in large cities still the quantitative deficit by 2030.288 represent less than 8 percent of total transportation. 316. Bus transportation in Mexico City is broadly represen- Coping with persistent Low- tative of public transportation in nearly every urban area Quality Public Transportation and of the country. Bus networks consist of numerous, largely Limited Mobility Alternatives informal private operators providing low-quality bus ser- 314. In many Mexican cities, employment is concentrated vices under vague contractual conditions. They rarely em- in the urban core, yet a large share of the population lives ploy formal labor and contribute little or no tax revenue. in peripheral areas, leading to intense traffic congestion. An The Bus business model is fundamentally inconsistent estimated 30 percent of residents of the greater Mexico with the provision of modern, clean, safe, and affordable City area spend over 120 minutes each day commuting, re- transit services. While per-trip fares are often low—e.g., ducing productivity by an estimated 3.3 million hours per US$0.23 in Mexico City—Buses are not integrated with day. Rising rates of private vehicle ownership are exacer- other transit services, and most trips require one or more Figure 54: Transportation Breakdown by City Type 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% Big Near the border North Center South Tourist National Average Type of cities Public Transportation Taxis Cars Source: Estimated data from questionnaires administered in 40 of Mexico’s 93 cities (Molinero, 2013). MEXICO POLICY NOTES 125 287 Data from the Registro Único de Vivienda. http://sniiv.conavi.gob.mx/oferta/cuboregistro.aspx. 288 Based on 2014 estimates from CONAVI. 289 INEGI, 2010; Corredor Insurgentes S.A. (CISA), 2012. Chapter 12 transfers, each with an additional full fare. Consequently, 319. Urban transportation alternatives have garnered lit- low-income households often spend a large share of their tle attention. While public transportation is primarily the income on transportation. In the greater Mexico City area, responsibility of state governments, alternative mobil- households in the lowest income quintile spend an aver- ity options such as non-motorized transportation (e.g., age of 18 percent of their income on transportation, and walking and cycling) are under the purview of municipal this share rises to 25 percent in the urban core. Informal governments. Limited normative and financial incentives Bus operators have limited access to capital and often ne- for cooperation among different levels of government, in- glect fleet maintenance or renewal, while the authorities cluding joint project planning, combined with inadequate rarely enforce regulatory standards. technical capacity at the local level have contributed to underinvestment in walkable areas, bike lanes, and other 317. Despite a number of important achievements, the Fed- forms of non-motorized transportation. Travel-demand eral Program to Support Mass Transit (Programa de Apoyo management initiatives, grants for cleaner technologies, Federal al Transporte Masivo, PROTRAM) is struggling to and transit-oriented development interventions remain keep pace with the demand for BRT projects. Launched limited. in 2008, PROTRAM is designed to mitigate the negative effects of increasing motorization, strengthen local gov- ernmental capacity to manage urban transportation, and Reverting the Neglect of Solid offer financial support to local governments to develop Waste Management mass-transit infrastructure. By providing financial and 320. Inadequate solid waste management in Mexican cities technical support to state and municipal governments, has major economic, environmental, and public health con- PROTRAM has contributed to important improvements sequences, yet it receives far less attention than other mu- in transportation quality, with positive economic, social, nicipal policy areas. In 2015, Mexican cities produced 53.1 and environmental implications. PROTRAM’s current na- million tons of solid waste, up 61.2 percent from 2003. In- tionwide portfolio comprises a pipeline of more than 46 dustrial development and urbanization are continuing to projects in different stages of development: identification, increase in solid waste output.290 According to the Secre- planning, evaluation, implementation, and operation. The tariat of Environment and Natural Resources (Secretaría del total cost of this portfolio is US$6.4 billion, of which PRO- Medio Ambiente y Recursos Naturales, SEMARNAT), the cost TRAM will finance US$1.2 billion directly, and the remain- of inadequate solid waste management is estimated to ing US$5.2 billion will be raised from private and public amount to 0.5 percent of GDP.291 This cost partially reflects sources. PROTRAM’s investments have not achieved their increased health expenditures and lost productivity result- full potential, and the financial sustainability of its projects ing from exposure to disease, and solid waste is a disease remains in doubt. Many PROTRAM-supported projects vector to which low-income households are particularly suffer from insufficient cashflow, and service quality of- exposed. Additional costs arise from water, soil, and air ten deteriorates after a few months of operation. Private pollution at improperly managed disposal sites, flooding equity investment is low, as investors are reluctant to take caused or exacerbated by waste-clogged drains, and lost on project risks. Furthermore, risk analyses are rarely con- revenue from the tourism sector and other economic ac- ducted during project preparation, making it difficult to tivities. In addition, inadequate solid waste management persuade private stakeholders that a project is viable, let is among Mexico’s five largest sources of greenhouse gas alone profitable. emissions. 318. Subnational transit authorities face governance 321. All aspects of solid waste management require at- challenges, institutional capacity constraints, fiscal lim- tention and improvement. The 2003 National Law on itations, jurisdictional overlaps, and an ambiguous rela- Solid Waste Management and the 2008-2012 National tionship with the federal government. Subnational gov- Program for Prevention and Integrated Solid Waste Man- ernments often lack sufficient financial resources, which agement established standards and principles for solid leads them to use public-private partnership schemes waste management at the federal, state, and local levels. as a funding mechanism rather than a value- for-money However, solid waste management remains highly inef- strategy to leverage private equity and project revenues ficient. Mexico currently recycles just 10 percent of the to finance infrastructure development. This focus on 120 million tons of waste it produces each year, while 78 funding is problematic, because subnational govern- percent is sent for final disposal, and the destination of ments do not prioritize risk allocation. Properly identify- the remaining ing design, social, political and regulatory, commercial, operational, and fiscal risks is the first step to achieving 322. 12 percent is unknown. Only 11.5 percent of final dis- bankability and improving the private sector’s willing- posal sites comply with Mexico’s Environmental Law, and ness to participate in transportation projects that ad- an average of just 84 percent of municipal waste is collect- vance social policy objectives. ed. Among the governments of Mexico’s 31 states and the MEXICO POLICY NOTES 126 290 Informe de la Situación del Medio Ambiente en México, SEMARNAT 2015. 291 SEMARNAT, 2014. Figure 55: Solid Waste Production by State, 2013 (tons) Chapter 12 Generación de RSU (miles de t) 228 - 500 501 - 1,000 1,001 - 1,500 1,501 - 2,000 2,001 - 6,000 6,001 - 6,798 Source: Dirección General de Equipamiento e Infraestructura en Zonas Urbano-Marginadas, Sedesol, México. 2013 Federal District, only 12 perform selective collection292 of 325. Municipalities are primarily responsible for solid waste municipal waste. Most of Mexico’s solid waste manage- management, but they often lack the knowledge, tech- ment indicators are low by the standards of comparable nical capacity, resources, and political will to address key countries in the region, and all are well below the average challenges. Mayors serve a single three-year term, which for OECD member states. limits the types of support and collaboration that can be pursued by federal and municipal governments. In prin- 323. At the federal level, a lack of interinstitutional coor- ciple, states enforce municipal compliance with environ- dination inhibits solid waste management. SEMARNAT is mental laws and encourage cooperation among adjacent responsible for establishing the policy framework and municipalities on environmental and waste management channeling federal resources to states and municipalities. issues, yet in practice both regulatory enforcement and in- However, SEMARNAT faces numerous challenges, includ- ter-municipal cooperation are rare. And while 65 percent ing: (i) a lack of staff; (ii) inadequate resources from the of Mexican states have formulated regulatory frameworks federal government; (iii) weak technical capacity; (iv) the for solid waste management, these frameworks are not low priority accorded to solid waste management at the standardized, and compliance is weak. In consultations national level; and (v) limited collaboration with state and with World Bank staff, SEMARNAT has indicated that a municipal governments on solid waste management is- lack of reliable information from states and municipalities sues. affects its ability to assess the current situation and de- velop a clear strategy and policies for the sector. Finally, 324. The National Bank for Public Works and Services inadequate cost-recovery schemes (compared with other (Banco Nacional de Obras y Servicios Públicos, BANO- sectors) limit local investment, leaving solid waste man- BRAS) administers the National Infrastructure Fund (Fondo agement projects highly dependent on federal technical Nacional de Infraestructura, FONADIN), which provides and financial support. funding for municipal solid waste infrastructure through its Solid Waste Program (Programa de Residuos Sólidos, PRORESOL). However, since its inception in 2002, PRORE- Addressing Inefficient and Costly SOL has financed just three municipal projects due to: Energy Use at the Municipal Level (i) complex and cumbersome internal regulations and 326. Energy inefficiency constrains productivity and con- approval criteria that have not been specifically adapt- tributes to environmental degradation. From 2000 to 2015, ed to the solid waste management sector; (ii) a lack of Mexico’s energy consumption grew by 30 percent, driven qualified technical staff; and (iii) weak collaboration with by transportation, industry, and the residential sector. SEMARNAT in leveraging complementary financing. Electricity consumption by end users increased by 60 per- PRORESOL also lacks a strategy for social inclusion and a cent between 2000 and 2015.293 The transportation sector plan to promote recycling, reutilization, and waste min- relies heavily on fossil fuels and is responsible for 46 per- imization, and it has been unsuccessful at forming part- cent of end consumption. Given the projected growth of nerships with industrial, commercial, or tourism-related the transportation sector, transitioning to a cleaner energy firms. Consequently, although about US$100 million has mix will require the adoption of new technologies to re- been invested in the sector over the last two years, little duce reliance on fossil fuels. The industrial sector accounts MEXICO progress has been achieved. for 30 percent of end consumption. It consumes electricity POLICY NOTES 292 Selective collection refers to the mandatory separation of organic and inorganic. 293 SENER. 2016. Estrategia de Transicion para Promover el Uso de Tecnología y Combustibles más Limpios. 127 Chapter 12 Figure 56: Final Energy Consumption, Baseline Scenario versus Transition to Energy-Efficient Technologies, 2010-2050 (petajoules) 8000 7000 6000 5000 Petajoules 4000 3000 2000 1000 0 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032 2033 2034 2035 2036 2037 2038 2039 2040 2041 2042 2043 2044 2045 2046 2047 2048 2049 2050 Escenario de Base Escenario de Transición Source: SENER, CONUEE and coke, and the most energy-intensive subsectors are by 78 percent. Replacing existing technology with more iron, steel, and cement production. The residential and energy-efficient alternatives could reduce the amount commercial sectors together account for 19 percent of of electricity consumed by street lighting alone by 36 to end consumption, and the residential sector consumes 74 percent.295 As cities in Mexico continue to experience natural gas, biomass, and electricity. rapid demographic and economic growth, municipalities will face increased pressure to expand high-quality and 327. Mexico’s energy consumption is expected to double by affordable public services. Efforts to improve energy effi- 2050. As the country continues to urbanize, it will need to ciency at the municipal level have thus far yielded limited build more buildings, provide more services, and transport results. These efforts have focused mainly on retrofits to more people and goods in the coming decades. Energy street-lighting and water-pumping systems. The Nation- should be an integral part of long-term transportation and al Commission for the Efficient Use of Energy (Comisión urban development policies, and policymakers should Nacional para el Uso Eficiente de la Energía, CONUEE) and strive to avoid committing to investments in inefficient BANOBRAS have supported a municipal street-lighting fossil-fuel-based transportation and energy-intensive ur- retrofit program since 2010. The Secretariat of Energy (Sec- ban development models. Measures to increase energy retaría de Energía, SENER) provided direct support of up to efficiency and promote electric and nonmotorized trans- 15 percent of the total investment cost to 27 street-light- portation would help Mexico reach its target of reducing ing subprojects, generating an average energy savings of energy consumption by 40 percent by 2050. 38 percent.296 328. Investing in energy efficiency could significantly re- 329. Municipalities implementing energy-efficiency mea- duce municipal energy expenditures and open fiscal space sures, particularly street-lighting and water-pumping proj- to expand public service provision. Municipal service pro- ects, face numerous challenges. These include: (i) a lack of viders have many opportunities to adopt new technolo- information regarding the available technologies, their gies for street lighting, interior lighting, air conditioning, prospective benefits, and the methods of financing and water pumping and heating, inter alia,294 as well as test implementing them; (ii) limited capacity to identify, de- operational and financial mechanisms for energy efficient sign, finance, implement, monitor, and verify savings from upgrading. After salaries, Mexican municipalities’ largest energy-efficiency projects, due in part to the absence of expenditures are on street lighting, water services, and metering for most street-lighting infrastructure; (iii) the wastewater treatment. In 2014, street lighting account- short time horizon of municipal governments and the me- ed for 58 percent of total public electricity consumption dium-to-long term nature of energy-efficiency gains; (iv) at the municipal level, while water-pumping and related restrictive budgeting procedures that prevent municipal activities comprised the remaining 42 percent. These sub- governments from retaining operational cost savings to sectors have grown substantially since 2002: electricity make debt-service payments in subsequent budget years; consumption for street lighting increased by 32 percent, (v) a procurement process that focuses on upfront costs MEXICO and electricity consumption for water services increased rather than lifecycle costs; and (vi) limited access to financ- POLICY NOTES 128 294 Ibid. 295 National Energy Efficiency Program, Balance 2010-2016, CONUEE. 296 Ibid. ing, especially among marginally creditworthy municipali- oratively explore innovative approaches combining direct Chapter 12 ties, as well as constrained debt capacity and difficulty ob- support with more efficient urban regulations to create taining long- term credit to cover initial investment costs. more affordable units in urban centers. For example, cre- The Government is piloting an innovative operational and ating denser mixed- use developments could spread high financing mechanism to support energy efficiency at the real estate costs across a greater number of units and municipal level. The pilot program (PRESEM) covers pub- households at different income levels, boosting the supply lic facilities, including street-lighting networks, municipal of affordable housing in Mexican cities. buildings, water utilities, public schools, and public hos- pitals and is designed to address key barriers to energy ef- 333. Support the development of robust urban-planning ficiency, foster the growth of markets for energy-efficient institutions at the local level. The benefits of increased technology, provide budgetary relief to public entities, population density can only be realized if local institu- enhance the quality of energy services, and contribute to tions have the capacity to deliver services effectively and meeting Mexico’s environmental goals297. to allocate or attract the necessary investment to serve a denser population. Capacity constraints and limited re- sources at the state and municipal levels have restricted POLICY OPTIONS local urban-development and land-use planning to the preparation of specific investment projects and the cre- ation of intricate land-use regulations. Moreover, these Renewing Government Efforts to limited plans and regulations are prepared without a com- Promote Compact and Inclusive prehensive assessment of population growth, demand for Urban Development housing and basic services, land-use and pricing trends, 330. The government should continue and strengthen its growth corridors, social issues, or the government’s own policies to provide differentiated up- front housing subsidies implementation capacity. The federal government can depending on location. Supporting social housing using a build the planning capacity of local authorities by offer- larger share of its existing envelope of subsidies in strate- ing incentives to create long-term strategic visions for city gic locations could help bolster the supply of affordable development and to better integrate land-use planning, housing in the core of urban centers and reduce pressure housing development, and transportation investment. to further expand housing developments on the urban Moreover, well-functioning municipal cadastral systems periphery. (including fiscal cadasters) will be necessary to raise more local revenue and face the growing local service delivery 331. Targeted reforms could expand CONAVI´s housing pro- pressures (see Policy Note on Subnational Finances). gram to a larger share of the population. These include: (i) diversifying available housing solutions by enabling eligi- Further Strengthening ble households to acquire new or existing units, improve Transportation and Mobility their houses, purchase serviced plots, self- construct, or Policies and Institutions rent; and (ii) adjusting program parameters to focus ben- efits on the groups that are most in need of housing sup- 334. Strengthen the regulatory and institutional framework port. that governs bus services and introduce financial instru- ments to support fleet renewal. Most Mexican cities have 332. CONAVI and the Secretariat of Agrarian, Land, and declared that replacing old transit vehicles is an important Urban Development (Secretaría de Desarrollo Agrario, Ter- policy goal. Achieving this goal will require: (i) an analysis ritorial y Urbano, SEDATU) could partner with municipali- of alternative business models, including private-sector ties to reduce regulatory constraints and strengthen urban debt financing, the use of special purpose financial vehi- planning at the local level. Demand-side support should be cles, and project financing, and (ii) an analysis of the reg- complemented by regulatory reforms and fiscal and land- ulatory framework for formal bus-service providers and use incentives. CONAVI is responsible for setting Mexico’s the range of available policy tools and incentives, includ- housing policy, while local governments are tasked with ing travel-demand management and land-value capture formulating and implementing urban development plans mechanisms. and other land-use strategies. Local governments can designate areas suitable for housing development and 335. Revamp PROTRAM by reviewing its internal processes regulate housing standards, densities, zones and uses, and structure. Internal procedures for approving federal permits and licenses, and parking rules, inter alia. These support for mass-transit interventions should formalize regulations directly impact developer costs and housing the role of PROTRAM’S decision-making body, the Consul- affordability. CONAVI and local governments could collab- tative Working Group, and consider ways to involve other MEXICO POLICY NOTES 297 PRESEM is a World Bank financed project. It supports an energy-efficiency fund operated by the Electricity Energy Savings Trust Fund (Fideicomiso para el Ahorro de Energía Eléctrica, FIDE). Under the program, FIDE, SEDER, and grant beneficiaries sign energy-savings agreements for eligible investment projects. PRESEM’s methodology is designed to account 129 for the limited technical capacity and financial constraints of prospective investors. In December 2017, the bidding process was successfully completed, and the first contract was awarded for a street-lighting project in the Municipality of Leon. Competition during the tendering process resulted in a final cost that was 43 percent lower than originally estimated based on list prices, revealing the considerable gains in value-for-money offered by market-based mechanisms for promoting energy-efficient investments. Chapter 12 ministries in its deliberative process.298 PROTRAM’s man- a robust technical and administrative team; (iii) creating agement unit also requires additional human resources a joint operating platform with FONADIN/BANOBRAS to to handle its large and growing agenda and to monitor prioritize and finance municipal solid waste management project implementation in detail. Investing in tools to projects; (iv) establishing a national commission for sol- gather and analyze transportation data could advance the id waste management comprising representatives from decision-making process for federal support. SEDESOL, SEDATU, FONADIN, and other key agencies; and (v) strengthening the role of the Federal Prosecutor for En- 336. PROTRAM’s funding should be increased to cover the vironmental Protection (Procuraduría Federal de Protección program’s current and future scope. The inclusion of me- al Ambiente, PROFEPA) in enforcing solid waste manage- dium-sized cities will add at least 20 cities to PROTRAM’s ment regulations at the state and local level and across current pipeline. However, the growing needs of FONA- both the public and private sectors. BANOBRAS could DIN’s other supported sectors may reduce the resources enhance its effectiveness by: (i) streamlining internal pro- available for PROTRAM. In this context, PROTRAM should cedures for approving solid waste management projects; explore alternative sources of funding to support trans- (ii) designing internal eligibility criteria that differ from the portation projects in medium-sized cities. criteria used for large transportation and water projects; (iii) adopting flexible financing schemes for private-sector 337. Strengthen the capacity of subnational governments participation; and (iv) developing a joint operating plat- to improve mobility. The federal government should pro- form with SEMARNAT to prioritize and finance municipal vide financial support and technical assistance to build solid waste management projects. Meanwhile, the federal the planning and coordination capacity of state govern- government should prioritize solid waste management ments. This measure will increase the prospects of success and allocate the resources necessary to address current of a National Urban Mobility Plan. It would also allow state and future challenges. governments to coordinate with federal and municipal entities to promote housing policies and programs that in- 339. State governments could take a more active role in corporate access to both mass transit and rapid transpor- promoting, financing, and supporting solid waste manage- tation services to reduce neighborhood segregation, pro- ment projects in partnership with municipal authorities. mote equitable growth, alleviate pollution, and boost pro- To collaborate effectively, states need to strengthen their ductivity in Mexican cities. At municipal level there should capacity to promote, co-finance, and oversee solid waste be a greater focus on Transit Oriented Design (TOD) poli- management projects. In addition, municipal govern- cies to promote the creation of compact, walkable, pedes- ments will need to: (i) more accurately assess the costs trian-oriented, mixed-use communities centered around and financial implications of solid waste management mass transit stations. This would increase the efficiency of systems and cost-recovery schemes; (ii) strengthen their transportation, reduce the need for long commutes, and technical teams and build the overall capacity of their hu- improve quality of life. Attention at the local level should man resources; (iii) take advantage of new technological also cover travel demand management initiatives, in which and operating approaches to solid waste management; the private sector plays a very important role. Some of and (iv) prioritize solid waste management as a matter of these initiatives are: alternative work schedules, telecom- municipal public policy. muting options, car sharing, bus transport for employees, and company incentives to use alternative transportation 340. Better equipment and infrastructure will be needed to systems, among others. The use of technology to improve achieve SEMARNAT’s emissions- reduction targets. Invest- sustainability and efficiency in transport is also essential. ment should focus on the construction or rehabilitation of Some of the latest developments in intelligent transport final disposal facilities, the closure of existing open dumps, technologies include: semi-autonomous vehicles, techno- and the installation of biogas-capture technologies and plogy enabled models of taxi services, mobile applications waste-to-energy equipment. Funding should be provided to support shared us mobility, enhanced traveler informa- to selected municipalities to develop efficient integrated tion applications, adaptive traffic signal control, dynamic solid waste management systems that include transfer and lane reversal, etc. separation stations, containerization, recycling equipment, composting plants, and collection systems, inter alia. Overhauling Solid Waste 341. Boosting the participation of the private sector in sol- Management id waste management through public-private partnerships 338. Effectively managing the large and growing volume could augment the limited resources of the federal, state, of solid waste produced by Mexican cities will require in- and municipal governments. There is a great opportunity stitutional capacity-building at all levels of government. for PPPs in solid waste management as several municipali- SEMARNAT can strengthen its capacity for solid waste ties are already working with private investors to structure management by: (i) fully assessing the costs associated projects. Encouraging greater private-sector engagement MEXICO POLICY NOTES with improper solid waste management; (ii) establishing will require: (i) creating an open and fair platform for na- 130 298 For example, the Ministry of Energy could be engaged in future electro-mobility initiatives. tional and international companies to compete for solid level would help shorten payback periods and encourage Chapter 12 waste management projects;299 (ii) fostering an enabling investment in more sophisticated (and costly) forms of en- environment for public-private partnerships; and (iii) ergy-efficient infrastructure and technology. incentivizing private firms to introduce innovative oper- ational systems and technologies for solid waste manage- 345. At the federal level, a new publicly financed energy-ef- ment. ficiency program could help overcome the challenges faced by current initiatives, demonstrate the benefits of energy 342. Behavioral change is needed to improve solid waste efficiency, and help develop a market for private invest- management at all three levels of government. Public ment in energy-efficient public works. This program could outreach campaigns promoted in partnership with civil incorporate representatives from SENER, the Federal Elec- society groups and NGOs can both facilitate the transi- tricity Commission (Comisión Federal de Electricidad, CFE), tion to improved solid waste management systems and CONUEE, and FIDE. It could focus on improving the institu- publicize their benefits. Cross-cutting issues such as recy- tional framework for energy-efficient investment, increas- cling, waste minimization, waste treatment, and e- waste ing transparency, strengthening procurement processes, management, as well as cost-recovery mechanisms, could and creating financial systems that verify repayment be incorporated into a national plan to comprehensively through achieved savings. The program’s goals could improve solid waste management. include: (i) establishing clear rules for determining ener- gy- consumption baselines; (ii) testing project budgeting and financing systems; (iii) enhancing technical capacity Leveraging the Power of Local for procurement at the federal and local levels, (iv) im- Governments to Create More proving systems to monitor and verify energy savings; and Energy-Efficient Cities (v) building a credible record of municipal repayments. 343. Strengthen the capacity of local governments to create Facilitating a sufficient number of projects to enable the an enabling environment for investment in energy-efficient private sector to accurately assess costs, risk profiles, ex- infrastructure. The federal government should collaborate pected savings, and contractual mechanisms could enable with local authorities to build their technical and financial the development of a dynamic and sustainable market for capacity to identify, vet, manage, monitor, evaluate, and energy-efficient investment. expand energy-efficiency projects. Collecting data on energy consumption by sector and assessing energy effi- 346. Build private-sector confidence in energy-efficient in- ciency at the municipal level will help local governments vestment. The government can draw on the experience understand the costs and benefits of alternative projects of the municipal energy efficiency pilots as they strive to and policies, develop sound project-vetting criteria, and raise awareness of the benefits of energy-efficient invest- prepare a robust project pipeline. ment, balance the interests of multiple stakeholders, and engage the private sector in expanding energy-efficient 344. The federal government should provide non-reim- infrastructure nationwide. As Mexico is widely regarded bursable direct support to subnational authorities to incen- as a leader in economic development policy, especially tivize investment in energy-efficient projects. The policy in Latin America, the lessons and insights generated by discontinuity caused by the rapid turnover of municipal the pilots could inform the design and implementation of administrations is a major barrier to implementing ener- similar mechanisms for scaling up energy-efficient invest- gy-efficient investments. Direct support from the federal ment across the region. MEXICO POLICY NOTES 299 Currently, four to five national firms receive the contracts for all solid waste management projects 131 Chapter 12 References: INEGI. 2016. Encuesta Nacional Ingreso y Gasto de Hogares 2016. Mexi- enda 2013–2018. Mexico City, Mexico: SEDATU. http://www. co City, Mexico. Mexico Informal Transit Reform Support (P156729) economia.unam.mx/cedrus/descargas/PNDUyV_PNDUV OECD. 2015. OECD Urban Policy Reviews: Mexico—Transforming _Corregido.pdf Urban Policy and Housing Finance. Paris: OECD. SENERMEX. 2013. Proyecto de Transporte Masivo de Pasajeros en PAD Improving Access to Affordable Housing Project (P157932) la Modalidad de Tren Ligero entre los Municipios de Zapopan, PAD Municipal Energy Efficiency Project (P149872) Guadalajara y Tlaquepaque, Jalisco. México. PAD Reform of the Federal Support Program for Mass Transit World Bank, 2016. Mexico Urbanization Review- Managing Spatial (P165724) growth for Productive and Livable Cities in Mexico PCN National Integrated Solid Waste Management (P154801) SEDESOL and CONAPO (Consejo Nacional de Población). 2012. SEDATU (Secretaria de Desarrollo Agrario, Territorial y Urbano). Sistema Urbano Nacional. Mexico City, Mexico: SEDESOL. 2013. Programas Nacionales de Desarrollo Urbano y de Vivi- MEXICO POLICY NOTES 132 13. Further strengthening fiscal sustainability with equity F iscal policy has been a core pillar to maintain macroeconomic stability in Mexico. In the context of the fall in commodity prices, the country pursued prudent policies, including through a comprehensive tax reform in 2013 and measures to contain and enhance the effi- ciency of public spending taken between 2015 and 2017. Mexico’s fiscal stance remains strong and sustainable, but it faces secular pressures in non-discretionary spending driven by a combination of demographic change, policy commitments and systemic budgetary rigidities. As population’s aspirations rise, there are higher pressures on public service delivery. Moreover, higher growth rates over the medium term require a higher level of public infrastructure investment (comple- mented with private resources). To open the fiscal space for these pressures, efforts will need to continue to identify technical and allocative efficiencies and fiscal savings on the spending side, while also planning the next wave of tax reforms. Tax collection in Mexico remains below the LAC regional average and is the lowest among OECD countries. Even after the highly positive 2013 reforms, extensive tax expenditures (e.g., exemptions, zero rates, deductions) continue to impose major fiscal costs and could be significantly reduced while guarding for equity impacts. There is a growing digital economy that remains untaxed, presenting an opportunity for a new revenue base that did not exist years ago. Fiscal savings on the spending side can be harnessed while improving efficiency and not harming equity in public sector procurement and by streamlining overlapping transfer programs. There are also efficiency gains in reducing fragmentation in the health sector. Efficiency in managing public infrastructure is key as more fiscal space is opened for this needed MEXICO outlay. More broadly, significant progress has been made in establishing a solid fiscal responsi- POLICY NOTES bility framework, though enhancement to that framework coupled with the establishment of an independent fiscal council could signal a continued strong commitment to fiscal discipline. 135 Chapter 13 spending pressures and create the fiscal space necessary CONTEXT AND RECENT to finance additional public investment while reducing DEVELOPMENTS the public debt-to-GDP ratio. A subsequent sharp drop in global oil prices as of the second half of 2014 and a 347. Fiscal policy has been a core pillar to maintain mac- continued decline in the volume of oil production led to roeconomic stability in Mexico. The fiscal authorities have falling oil revenue largely offsetting the significant rise in been successful in accommodating public expenditures non-oil revenue. to available revenue. Robust performance of public revenue, combined with moderate additional deficit 349. Measures to contain and enhance the efficiency financing, enabled the government to substantially in- of public spending were taken between 2015 and 2017. crease public expenditures over the past 15 years (Fig- These measures included a significant containment of ure 57). Following the global financial crisis, a widening the public-sector wage bill and a consolidation in federal gap between revenues and expenditures increased the programs for current spending subsidies and transfers. public debt-to-GDP ratio. More recently, falling oil reve- These public spending adjustments in response to fall- nue required significant fiscal adjustments, which have ing oil revenue present a valuable opportunity to lock-in been applied adequately: (i) by fully implementing a tax efficiency improvements. Despite the needed measures, reform designed in 2013 to keep up with growing spend- the consolidation also relied heavily on cuts to public ing pressures; and (ii) by taking measures subsequently investment, with potentially negative implications for to control and reduce spending. The cumulative effect of medium-term growth. Public debt in Mexico started to fiscal consolidation enabled the government to achieve stabilize and more recently to decline, leading the way a primary surplus in 2017 and to set the debt-to-GDP among emergingand developing economies. All this oc- ratio on a declining path. These actions permitted the curred while the economy continued to grow close to its country to weather a difficult time for the Latin American average pace. region and many commodity-exporting emerging mar- ket economies (Figure 58). 350. Mexico’s fiscal stance remains strong and sustainable, but managing secular public spending pressures and signif- 348. The comprehensive tax reform package of 2013, icant investment needs will pose an increasingly complex helped to boost non-oil tax revenue by more than 3 per- challenge over the medium term. Mexico faces a secular centage points of GDP. Successive governments revised increase in non-discretionary and inflexible expenditures the tax regime in 2007, 2009, and 2013, eliminating driven by a combination of demographic change, policy exemptions, introducing new taxes, and increasing tax commitments and systemic budgetary rigidities. As popu- rates. These changes, coupled with improvements in tax lation’s aspirations rise, there are higher pressures on pub- administration, significantly boosted tax revenue. The lic service delivery. Moreover, to support higher growth relatively steady growth of tax revenue stands in contrast rates over the medium term, a significantly higher level of to the volatility and secular decline of oil revenue. The public infrastructure investment (complemented with pri- share of non-oil tax revenue in the federal public-sector vate resources) will be needed. To open the fiscal space for budget increased from 42 percent in 2008 to 58 percent these pressures, efforts will need to continue to identify in 2017, while oil revenue decreased from 37 percent technical and allocative efficiencies and fiscal savings on to 17 percent over the same period. The 2013 reforms the spending side, while also planning the next wave of were designed to raise revenue to address rising public tax reforms. Figure 57: Federal Public-Sector Revenue and Expenditure and Federal Public-Sector Debt, 2000-2017 (% of GDP) 28% 60% 26% 50% 24% 22% 40% 20% 30% 18% MEXICO POLICY NOTES 16% 20% 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 Expenditure Revenue Net debt Gross debt 136 Source: Bank staff estimates based on SHCP, INEGI and IMF Figure 58: General Government Fiscal Balances and Gross Government Public Debt, 2012-2017 Chapter 13 (% of GDP) 4 65 2 60 0 55 -2 50 -4 45 -6 40 -8 35 2012 2013 2014 2015 2016 2017 2012 2013 2014 2015 2016 2017 Mexico, fiscal balance LAC, fiscal balance Mexico LAC Emerging and developing, fiscal balance Emerging and developing economies Source: IMF, World Bank estimates 353. Moreover, Mexico’s tax structure relies heavily on income KEY CHALLENGES taxes compared to relevant peers. Income tax revenue makes up nearly 42 percent of Mexico’s total tax revenue, well Domestic Revenue Mobilization. above the averages for both the LAC region and the OECD. By contrast, Mexico derives less than 40 percent of its rev- 351. In 2013, a comprehensive tax-reform program significant- enue from indirect taxes, whereas these taxes account for ly increased nonoil tax revenue. The reforms eliminated tax over half of revenues in LAC countries (Figure 58). Mexico’s benefits and preferential tax regimes, limited tax deductions, tax structure reflects differences in its revenue-generating introduced dividend and capital gains taxes, raised the top capacity and has economic and distributional implications. marginal income tax rate, and replaced the tax regime for For example, due to broad exemptions and zero-rating in small businesses. The reforms also eliminated a preferential the VAT regime, Mexico collects only 31.5 percent of the rev- VAT rate that had been applied in border regions. They intro- enue that it could theoretically collect if VAT was applied at duced new excise taxes on carbon, sugar-based beverages the standard rate to all goods and services. By contrast, this and high-calorie foods. The negative excise tax on domestic VAT Revenue Ratio is 42.6 percent in Colombia, 55.1 percent fuel sales was phased out in the context of low international in Peru and 64.4 percent in Chile. Furthermore, Mexico’s oil prices and energy-sector reforms. The federal govern- personal income tax reduces gross income inequality more ment’s non-oil tax revenue rose from 10 percent of GDP in than any other system in Latin America.300 2013 to 13.1 percent in 2017. Rising income tax collection contributed 1.4 percentage points of GDP to the increase 354. Even after the positive 2013 reforms, extensive tax-ex- in non-oil tax revenue, fuel excise taxes contributed 1.0 per- penditures continue to impose major fiscal costs. Tax ex- centage point, VAT contributed 0.3 percentage points, and penditures include exemptions, deductions, deferrals, and other excise taxes contributed the remaining 0.2 percentage preferential rates applied to specific activities or types of points. The total increase in non-oil tax revenue largely offset taxpayers. Forgone revenue through tax expenditures is a decline in oil revenue equal to 4 percentage points of GDP. estimated at 3.7 percent of GDP. Tax expenditures are in- curred, inter alia, through exemptions and zero-rating in 352. Nevertheless, tax collection in Mexico remains below the VAT regime (1.5 percent), since 2017, through a dis- the regional average and is the lowest among OECD coun- count to the excise tax on fuel (0.7 percent), exemptions tries. International comparisons of tax collection employ a to wage income (0.4 percent of GDP) and pension income broad concept of tax revenue, which includes social secu- (0.2 percent), a negative income tax designed to encour- rity contributions and subnational taxes. Under this defi- age formal employment (0.2 percent). nition, Mexico’s tax-to-GDP ratio increased from 13.4 per- cent in 2010 to 17.2 percent in 2016. However, during this 355. In addition, tax evasion continues to inflict significant period average tax-to-GDP ratios also increased among revenue losses, although they have been declining in recent other countries in the region, and Mexico’s tax burden re- years. These losses were estimated to be close to 2.5 per- mains below the Latin America and the Caribbean (LAC) cent of GDP in 2016, of which income tax evasion amount- regional average of 22.7 percent. Mexico’s tax-to-GDP ratio ed to 1.5 percent of GDP and VAT evasion amounted to is also the lowest among OECD countries, amounting to MEXICO 1.0 percent. However, tax evasion appears to have fallen just half the OECD average of 34.3 percent. POLICY NOTES 300 OECD et al. (2018) 137 Chapter 13 Figure 59: Tax Structures in Mexico, Latin America and the Caribbean, and the OECD, 2016 % of GDP % of total tax revenue 35 100 2.9 5.8 6.3 7.7 30 4.2 14.7 12.4 21.2 80 25 20.0 6.7 23.9 1.6 60 20 29.3 1.2 5.2 9.0 13.9 25.8 15 2.4 6.2 40 4.1 15.9 10 2.2 3.7 11.5 20 41.7 5 34.1 7.3 6.0 27.3 0 0 Mexico LAC OECD Mexico LAC OECD Other taxes Other goods and services VAT Social security Income Source: OECD et al. (2018) significantly over the past few years, likely due to improve- and services to final consumers in Mexico. Coupled with ments in tax administration. rapid increases in fixed internet and mobile broadband penetration, these market dynamics mean that more and 356. Liberalization of domestic fuel prices facilitates a more consumers are in a position to reap the benefits of change to a more efficient fixed excise tax and lock in reve- the digital economy. Typically, digital services consumed nue levels. Prior to 2017, when the government liberalized in Mexico are subject to VAT, whether provided by a for- the retail fuel sales market, the domestic sale prices for eign or domestic supplier, with the obligation to collect gasoline and diesel were administratively determined. As and remit the VAT on the seller or the resident importer the final sales prices were fixed, any change in the under- in the case of purchases from abroad. Since in the case of lying international reference prices was reflected in a vari- digital services there may be no business need for foreign ation in the (implicit) excise tax. High oil prices between suppliers to establish a physical or legal presence in the 2006 and 2014 effectively led to a negative fuel excise country, there is not necessarily an entity that charges tax, whereas a subsequent drop in international oil prices and remits the VAT owed on digital services imported by turned the tax positive again, generating average revenue final consumers, and that revenue is effectively foregone equal to 1.3 percent of GDP in 2015 and 2016. Upon the by the Mexican authorities. Moreover, since Mexican- res- gradual liberalization of the fuel retail sales market during ident firms selling digital services in the domestic market 2017, the variable fuel excise tax was replaced by a fixed are obliged to collect VAT on their sales, they may be at a per-unit excise tax. To moderate the initial effect and to pricing disadvantage vis-à-vis non-resident competitors. smooth the impact of variations in international reference prices on retail sales prices, the authorities adjusted fuel excise taxes on a weekly basis in 2017 and 2018 by apply- Expenditure policy ing a variable discount to the excise tax, which resulted in a costly tax expenditure of 0.7 percent of GDP in 2017. 358. Mexico’s federal public-sector expenditures have in- creased substantially over the years, and pressures will con- 357. Technological progress poses new challenges to in- tinue to mount. The country experienced a secular increase clude a growing digital economy in the country’s tax base. in public spending driven by a combination of policy de- Following global trends, Mexico’s digital economy is grow- cisions, demographic trends, legacies, and other spending ing rapidly, supporting economic development more rigidities. This has resulted in a decade-long rise in health, broadly. Although from a low base, between 2010 and education, social protection, and public security spending. 2016, the share of the adult population that had ordered The number of people aged 65 and older will more than goods or services online increased by a factor of 4.5, the double to 18.4 million over the next two decades and pen- largest such increase of all OECD countries. Meanwhile, sion costs on the federal public-sector budget, due to this it has been estimated that Mexico’s business-to-consum- increase in the number of elderly combined with the favor- er (B2C) e-commerce market could grow to $40.8bn by able treatment of this cohort in past pension reform, is pro- 2019, of which nearly two thirds would consist of digital jected to increase by about 1 percent of GDP per decade services.301 New market entrants armed with disruptive before ultimately peaking between 2040 and 2050. Popu- MEXICO technologies can increase the choice, improve the qual- lation aging is also intensifying pressure on Mexico’s public POLICY NOTES ity, and reduce the price of inputs to firms and of goods healthcare system by shifting the disease burden to chron- 138 301 2016: The Tipping Point for E-Commerce in Mexico, A.T. Kearney, 2016 Table 2 : Public Expenditure, 2014-2017 (% of set an explicit annual target for the Public-Sector Borrow- Chapter 13 GDP) ing Requirement (PSBR), which is more comprehensive measure of the federal public- sector balance; (ii) capped 2014 2017 change current expenditure growth to avoid a procyclical expan- Total 25.9% 23.8% -2.1% sion of public spending in times of revenue buoyancy; and (iii) specified the amount of resources to be transferred Wages and salaries 5.8% 5.3% -0.6% annually to the budget from a newly created sovereign Other operating costs 5.8% 5.7% -0.1% wealth fund, the Mexican Petroleum Fund (Fondo Mexica- no de Petróleo, FMP). Between 2015 and 2017, the govern- Subsidy and transfer programs 3.8% 3.1% -0.7% ment met its annual targets for the PSBR and respected Physical investment 4.7% 2.6% -2.1% the cap on current expenditures. Nevertheless, use of the escape clause to the balanced-budget rule in 2014302 Other capital expenditures 0.4% 1.0% 0.6% pushed the PSBR targets for 2015 and 2016 to levels that Participaciones 3.3% 3.5% 0.2% were incompatible with reducing the debt-to-GDP ratio. Interest payments 2.0% 2.5% 0.5% Moreover, the frequent use of this clause may indicate that the framework has some limitations enabling a count- Source: Bank staff estimates based on SHCP and INEGI er-cyclical stance in the future. ic degenerative diseases, with population aging account- ing for nearly 60 percent of the increase in public health spending to 3.1 percent of GDP in 2015 and an increase by POLICY OPTIONS 0.5 percent of GDP over the previous decade. These long- term trends underscore the importance of improving the efficiency and effectiveness of public expenditures. Tax Policy 362. Revenue-enhancing tax reforms could ease fiscal pres- 359. Expenditure rationalization over the past few years sures, finance new government programs, and address infra- took place across several categories, but it hit hard public structure gaps as well as competitiveness, productivity and investment. Spending containment and cuts took place equity challenges. on subsidy, transfer programs, and the public-sector wage bill. However, general government public investment (in- 363. Reducing tax expenditures (i.e., eliminating exemp- cluding federal investment projects and transfers for state tions, deductions, and preferential rates in the income and municipal investments) was affected significantly, tax and VAT regimes) would broaden the tax base and dropping from 4.7 percent of GDP in 2014 to 2.6 percent increase revenue collection. The government could use in 2017 (Table 1). These cuts also included PEMEX capital the additional revenue to ease fiscal pressures, finance expenditures, potentially affecting its future production additional spending (including compensatory programs), capacity. lower statutory tax rates, and/or increase the personal in- come tax threshold. The proliferation of special provisions 360. The Mexican budget contains a relatively high degree in both the income tax and VAT regimes pushes Mexico’s of expenditure rigidity that limits the ability of policymakers tax revenue below the regional average. Though typically to rapidly shift budget resources in response to changing cir- justified by worthwhile objectives, tax exemptions, deduc- cumstances. Budget rigidities come as legal and contrac- tions, and preferential regimes reduce fiscal revenue, dis- tual obligations, such as pension payments, wages and tort market prices, encourage tax evasion and avoidance, salaries of public sector workers, and transfers to subna- and in many case do not meet intended objectives. tional governments, other entitlements that are extremely difficult to alter in practice, and interest payments. The 364. When designing reforms that broaden the tax base, lack of flexibility in the level and composition of spending policymakers should consider including efficient com- may undermine the quality of fiscal adjustment efforts in pensatory mechanisms to offset any negative impact on the future, as expenditure cuts will fall hardest on those poor households. Many exemptions, deductions, and sections of the budget that can be more easily modified, preferential regimes are designed to combat poverty or such as public investment. reduce income inequality. However, the rise of targeted cash-transfer systems has made these types of tax expen- ditures largely obsolete. As tax expenditures are reduced Fiscal Framework or eliminated, cash transfers can be used to compensate poor households for any increase in their tax burden. 361. Recent modifications to the Federal Budget and Fiscal Responsibility Law strengthened the fiscal policy frame- 365. The discount on the fuel excise tax could be phased out work, but some enhancements to the framework may be and be replaced with more permanent approach. Imposing MEXICO warranted. Amendments to the LFPRH in 2014, interalia (i) a fixed excise tax on fuel while allowing its sales price to POLICY NOTES 302 The escape clause to the balanced budget rule had previously been invoked in 2010 following a sharp contraction in economic activity because of the global financial crisis. The use of the escape clause allows an increase in deficit financing, which should be phased out gradually over a three-year period. 139 Chapter 13 vary with market conditions would be consistent with in- tially negative implications for the efficiency of the public ternational good practices. The purpose of the excise tax administration, and the use of temporary employees to should be to internalize the social costs of fuel consump- perform staff functions represents a potential legal and tion, including air pollution, traffic congestion, and green- financial liability for the government. At the same time, house gas emissions. As part of the transition to retail fuel marginally accelerating the attrition of administrative staff price liberalization, the authorities have adjusted the fixed would enable the creation of new positions in the areas of excise tax on a weekly basis since 2017 by applying a dis- greatest need. As most of the federal workforce is assigned count to the legally established excise tax rate. While these to administrative roles, reintroducing the voluntary retire- discounts have smoothed the impact of changes in the ment program for administrative staff may create space international reference price on domestic retail sales pric- to introduce new positions to fill identified skills gaps. A es, they can only do so by reducing the applicable excise centralized payroll database would enable more frequent tax. Excessive price volatility may require more symmetric analyses of the federal workforce, enhance transparen- smoothing mechanisms. For example, Chile has established cy, facilitate comparisons between budgeted and actual a Mechanism for Stabilizing Fuel Prices (Mecanismo de Esta- compensation, and provide more detailed and disaggre- bilización de Precios de los Combustible, MEPO) based on a gated information to inform human resource policies. fuel excise tax with both a fixed and a variable component. Unlike the Mexican system, which applies only discounts 369. Public procurement could generate additional fiscal to the excise tax, the variable component of the MEPO in- savings. While a comprehensive analysis for the identi- cludes both discounts and surcharges. Under a perfectly fication of specific areas to gain fiscal savings in public designed smoothing mechanism, these excise surcharges procurement is needed, experience across countries show and discounts would cancel each other out over time. this is possible by mainstreaming framework agreements and better using the purchasing power of the public sec- 366. The tax authorities could raise revenues and level the tor. In the case of Mexico, this would need to be accom- playing field between domestic and non-resident suppliers panied by better linking public procurement processes to of digital services by fostering reforms to tax more actively expenditure budgeting and outlay systems. the digital economy. Some key reforms could include (i) en- list providers of platform-enabled services to report trans- 370. Further consolidation of public sector programs is need- actions of individual service providers, prefill tax declara- ed. Rationalization efforts will be necessary to reduce bene- tions or withhold and remit value added and income taxes ficiary duplication, redundancy, fragmentation and system- owed by these service providers, (ii) require foreign digital ic costs. A proliferation of programs occurred throughout service providers to final consumers to withhold and remit the public sector, though this phenomenon has been VAT on those transactions by facilitating the registration particularly marked in the areas of social protection, rural of these firms for VAT purposes, or by transferring the development and small and medium enterprise (SME) sup- responsibility for withholding and remitting the VAT on port. Highly effective programs could be scaled up, while these transactions to financial intermediaries, (iii) amend underperforming programs should be eliminated, and permanent establishment rules to include “significant eco- programs with similar objectives could be consolidated to nomic presence” in domestic legislation and renegotiate leverage economies of scale. This effort will also require the double taxation agreements accordingly to start collect- establishment of a single registry and identification num- ing income tax on income generated by virtual establish- ber system for all beneficiaries across all programs. Sector ments from their business in Mexico or, in the meantime, specific recommendations on enhancing the efficiency and introduce a free standing charge on the turnover of these effectiveness of public sector spending can be found in the companies withheld and remitted by businesses with a sectoral policy notes on Health, Education, Social Protec- permanent establishment in Mexico. tion, Productive Landscapes and Productivity as well as the Mexico Public Expenditure Review (World Bank, 2016). Expenditure policy 371. Efficiency in managing public infrastructure is key as 367. To accommodate medium-term fiscal pressures and more fiscal space is opened for this needed outlay. There has create fiscal space for public investment, further technical been significant progress in this area, particularly under the and allocative efficiency measures will be needed. leadership of SHCP work, but further efforts are needed to consolidate the strategic planning of public investment 368. Efforts to contain the growth of the wage bill should across areas. This would include having a cross-sectoral not exacerbate distortions by incentivizing the use of national master plan that establishes key federal priorities temporary workers or disparities in the central pay scales. and also takes into account the more sizable sub-national Asserting greater control over the temporary personnel projects that typically may have externalities across state budget would be an important first step in containing the jurisdictions. Building a master plan and a proper pipeline MEXICO growth of the wage bill. Increased demand for temporary of projects requires significant effort and clear and trans- POLICY NOTES staff is partly the result of tighter controls on the creation parent criteria, as these plans typically go beyond a single of permanent positions, as well as the assignment of new administration. This planning, would enable a more strategic 140 mandates and responsibilities to federal institutions. targeting of the limited budget for investments and a better However, over-reliance on temporary workers has poten- use of additional resources as the fiscal space opens for more infrastructure spending. The whole cycle of the Public Invest- its pro-cyclicality power while preserving sustainability Chapter 13 ment Management (PIM) would also need to be streamlined. would be a combination of an adjusted expenditure rule This includes increased coordination among federal agen- (i.e, some adjustments to the existing expenditure rule) cies, with a fully empowered lead planning entity, and better and an explicit ceiling on the public debt-to-GDP ratio. coordination between federal and sub-national entities. This could be complemented with corrective measures at certain thresholds of debt to GDP below the ceiling. Yet an- 372. Promoting coordination and transparency in evaluat- other alternative would be to simply add the debt ceiling to ing and reporting on PPPs can help maximize value-for-mon- the existing framework. The authorities could also consider ey and reduce fiscal risks. The decline in public investment lowering the threshold for longer-term oil-rent savings now has led to an increased reliance on private sector partici- that oil revenue has fallen to less than half its previous lev- pation for the delivery of public infrastructure projects. In el. The FMP saves all revenue from oil and gas exploration 2017, total private participation in infrastructure reached and extraction contracts above the current threshold of 4.7 $8.6bn, surpassing the previous peak of $6.5bn seen in percent of GDP, while revenue below the threshold is trans- 2007. Once largely focused on toll roads, these invest- ferred to the Treasury. The current threshold reflects the ments have become more diverse. The mobilization of level of oil revenue received by the federal government in private participation in infrastructure could be expanded 2013. The FMP began operating in 2015, when oil revenue even further to support the public sector in this area. But had already declined from its 2013 level. Less than 2 percent a number of issues need to be streamlined. At present, the of GDP in oil revenue accrued to the Treasury each year, and authorities are obliged under the PPP Law to prepare a the FMP therefore accumulated no savings between 2015 comparative assessment of financing a project via PPP vis- and 2017. Lowering the threshold would attenuate the dis- a-vis financing it through public procurement. To improve tortive impact of oil revenue on the budget, which is likely transparency and accountability, consideration should to become more pronounced as oil revenue rebounds. be given to publishing these assessments alongside the cost-benefit analyses already made available on the Obra 375. Mexico’s fiscal framework could be complemented by Pública Abierta database. In addition, systematic reporting the establishment of an independent fiscal council that pro- on longer-term budget commitments, terms in the con- motes sound fiscal policies through independent oversight. tracts that could impact future payments or revenues, Fiscal councils have been adopted by different countries and other contingent liabilities should be part of decision around the world in addition to fiscal rules over the past few making process and the periodic budget documentation. decades as an institutional device to strengthen the cred- Given the decentralized nature and limited coordination ibility of governments’ commitment to sustainable public of the investment process, it is harder for suitable private finances. Despite a diversity of institutional arrangements, entities to find opportunities for PPPs. A clearer and de- a fiscal council is generally defined as a permanent agency fined pipeline of projects, with some centralized reporting with the mandate to assess publicly and independently would help, among other measures, in reducing informa- government’s fiscal policies and performance against mac- tion asymmetries and coordination problems. roeconomic objectives related to long-term fiscal sustain- ability and medium-term macroeconomic stability. Fiscal Framework 373. It would be prudent to maintain a target of the PSBR (maximum) at 2.5 percent of GDP. Given the rising costs of debt and with current macroeconomic assumptions, Mex- ico’s debt- stabilizing primary surplus would need to be around 0.8 of GDP.303 374. Significant progress has been made in establishing a solid fiscal framework, though some adjustments to the framework may signal a continued strong commitment to fis- cal discipline. The 2014 LFPRH called for an annual numer- ical target and an indicative multi-year path for the PSBR to complement the traditional budget balance. Without changing the current framework, a positive development would be to establish the fixed numerical cap of the PSBR (consistent with a downward-sloping debt path) in legisla- tion, as a signal of a continued strong fiscal policy anchor moving forward. This could be complemented with tighter criteria on the use of the escape clause. A deeper adjust- MEXICO ment to the fiscal rule framework, with the aim to enhance POLICY NOTES 303 Estimates of the fiscal adjustment necessary to maintain medium-term debt sustainability are highly sensitive to the underlying assumptions regarding the current debt-to-GDP ratio, the interest rate, and the economic growth rate. 141 Chapter 13 Secretaría de Hacienda y Crédito Público (SHCP). 2017. “Pre- References supuesto de Gastos Fiscales 2017.” Mexico: SHCP. https:// Cetrángolo, Oscar, Juan Pablo Jiménez and Ramiro Ruiz del Cas- www.gob.mx/shcp/documentos/presupuesto-de-gastos-fis- tillo. 2010. “Rigidities and fiscal space in Latin America: a com- cales-2017 parative case study.” Santiago, Chile: La Comisión Económica Valencia, Fabian. 2015. “Strengthening Mexico’s Fiscal Framework” para América Latina y el Caribe (CEPAL). IMF Country Report No. 15/314. Washington D.C. IMF http:// Escolano, Julio. 2010. “A Practical Guide to Public Debt Dynamics, www.imf.org/en/Publications/CR/Issues/2016/12/31/Mexi- Fiscal Sustainability and Cyclical Adjustment of Budgetary Ag- co-Selected-Issues-43399 gregates.” Washington, DC: IMF. World Bank (2016), “Mexico Public Expenditure Review.” Washington OECD, et al. (2018), Revenue Statistics in Latin America and the Carib- D.C. World Bank http://documents.worldbank.org/curated/ bean 2018, OECD Publishing, Paris. http://dx.doi.org/10.1787/ en/284151472615491033/Mexico-Public-expenditure- review rev_lat_car-2018-en-fr MEXICO POLICY NOTES 142 14. Improving the Efficiency and Equalization Effects of the Fiscal Federalism Framework M exico’s fiscal federalism arrangements underpin the provision of public goods and ser- vices by subnational governments (SNGs). Because the decentralization of expenditure responsibilities was not fully supported by a significant decentralization of own revenues, among other factors, Mexico’s intergovernmental fiscal model is characterized by large vertical gaps (between the expenditure needs and own-source revenue capacity of SNGs) and horizontal gaps (among states’ capacity to cover their spending needs). A first key step would be to help nar- row the vertical gap, through improved collection of key existing levies such as the property tax, which collects much less than in comparable countries. Other policies could follow, including the creation of SNG surcharges (piggybacks) on federal income or additional indirect taxes. A second key step, would be to enhance the efficiency of federal transfers and place a stronger emphasis on equalization across states. The “Aportaciones” transfers (primarily focused on health and educa- tion) could be allocated with stronger spending needs criteria and more purposeful equalization adjustments. Over the medium term, the unconditional revenue-sharing transfer (Participaciones) could be further enhanced to more strongly reflect local expenditure needs, fiscal capacity, and fiscal effort. The government could reduce or re-shape existing discretionary transfers to sub- national entities such as Ramo 23, turning them into formula based capital transfer or match- ing grant to support strategic investments. The government has begun implementing a new and highly improved legal framework for fiscal discipline of subnational governments. Consolidating this framework will require strengthening enforceability mechanisms, improving SNGs’ financial MEXICO POLICY NOTES reporting and accounting systems, making specific technical adjustments to the regulation, and eliminating SNGs’ recourse to extraordinary transfers from the federal government. 145 Chapter 14 goods and services. Improvements in subnational tax-reve- CONTEXT AND REFORM nue collection have been modest, and a heavy dependence PROGRESS on intergovernmental transfers weakens SNGs’ incentives to improve their own-source revenue capacity. In addition, 376. Mexico’s subnational governments (SNGs) play a crit- despite improvements, intergovernmental transfers still ical role in public spending. Throughout the 1990s and have limited equalization impact, perpetuating the large 2000s, the federal government transferred substantial re- regional disparities in public spending across states. sponsibility for providing essential public goods and ser- vices, such as education, health, and basic infrastructure, 380. This policy note outlines reform options to improve to SNGs. Whereas SNGs were responsible for 20 percent of SNG tax collection, enhance expenditure efficiency, and all public spending in the 1990s, by 2016 this share had further adjust intergovernmental transfers to promote equi- risen to over 50 percent. Consequently, ensuring that pub- table service delivery. The proposed option for reform are lic spending effectively supports economic growth will designed to consolidate the LDFEFM, strengthen regulato- require improving expenditure efficiency among SNGs. ry compliance among SNGs, and reinforcing subnational Moreover, equalizing the spatial distribution of fiscal re- fiscal sustainability. The policy options described below sources among SNGs could help alleviate Mexico’s large focus on increasing the own-source revenue capacity of regional socioeconomic disparities, improving fiscal equi- subnational governments, improving the ability of in- ty and service delivery to reduce poverty and foster long tergovernmental transfer mechanisms to promote fiscal term growth. equalization, sharpening expenditure-efficiency incen- tives among SNGs, and strengthening the effectiveness of 377. During the last decade, the government implemented the LDFEFM in controlling SNG borrowing. multiple reforms to the fiscal decentralization framework designed to increase SNGs’ tax revenues, strengthen the equalization effect of intergovernmental transfers, promote expenditure efficiency, and KEY CHALLENGES 378. tighten fiscal discipline among SNGs. In 2007, the 381. The large vertical fiscal gap is a distinctive feature of government adjusted the assignment of tax bases, and Mexico’s fiscal decentralization framework. Since the early modified the distribution criteria for unconditional reve- 1990s, responsibility for providing education, healthcare, nue-sharing transfers (Participaciones) and several of the public security, and local infrastructure has been pro- major conditional earmarked transfers (Aportaciones) that gressively transferred to state governments, while mu- support decentralized service delivery. The changes to the nicipalities have become responsible for implementing distribution formula for Participaciones favored regional the bulk of basic social services and urban infrastructure fiscal equalization and introduced incentives to improve investment. In the 2000s, the rapid expansion of public ed- SNGs’ tax collection. To promote expenditure efficiency ucation and health services, coupled with new demands and fiscal equalization, Aportaciones were modified to in- on the social protection and security sectors, accelerated corporate demand-side distribution criteria. Further mod- expenditure decentralization. However, this process was ifications were made in 2014, including important adjust- not accompanied by a commensurate decentralization of ments to the transfer supporting decentralized education revenue authority. Whereas SNGs are responsible for over services. Meanwhile, the authorities devolved responsibil- 50 percent of total spending, they collect about 10 percent ity for property tax (Predial) and vehicle tax (Tenencia) to of total tax revenue—the largest vertical fiscal gap among state governments in 2012 and established an excise tax OECD countries (Figure 60). on gasoline to be collected by the federal government and transferred to state governments. Between 2015 and 382. While gaps between expenditure and revenue decen- 2017, the government passed a landmark Fiscal Respon- tralization are common in federal republics or other decen- sibility Law for Federative Entities304 and Municipalities tralized public administrations, Mexico’s vertical fiscal gap is (Ley de Disciplina Financiera para Entidades Federativas y unusually large and potentially problematic. Transfers from Municipios, LDFEFM) and its implementing regulations in the federal government account for about 90 percent of an effort to ensure fiscal sustainability among SNGs. The SNG revenues. An excessive reliance on transfers weakens implementation of the new regulatory framework for con- the incentives for SNGs to collect own-source revenues, as trolling SNG indebtedness began in 2018. The statutory the costs of public goods are not fully internalized. Con- requirements of the LDFEFM become binding of state sequently, SNG tax collection remains modest despite the governments in 2018 and local governments in 2019. expanded tax bases granted to SNGs in 2007. 379. Despite these reforms, Mexico’s intergovernmental fis- 383. Mexico’s system of fiscal federalism is also marked by cal framework remains characterized by a large asymmetry large horizontal fiscal imbalances, with important implica- MEXICO POLICY NOTES between revenue and expenditure decentralization, as well tions for the regional distribution of public goods and ser- as substantial regional disparities in the provision of public vices. The relative economic strength of different regions 146 304 The phrase “federative entities” refers to Mexico’s 32 states. Figure 60: Vertical Fiscal Gaps in Selected Figure 61: Per Capita Own-Source Revenues by Chapter 14 Countries State PER MEX GBR KOR NOR BEL DNK POL 1,499 LUX ITA 2,341 IRL 2,350 CZE SVN 1,418 SVK PRT 1,447 FIN BRA ESP SWE 1,064 ISR 737 2,288 FRA AUT 445 CAN 882 863 DEU ISL USA UAE -20 0 20 40 60 80 Source: OECD. World Bank staff calculations Source: Ministry of finance, SHCP. World Bank and municipalities directly influences local revenue ca- mies of scale and institutional incentives, major tax bases pacity. Per capita state and municipal tax collection in the such as personal and corporate income taxes, social securi- northeast and center-north regions is 4.5 times the level ty contributions, and broad-based indirect taxes are more of the southwest (Figure 61). Mexico City collects 15 times efficiently administered at the federal level. Consequently, more state and municipal revenues per capita than Chi- state governments are relegated to minor tax bases such apas, Oaxaca, and Tlaxcala and over 10 times more than as payrolls, motor vehicle, and lodging taxes. Muncipali- Guerrero and Zacatecas, the states with the lowest levels ties have also narrow tax bases including the property of subnational tax collection. tax (Predial) and a real estate transfer tax (Impuesto sobre Adquisiciones Inmuebles – ISAI). Mexico’s property tax (Pre- dial) revenues are low compared to those of other regional Subnational Taxation and OECD countries. Revenue collected from these taxes 384. The federal government has expanded the tax base of amount to 0.3 percent of GDP, below the 0.5 percent level states over the last years. The gradual adoption of the pay- observed in Argentina, Colombia, and Chile and far below roll tax (Nómina) by state governments, the devolution of the OECD average of 1.9 percent. the tax on motor vehicles (Tenencia) from the federal to the state level, and the imposition of state taxes on specific ser- 387. Limited institutional and technical capacity reduce vices have expanded state governments’ own-source reve- the efficiency of property-tax collection. Municipalities nue capacity. In 2012, the federal government transferred with modest administrative resources have difficulty the authority to collect the Tenencia to the states, including maintaining up-to- date property registries, or “cadas- the ability to set the tax rate, but inter-state tax competi- ters.” Municipal tax-administration units often have few tion greatly reduced this measure’s revenue impact. Most staff with limited technical qualifications, and high fixed states slashed Tenencia rates, established exemptions costs increase the financial burden of maintaining ca- and subsidies that eroded the tax base, or eliminated the dasters in small municipalities. Finally, efforts to update Tenencia altogether. Overall, the devolution of the tax to property values and collect property taxes often face po- the state level reduced Tenencia revenue by 30 percent. litical resistance. 385. However, low levels of subnational tax collection are a persistent feature of Mexico’s fiscal federalism system. Intergovernmental transfers Subnational tax revenues amount to less than 1 percent 388. Intergovernmental transfers are aimed to help bridge of GDP, with state taxes totaling 0.7 percent and municipal the vast vertical imbalances (i.e., gap between the expendi- taxes 0.3 percent. This level is far below both the average ture responsibilities of SNGs and their limited own-source of over 10 percent for large federations globally and the revenue capacity). Participaciones are the government’s OECD average of 9 percent. It also fares below LAC federal primary mechanism for reducing vertical and horizontal republics like Argentina (5 percent of GDP) and Brazil (11 fiscal imbalances. Through Participaciones, the federal percent of GDP). government shares more than 20 percent of its tax rev- MEXICO POLICY NOTES enues (or around 4 percent of GDP) with SNGs. Most of 386. Low subnational revenue levels reflect the limited tax the remaining gap is covered by Aportaciones, a set of powers and weak collection efforts of SNGs. Due to econo- conditional earmarked transfers that amount to about 5 147 Chapter 14 Figure 62: Property Taxes in Selected Countries (% of GDP) 1.2 1 0.8 0.6 0.4 0.2 0 Argentina Brazil Chile Colombia Guatemala Mexico Paraguay Peru Uruguay LAC OECD Average 2008 2010 2013 Source: OECD percent of GDP, which finance basic education, healthcare, sition to an equal per-capita transfer and makes difficult security, and social infrastructure. Finally, voluntary grants to indetify what are the factors driving the horizontal dis- negotiated between federal agencies and SNGs amount tribution of Participaciones. In fact, the overall distribution to about 0.5 percent of GDP each year. of Participaciones remains regressive, as reflected by the direct relationship between per capita transfers and per 389. But they have very limited equalization power to re- capita GDP. In 2016, the amount of Participaciones per cap- duce horizontal imbalances (among SNGs) despite positive ita received by Campeche was twice the amount received reforms. Established in 1978, the Participaciones system by Chiapas, Guerrero, and Oaxaca. was originally designed to compensate SNGs for the cen- tralization of tax bases that were previously within the 391. Aportaciones were primarily designed to cover the purview of state governments. The distribution formula costs associated with the decentralization of public ser- of Participaciones was revised in the early 1990s and the vice delivery. Because they were intended to facilitate reimbursable or devolution principle was watered down. the transfer of responsibility for providing public goods In 2007 a new formula was introduced and was based on and services from the federal government to SNGs, the the state-level economic growth as a proxy for federal col- distribution criteria for Aportaciones were driven by sup- lections in each state (devolution component) and for the ply-side considerations, and an inadequate emphasis on growth of states’ own-source tax collection (tax collection transparency, accountability, and results undermines their effort incentive) and the level of states’ own-source tax efficiency. Moreover, because the distribution criteria were collection (compensate for revenue efforts made by the defined at the time when the various services were decen- states in the past). A “hold harmless” clause was also intro- tralized, the allocation of Aportaciones has a strong inertial duced to ensure that no subnational entity will suffer an component, and costs considerations are often absent absolute loss relative to the level of transfers it previously from their distribution criteria. Indeed, Aportaciones have received. increased service-delivery costs without generating corre- sponding improvements in service quality. 390. Although, that the formula aimed to reward economic activity as a proxy of federal collections in each state and the 392. Recognizing these limitations, the distribution formu- states’ own tax effort, the effectiveness of these terms is mut- la was reformed for the largest Aportaciones in an effort ed because these components are weighted by the states’ to improve their efficiency and equalization effect. The population shares, making unclear the distributive impact distribution criteria for the Earmarked Transfer Fund for of Participaciones. Indeed, because each state’s population the Education Payroll and Operating Expenses (Fondo share directly influences all variables in the 2007 formula, de Aportaciones para la Nómina Educativa y Gasto Opera- relative population size drives the distribution of Partici- tivo, FONE)305 were modified in 2007 to introduce some paciones. The population-based component of Participa- demand-side considerations. The new FONE distribution ciones has a modest equalizing effect, as a large share of formula included the number of students enrolled in ba- federal taxes is collected in the most-developed regions sic education, as well as indicators of education quality. and distributed across the country on an equal per capita A compensatory component was also added the formu- basis. However, the revised formula had not had strong la to provide additional funds to states with per-student equity impacts, as the distribution criteria had not taken transfers below the national average, and the new formula into consideration indicators that reflect expenditure included a factor designed to encourage SNGs to invest MEXICO need, economic conditions or the low revenue capacity. their own resources in education. In 2014, under pressure POLICY NOTES Moreover, the hold harmless clause slows down the tran- from states that had had their transfers reduced by the 148 305 FONE was previously known as the Earmarked Transfer Fund for Basic Education (Fondo de Aportaciones para la Educación Básica, FAEB). Figure 63: Per Capita Total Revenues (after transfers) by State Chapter 14 [12,488 -16,282] 14,232 [16,283 -20,076] [20,077 -max] 21,506 15,486 23,413 15,522 16,355 18,269 18,249 15,204 21,710 16,523 17,565 16,610 14,675 12,854 15,824 17,347 14,797 12,488 20,305 20,350 17,228 17,471 16,942 Source: Ministry of Finance, SHCP. World Bank. new formula, and recognizing that the reforms had failed 395. While some Aportaciones have clear equalizing effects, to eliminate perverse incentives, the government modi- the overall system of aportaciones does little to correct hor- fied the FONE distribution formula to include both a sup- izontal fiscal imbalances. The Earmarked Transfer Fund for ply-side component (the teacher payroll and school oper- Social Infrastructure (Fondo de Aportaciones para la Infrae- ating costs) and a demand-side component (the number structura Social, FAIS) finances investment in social infra- of students in each state). structure by state and local governments via the Social Infrastructure Fund of the States (Fondo para la Infraestruc- 393. Much like FONE prior to the 2007 reforms, the distri- tura Social de Estados, FISE) and the Social Infrastructure bution formula for the Earmarked Transfer Fund for Health Fund of Municipalities (Fondo para la Infraestructura Social Services (Fondo de Aportaciones para los Servicios de Salud, de Municipios, FISM). FISM represents about 88 percent of FASSA) is driven by supply-side factors, including the num- FAIS. FAIS distributes 2.5 percent of federal taxes (equiva- ber of decentralized federal health workers and the oper- lent to 0.3 percent of GDP) to the states through a formula ating costs of federal health facilities in each state. While based on poverty indicators and unmet basic needs. State FASSA also incorporates equalization criteria, the distribu- governments then distribute funds to municipalities ac- tion of resources is largely defined by historical budget al- cording to a similar formula. However, while FAIS clearly locations, which in turn reflect the supply-side conditions favors less-developed states, it does not necessarily favor that prevailed when health services were decentralized. less-developed municipalities, as a relatively poor munic- Because wealthier regions tend to have larger numbers ipality in a relatively wealthy state could receive less than of federal health workers, more-qualified health workers, a relatively wealthy municipality in a relatively poor state. and better health facilities, FASSA may have exacerbated Moreover, the loose definition of “investment in social in- regional disparities in health services.306 frastructure” and the fragmentation of resources reduce the efficiency of FAIS transfers, and a World Bank evalua- 394. However, FASSA’s equalizing effects should be assessed tion of FAIS found that it has had a limited impact on mon- within the context of other sources of health financing that etary and nonmonetary poverty indicators.307 In addition, interact with FASSA, such as the universal health program, the Earmarked Transfer Fund for Strengthening Federative Seguro Popular. Seguro Popular distributes an amount Entities (Fondo de Aportaciones para el Fortalecimiento de equal to 0.4 percent of GDP among states based on the Entidades Federativas, FAFEF) distributes 1.4 percent of size of their uninsured populations. The establishment of federal taxes (or 0.2 percent of GDP) to the states accord- Seguro Popular and its rapid expansion is gradually cor- ing to the inverse of their average per capita economic recting the unequal and inertial regional distribution of output. However, because FAFEF resources finance SNG FASSA transfers. debt obligations, pension liabilities, and institutional and MEXICO POLICY NOTES 306 For more details on Seguro Popular, see Policy Note X on the health sector. 307 World Bank, (2017). El Efecto del Fondo de aportaciones para la Infraestructura Social FAIS en el Desarrollo Regional de Mexico. Mimeo. Washington DC: The World Bank. 149 Chapter 14 Figure 64: State and Local Government Debt, Figure 65: The Aggregate SNG Debt Stock, 2005- 2016 (MXN billions) 2005-2016 (% of GDP) 600 3.5 53 3.1 3.1 3.1 3.1 500 54 3.0 2.8 2.9 56 52 2.5 50 2.5 400 Local Government Debt 2.2 50 2.0 300 38 1.6 1.6 1.7 1.7 31 482 515 1.5 200 430 453 18 24 385 1.0 341 13 16 277 100 221 179 0.5 134 145 169 0 0.0 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 Source: Ministry of Finance, SHCP Source: Ministry of Finance, SHCP technical capacity-building, the fund has no direct impact and 7 had debt levels exceeding 100 percent. Moreover, on regional fiscal disparities. an increasing number of SNGs have faced liquidity prob- lems and financing gaps requiring extraordinary transfers 396. The discretionary use of extraordinary federal trans- from the federal government. fers (e.g., Ramo 23) may help undermine the credibility of the subnational fiscal-discipline framework. Registered 399. The new subnational fiscal responsibility law provides under Ramo 23, extraordinary federal transfers managed the cornerstone for an institutional framework to control through agreements between the Federation and state SNG borrowing. In 2015, the government passed a Con- governments create an ad hoc system, with limited effi- stitutional amendment granting the federal government ciency and transparency in the intergovernmental transfer the authority to regulate SNG borrowing.308 This paved the system. They also intensify budgetary unpredictability, way for the passage of the LDFEFM in 2016, with supple- and may foster soft budget constraints. mentary regulations approved in March 2017. The LDFEFM provides a sound basis for controlling SNG borrowing and promoting responsible fiscal management. The fiscal rule Subnational Borrowing included in the law is simple and transparent, and compli- 397. Mexico’s subnational public debt stock has increased ance is easily verified, which should help stabilize SNG debt significantly over the past decade. In the wake of the glob- over the medium term. The LDFEFM classifies SNG debt al financial crisis, the Mexican economy suffered a sharp levels as either “sustainable,” “under observation,” or “high,” contraction that negatively affected subnational fiscal and it defines a fiscal deficit or net financing ceiling for and debt dynamics. Between 2008 and 2013, faced with each category. The LDFEFM also allows the federal govern- rising spending obligations and a limited capacity to raise ment to participate in subnational fiscal adjustments and own-source revenues, SNGs increased their borrowing to debt-restructuring operations by providing partial federal finance widening deficits (Figure 64). The aggregate SNG guarantees on SNG debt to reduce debt-service costs. debt stock rose from 1.7 percent of GDP in 2008 to 3.1 per- cent in 2013, where it stabilized (Figure 65). 400. However, implementation challenges could weak- en the LDFEFM’s enforceability and undermine its effec- 398. While the total subnational debt stock remains low and tiveness. The federal government’s limited ability to is not a source of systemic risk, an increasing number of states observe and enforce compliance with the LDFEFM— have experienced various types of fiscal distress, including combined with SNGs’ variable levels of technical and rising debt levels, shrinking fiscal space, and liquidity prob- institutional capacity, data constraints, inconsistencies lems. Poor revenue performance and mounting recurrent in public finance accounting rules and practices, and expenditure pressures have resulted in declining fiscal bal- weaknesses in public financial management, especially ances and high debt levels, while rising debt obligations at the local level—could prevent the full implementa- have further narrowed the fiscal space for investment. Be- tion of the new framework. Moreover, the LDFEFM is a tween 2008 and 2012, the number of state governments numerical fiscal rule, and as such the government will with debt levels exceeding 50 percent of non-earmarked face incentives to use creative accounting to remain revenues rose from 7 to 12, and the number with debt within its parameters. levels exceeding100 percent of non-earmarked revenues rose from 0 to 5. By 2016, 18 state governments had debt 401. The government may need to adjust the technical MEXICO levels exceeding 50 percent of non-earmarked revenues, design of the LDFEFM’s regulatory framework. The law’s pri- POLICY NOTES 150 308 Before the Constitutional amendment, subnational governments had full autonomy over their debt policies, and the federal government had no authority to regulate them. Figure 66: State Government Debt as a Share of Non- Earmarked Revenues, 2008-2016 (%) Chapter 14 Zacatecas Yucatán 2008 Veracruz 2012 Tlaxcala Tamaulipas 2016 Tabasco Sonora Sinaloa San Luis Potosí Quintana Roo Querétaro Puebla Oaxaca Nuevo León Nayarit Morelos Michoacan E. de Mexico Jalisco Hidalgo Guerrero Guanajuato Durango Distrito Federal Chihuahua Chiapas Colima Coahuila Campeche Baja California… Baja California Aguas Calientes 0% 50% 100% 150% 200% 250% 300% Source: Ministry of Finance, SHCP mary purpose is to control the growth of borrowing and framework for cooperation between state and municipal mitigate financial stress among subnational governments. revenue agencies, or by developing administrative coop- However, the thresholds for classifying the ratio of debt eration agreements for outsourcing tax administration to to non-earmarked revenue as either “under observation” regional agencies. or “high” were set at such an elevated level that only one state government was prevented from incurring a fiscal 403. Creating national or state fiscal cadasters (while improv- deficit and engaging in net borrowing in 2018. In addition, ing the existing ones in larger cities) could greatly improve some provisions of the LDFEFM may be difficult to comply the efficiency of property-tax administration. Technological with and monitor. advancements such as satellite imaging have facilitated the creation of more accurate and comprehensive proper- ty cadasters. Establishing national or state cadasters could POLICY OPTIONS leverage economies of scale and reduce the influence of local interest groups over property taxation. Colombia, Uruguay and Spain have already created national or state Increasing Own-Source Revenue cadasters. While a national or state agency would maintain Collection in Subnational the cadaster, property taxes would still be set and collect- Governments ed locally, and a regular schedule for reassessing property 402. The government could prioritize efforts to strengthen values could improve revenue collection.309 the administration of property taxes. Whereas small and medium size municipalities lack the resources to en- 404. Improvements in SNG tax administration could be sure efficient predial collection, their states tend to have accompanied by the expansion of their tax bases in the fu- a greater capacity for tax administration. With support ture. While improving the collection efficiency of existing from the federal government, administrative cooperation subnational taxes could boost SNG revenues, expanding agreements between state and municipal revenue agen- the tax bases available to state and municipal govern- cies have yielded positive results. The government could ments could greatly enhance their own-source revenue build on these recent successes by establishing a national capacity and ease political pressure for additional federal MEXICO POLICY NOTES 151 309 For example, Guanajuato’s predial law requires that municipalities update property values in the cadaster every two years. The Municipality of León adopted an innovative strategy in which it automatically charges an additional 0.234 percent of the value of properties that were assessed within the past two years and as much as 0.681 percent for properties that have not been revalued for several years. Chapter 14 transfers. Allowing states to impose a surcharge on federal grants instead of ad-hoc transfers could help to improve taxes could greatly increase subnational revenues at a very accountability at the subnational level. Other options modest administrative cost. For example, each state could would be to consolidate resources from Ramo 23 and FAIS, impose a 1-3 percent flat-rate surcharge on the federal to enhance the latest and use its distribution criteria or the income tax. Restricting the surcharge to between 1 and establishment of a capital transfer with clear equalizing 3 percent would help prevent a “race to the bottom” in criteria. which states attempt to outcompete each other for invest- ment by lowering their tax rates. The Tax Administration 409. Over the medium term, the effectiveness of Participa- Service (Servicio de Administración Tributaria - SAT) could ciones transfer could be improved by dividing their pool of collect the surcharge, easing the administrative burden on funds into 2 pools: a pure revenue-sharing component and the states. State governments could also impose surcharg- an equalization-transfer component. The formula for the es on other federal taxes, such as excise taxes. pure revenue-sharing component could be designed to ensure that transfers reflect the amount of tax revenue collected within the administrative boundaries of SNGs Improving the Efficiency and (i.e., a derivation or origin-basis criterion). As most federal Enhancing the Equalization Effect taxes that finance Participaciones cannot be apportioned of Intergovernmental Transfers to states (e.g., VAT and corporate income tax), states’ eco- 405. In terms of the Aportaciones transfers, the FONE re- nomic output can be used to approximate the federal sources could be distributed with a stronger emphasis on a revenue collected in each state. The distribution criteria for per-student and needs-based allocation. Thus, the modified the equalization component could be designed to cover proposed formula could reflect a core per-capita compo- the gap between the expenditure needs and fiscal capac- nent that is adjusted for the costs of delivering a standard ity of SNGs. Expenditure needs (not actual expenditures) service across the country (favoring poorer regions). It are defined as the amount each SNG would need to spend could also include an equalizing factor designed to accel- to provide a standard level of public services based on the erate the convergence of education indicators of lagging size of its population, local socioeconomic conditions, and states. This proposed system would rely on distribution the costs of providing the necessary public services. Reve- criteria similar to those used prior to the 2007 reform, but nue capacity (not actual revenue collection) is the ability of simpler and with a much stronger equalization compo- a government to raise own-source revenues based on an nent. average level of administrative effort, adjusted for the size of the government’s assigned tax bases and the funds re- 406. Prospective adjustments to other Aportaciones could ceived through pure revenue-sharing Participaciones. This yield additional efficiency gains, but require further assess- equalization transfer would increase with SNG expenditure ment. Merging FISE with FISM could reduce the fragmen- needs and decrease with SNG revenue capacity. Allocating tation of the Aportaciones system. Because FISE represents equalization transfers according to expenditure needs and a very small share of FAIS, and municipal governments im- revenue capacity would eliminate perverse incentives that plement most social infrastructure projects, which mainly encourage excessive or inefficient spending and discour- deliver local benefits with limited spatial spillovers, merg- age collection efficiency. This formula would also need a ing the two transfers could yield a modest efficiency gain. very strong coefficient to reflect fiscal effort (i.e., improve- The authorities would need to simplify the two-stage dis- ments in collection above the average of what would be tribution rule to prevent two municipalities with identical expected given the revenue capacity of the state. conditions from receiving different amounts just because they are located in different states. Strengthening Subnational Fiscal 407. FASSA could be distributed on per uninsured person Discipline basis. While Seguro Popular is gradually improving both, 410. While a thorough evaluation of the LDFEFM would be the coverage of uninsured population and the regional premature, the law appears to be strengthening fiscal dis- equalizing impact of federal transfers to the health sector, cipline among SNGs. Since 2014, improved fiscal balances a new distribution criterion for FASSA transfers may accel- and an increased awareness of borrowing controls have erate progress in these fronts. As a transition, the federal helped stabilize the debt-to-GDP ratios of SNGs. The rig- government could distribute the same amount of resourc- orous application of the LDFEFM regulatory framework re- es in nominal terms in a given year (harm hold clause) and mains critical to ensure fiscal discipline at the subnational the annual increase in the transfer could be distributed on level. equal per uninsured person basis. 411. The LDFEFM’s technical design may require some mi- 408. The government could reduce or re-shape existing nor adjustments. Because the current indicative thresh- discretionary transfers to subnational entities such as Ramo olds for the ratio of debt to non-earmarked revenue are MEXICO POLICY NOTES 23. Ramo 23 resources could be transformed into a capital too high to serve as a binding constraint on most state grant which can be used by the federal government to fi- governments, the LDFEFM does not adequately restrain nance investments by SNGS that have strategic national SNGs from incurring fiscal deficits or from increasing their 152 or/and regional spillovers. Using formula based matching public debt stocks to levels that could generate liquidity or solvency problems. Reducing these thresholds would fiscal framework. Additional federal technical assistance Chapter 14 increase the effectiveness of the LDFEFM. An assessment to SNGs will be critical to close data gaps and harmonize of the LDFEFM and its implementing regulations could accounting standards. provide a sound analytical basis for further revising and rationalizing the fiscal framework. 413. Eliminating the discretionary use of extraordinary fed- eral transfers could enhance the credibility of the fiscal dis- 412. Effectively controlling subnational indebtedness will cipline framework for SNGs. Allowing the federal govern- require improvements in the government’s financial report- ment to transfer extraordinary resources to SNGs weakens ing and accounting systems. While the General Government fiscal discipline. By contrast, the federal government’s abil- Accounting Law (Ley General de Contabilidad Gubernamen- ity to issue a partial guarantee on subnational debt within tal) has increased financial transparency, data constraints the context of a fiscal adjustment or debt-restructuring and incomplete accounting standardization across SNGs program provides a much more direct and transparent limit the ability of creditors to assess financial risk and way for the federal government to support SNGs that are prevent policymakers from addressing weaknesses in the experiencing financial distress. MEXICO POLICY NOTES 153