Bhutan Labor Market Assessment Report Jumana Alaref, Laurine Martinoty, Mariana Viollaz, Esther Bartl, Phillippe Leite, and Alvin Etang Ndip © 2024 The World Bank 1818 H Street NW, Washington, DC 20433, USA. Telephone: 202-473-1000; Internet: www.worldbank.org. Some rights reserved — This work is a product of the staff of The World Bank with external contributions. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denomina- tions, 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 construed or considered to be a limitation upon or waiver of the privileges and immu- nities of The World Bank, all of which are specifically reserved. Rights and Permissions — This work is available under the Creative Commons Attribution 3.0 IGO license (CC BY 3.0 IGO) http://creativecommons.org/licenses/by/3.0/igo. Under the Creative Commons Attribution license, you are free to copy, distribute, transmit, and adapt this work, including for commercial purposes, under the following conditions: Attribution — Please cite the work as follows: Jumana Alaref, Laurine Martinoty, Mariana Viollaz, Esther Bartl, Phillippe Leite, and Alvin Etang Ndip. 2024. “Bhutan Labor Market Assessment Report.” World Bank, Washington, DC. License: Creative Commons Attribution CC BY 3.0 IGO. Translations — If you create a translation of this work, please add the following disclaimer along with the attri- bution: This translation was not created by The World Bank and should not be considered an official World Bank translation. The organizations 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 attribu- tion: 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 content contained within thecwork. 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 re‑use a component of the work, it is your responsibility to determine whether permission is needed for that re‑use and to obtain permission from the copyright owner. Examples of components can include, but are not limited to, tables, figures, or images. All queries and rights and licenses should be addressed to World Bank Publications, The World Bank, 1818 H Street NW, Washington, DC 20433, USA; e-mail: pubrights@worldbank.org Contents Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vii About the Authors and Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . viii Abbreviations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . x Executive Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1 Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 1.1 Objectives of the report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 1.2 Structure of the report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 2 Profile and Challenges Facing Workers in Bhutan . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 2.1 Evolution of the labor market over the last 10 years . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 2.2 Labor force participation rates for men and women along the life cycle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 2.3 Unemployment among youth and educated workers in 2022 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 2.4 Current state of employment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30 2.5 Quality of employment: Hours worked, wages, informality, and a job quality index . . . . . . . . . . . . . . . . . . . 35 2.6 Internal mobility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46 3 Firm Dynamics in Bhutan and Its Alignment with Labor Supply . . . . . . . . . . . . . . . . . . . . . . . . . . . 48 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48 3.1 Profile of firms in Bhutan . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48 3.2 Labor demand prospects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54 3.3 Barriers to firm growth and business management practices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70 4 Bhutan’s Employment Support Programs and Delivery System . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73 4.1 Objectives of employment support programs and their rationale for Bhutan . . . . . . . . . . . . . . . . . . . . . . . . . 73 4.2 Bhutan’s employment support programs and delivery system . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76 4.3 Conclusion: Ways to strengthen employment support programs in Bhutan and international best practices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85 iii 5 Policy Directions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92 Appendixes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98 A Data Sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98 B Supplementary Figures and Tables, Chapter 2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99 C Public-Private Wage Differentials in Bhutan . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143 D Supplementary Figures and Tables, Chapter 3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 148 E Overview of Selected ALMPs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162 F Supplementary Tables, Chapter 4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 164 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 166 Boxes 2.1 Blinder-Oaxaca decomposition methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40 2.2 Job quality index—dimensions and methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44 4.1 Objectives of active labor market programs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75 4.2 12th Five-Year Plan (2018–23) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78 4.3 The TVET system in Bhutan . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81 4.4 Global evidence from Latin America’s Jovénes programs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86 4.5 The Republic of Korea’s TVET system . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87 4.6 Bangladesh’s microfinance system . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88 4.7 Green Jobs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89 4.8 Documenting global best practices in labor interventions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91 Figures 1.1 Impact of hydropower on GDP, 1985–2021 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 1.2 Poverty rate (US$3.65/day), 2017 and 2022 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 1.3 Human Development Index, 2010–21 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 1.4 Contributions of TFP, capital, and labor to economic growth, 2001–19 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 1.5 Monthly migration through Paro International Airport, 2015–22 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 1.6 World Bank’s Human Capital Index, Bhutan and selected countries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 1.7 Education and maternal and child health incomes, Bhutan and LMICs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 1.8 Population pyramid, Bhutan . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 2.1 Working-age population and ratio of individuals ages 15–64 to dependents under 15 and above 64, 2013–22 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 2.2 Education level of the working-age population, 1950–99 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 2.3 Labor force participation (LFP) rate, by gender and location, 2013–22 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 2.4 Labor force participation (LFP) rate, by gender and education, 2013–22 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 2.5 Employment rates, overall and by gender, 2013–22 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18 2.6 Employment rates, by location, 2013–22 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 iv 2.7 Unemployment rate, overall and by gender, 2013–22 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20 2.8 Unemployment rate, overall and by location, 2013–22 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 2.9 Labor force participation rate, by gender and life cycle, 2013—22 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22 2.10 Labor force participation rate, by gender, location, and life cycle, 2022 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22 2.11 Labor force participation rate of prime-age individuals (25–54), by gender, location, and education, 2022 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 2.12 Labor force participation rate, by gender, location, and marital status, 2022 . . . . . . . . . . . . . . . . . . . . . . . . . 24 2.13 Motives for inactivity, by gender and age group, 2022 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24 2.14 Female labor force participation, by presence of children, 2022 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25 2.15 Female labor force participation, by gender and dependent household members, 2022 . . . . . . . . . . . . . . . 25 2.16 Reasons for being not in education, employment, or training (NEET), by gender (ages 15–24), 2022 . . . . . 26 2.17 Unemployment rate, overall and by age, education, location, and gender, 2022 . . . . . . . . . . . . . . . . . . . . . . 27 2.18 Unemployment duration, 2021 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28 2.19 Self-reported reason for unemployment, youth and nonyouth, 2021 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28 2.20 Sector preference of job-seekers, by age group, 2021 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 2.21 Reasons for sector preference, by age group and sector, 2021 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 2.22 Average real monthly wage and reservation wage, by education and age, 2022 . . . . . . . . . . . . . . . . . . . . . . . 30 2.23 Employment rate, overall and by age, education, location, and gender, 2022 . . . . . . . . . . . . . . . . . . . . . . . . 31 2.24 Structure of employment, 2022 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33 2.25 Share of industries in total employment and its evolution, 2013–22 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34 2.26 Relative employment growth and productivity, 2013–21 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35 2.27 Number of hours worked and standard deviations, by demographic characteristics, economic sector, employment type, and gender, 2022 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 2.28 Average monthly and hourly wages, overall, and by education, gender, and area, 2022 . . . . . . . . . . . . . . . 38 2.29 Distribution of total employment types, by location, 2022 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41 2.30 Weekly working hours and real hourly wages of employees in companies, businesses, or nongovernmental organizations, by formality status, 2017 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42 2.31 Formal and informal jobs and attached benefits, 2017 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43 2.32 Job quality index, by location and gender, 2018–22 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44 2.33 Reason for moving reported by Bhutan-born working-age individuals, 2019 . . . . . . . . . . . . . . . . . . . . . . . . 45 2.34 Current labor market status of Bhutan-born working-age migrants, 2019 . . . . . . . . . . . . . . . . . . . . . . . . . . . 46 3.1 Profile of firms by economic sector, region, size, and employment share, 2018 and 2022 . . . . . . . . . . . . . . . 50 3.2 Average number of workers hired and exiting firms, 2019–21 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51 3.3 Net job creation, by year and region, 2019–22 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51 3.4 Net job creation, by year and economic sector, 2019–21 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52 3.5 Labor productivity and net job creation rate in 2021 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53 3.6 Distribution of current employment and expected labor demand, by occupation, 2022 . . . . . . . . . . . . . . . 56 3.7 Expected labor demand for services and sales workers and craft and related trades workers, by education, 2022 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57 3.8 Comparison of the expected labor demand and the current labor force and inactive population, by education, 2022 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58 3.9 Percentage of firms reporting hiring difficulties, by reason, 2022 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59 v 3.10 Percentage of firms facing worker shortages, by occupation and education level, 2022 . . . . . . . . . . . . . . . . 61 3.11 Ratio of number of job-seekers to number of workers, by education and region, 2022 . . . . . . . . . . . . . . . . 62 3.12 Distribution of foreign workers, by occupation, 2022 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64 3.13 Employees’ perceptions of their digital and information technology skills, 2022 . . . . . . . . . . . . . . . . . . . . . 66 3.14 Major constraints to growth, by firm size, 2022 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68 3.15 Correlations between salary increment/employee promotion systems and hiring difficulties, worker shortages. and retention challenges across economic sectors, 2022 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71 4.1 Pillars of the employment delivery system in Bhutan . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76 4.2 Total number of TVET graduates, by year and gender, 2018–22 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82 4.3 Share of TTI graduates, by gender, 2018–22 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82 Tables ES.1 Policy directions to address challenges in the labor market . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 3.1 Distribution of expected vacancies over the next one or two years, by occupation, 2022 . . . . . . . . . . . . . . . . 55 3.2 Percentage of firms facing hiring difficulties, 2022 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59 3.3 Types of hiring difficulties, by occupation, 2022 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60 3.4 Impacts of worker shortages on firm performance, 2022 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63 3.5 Level of connection with potential partners for providing training, 2022 . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64 3.6 Training needs for the next five years, 2022 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65 3.7 Training critical for the current occupation and funding plans according to employees, 2022 . . . . . . . . . . 65 3.8 Importance of factors for business expansion or diversification plans, 2022 . . . . . . . . . . . . . . . . . . . . . . . . . . 67 3.9 Constraints in the management of firms, 2022 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68 3.10 Compliance with labor regulations, by firm size, 2022 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69 4.1 Overview of selected ALMPs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79 4.2 Number of regular graduates in TTIs and IZCs, by year and gender . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82 5.1 Mapping of policy directions according to four broad labor market challenges . . . . . . . . . . . . . . . . . . . . . . . . 96 vi Acknowledgments This World Bank report was prepared by a team led by Jumana Alaref. Chapter authors are Jumana Alaref, Laurine Martinoty, Mariana Viollaz, and Esther Bartl. Alvin Etang Ndip and Phillippe Leite served as core members of the team and contributors to various chapters. Nazia Moqueet provided research assistance and support through- out the study, and Elfreda Vincent, Sofia Said, Dorji Drakpa, and Tshering Yangki offered administrative support. Sabra Ledent edited the report, and Sean Willmott served as its graphic designer. The team would like to thank the following World Bank colleagues for their help. Abdoulaye Seck, Nicole Klingen, Stefano Paternostro, Cem Mete, S. Amer Ahmed, and Adama Coulibaly provided feedback, guidance, and support from management. Suhail Kassem, Melanie Simone Trost, and Mauro Testaverde served as peer reviewers at the decision review stage of the report. Ashiq Aziz, Sonam Choden Wangdi, and Joachim Vandercasteelen provided input, feedback, and support at various points during preparation of the report. The team is also grateful to the external stakeholders who offered feedback and assistance at different points during the report preparation process. From the Royal Government of Bhutan were Duptho Wangmo, Tenzin Choden, and Ugyen Namgyel from the Ministry of Education and Skills Development (MoESD), and Dasho Tashi Wangmo, Director General Kunzang Lhamo, Dil Maya Subba, Jamyang Tshomo, Jigme Thinley, Ridgen Wangchuk, and Tshering Yangki from the Ministry of Industry, Commerce, and Employment (MoICE). The report benefited from the generous funding of the World Bank’s Rapid Social Response Program (RSR), which is gratefully acknowledged. The team apologizes to any individuals or organizations inadvertently omitted from this list. It is grateful to all who provided guidance and assistance for this report. vii About the Authors and Contributors Jumana Alaref is a senior economist in the Social Protection and Jobs Global Practice at the World Bank. Her work focuses on strengthening social safety nets, improving female employment, investing in active labor market pro- grams, and overall system strengthening. In South Asia, she has directed her efforts toward helping countries to build a delivery system for employment support services, as well as to enhance their analytical and informational capacity through the establishment of labor market information systems. She has also contributed to the develop- ment of national labor market strategies in both Saudi Arabia and Kuwait. And she has co-led impact evaluations on labor market programs in Lebanon and Tunisia. She holds a master’s degree in public policy from the Univer- sity of Chicago. Laurine Martinoty has served as an assistant professor at Université Paris 1 Panthéon Sorbonne since September 2016. She is a member of the research group on international economics and labor markets within the Centre d’Economie de la Sorbonne. She has been focusing on household responses to shocks in labor demand in econo- mies at all income levels. She currently works on family formation and labor market opportunities in Europe, with an emphasis on the role of economic shocks and gender attitudes. She is a consultant at the World Bank. Mariana Viollaz is a senior researcher at the Center for Distributive, Labor and Social Studies (CEDLAS) of the Universidad Nacional de La Plata (UNLP) in Argentina. She has a PhD in economics from UNLP and has been a postdoctoral fellow at the School of Industrial and Labor Relations at Cornell University. Her research focuses on gender and labor economics in developing countries. She is a consultant at the World Bank and the Inter-Ameri- can Development Bank. Esther M. Bartl is a PhD student in development economics at the University of Sussex, UK, and a consultant at the World Bank. Her research focuses on labor markets and international labor migration in low- and middle-income countries, in particular South Asia and Central Asia. She holds a master’s degree in international economics from SAIS, Johns Hopkins University. Phillippe Leite is a senior social protection economist in the South Asia unit of the Social Protection and Jobs Global Practice at the World Bank. He is also a member of the global lead teams on social safety nets and delivery systems. He joined the social protection and labor network team in Africa and participated in the design of social programs in, among other countries, Brazil, Colombia, Mali, Mexico, the Republic of Congo, and Senegal. He pre- viously worked for the Development Research Group on determinants of poverty and inequality, poverty maps methodology, and microeconometric simulation models. He holds a BA and MS in statistics (sampling and model- ing) from ENCE/Brazil and an MS and PhD in economics from École des Hautes Études en Sciences Sociales, Paris. viii Alvin Etang Ndip is a senior economist in the Poverty and Equity Global Practice at the World Bank. At the Bank, he has led country-level poverty engagement and regional and global initiatives. One highlight of his recent accom- plishments is his contribution to Sudan’s reengagement, including by supporting preparation of the government’s poverty reduction strategy paper, thereby enabling Sudan to reach the Heavily Indebted Poor Countries (HIPC) decision point in June 2021 and providing a path to the country’s debt relief. He had previously managed the World Bank’s Listening to Africa Initiative on using mobile phone panel surveys to monitor welfare. Previously a postdoc- toral associate at Yale University, he holds a PhD in economics. Nazia Moqueet is a social protection specialist in the Social Protection and Jobs Global Practice at the World Bank. Her work includes research and operations related to social safety nets and economic inclusion, with a focus on female labor force participation in South Asia. Prior to the World Bank, she worked at BRAC, where she provided technical assistance to governments and nongovernmental organizations on designing and implementing the graduation approach to building sustainable livelihoods and resilience among vulnerable populations in the Arab Republic of Egypt, Guinea, Pakistan, Kenya, and the Republic of Yemen. She holds a master’s degree in develop- ment economics from George Washington University. ix Abbreviations ALMP active labor market program BLFS Bhutan Labor Force Survey BLSS Bhutan Living Standards Survey CCD Critical Capability Development (program) CEM Country Economic Memorandum (World Bank) COVID-19 coronavirus disease 2019 CSI Cottage and Small Industries CST Critical Skills Training (program) DWPSD Department of Workforce Planning and Skills Division (MoESD) ECCD early childhood care and development ES Establishment Survey ESC employment services center FY fiscal year GDP gross domestic product GNI gross national income JQI job quality index IT information technology IZC institute for Zorig Chusum KRIVET Korea Research Institute for Vocational Education and Training KSQA Korea Skills Quality Authority LMIS labor market information system M&E monitoring and evaluation MFI microfinance institution MIS management information system MoAF Ministry of Agriculture and Forests MoESD Ministry of Education and Skills Development MoICE Ministry of Industry, Commerce, and Employment MoLHR Ministry of Labor and Human Resources MSME micro, small, and medium enterprises NAS National Account Statistics NEET not in education, employment, or training NFE nonformal education OLS ordinary least squares PPP purchasing power parity RGoB Royal Government of Bhutan x RUB Royal University of Bhutan SDG Sustainable Development Goal SDP Skills Development Plan SOE state-owned enterprise SSDP Special Skills Development Program TFP total factor productivity TTI Technical Training Institute TVET technical and vocational education and training VSDP Village Skills Development Program YELP Youth Engagement and Livelihood Program xi Executive Summary Between 2001 and 2019, Bhutan made significant eco- Key messages nomic progress with an average annual growth rate of 7 percent. This progress was accompanied by significant Message 1: Workers in Bhutan face many challeng- improvements in monetary and nonmonetary stan- es, including limited inclusion of women in mean- dards of living. In recent years, the Royal Government ingful employment and persistence of low-pro- of Bhutan (RGoB) has committed to advancing reforms ductivity agricultural employment. Employment through the implementation of its 12th Five-Year Plan quality outside of the public sector remains weak, (2018–23), which focuses on creating gainful employ- leading to public sector queuing, rising unemploy- ment among urban workers, and a record number ment and enhancing private sector diversification. of Bhutanese migrating abroad. However, recent shocks from the COVID-19 pandemic and global macroeconomic volatilities disrupted these Young women (ages 15–24) invest less in higher edu- efforts. The recovery could be further complicated by cation than men, and they have a 1.5 times higher the country’s pressing structural challenges related to a risk of being not in education, employment, or train- lack of economic diversification away from the hydro- ing (NEET), notably to fulfill household responsibili- power-led growth model, vulnerabilities to shocks, ties. The labor force participation of women of prime and weak productivity gains. In addition, human working age is strongly correlated with their marital capital remains low. A child in Bhutan will be only 48 status, the presence of young children, as well as labor percent as productive in adulthood as he or she could force participation of other women in their family and have been with a complete education and better health community. They tend to work in low-productivity care. The record number of Bhutanese migrating sectors such as agriculture, manufacturing, and ser- abroad with the reopening of the country’s borders in vices or as self-employed or family workers, and they mid-2022 is fueling further concerns by policy makers have limited access to private employment or public about the country’s development prospects. sector jobs. Although women’s labor force participa- tion increased substantially during the pandemic, it fell This report examines the labor market in Bhutan with in 2022 back to its pre-pandemic levels at 53 percent, the objective of identifying the most pressing chal- which could be indicative of a discouragement effect. lenges at the pandemic recovery stage and ways to mit- igate them. It is hoped that the findings can be used Over 40 percent of workers in Bhutan remain engaged to support the RGoB in its implementation of its 13th in low-productivity agricultural employment. This Five-Year Plan (2024–29). This report analyzes data finding is reflected in the low-skill nature of the labor from three sources: (1) the Bhutan Labor Force Survey market, where 37 percent of employed workers have (BLFS), 2013–22; (2) the 2022 Bhutan Establishment no formal schooling. This outcome is a consequence Survey; and (3) administrative program–level data of the hydropower-led growth model, which is not from the Ministry of Education and Skills Development labor-intensive. It thus does not tempt workers to leave (MoESD) and the Ministry of Industry, Commerce, and agriculture, nor does it directly affect agricultural pro- Employment (MoICE)—see appendix A. ductivity. Therefore, the labor demand in agriculture Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 1 remains unaffected1 and structural transformation sector jobs over private sector ones. In addition, a large occurs slowly. Between 2013 and 2022, the share of number of Bhutanese workers have migrated since agriculture in total employment fell from 57 percent 2022 or have plans to migrate. The average number of to 44 percent. Meanwhile, public sector employment those migrating increased significantly to more than grew to absorb educated and urban workers, com- 5,000 a month in early 2023, compared with less than prising 25 percent of total employment. As a result, 500, on average, one month prior to the pandemic. One the labor market in Bhutan is heavily segmented along out of 10 NEET individuals plans to migrate abroad. the lines of gender, location, and education. Men and high-skilled workers in urban areas dominate public Message 2: The private sector in Bhutan faces the sector employment, while female, low-skilled, and twin challenges of accelerating job creation in rural workers are more likely to remain employed in productive sectors that can absorb the increasing- agriculture as self-employed or family workers, with ly educated workforce, while also improving the limited options for upward mobility. allocation of labor to fill existing vacancies in low- and semi-skilled positions. Outside of the public sector, poor employment quality persists in the nonagricultural economic sectors in The underdeveloped private sector is dominated by urban centers. An assessment of job quality through low-productivity microenterprises—one of the main the lens of hours worked finds that overwork (working reasons for the high unemployment rate among skilled more than 48 hours a week) is prevalent and affects 63 professionals. It also contributes to the low quality of percent of the workforce. Overwork is concentrated employment outside the public sector and may explain in certain economic sectors where vulnerable types in part the speed and scale of outmigration in recent of employment (such as self-employed and family months. Over 95 percent of registered firms in the workers) tend to dominate. These sectors include private sector have five employees or fewer, and they construction, wholesale and retail trade, transporta- do not grow over time. Close to 75 percent of firms tion, and accommodation and food services. A look with fewer than five employees have been operating at employment quality through the lens of the bene- for one to nine years. Firms are also geographically fits attached to a job reveals that one out of three sal- concentrated (69 percent of them are in Thimphu, aried employees have no written contract from their Gelephu, or Punakha) and are not sufficiently diver- employer, accounting for over 11 percent of workers sified in terms of economic activity (80 percent are in Bhutan (and 23 percent of workers in urban areas). in the wholesale and retail trade and accommodation Manufacturing, construction, wholesale and retail and food services sectors). These dominant economic trade, and accommodation and food services (where sectors have, on average, low labor productivity. overwork is more prevalent) account for 55 percent of nonwritten contracts. Meanwhile, the private sector is undergoing labor shortages, despite rising unemployment. Labor short- In response to the limited attractive options for the ages stem from two imbalances in the labor market. fast-growing cohorts of educated workers outside of First, the profiles of available or future vacancies in the public sector, the rise of unemployment in urban the private sector do not match those of the current areas that began in 2019 continued in 2022. Although job-seekers in Bhutan. Today, most firms are hiring there is no evidence that the unemployed have unre- to fill low-skilled occupations, and about one-third alistic wage expectations, the majority prefer public of firms expect to have new vacancies in the next one 1.  This argument is elaborated on in the forthcoming Country Economic Memorandum: Bhutan (World Bank, forthcoming a). 2 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice or two years. Firms will need workers with a low to that hiring difficulties negatively affect firm perfor- medium level of education and specific technical skills mance and potential for growth. (through certificates or diplomas) to mainly fill posi- tions in services and sales and craft and related trades. Firms also face many other barriers to their growth On the supply side, only 10 percent of job-seekers have that relate to investment climate factors and labor reg- no education, and the majority have completed a sec- ulations. The extent to which some of those barriers ondary education. There is also a severe shortage of are binding varies by firm size, which may require a job-seekers with certificates. Second, a large number tailored approach to supporting private sector devel- of low- or semi-educated workers who could fill the opment. Among smaller firms, access to finance, vacancies are either outside the labor force (mostly markets, and raw materials appears to be a major con- women), or are engaged in low-productivity liveli- straint. According to the 2017 “Investment Climate hood activities (as self-employed workers). There is Assessment of Bhutan” (Santini, Tran, and Beath 2017), also evidence that some Bhutanese workers apply high some Bhutanese firms are unable to access a loan due reservation wages to jobs in certain vacant occupa- to lack of credit information and due to the complex, tions, particularly in the construction sector. There- unpredictable, and ineffective restructuring and insol- fore, making the available low-skilled positions more vency regime. Access to markets is also hampered by attractive and accessible to Bhutanese workers, espe- trade and logistical deficiencies and limited foreign cially women in urban areas, is an important priority. investment. For larger firms, stringent labor regula- tions, high worker turnover, and market wage level As a result of labor shortages, approximately one- are more prominent difficulties. Finally, private sector third of firms in Bhutan faced hiring difficulties in activity may be hampered by a competition policy that 2022, mostly due to few or no applicants. Findings encourages the dominance of state-owned enterprises show that labor shortages may have a spatial dimen- in key economic sectors. sion. Regions that report the most hiring difficul- ties (Samdrup Jongkhar and Trashigang) also have Message 3: Bhutan’s employment support pro- tight labor markets (that is, they have more employed grams and delivery systems face gaps in address- workers than job-seekers for the education categories ing some of the challenges related to the activation in demand). Although cross-dzongkhag (district) mobil- of women, limited job-relevant skills, and the ity for work is common and could alleviate some of the difficulties firms face in accessing trained labor. observed spatial mismatches, it is less common among low-skilled workers. Reasons for that could include Although Bhutan has improved its employment pro- mobility, financial, skilling, or informational barri- grams and policies over the last few years, signifi- ers that may prevent workers from moving to regions cant gaps remain. Most active labor market programs where some of those wage opportunities exist. Hiring remain small and do not specifically focus on the acti- low-skilled workers from neighboring countries could vation of women by alleviating some of the childcare or address worker shortages, but only 7 percent of firms mobility constraints they face. Although the technical hired at least one foreign worker in 2022, and that per- and vocational education and training (TVET) system is centage has declined over time. In addition, over 90 relatively well established, it remains too small to meet percent of firms noted that they have no links to public the existing labor demand. For example, there were vocational or training institutions to help meet their only 401 TVET graduates in 2022. Tracer surveys reveal demand for skilled workers with technical and certif- that over 45 percent of graduates remain unemployed icate-level education. A sizable share of firms noted one year after completing training and that employers have no links to most TVET institutes (MoESD 2023a). Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 3 Meanwhile, Bhutan’s five employment services centers sector, as well as returns to education and occupa- (ESCs) are not empowered enough to play a pivotal role tional skills. in job placement and matching. Most ESCs have capac- ity gaps that limit the services offered. They also lack Message 4: Addressing Bhutan’s labor market proper equipment and resources. Finally, although challenges requires pursuing vertical growth Bhutan has a new labor market information system policies that are sector-specific to support private (LMIS), it is still in its early stages and does not yet gen- sector development and job creation. In parallel, erate comprehensive data on the labor market. The horizontal reforms that improve the business en- key challenges include a lack of clarity on the extent to vironment, strengthen human capital accumula- tion, and increase the effectiveness of employment which data sharing arrangements are in place between support programs are needed. the entity administering the system and the ministries that collect and store labor market data. In addition, Table ES.1 summarizes the policy directions suggested there are information gaps in the LMIS in key areas, to address Bhutan’s the labor market challenges. such as vacancies and their distribution by occupation/ Table ES.1. Policy directions to address challenges in the labor market Challenge Underlying constraint Suggested policy direction Limited produc- Bhutan’s hydropower-led Pursue a vertical approach to promoting the growth of promising job-rich tivity and job growth model has had sectors. At the same time, implement horizontal reforms across all sectors to creation in the negative implications for improve the productivity of small firms and support their growth. These reforms private sector, the development of pro- include (1) strengthening entrepreneurship by facilitating access to finance, which partly ductive sectors outside of mentorship, and links to supply chains and markets; (2) governance reforms explain the speed agriculture and the public related to the investment climate, foreign direct investment, and the efficiency and scale of recent sector. of state-owned enterprises; and (3) labor market reforms to promote flexible outmigration labor regulations that can support worker mobility and firms’ access to labor, as well as a functional labor market information system (LMIS) that can regularly identify skills in the labor market and support hiring needs for start-ups. Limited The productivity of Targeted government resources are needed to ramp up investments in early human capital Bhutan’s workforce is childhood education, nutrition, and development. Low-income young mothers accumulation undermined by low benefiting from employment support could be linked to other human capital levels of human capital programs to strengthen maternal and infant health. that stem in part from unequal and inadequate access to quality foun- dational human capital services, especially for disadvantaged families in rural areas. Limited human The productivity of agri- In rural areas, the provision of coordinated economic inclusion services could capital utilization cultural workers is low. support improvements in agricultural productivity. In urban areas, the following In urban areas, there is a efforts are needed to strengthen employment support programs: (1) reorient suboptimal allocation of the technical and vocational education and training sector to improve the links labor to meet employers’ with the private sector; (2) allocate the appropriate resources to improve the hiring needs. capacity of employment services centers to establish relationships with local employers, engage in vacancy collection, and provide services such as on-the- job assistance, counseling, and mobility support for low-skilled workers; (3) implement programs to bridge the gap between labor supply and demand, such as on-the-job training; and (4) enhance the capacity of the existing LMIS to reduce data gaps and better understand the profiles of workers and how they align with the skills and occupations demanded by the private sector, thereby reducing skill mismatches. 4 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice Chapter 1 Overview Jumana Alaref, Alvin Etang Ndip, and Phillippe Leite macroeconomic volatilities disrupted these efforts, Introduction despite a swift government response in the form of easing social and mobility restrictions and continued Globally, labor markets and productivity growth have fiscal support. In addition, the RGoB continues to face been hit by multiple crises, including most recently the several structural challenges that could further com- macroeconomic volatility and economic fallout associ- plicate the recovery process. ated with the COVID-19 pandemic. Average economic growth worldwide could slump to a three-decade low A key structural challenge facing Bhutan is the reli- through 2030 because overlapping crises of the past ance of its growth model on the hydropower sector. few years have disrupted economic growth trajectories Bhutan’s economy grew at an average annual rate of 7 and slowed employment and productivity, which are percent between 2001 and 2019, driven by the expan- essential for income growth and higher wages (Kose sion of the hydropower sector since the mid-1980s. and Ohnsorge 2023). As a result, countries are facing Figure 1.1 reveals that the growth of Bhutan’s gross the immediate challenge of rethinking their national domestic product (GDP) has been heavily driven by its policy plans and reformulating their labor market pri- hydropower sector, and episodes of high growth have orities to ensure a quick recovery. At the same time, largely coincided with the launch of hydropower proj- they need to tackle the long-running structural chal- ects.2 Hydropower accounts for more than a third of lenges to boost investment, productivity, and eco- the country’s goods exports and domestic revenue and nomic growth. constitutes 26 percent of its GDP. Bhutan, a small mountainous, landlocked country Growth in the non-hydro sector averaged 7 percent (population, about 770,000) bordered by India and from 2001 to 2019 (lower than the 9 percent growth China is no exception. In recent years, the Royal Gov- of the hydro sector during the same period). It was ernment of Bhutan (RGoB) has advanced structural driven by the services, construction, and manufactur- reforms through implementation of its 12th Five-Year ing industries. Growth in services was driven by public Plan (FYP) (2018–23). The plan placed a strong empha- administration (including health and education), trade sis on creating productive, gainful employment and and transport (accounting for 75 percent of growth), on enhancing private sector diversification. However, followed by the financial sector (including real estate). recent shocks from the COVID-19 pandemic and global Since 1974, Bhutan has targeted high-value tourism to 2.  The commissioning of the Chhukha and Tala hydropower plants resulted in discrete jumps in the growth rate. Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 5 Figure 1.1. Impact of hydropower on GDP, Figure 1.2. Poverty rate (US$3.65/day), 2017 1985–2021 and 2022 Percent Percent 35 Chhukha (1987) 12 30 10.7 25 Tala (2007) 10 20 15 Mangdechhu 10 (2019) 8 7.4 5 0 6 -5 -10 4 -15 2.4 -20 2 1.7 1985 1988 1991 1994 1997 2000 2003 2006 2009 2012 2015 2018 2021 0.7 0.5 GDP growth 0 Urban Rural Bhutan GDP growing, excluding electricity and water 2017 2022 Source: World Bank, forthcoming a. Based on national sources. Source: World Bank, forthcoming b. Based on Bhutan Living Standards Survey (BLSS), 2017 and 2022. minimize the impacts of growth on the environment. significantly improve services, education, and health. The hotel and restaurant sector grew at an average rate In addition, GDP per capita between 1980 and 2017 of 16 percent a year over 2001–19. grew by 7.5 percent annually—one of the highest growth rates in the world. Human development indicators Hydropower-led economic growth has contributed to improved over time as measured by life expectancy at substantial and inclusive poverty reduction over the birth, years of schooling, and gross national income last two decades. Bhutan’s per capita income increased per capita (figure 1.3). Educational attainment steadily threefold in purchasing power parity (PPP) terms increased in both urban and rural areas, and expected between 2000 and 2019. As a result, extreme poverty, years of schooling increased from 12 years in 2010 to 13 based on the US$2.15 a day poverty line, was eliminated years in 2021. Life expectancy at birth increased from by 2022, and there was a vast reduction of people living 68 years in 2010 to 72 years in 2021. In addition, access below the US$3.65 a day poverty line (used for low- to electricity has become almost universal. With a gross er-middle-income countries) from 7 percent to less national income (GNI) per capita of US$3,040 in 2021, than 2 percent between 2017 and 2022 (figure 1.2). Based Bhutan is approaching the threshold for upper-mid- on the US$6.85 a day poverty line (used for upper-mid- dle-income status. dle-income countries), poverty declined from nearly 37 percent to 19 percent during this period. However, the current growth model has had negative implications on economic diversification and remains The hydropower-led growth model also delivered sub- unsustainable for many reasons. As typically observed stantial improvements in living standards over the in resource-rich economies, large foreign currency last two decades by creating fiscal space for invest- inflows during the construction and export phases of ing in human capital, thereby allowing Bhutan to hydropower projects contributed to an appreciation 6 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice Figure 1.3. Human Development Index, Figure 1.4. Contributions of TFP, capital, and 2010–21 labor to economic growth, 2001–19 Index Percent 10 0.68 9 8.6 0.66 8 7.0 0.64 7 5.8 0.62 6 5 0.6 4 0.58 3 0.56 2 1 0.54 0 0.52 2001-19 2001-08 2009-19 Total growth 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 Total factor productivity Labor Capital Interaction Source: United Nations Development Programme. Source: World Bank, forthcoming a. Note: The Human Development Index is a summary measure of the average achieve- ment in health, education, and standard of living. The health dimension is assessed by life expectancy at birth; the education dimension by mean of years of schooling for adults age 25 years and over and expected years of schooling for children of school entering age; and standard of living dimension by gross national income. of the real exchange rate. This appreciation has placed in driving aggregate growth, making significantly upward pressure on domestic prices, with adverse smaller contributions than capital stock and labor impacts on the competitiveness of non-hydro trad- (figure 1.4). Limited productivity gains have slowed able sectors (the “Dutch Disease”)—see World Bank structural transformation. Labor remains predom- (forthcoming a); Boyreau and Rama (2015). In addi- inantly employed by the low-productivity agricul- tion, the capital-intensive nature of the hydropower tural sector. The second major employer is the public sector limited employment opportunities—the sector sector, which the government used to create employ- employed less than 1 percent of the labor force (World ment opportunities for the growing share of educated Bank, forthcoming a). In fact, a primary driver of workers in urban centers, leading to greater pressure poverty reduction in recent years is the significant on an already small fiscal space. improvement in living standards among rural house- holds, which stems more from wealth redistribu- Limited private sector development may also be under- tion and social transfers than improvements in labor mined in Bhutan by its large state-owned enterprise market outcomes (World Bank, forthcoming b). (SoE) sector. The sector accounts for about 20 percent of total public sector employment. The total assets of As a result, private sector development remains SoEs amounted to 171 percent of GDP in 2020, and they limited, and Bhutan suffers from weak productivity remain important actors with a presence in the stra- gains. Despite a modest acceleration in recent years, tegic and economic sectors, including power, tele- total factor productivity (TFP) plays only a minor role communications, transport, manufacturing, finance, Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 7 trade, agriculture, and natural resources. The SOE among those ages 20–24), and those with a postsec- sector has been a critical part of the country’s develop- ondary education and above (12 percent). The average ment strategy since the 1960s because of its important number of Bhutanese migrating via Paro International role in providing critical services. With its geography Airport increased significantly with the reopening of and small population, Bhutan finds the provision of the borders in mid-2022—to more than 5,000 a month infrastructure and services costly and economies of in early 2023, compared with less than 500, on average, scale difficult to achieve. However, the sector may be a month prior to the pandemic (figure 1.5). This migra- distorting the playing field with the private sector, so tion has raised concerns among policy makers about a that it is unable to operate at its optimal level of effi- brain drain, which can hinder development prospects ciency (World Bank, forthcoming c). (World Bank, forthcoming a). The limited jobs available in the private sector became Bhutan’s structural challenges affect its ability to with- more pressing in the aftermath of the COVID-19 pan- stand external macroeconomic volatility and climatic demic, and that is likely contributing to the alarmingly shocks. Although it has maintained macro-fiscal sus- high outmigration of some professionals in Bhutan.3 tainability over the last two decades, supported by Unemployment increased from an average of 3 percent large hydro revenues and external grants, the external in 2015–19 to 6 percent in 2022. Unemployment is sig- shocks of the pandemic and the global ramifications nificantly higher among urban women (15 percent), of the Russian Federation’s invasion of Ukraine have youth (37 percent of those ages 15–19 and 27 percent disrupted Bhutan’s growth trajectory and exacerbated Figure 1.5. Monthly migration through Paro Figure 1.6. World Bank’s Human Capital Index, International Airport, 2015–22 Bhutan and selected countries Number of people Percent 60 60 60 6,000 60 5,500 50 50 50 48 5,000 4,500 40 40 40 4,000 3,500 30 3,000 2,500 20 2,000 1,500 10 1,000 500 0 Pakistan South Africa Bhutan Jordan Bangladesh Chile Peru Thailand 0 Jan-15 Jun-15 Nov-15 Apr-16 Sep-16 Feb-17 Jul-17 Dec-17 May-18 Mar-19 Aug-19 Jan-20 Nov-20 Apr-21 Oct-18 Sep-21 Feb-22 Jul-22 Dec-22 Source: World Bank, forthcoming a. Source: World Bank 2020b. That said, a combination of push and pull factors could be driving outmigration, and this phenomenon may not be limited to high unemployment. For 3.  example, anecdotal evidence suggests that outmigration is high among employed workers because a high share of public sector employees are migrating in response to the high wage differential between opportunities in the domestic market and abroad. Without comprehensive data on the profile of recent migrants, the team cannot make a further assessment. 8 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice structural challenges. The trade and tourism depen- Low levels of human capital also stem from unequal dence of the small landlocked economy has left it and inadequate access to quality foundational human susceptible to the pandemic-induced shocks.4 In addi- capital services. For example, pre-primary enrollment tion, with its vulnerable mountain terrain and vola- remains low, at 28 percent, with high regional dispar- tile ecosystems, the country is susceptible to a variety ities (figure 1.7, panel a).6 When years of schooling are of natural hazards. These include earthquakes, glacial adjusted to quality of learning, the years of schooling lake outburst floods, as well as seasonal hazards such as standards are, on average, six, which is a level consider- landslides and flash floods during the monsoon season ably lower than those of comparable countries (figure and forest fires during the dry winters. 1.7, panel b). Moreover, the demand for and quality of health services related to maternal, newborn, and child Bhutan’s vulnerability is further exacerbated by chronic health and nutrition services continue to lag, especially human capital challenges that persist despite progress among low-income households and in rural areas. over the years and significant spending on health and Stunting, for example, is high on average, with over a education (figure 1.6). A child in Bhutan will be only 48 fifth of children under age five suffering from stunting percent as productive in adulthood as he or she could (WFP 2022). Figure 1.7, panel c, presents several lagging have been with a complete education and better health maternal and child health outcomes that are below the care. Low human capital is driven by chronic poverty Sustainable Development Goals (SDGs). that remains more prevalent in certain pockets, such as rural areas and across dzongkhags (districts). The These challenges will make it difficult for Bhutan to poverty rate (less than US$3.65 a day) is as high as reap the demographic dividend. Figure 1.8 shows that 11 percent in Zhemgang and 4 percent in Samdrup Bhutan has a favorable age structure as half of the pop- Jongkhar.5 The share of the per capita real expendi- ulation is under the age of 29. However, the dividend ture of the richest quintile in Bhutan is more than four may not be fully realized when a large segment has low times higher than that of the poorest, suggesting that levels of human capital, and many working-age indi- progress could still be made in terms of shared pros- viduals are not employed productively. They cannot perity (World Bank, forthcoming b). therefore use their skills and protect themselves from various shocks along the life cycle. 4. After contracting by 2.3 in fiscal 2019/20 and 3.3 percent in fiscal 2020/21, respectively, Bhutan’s economy is recovering from the pandemic. Output grew by 4.6 percent in fiscal 2022/23, supported by the reopening of borders to tourism in September 2022. 5.  The poverty rate is, however, less than 0.2 percent in Paro, Punakha, Thimphu, and Trashi Yangtse. 6. According to the Bhutan Living Standards Survey 2017, 22 percent of urban children have attended primary, early childhood care and development (ECCD), or daycare, compared with only 13 percent of rural children. Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 9 Figure 1.7. Education and maternal and child health outcomes, Bhutan and LMICs a. Pre-primary enrollment, Bhutan and LMICs b. Learning-adjusted years of schooling, Percent Bhutan and selected countries 70 Percent 60 10 9.5 60 8.9 8.5 8.1 50 8 6.2 6.3 5.8 40 6 4.8 30 28 4 22 20 13 2 10 0 Pakistan South Africa Bangladesh Bhutan Jordan Peru Thailand Chile 0 Bhutan LMICs Urban Rural Source: World Bank 2020b. Note: LMICs = low- and middle-income countries. Source: World Bank 2020b. c. Maternal and child health outcomes, Bhutan Percent 100 80 60 40 20 0 Maternal Under-5 Infant mortality ratio mortality rate mortality rate (per 100,000) (per 1,000) (per 1,000) Bhutan SDG target Source: Annual health bulletin, 2020. 10 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice Figure 1.8. Population pyramid, Bhutan and programs aimed at addressing some of the chal- lenges faced by workers and firms. Age 90-94 The report uses three data sources: (1) yearly cross-sec- 75-79 tional, nationally representative data (2013–22) from the Bhutan Labor Force Survey (BLFS), conducted by 60-64 the Bhutan National Statistics Bureau; (2) the nation- ally and regionally representative Establishment 45-49 Survey (ES) conducted in 2022 by the Ministry of Labor and Human Resources (MoLHR);8 and (3) administra- 30-34 tive program–level data from the Ministry of Educa- tion and Skills Development (MoESD) and the Ministry 15-19 of Industry, Commerce, and Employment (MoICE). 0-4 Data are analyzed by applying statistical methods that are mostly standard in microeconometric research. 7 6 5 4 3 2 1 0 1 2 3 4 5 6 7 The report also synthesizes the existing literature Share of population, percent for Bhutan and draws on published statistics for the Male Female country, such as the 2022 National Accounts Statistics and 2018 Economic Census, among others. Appendix A Source: UNDESA 2022. provides information on the two primary data sources, the ES and BLFS. 1.1 Objectives of the report This report has three limitations. First, it does not With a renewed push to address these structural chal- address granular, firm-level productivity and its rela- lenges, the RGoB is currently working on its 13th FYP, tionship to the capacity of firms to create more, better, and strategies related to improving employment out- and more inclusive jobs in view of the absence of pro- comes and productivity will likely be at its forefront. ductivity data in the ES. Second, it does not include a review of Bhutan’s labor regulations and their effects The objective of this report is to support the RGoB labor on the capacity of the private sector to create jobs. market agenda and inform the design of the programs, Although chapter 3 looks at firms’ perceptions of the policies, and strategies associated with the 13th FYP. To extent to which labor regulations act as a barrier to this end, the report provides an update on the labor growth, a deeper review entails a separate, stand-alone market in Bhutan that reflects the impact of various assessment. And, third it does not include a detailed shocks in recent years.7 It examines the labor market analysis of economywide or sector-specific constraints before, during, and after the pandemic to improve to job creation related to connectivity and logistics, lib- understanding of the most pressing challenges at the eralization of trade and foreign direct investment, land recovery stage, and it looks at ways to mitigate them. policies, and distortive subsidies because this report The report also assesses Bhutan’s employment systems does not aim to be exhaustive in its review of all job creation constraints. 7.  The last update on the labor market was prepared by the World Bank in 2016 (World Bank and MoLHR 2016). 8. The 2022 ministerial reform led to the dissolution of the Ministry of Labor and Human Resources (MoLHR) in January 2023. The Departments of Labor, Employment, and Entrepreneurship were moved to the newly established MoICE and the Department of Skills to MoESD. Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 11 1.2 Structure of the report 1. Active labor market programs (ALMPs), which aim to bring inactive women into employment and Chapter 2 uses 10 waves of the BLFS to analyze the evo- engage employers more strongly with job-seekers. lution of labor supply from 2013 to 2022 with a focus These typically include skilling programs, on-the- on labor force participation, skills composition, unem- job training, and hiring incentives for employers ployment, and employment. It also addresses con- such as wage subsidies. straints to labor force participation and highlights 2. Technical and vocational education and training the unemployment challenges facing youth and edu- (TVET), which has an important role in training cated workers. The chapter concludes with an assess- workers at the postsecondary level and delivering ment of employment quality from the perspective of high-quality courses that are certified and strongly hours worked, wages, and informality and the extent linked to private sector demand. of worker mobility in Bhutan. 3. Employment services centers (ESCs), which serve as the bridge between labor supply and labor demand Chapter 3 explores the degree to which the challenges by actively aiding in job placement, job matching, facing workers stem from limited productive opportu- and implementing some ALMPs. nities created by the private sector. It analyzes the char- 4. A well-functioning and up-to-date labor market acteristics of registered firms in Bhutan as well as past information system (LMIS) that can both support patterns of job creation and destruction. It also exam- evidence-based policy making on workforce train- ines key occupation and education categories expected ing needs and employment promotion services. to be in demand in the near future and compares them It can also support better matches between labor with the profile of the labor force discussed in chapter demand and labor supply by producing data on the 2. In doing so, it highlights potential skill mismatches profile of workers, as well as on the skills and occu- in the labor market, and as a result, the hiring difficul- pations demanded by the private sector. ties and supply shortages faced by firms. Finally, the chapter looks into firm business management prac- tices and the difficulties firms encounter in expanding Chapter 5 concludes the report by offering directions their growth. for orienting public policies and programs to address the constraints facing workers and employers in Chapter 4 examines the extent to which programs and Bhutan. delivery systems are effective in addressing some of the challenges outlined in chapters 2 and 3 facing both Finally, the appendixes provide details on the sources workers and firms. The chapter considers four key of data used in this study, supplementary figures and components in its review: tables for chapter 2, and additional information on the public-private wage differential in Bhutan. 12 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice Chapter 2 Profile and Challenges Facing Workers in Bhutan Laurine Martinoty and Jumana Alaref Introduction 2.1 Evolution of the labor market over the last 10 years In this chapter, 10 waves of the Bhutan Living Stan- dards Survey (BLFS) are used to analyze the evolution of Bhutan’s labor supply from 2013 to 2022. Section 2.1 Since 2013, the size of the working-age popu- discusses trends in key labor market indicators over lation has fallen, but the demographic divi- the last 10 years, covering the COVID-19 pandemic and dend has remained the same. its aftermath. These indicators include the growth of Bhutan’s working-age population in terms of size and Although the size of the working-age population has education gaps, labor force participation, unemploy- been decreasing since 2013, the number of individuals ment, and employment. Section 2.2 focuses on current ages 15–64 in the workforce relative to the number of labor force participation over the full life cycle of men dependents under age 15 or above age 64 has remained and women, and it explores the differences by gender, stable. According to figure 2.1, in 2013, 529,000 persons age, and location. Section 2.3 examines the unemploy- were of working age, declining to roughly 485,000 in ment challenges among young and educated workers 2022. At the same time, the demographic dividend— and the drivers of them. Sections 2.4 and 2.5 detail that is, the share of the working-age population rela- the current sectoral and occupational composition tive to younger or older dependents—remained stable of Bhutan’s employment and measures the quality of over the years. employment in terms of hours worked, wages, and informality. Lastly, section 2.6 explores the extent of worker mobility in Bhutan to help promote better employment opportunities. Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 13 Figure 2.1. Working-age population and ratio of individuals ages 15–64 to dependents under 15 and above 64, 2013–22 Working-age population (thousands) Ratio, population ages 15-64 to population <15 years and >64 years 600 557 559 570 2.50 529 538 497 488 490 485 482 500 2.25 2.14 400 2.04 2.01 1.97 1.97 1.98 1.97 1.97 300 2.00 1.90 1.90 200 1.75 100 0 1.50 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 Source: Bhutan Labor Force Survey, 2013–22. Over the years, the skill level of working-age accessing education: the odds ratio for 2022 is only 1.2 men and women has increased rapidly, faster for men and women ages 23–32. Figure B.1 in appendix in urban areas than in rural areas. B shows that between 2013 and 2022 the share of men and women without a diploma fell by 10 percentage The skill level of the working-age population, both points, while the share of those with a tertiary degree men and women, increased rapidly between 2013 and rose by a factor of two. 2022. Figure 2.2, panel a, displays the distribution of education levels by sex and birth cohort in 2022 and The rural-urban gap in education persists, with rates in suggests that the closing of the gender gap in educa- urban areas rising faster than in rural areas. Figure 2.2, tion is a slow but positive phenomenon. Men born in panel b, reveals that the share of uneducated men and 1950–59 and 1960–69 were six times and four times women was originally higher in rural areas (87 percent) more likely, respectively, to be educated than women than in urban areas (67 percent) with a relative risk of of the same generation. For men born in 1970–79 and 1.3 for rural dwellers born between 1950 and 1959.9 1980–89, this relative chance decreases—they are 2.4 This share declined at a slower pace in rural areas (28 and 1.5 times more likely, respectively, to graduate percent in rural areas versus 11 percent in urban areas than women of their generation. Men and women of for the birth cohort 1990–99, with an increased rela- the most recent cohort have almost the same chance of tive risk of 2.5). The education gap has closed for the The relative risk is the ratio of the probability of an outcome in a group to the probability of an outcome in another group. Presented here is a comparison 9.  of the probability of rural and urban dwellers being uneducated at different points in time: 0.87/0.67 ≈ 1.3 for the older cohort and 0.28/0.11 ≈ 2.5 for the younger cohort. A relative risk above 1 indicates that rural dwellers are more uneducated than urban dwellers, and the fact that the relative risk increases over time shows that the gap in relative education increased between these two groups, even though the probability of being uneducated decreases in time for both rural and urban men and women. 14 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice Figure 2.2. Education level of the working-age population, 1950–99 a. By gender and birth cohort b. By rural/urban location and birth cohort Percent Percent 100 100 80 80 60 60 40 40 20 20 0 0 Rural Urban Rural Urban Rural Urban Rural Urban Rural Urban Male Male Male Male Male Female Female Female Female Female 1950-59 1960-69 1970-79 1980-89 1990-99 1950-59 1960-69 1970-79 1980-89 1990-99 None/NFE Primary/ECCD Secondary None/NFE Primary/ECCD Secondary Tertiary Monastic Tertiary Monastic Source: Bhutan Labor Force Survey. Note: ECCD = early childhood care and development; NFE = nonformal education. middle-skill level as evidenced by the young cohort, by a rapid increase from 2019 to 2021 and a decline whose rural and urban members are almost equally in 2022. The labor force participation rate in Bhutan likely to be high school graduates. However, even for trended upward from 2013 to the height of the pan- this young cohort, the proportion of individuals with demic, after which it dropped from 69 percent in 2021 only a primary school diploma remains twice as high in to 63 percent in 2022. rural areas as in urban areas, and they are two to three times less likely to be university graduates. The rising labor force participation rate between 2019 and 2021 and its decline in 2022 was mostly driven by women. According to figure 2.3, panel b, the male During the pandemic, labor force partici- labor force participation rate grew slowly but steadily pation rose, driven mostly by the entry of from 2013 to 2022. The female labor force participa- women in both urban and rural areas. In tion rates ebbed and flowed between 2013 and 2018, 2022, the participation rate dropped back to ranging between 54 and 59 percent, and increased its pre-2019 level. to 65 percent in 2021. In 2022, 53 percent of women in Bhutan were in the labor force, compared with 73 Between 2013 and 2018, the labor force participation percent of men. The gap between women’s and men’s rate10 in Bhutan was stable at 63–65 percent, followed labor supply declined between 2019 and 2021, as The definition of labor force participation rate does not exclude subsistence activities for household consumption. The 2021 BLFS includes a new question 10.  for households in agricultural farming (“Are these products intended mainly for sale or for family consumption? [1] only for sale [2] mainly for sale [3] mainly for family consumption [4] only for family consumption.” It aligns more closely with International Labour Organization (ILO) recommendations on excluding consumption activities from labor force participation. This definition is not used in this chapter because some questions around inactivity motives were not captured. The difference in participation across definitions is not significant (see figures B.2 and B.3 in appendix B). Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 15 women increased their participation rates. However, between 2019 and 2022, driven by rural men catching the gap increased to 20 percentage points in 2022 as up with urban men and increasing their labor supply. the supply of women’s labor went down and men’s By contrast, women’s labor supply saw substantial fluc- remained constant at high levels. tuations. The trend in the growth of women’s labor supply from 2013 to 2022 followed the same pattern The rise and subsequent fall of the female labor force in both urban and rural areas, as shown in figure 2.3, participation rate applies to both urban and rural areas. panel d. Both rates increased between 2019 and 2021, Figure 2.3, panel c, shows that the male labor force par- although the labor supply of urban women increased ticipation rate in rural and urban areas did not fluctu- at a faster rate. In 2022, women’s labor supply rates in ate significantly between 2013 and 2022. Meanwhile, both areas fell back to their pre-2019 levels. the gap in participation rates between both areas fell Figure 2.3. Labor force participation (LFP) rate, by gender and location, 2013–22 a. Overall LFP rate b. LFP rate, by gender Participation rate (%) Participation rate (%) 80 80 70 70 60 60 50 50 40 40 30 30 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 Female Male c. LFP rate, males, by location d. LFP rate, females, by location Participation rate (%) Participation rate (%) 80 80 70 70 60 60 50 50 40 40 30 30 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 Rural Urban Rural Urban Source: Bhutan Labor Force Survey, 2013–2022. 16 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice Men’s labor force participation rate for all educa- transitions from inactivity to participation accounted tion groups trended upward from 2018 to 2022, while for 3.5 percent of total participation in 2020, com- women’s labor force participation rate climbed for pared with 2.3–2.4 percent in 2018 or 2019, suggest- all education groups before it fell in 2022 for those ing an additional worker effect. However, as shown with a secondary education and below. For males in table B.1 in appendix B, women entering the work- (figure 2.4, panel a), the labor force participation rate force in 2020 were younger, better-educated, and with increased over the years at all education levels. For fewer family responsibilities than women entering the females (figure 2.4, panel b), the rate increased for all workforce in 2018 or 2019, contradicting the additional education groups until 2021, when it dropped back worker hypothesis. to its pre-COVID-19 levels, apart from those with ter- tiary education who saw their participation rate grow until 2022. This rise may indicate that highly edu- The employment rate between 2013 and 2022 cated women have become more attached to the labor for both men and women was similar in market, whereas women with lower levels of education trends to the labor force participation rate. left the labor market in the aftermath of the pandemic. During the pandemic, the increase in employ- ment was enabled by the significant increase There is limited evidence to suggest that the increase in labor force participation. in women’s labor force participation during the pan- demic stemmed from an additional worker effect—that Similar to the labor force participation rate,11 the is, an increase in the labor supply of married women employment rate among the working-age popula- after their husbands lost their jobs because of COVID- tion rose during 2019–2021 before declining in 2022. 19. Data on past employment histories reveal that Figure 2.5, panel a, shows that from 2018 to 2021 the Figure 2.4. Labor force participation (LFP) rate, by gender and education, 2013–22 a. LFP rate, males, by education b. LFP rate, females, by education Participation rate (%) Participation rate (%) 100 100 80 80 60 60 40 40 20 20 0 0 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 None/NFE Primary/ECCD None/NFE Primary/ECCD Secondary Tertiary Secondary Tertiary Source: Bhutan Labor Force Survey, 2013–22. 11.  The labor force participation rate and the employment rate in Bhutan follow the same trends because the overall unemployment rate is not very high. Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 17 Figure 2.5. Employment rate, overall and by gender, 2013–22 a. Overall employment rate b. Employment rate, by gender Employment rate (%) Employment rate (%) 80 80 70 70 60 60 50 50 40 40 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 Female Male Source: Bhutan Labor Force Survey, 2013–22. employment rate among the working-age popula- At the intensive margin of employment, tion increased by 5 percentage points, reaching 66 only a few adjustments were evident during percent in 2021, before dropping to pre-2019 levels, the pandemic because the number of hours 59 percent in 2022. Figure 2.5, panel b, shows that the worked, proportion of part-time jobs, type male employment rate has been consistent over the of jobs, and monthly wages were only mildly years at about 70 percent, whereas the female employ- affected. ment rate increased during the pandemic, to almost 60 percent, and dropped back to 49 percent in 2022. Workers adjusted slightly at the intensive margin during the pandemic (see figure B.4 in appendix B). In 2020, Female employment in both urban and rural areas they worked, on average, two to three fewer hours per increased between 2019 and 2021, before declining week than in 2019. The number of hours worked was again in 2022, whereas male employment saw limited still significantly lower in 2021 than in 2019, but the gap fluctuations in line with the male labor force partici- had declined by a half to two-thirds. The likelihood of pation rate. Figure 2.6 shows that for men, the urban being employed part-time (less than 35 hours a week) employment rate went down slightly, to below 70 increased slightly at the beginning of the pandemic. percent, during 2019–2021, whereas the rural employ- As shown in figure B.5 in appendix B, the probability ment rate remained on an upward trajectory and of working less than 35 hours increased (1–4 percent- increased to 71 percent in 2022. On the other hand, the age points from 2019 to 2020) but it decreased in 2021 employment rate for women in both urban and rural relative to 2019 (3–4 percentage points for women and areas increased between 2019 and 2021. The urban men). On the other hand, the proportion of individu- employment rate for women was 40 percent in 2022, als willing to work more hours conditional on working or considerably lower than the rural employment rate less than 35 hours increased between 2019 and 2021, for women, 55 percent. suggesting discontentment with the lower number of hours worked, particularly for men. 18 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice Figure 2.6. Employment rate, by location, 2013–22 a. Employment rate, males b. Employment rate, females Employment rate (%) Employment rate (%) 80 80 70 70 60 60 50 50 40 40 30 30 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 Rural Urban Rural Urban Source: Bhutan Labor Force Survey, 2013–22. The effect of the pandemic on employment was not 2019. By contrast, women working outside of agricul- uniform across job types. Self-employed workers’ ture earned 4 percent less (450 Nu) than in 2019. These share of total employment rose, whereas the shares of differences did not persist in 2021. employers and family workers fell. In the agricultural sector, the share of self-employed workers increased relative to the share of family workers. This increase Unemployment rates were stable between was entirely driven by the changing nature of employ- 2013 and 2019, but rose during the pandemic ment for women (up 6 percentage points in the share of and continued to rise in 2022. female self-employed workers and down 7 percentage points in the share of female family workers as shown The unemployment rate oscillated between 2 percent in figure B.6 in appendix B). In nonagricultural sectors, and 3.5 percent over 2013–19 and increased to 5 the structure of employment also changed in favor of percent in 2020–21. Figure 2.7, panel a, shows that self-employed workers (up 2.5–3 percentage points), the unemployment rate was fairly stable at very low while the relative importance of employers decreased levels between 2013 and 2019, averaging 3 percent. It (down 0.7 percentage points, or a 38 percent decline increased by 2 percentage points once the pandemic from the original share of total employment in 2019), began and reached 5 percent in 2020 and 6 percent in suggesting that firms shrank in size. These variations 2022. were driven by changes in the nature of jobs held by males (see figure B.7 in appendix B). Altough the unemployment rate increased during the pandemic for both women and men, women’s Monthly earnings increased for men in the agricul- unemployment rate grew at a higher rate, owing in tural sector during the pandemic. As shown in figure part to their higher labor force participation during B.8 in appendix B, in 2020, in real terms, men working that period. Figure 2.7, panel b, shows that after in agriculture earned 15 percent more (750 Nu) than in 2019 women’s unemployment rate increased by 5 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 19 Figure 2.7. Unemployment rate, overall and by gender, 2013–22 a. Overall unemployment rate b. Overall unemployment rate, by gender Unemployment rate (%) Unemployment rate (%) 12 12 10 10 8 8 6 6 4 4 2 2 0 0 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 Female Male Source: Bhutan Labor Force Survey, 2013–22. percentage points, and stood at 6 percent in 2022, a percentage point increase in their unemployment rate record high since 2013. The unemployment rate for after 2019. men increased after 2019 by 2 percentage points and was at 4 percent in 2022, which is also a record high Rising unemployment in 2022 could explain why labor since 2013. The faster growth in women’s unemploy- force participation rates went down after the pan- ment rate during the pandemic is due to the growth demic (discouraged worker effect). In other words, in their labor force participation rate as noted earlier. many workers (especially urban women) who joined Figure B.9 in appendix B indicates a higher rate of the labor market during the pandemic may have unemployment among those ages 15–24 and those dropped out in 2022 because they could not find jobs. with tertiary education during the pandemic, which is Meanwhile, the fact that unemployment increased in covered in greater detail in section 2.3. 2022 even as labor force participation rates went back to their pre-2019 levels is indicative of job losses among The unemployment rate continued to grow in 2022 workers previously employed. Table B.2 in appendix B even as the overall labor force participation rate went compares the characteristics of job-seekers before and down, especially in urban areas, suggesting that unem- during the pandemic and documents which category ployment during the recovery phase is an urban chal- of workers was particularly exposed to unemploy- lenge. According to figure 2.8, between 2019 and 2022 ment. The share of unemployed who had ever worked the rural unemployment rate remained low overall, before increased sharply, from one-third of the unem- below 4 percent. On the other hand, the unemploy- ployed in 2019 to half in 2020–21. In addition, condi- ment rate in urban areas increased by 5 percentage tional on having worked before, the vast majority of the points between 2019 and 2022, driven mostly by urban unemployed were working in the private sector (two- women. Women in urban areas experienced an 8 thirds in 2019 and three-quarters in 2020). In terms of industries and occupations, the share of unemployed 20 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice Figure 2.8. Unemployment rate, overall and by location, 2013–22 Unemployment rate (%) 20 15 10 5 0 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 Female: rural Female: urban Male: rural Male: urban Source: Bhutan Labor Force Survey, 2013–22. from hotel and restaurants, working in services and between the ages of 25 and 64, labor supply moved sales, increased the most (100 percent and 30 percent, upward between 2019 and 2021 before it declined in respectively). The share of administration employ- 2022. Notably, for the female age groups 35–39, 40–44, ment in total unemployment increased sharply, from and 45–49, labor force participation rates were above 2 percent of total unemployment in 2019 to 17 percent 80 percent between 2019 and 2021 before dropping in 2020. back to below 70 percent in 2022. Although men in both urban and rural areas participate 2.2 Labor force participation rates in the labor market in equal numbers, rural women for men and women along the life remain more active than urban women. In 2022, there cycle was an 11 percentage point difference in the women’s labor supply in rural areas (58 percent) and urban areas (47 percent). By contrast, 73 percent and 74 percent of The gender gap in labor force participation is rural men and urban men, respectively, participated pronounced across the life cycle and in both in the labor market in 2022. An ordinary least squares urban and rural areas. (OLS) regression analysis on the determinants of labor force participation, calculated for women and men, The difference in male and female labor force partici- shows that location matters much more for women’s pation is evident across the full life cycle. In figure 2.9, labor supply than men’s (see tables B.3a and B.3b in panel a, the male labor force participation rate, which appendix B). In 2022, rural women were 11–13 percent- is highest for those between the ages of 25 and 64, age points more likely to participate in the labor force remained roughly constant at high levels between 2013 than urban women. Rural men were also more likely to and 2022. By contrast, the rate varied for women (figure participate than urban men, but the participation gap 2.9, panel b). For all age groups, particularly for those is much lower, 2–4 percentage points. Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 21 Figure 2.9. Labor force participation rate, by gender and life cycle, 2013–22 a. Males b. Females Participation rate (%) Participation rate (%) 100 100 80 80 60 60 40 40 20 20 0 0 15-19 20-24 25-29 30-34 35-39 40-44 45-49 50-54 55-59 60-64 65+ 15-19 20-24 25-29 30-34 35-39 40-44 45-49 50-54 55-59 60-64 65+ Age group Age group 2013 2014 2015 2016 2017 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2018 2019 2020 2021 2022 Source: Bhutan Labor Force Survey, 2022. The gaps in labor force participation rates by gender Figure 2.10. Labor force participation rate, by and location are pronounced across the full life cycle. gender, location, and life cycle, 2022 Figure 2.10 reflects two key messages. First, young Participation rate (%) people between the ages of 15 and 24 and those older 100 than 65 participate in lower numbers in the market, consistent with trends elsewhere—that is, most workers 80 across both genders and location participate while in their prime age (between 25 and 54). The OLS regres- sion analysis in figure B.10, panel a, in appendix B con- 60 firms that the inverse U shape of participation with respect to age holds when everything else is equal, so 40 it is not driven by changes in the education or other household and local labor market characteristics. 20 Second, urban women are the group with the lowest participation rates across the full life cycle when com- pared with rural women or rural men and urban men. 0 15-19 20-24 25-29 30-34 35-39 40-44 45-49 50-54 55-59 60-64 65+ In the 40–44 age group, there is a 42 percentage point difference between rural men and urban women, a 41 Age group percentage point difference between urban men and Female: rural Female: urban urban women, and a 22 percentage point difference Male: rural Male: urban between rural women and urban women. Source: Bhutan Labor Force Survey, 2022. 22 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice The labor force participation rate of prime- rate for women was recorded for those with a tertiary age individuals differs strongly by gender: education—79 percent in urban areas and 83 percent male participation rate is high, whereas in rural areas. In urban areas, women tend to partici- female participation is structured according pate more if they are educated: 40 percent of women to level of education, place of residence, fam- without an education participate in the labor market, ily constraints, and norms. but their participation increases by 7 percentage points if they graduated from primary school and by another The participation rate of prime-age men is high for all 16 percentage points if they are secondary school grad- locations and education levels, whereas women are uates. Education matters less for participation in rural more likely to participate at higher levels of education. areas: about three out of four women participate in According to figure 2.11, male participation is lowest the labor market, regardless of the highest degree for the most educated in both urban and rural areas. In attained (excluding tertiary education).12 The OLS esti- rural areas, it is 92 percent for tertiary-educated and 96 mates reported in figure B.10, panel b (and table B.3 in percent for men without an education. In urban areas, appendix B) show that women are more likely to par- the participation gap is higher—90 percent for those ticipate at higher levels of education, whereas educa- with a tertiary education, compared with 95 percent tion has a negative correlation with male labor force for those with no education. By contrast, the highest participation. Figure 2.11. Labor force participation rate of prime-age individuals (25–54), by gender, location, and education, 2022 Participation rate (%) 41 Urban 47 63 79 Female 75 Rural 77 68 83 95 Urban 98 94 90 Male 96 Rural 97 95 92 0 10 20 30 40 50 60 70 80 90 100 None/NFE Primary Secondary Tertiary Source: Bhutan Labor Force Survey, 2022. Note: NFE = nonformal education. The important role that education plays in the labor force participation rate of prime-age women, especially in urban areas, and the small role it plays 12.  in prime-age men’s, is demonstrated in the OLS regression estimates. The estimates reported in tables B.3a and B.3b in appendix B show that, compared with those having no education, women with a primary education have a 7 percentage point higher probability of participating and women with a sec- ondary diploma an 11 percentage point higher probability of participating. Meanwhile, women with a tertiary diploma have a 16 percentage point higher probability of participating. For men, education matters less in the decision to participate in the labor market. The two are negatively correlated: all else being equal, the participation rate is lower for the most educated men. Figure B.10, panel b, in appendix B reports the predicted participation rates of men and women according to their location and diploma, based on the OLS estimates in tables B.3a and B.3b in appendix B. Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 23 Figure 2.12. Labor force participation rate, by Figure 2.13. Motives for inactivity, by gender gender, location, and marital status, 2022 and age group, 2022 Participation rate (%) Inactivity rate (%) 97 100 100 96 83 80 80 80 76 77 72 60 60 53 40 40 20 20 0 15-24 55-64 75+ 15-24 55-64 75+ 0 25-54 65-74 25-54 65-74 Urban Rural Urban Rural Female Male Female Male Studies Family duties Old age Single Married Other motives Illness/injury/disability Source: Bhutan Labor Force Survey, 2022. Source: Bhutan Labor Force Survey, 2022. The female labor force participation rate is negatively participation by 10 percentage points, and having a affected by the presence of children in the household. child ages 3–5 years decreases participation by 6 per- Figure 2.14 compares the participation rate of women centage points. during their fertility years, according to whether they have at least one child, at least one child under age The presence of dependent adults is positively asso- 6, children over age 6, or no children. Although all ciated with female labor force participation, likely women under age 45 have a lower labor force partic- because they may help care for younger dependents. ipation when they have at least one child under age The impact of the elderly on female and male participa- 6 (48 percent for the mothers ages 20–24 versus 70 tion is an open empirical question because the elderly percent for women of the same age without children), need care, but they can also provide care for children. this participation gap diminishes with age and disap- Figure 2.15 plots the participation rate of prime-age pears at age 45, suggesting that career and family are women and men in the presence of dependent adults— more compatible for older, more educated women that is, household members who report being inac- than for younger women. Mothers whose children tive because of their old age or because of an illness, have all reached school age are also less likely to par- injury, or disability. The participation rate of women ticipate in the labor market than women without any is significantly higher in households in which an adult children under age 15, albeit to a smaller extent. These is not working because of old age (76 percent versus 67 findings align with the OLS estimates in figure B.11 and percent) or because of an illness, injury, or disability table B.3a in appendix B, which show that having a (72 percent versus 67 percent). By contrast, the partici- child under age 2 decreases the probability of women’s pation of prime-age men is not affected by the present 24 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice of dependent adults. This finding is in line with the a woman enters the labor market. A 1.0 percentage OLS estimates in tables B.3a and B.3b in appendix B point increase in the inactivity rate is associated with showing that the presence of adults age 65 and over has a 1.2 percentage point lower probability of participat- no impact on men’s labor supply but can be linked to ing. More surprising, local wages (measured at the higher female participation (by 3.4 percentage points). dzongkhag-gender-education level) are negatively asso- ciated with female participation: a 10 percent increase Female labor force participation is influenced by the in the hourly wage is associated with a 1.3 percentage local norms in their households and communities. In point decrease in female participation. Possibly the appendix B, the OLS estimates from table B.3 plotted wages of men and women are correlated, and women in figure B.12 show that, everything else being equal, respond to men’s higher wages by withdrawing from the higher the share of working women in the house- the labor market. Alternatively, the decrease could be hold, the higher is their probability of participating attributed to women choosing activities based on their (that is, a 10 percentage point increase in the share is ability, which leads to both higher wages and lower associated with a 0.6 percentage point higher proba- participation rates. For men, what matters is the local bility of participation). In addition, women seem to be unemployment rate, which creates a discouragement responsive to the community norms on the participa- effect: for each percentage point increase in the local tion of women, even after conditioning for age, edu- unemployment rate, the inactivity rate increases by 1.4 cation, and region fixed effects. The higher the female percentage point. inactivity in an area, the lower is the probability that Figure 2.14. Female labor force participation, Figure 2.15. Labor force participation rate, by by presence of children, 2022 gender and dependent household members, 2022 Participation rate (%) 90 Participation rate (%) 100 95 94 95 80 80 76 72 70 67 60 60 40 50 20 40 20-24 25-29 30-34 35-39 40-44 45-49 Age group 0 At least one child under 6 Only children 6-15 Ill Old None Ill Old None No child under 15 Female Male Source: Bhutan Labor Force Survey, 2022. Source: Bhutan Labor Force Survey, 2022. Note: III = inactive due to illness/injury/disability. Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 25 The econometric results are consistent with find- young women, compared with one in seven young ings from a recent qualitative study of the constraints men (see figure B.13 in appendix B). The proportion of faced by youth and women in accessing employment NEET is also much higher in urban areas than in rural opportunities (Etang and Choki, forthcoming). From areas—15 percent of young people between the ages of focus group discussions and key informant inter- 15 and 24 are NEET in rural areas, compared with 24 views emerged the challenges and constraints to par- percent in urban areas. ticipating in the labor force and being employed. They included inadequate or mismatched education Among women ages 15–24, one in three is NEET systems, lack of career guidance, gender stereotyping because of family duties (figure 2.16). Women in this (cultural norms), and lack of adequate facilities and group are also twice as less likely as men to be waiting support for female employees. for academic results and job interviews or planning a business. As shown in figure 2.16, only 7 percent of men are NEET because of a lack of interest in working. The Young women are more likely than men to majority of men in this age group attribute their NEET be not in education, employment, or training status to being unemployed (51 percent). The second (NEET) due to household responsibilities and most important reason for being NEET is plans for local norms. migration (15 percent), which applies as well to a sub- stantial share of NEET women (12 percent). Inactivity The pronounced gap between men and women who rates increase sharply for both men and women after are NEET underscores the extent to which young age 65, with retirement facilitated by the presence of women are not building the skills they need to partic- other relatives in the household (see figures B.14 and ipate in the labor market in the future. Young women B.15 in appendix B). are more likely than men to be NEET—one in four Figure 2.16. Reasons for being not in education, employment, or training (NEET), by gender (ages 15–24), 2022 Share of NEET respondents (%) 60 51 50 41 40 33 30 20 15 12 9 10 7 7 5 4 4 4 2 2 3 1 0 Female Male Unemployment Waiting for results Waiting for job/planning business Family duties Illness/injury/disability Not interested/no need Planning migration Other Source: Bhutan Labor Force Survey, 2022. 26 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 2.3 Unemployment among youth The duration of unemployment among the youth is and educated workers in 2022 similar to that of prime-age workers. According to figure 2.18, similar shares of young and prime-age workers remain unemployed for one to five months, Young and educated workers suffer from while for a higher share of prime-age workers long- unemployment rates that exceed the term unemployment extends to more than two years. national average. The short duration of unemployment could stem from the fact that unemployment increased sharply during Prior to 2019, young workers between the ages of 15 the COVID-19 pandemic, and the average duration of and 24 faced a higher unemployment rate than other unemployment decreased. groups, and their unemployment rate continued to climb after 2019, as shown in figure B.9, panel a, in A skill mismatch is the main reason 46 percent of appendix B. The rate climbed 11 percentage points young job-seekers believe they are unable to find between 2019 and 2020 and by 8 percentage points employment (figure 2.19). Recent graduation is the between 2020 and 2022. The overall unemployment second most-cited reason, which is consistent with rate reached 6 percent in 2022 (figure 2.17). Unemploy- the short duration of unemployment. The reasons for ment in the following groups exceeded the national unemployment evolved with the COVID-19 pandemic. average: urban women (15 percent), youth (37 percent During the pandemic, the mismatch in skills cited as among those ages 15–19 and 27 percent among those the main reason for employment (47 percent in 2019) ages 20–24), and those with a secondary education and decreased by 10 percentage points, while unemploy- above (12 percent). ment due to termination of a contract (5 percent in 2019) reached 25 percent in 2020–21 (see table B.2 in appendix B). Figure 2.17. Unemployment rate, overall and by age, education, location, and gender, 2022 Unemployment rate (%) Overall 6 15-19 37 20-24 27 25+ 3 No education 1 Primary 3 Secondary 11 Tertiary 12 Rural women 4 Rural men 3 Urban women 15 Urban men 7 0 5 10 15 20 25 30 35 40 Source: Bhutan Labor Force Survey, 2022. Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 27 Figure 2.18. Unemployment duration, 2021 Figure 2.19. Self-reported reason for unemployment, youth and nonyouth, 2021 Share of workers (%) Share of job-seekers (%) 40 50 46 34 39 40 38 30 27 23 30 22 21 25 19 20 17 14 20 12 11 13 11 10 10 6 5 5 4 4 2 0 0 Unemployed (15-24) Unemployed (25-54) Unemployed (15-24) Unemployed (25-54) <1 month 1-5 months 6-11 months Skill mismatch Recent graduation Terminated 12-23 months 2 years+ Resigned Family Other Source: Bhutan Labor Force Survey, 2022. Source: Bhutan Labor Force Survey, 2022. A skill mismatch affected more educated job-seekers level of education, such as services and sales, craft and because those employed are mostly low- to mid-skilled related trades, and elementary occupations. workers. Figure B.16 in appendix B displays for 2022 the share of total employment and unemployment of each educational group, as well as a breakdown of the Although there is no evidence that unem- inactive population ages 25–64 by educational group. ployed workers have a high reservation wage Fifty-eight percent of job-seekers have a secondary (the lowest wage at which a worker is willing degree, but, compared with the profile of employed to accept a particular type of job), the major- workers (and the inactive population), there is a clear ity prefer public sector jobs and have no supply shortage of uneducated job-seekers and an positive view of private sector employment. In addition, a sizable number have migrated oversupply of university-graduated job-seekers. For abroad or plan to. example, 10 percent of job-seekers have no educa- tion, compared with 37 percent of employed workers, and 24 percent of job-seekers are university gradu- Most workers prefer public sector employment, but a ates, compared with 9 percent of employed workers. higher share of youth (86 percent) prefer it than non- Chapter 3 provides a detailed discussion of the youth (75 percent)—see figure 2.20. The reasons for expected labor demand and notes that it is mostly con- preferring to work in the public sector are related centrated in occupations requiring a low to medium to job security and working conditions (figure 2.21). Eighty-three percent of youth prioritize job security, 28 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice compared with 74 percent of prime-age workers, High reservation wages or unrealistic expectations which could imply that queuing for public sector about the wage unemployed individuals could earn do employment is one of the reasons for unemployment. not explain unemployment because the average res- Working conditions are the second priority for both ervation wage is significantly lower than the observed age groups, and wages come third.13 The reputation of monthly wage of workers in similar age and educa- the public sector is also a critical factor, indicating that tion groups. Figure 2.22 shows that the average wage a social stigma is attached to working in other sectors expected by the unemployed is from 63 to 91 percent (UNDP 2022a). Among youth and nonyouth, the reason of the average wage of paid workers in the same age for preferring private sector employment is related to and education group. The only exception is young sec- personal interests, and few highlight other aspects of ondary graduates, whose reservation wage is signifi- the job (such as reputation, job security, wages, and cantly higher (855 Nu) than the real observed wage for working conditions) as the main reason for choosing workers with the same education. These unrealistic employment in the private sector, thereby suggest- wage expectations may explain in part why this partic- ing that private sector employment may be negatively ular age and education group has such a high unem- viewed by job-seekers. ployment rate (11 percent in 2022). Figure 2.20. Sector preference of job-seekers, Figure 2.21. Reasons for sector preference, by by age group, 2021 age group and sector, 2021 Share of workers (%) Share of workers (%) 100 100 86 83 80 77 74 80 75 68 60 51 51 60 40 38 40 37 32 3129 40 2624 23 20 20 10 10 19 4 20 0 10 0 5 Unemployed Unemployed Unemployed Unemployed 3 1 2 (15-24) (25-54) (15-24) (25-54) 0 Unemployed (15-24) Unemployed (25-54) Public service Private business Public service Private business Salary Working conditions Reputation Public company Agriculture Personal interests Job security Source: Bhutan Labor Force Survey, 2021. Source: Bhutan Labor Force Survey, 2021. This finding is consistent with that described later in this chapter that the wage differential is not large between the public and private sector (for women, 13.  it is negative). Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 29 Figure 2.22. Average real monthly wage and reservation wage, by education and age, 2022 a. Youth b. Prime-age workers Average real monthly wage for employed or reservation wage Average real monthly wage for employed or reservation wage for unemployed (Nu) for unemployed (Nu) None/NFE None/NFE Employed (15-24) 7,860 Employed (25-54) 7,717 Unemployed (15-24) 5,668 Unemployed (25-54) 6,992 Employed (15-24) 8,005 Employed (25-54) 9,151 Primary Primary Unemployed (15-24) 5,070 Unemployed (25-54) 7,600 Secondary Secondary Employed (15-24) 7,411 Employed (25-54) 10,693 Unemployed (15-24) 8,266 Unemployed (25-54) 8,443 Employed (15-24) 13,109 Employed (25-54) 16,416 Tertiary Unemployed (15-24) 11,664 Tertiary Unemployed (25-54) 12,399 0 5,000 10,000 15,000 20,000 0 5,000 10,000 15,000 20,000 Source: Bhutan Labor Force Survey, 2021. Note: NFE = nonformal education. Evidence suggests that a high share of the unemployed 2.4 Current state of employment have migrated in recent months, driven by both pull and push factors. As mentioned earlier, about 5,000 Bhutanese migrated monthly in the first part of 2023, Women, youth, and educated workers have and many workers who are NEET have plans to migrate. employment rates below the national aver- Australia is by far the most frequent destination. age, highlighting a high degree of underuti- Between January 1, 2018, and March 22, 2023, 13,583 lization of their human capital in the labor Bhutanese left for Australia through Paro International market. Airport. Since then, the monthly numbers have surged continually (Kuensel 2023). Many pull factors appear Urban women, rural women, youth, and workers with to be behind migration, such as higher incomes, better secondary education have employment rates below economic opportunities, greater financial security, the national average. Figure 2.23 shows that in 2022, and higher living standards in destination countries. although the employment rate (share of the work- Job insecurity, the negative impacts of reform, and the ing-age population employed) reached 59 percent, the high cost of living appear to be among the push factors following groups had lower employment rates: rural (Kuensel 2023). women (55 percent), urban women (40 percent), sec- ondary-educated workers (49 percent), and youth (39 percent of those ages 20–24). 30 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice Figure 2.23. Employment rate, overall and by age, education, location, and gender, 2022 Employment rate (%) Overall 59 15-19 5 20-24 39 25+ 69 No education 64 Primary 78 Secondary 49 Tertiary 61 Rural women 55 Rural men 71 Urban women 40 Urban men 68 0 10 20 30 40 50 60 70 80 90 Source: Bhutan Labor Force Survey, 2022. The labor market in Bhutan is mostly dom- The labor market is segmented by education levels. inated by agricultural and public sector Workers with no education dominate agricultural employment, and this dichotomy persists employment, and workers with a tertiary educa- along the lines of gender, location, and tion choose the public sector. Thirty-seven percent of education. employed workers in Bhutan have no formal school- ing (figure 2.24, panel b)—a finding consistent with Female, rural, and low-skilled workers are more likely the dominance of agricultural employment, where to work in agriculture as self-employed or family over 70 percent of workers have no formal school- workers, whereas male, urban, and high-skilled ing. Although the majority of workers with a tertiary workers opt for public sector employment. According education choose employment in the public sector, a to figure 2.24, panel a, 35 percent of male workers are high incidence of tertiary- and secondary-educated in the agricultural sector, and a significant share also workers also are private sector employees, employers, works in the public sector (31 percent). Meanwhile, self-employed, or family workers in the nonagricul- over 50 percent of women work in the agricultural tural sector. sector, with the majority serving as family and self-em- ployed workers. A much smaller share of women than The private sector hires workers in relatively more men work in the public sector (18 percent). In terms low- and middle-paying occupations than the public of location, 45 percent of urban workers are in the sector. Employers are mostly in high-skilled occupa- public sector, followed by 25 percent in the private tions, such as managers. As shown in figure 2.24, panel sector (figure 2.24, panel a). Self-employed workers in c, employees tend to access higher-paid occupations nonagricultural sectors make up 20 percent of urban when they work in the public sector. The proportion of workers. By contrast, over 60 percent of employment elementary occupations is almost twice as high in the in rural areas is in the agricultural sector as self-em- private sector as in the public sector (19 percent versus ployed workers or as family helpers. 12 percent). Meanwhile, 25 percent of employees in Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 31 the public sector are professionals, compared with 10 health and social work activities, and education. Public percent in the private sector. sector employees also work in state-owned enterprises for electricity, gas, steam, and air-conditioning, as well Self-employed workers in the nonagricultural sectors as in water supply, sewerage, waste management, and are mostly service and sales workers (41 percent), remediation activities, and in financial and real estate followed by workers in craft and related trades (22 activities. By contrast, private sector employment percent). More specifically, self-employment outside dominates construction (60 percent), administra- agriculture is mostly in wholesale and retail trade and tive and support services (hereafter “Administration, repair of motor vehicles (71 percent); transportation 60 percent), and arts, entertainment, recreation, and and storage (65 percent); and accommodation and other services (66 percent). food services (59 percent). In the remaining sectors— manufacturing, information and communication, as Despite the prevalence of public sector employment well as professional, scientific and technical activities— and the preference of youth for public sector jobs, the distribution of employment types is more uniform. apparently a large number of resignations among For example, within professional, scientific, and tech- public servants are linked to outmigration. Kuensel nical activities, 22 percent of workers are self-em- (2023) reports that voluntary resignations rose from an ployed, 45 percent are employees from the private average of 64 civil servants per month between January sector, and 33 percent are employees from the public 2015 and May 2022 to 234 per month between June sector. 2022 and February 2023. Several push factors related to the outmigration of public servants may be linked Public employment and private sector employment to lack of career progression, a poor system or poor dominate different industries. As shown in figure working environment, a lack of recognition, a greater 2.24, panel d, public employment dominates in public workload due to diminished staffing, and the negative administration and defense (hereafter “Public”) human impacts of government reform. 32 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice Figure 2.24. Structure of employment, 2022 a. Type of employment, by gender and b. Type of employment, by education level location Percent Percent Total None/NFE Rural Primary Urban Secondary Women Tertiary Men Monastic 0 20 40 60 80 100 0 20 40 60 80 100 Self-employed (nonagric.) Self-employed (agric.) Self-employed (nonagric.) Self-employed (agric.) Family worker (nonagric.) Family worker (agric.) Family worker (nonagric.) Family worker (agric.) Salaried worker (public) Salaried worker (private) Salaried worker (public) Salaried worker (private) Employer Employer Source: Bhutan Labor Force Survey, 2022. Source: Bhutan Labor Force Survey, 2022. c. Occupation, by type of employment d. Type of employment, by sector (excluding (excluding agriculture) agriculture) Percent Percent 100 100 80 80 60 60 40 40 20 20 0 Arts/other services Construction Hotels/restaurants Energy/water Finance/real estate Science Transportation Health Education Trade Manufacturing Administration Public Information/comm. 0 Self-employed Salaried Salaried Employer worker worker (public) (private) Army Managers Professionals Technicians Clerical Service/sales Self-employed Salaried worker (public) Craft Operators Elementary Salaried worker (private) Employer Source: Bhutan Labor Force Survey, 2022. Source: Bhutan Labor Force Survey, 2022. Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 33 Structural transformation has been slow in The agriculture and public administration sectors have Bhutan because the share of low-productivity lower than average labor productivity, and although agriculture in total employment remained the more productive sectors experienced rapid high between 2013 and 2022. employment growth between 2013 and 2022, they remained very small. By merging sectoral employment Between 2013 and 2022, the structure of employment from the BLFS with the most recent data from the remained quite stable, with agriculture and public national accounts, figure 2.26 plots how the relative administration accounting for the largest share of growth of 12 aggregated sectors between 2013 and 2021 employment. As shown in figure 2.25, which plots the correlates with their productivity in 2021. Employment share of each sector in total employment in 2013 and in agriculture clearly has negative labor productivity. 2022 with its growth rate, the sectors providing most of With the exception of finance, insurance, real estate, the jobs were agriculture, forestry, mining and quar- and other business services, the share of the most pro- rying (44 percent in 2022 and 57 percent in 2013), and ductive sectors in total employment grew after 2013. public administration and defense (12 percent in 2022 However, they remain too small to meaningfully alter and 9 percent in 2013). the employment landscape in Bhutan. Figure 2.25. Average real monthly wage and reservation wage, by education and age, 2022 a. Share, 2013 and 2022 b. Share growth rate, 2013–22 Share of total employment (%) Change in share of total employment (% 2013-22) Agriculture/mining 57 Agriculture/mining -22 44 6 Manufacturing 7 Manufacturing 12 1 Energy/water Energy/water 15 1 3 Construction 6 Construction 100 8 Trade 10 Trade 22 2 Transportation 3 Transportation 30 3 Hotels/restaurants 3 Hotels/restaurants 6 Information/comm. 1 Information/comm. 43 1 Finance/real estate 1 Finance/real estate 53 2 Science 0.1 Science 469 0.4 1 Administration 1 Administration -46 9 27 Public 12 Public 4 Education 6 Education 45 2 Health 2 Health 17 2 Arts/other services 3 Arts/other services 44 0 10 20 30 40 50 60 -50 0 50 150 250 350 450 2013 2022 Source: Bhutan Labor Force Survey, 2013–22. 34 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice Figure 2.26. Relative employment growth and productivity, 2013–21 3 Electricity/water Log (sector productivity/total productivity) 2 Finance/insurance/real estate Transportation 1 Mining/quarrying Wholesale/retail Public administration Manufacturing Construction 0 -50 -25 0 25 50 75 100 125 150 Education/health -1 Hotels/restaurants Agriculture -2 Other services -3 Change in employment share (Percent, 2013-21) Sources: Bhutan Labor Force Survey, 2021, and National Accounts Statistics (NAS) 2022. Note: The size of each circle reflects sector employment in 2013. Productivity is measured as the sector-level value added per worker in 2022. Improving agricultural productivity can acceler- 2.5 Quality of employment: Hours ate structural transformation in Bhutan because the worked, wages, informality, and a adoption of labor-saving technologies in agriculture job quality index can free excess labor for nonagricultural enterprises in the services and industry sectors. However, gaps remain large and stem primarily from inadequate Poor quality of employment, measured by access to irrigation, crop damages, labor shortages, working long hours (more than 48 a week) difficult transport and export logistics, and challeng- and underemployment, is prevalent mostly ing topography (World Bank, forthcoming a). Esti- in rural areas and among workers with low mates show that addressing issues related to irrigation education. could raise average yields for paddy, rice, and maize crops by about 6 percent each, and for cardamom by In 2022, over 62 percent of Bhutan’s workforce 11 percent. Protecting crops from damage could raise reported working more than 48 hours a week, and maize and paddy yields by 15 and 9 percent, respec- they are mainly those with the least education (67 tively. Removing the impact of labor shortages—for percent are not educated, compared with 42 percent example, through the adoption of labor-saving tech- of tertiary graduates). Regardless of the demographic nology—could increase paddy, cardamom, and areca profile of the worker, the primary self-reported reason nut yields significantly. for overworking is that the job requires it (see figure B.17 in appendix B). However, working longer hours Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 35 for additional income is more prevalent among indi- among workers in construction, wholesale and retail viduals with low education and those living in rural trade, transportation, and accommodation and food areas, who are likely to have a secondary occupation. services. Figure 2.27 shows the average number of hours The probability of working overtime to earn more also worked by demographic characteristics, economic increases with age, suggesting the presence of subsis- sector, employment type, and gender. The figure also tence issues for the working elderly. presents the dispersion of hours worked around the mean (that is, the extent to which the number of hours Underemployment (defined as working less than 35 worked varies within a given group), capturing the hours a week) is low overall in Bhutan. Women and extent of both under- and overemployment. Although workers with low levels of education are more likely demographic characteristics have little impact on the to work part time. Part-time employment accounts for average number of hours worked or their disper- 6 percent of all employment in Bhutan, but its preva- sion (figure 2.30, panel a), job characteristics matter. lence is higher among women (8 percent) than men (4 Workers in the financial, insurance, and real estate percent) as shown in figure B.18 in appendix B. In addi- sector and the health sector have the shortest work- tion, it is higher among workers with no education (7 week (45 and 47 hours, respectively) and a below-av- percent) than among those with a tertiary education (2 erage dispersion, suggesting that a short workweek is percent). Part-time work allows the elderly to remain the norm in these sectors. In other public-dominated active, accounting for 13 percent of total employment sectors, the workweeks may be longer, but are also con- for the 65+ age group. Part-time work is also more centrated around their average (50 hours in education, prevalent in rural areas. 52 hours in administrative and support service activ- ities, 52 hours in public administration and defense), Underemployment may not be a preference for some suggesting that under- and overemployment are not groups of workers and could likely reflect limited frequent. The workweeks are much more heteroge- employment opportunities. One out of 20 individu- neous across workers in manufacturing, wholesale and als working part-time reported being unhappy with retail trade, construction, and accommodation and the number of hours worked and would prefer to food services. Although in manufacturing, the weekly work more hours (figure B.19 in appendix B). The level number of hours is below average, the workweek is of dissatisfaction with hours worked is particularly far longer for workers in other sectors, ranging from important for men and for workers in urban areas. 56 to a maximum of 65 in wholesale and retail trade Individuals with a primary or secondary education are and accommodation and food services. The common more likely to work part-time than individuals with a feature of these sectors is that the dispersion in the tertiary diploma (figure B.19, panel b, in appendix B) number of hours worked is very large, suggesting the and also are more likely to want to work more hours, presence of under- and overemployment. suggesting that a part-time job has been imposed on rather than chosen by the least educated. These important sectoral differences are driven by their employment composition. With 51 hours worked (and a standard deviation of 12.5 hours), salaried Overwork is more prevalent in industries workers in the public sector have the shortest and most dominated by self-employment or private homogeneous workweek, while at the other extreme, businesses than in the public sector. nonagricultural self-employed and family workers work, on average, eight hours more (with a standard Underemployment dominates manufacturing, deviation of more than 20 hours). These differences whereas underemployment and overwork occur across job types are exacerbated for women, as shown 36 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice in figure 2.27, panel d. Compared with those for men, and the distribution of hours for salaried women is the working hours of women who are family workers even more concentrated around a lower average work- or self-employed workers are even more dispersed, week of 49 hours. Figure 2.27. Number of hours worked and standard deviations, by demographic characteristics, economic sector, employment type, and gender, 2022 a. By demographic characteristics b. By economic sector Hours worked per week Hours worked per week 60 70 50 60 50 40 40 30 30 20 20 10 10 0 Agriculture/mining Manufacturing Energy/water Construction Trade Transportation Hotels/restaurants Finance/real estate Science Administration Information/comm. Public Education Health Arts/other services 0 Urban men Total 15-19 20-24 25+ No education Primary Secondary Tertiary Rural women Rural men Urban women Standard deviation Standard deviation Source: Bhutan Labor Force Survey, 2022. Source: Bhutan Labor Force Survey, 2022. c. By employment type d. By employment type and gender Hours worked per week Hours worked per week 60 60 50 50 40 40 30 30 20 20 10 10 0 0 Total Family worker (nonagric.) Family worker (agric.) Salaried worker (public) Salaried worker (private) Employer Self-employed (nonagric.) Self-employed (agric.) Total Self-employed (nonagric.) Family worker (nonagric.) Family worker (agric.) Salaried worker (public) Salaried worker (private) Employer Self-employed (agric.) Standard deviation Standard deviation Female Male Source: Bhutan Labor Force Survey, 2022. Source: Bhutan Labor Force Survey, 2022. Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 37 Wages increase sharply with experience and to age and location. When other characteristics such as education, particularly for women and urban education are held constant, location does not matter workers, suggesting that human capital for the hourly wages of women and only slightly affects translates into better-quality employment. the wages of rural men. For both men and women, returns to age, which is a proxy for work experience, Older female and male workers and urban male are nonlinear. Figure B.20 in appendix B plots the pre- workers are more likely to have quality employment, dicted log hourly wage of men and women by age. The as measured by renumeration in the labor market. average working woman starts with a lower hourly Figure 2.28 shows that, on average, workers over age wage than the average young man, but her returns to 25 earn 27 percent more per month than those ages experience are higher. Another possibility is that only 15–19. Urban women and men earn 65 percent and 47 the best-paid women remain in the labor market, so percent more, respectively, than their rural counter- that by the age of 45 the average working woman earns parts. Table B.4 in appendix B presents an OLS model almost the same hourly wage as the average working on the determinants of real hourly wages for men and man. for women, thereby allowing examination of returns Figure 2.28. Average monthly and hourly wages, overall and by education, gender, and area, 2022 a. Average monthly wage b. Average hourly wage Average monthly wage (real 2010 Nu) Average hourly wage (real 2010 Nu) Overall 9,838 Overall 48 15-19 7,841 15-19 34 20-24 8,288 20-24 40 25+ 9,951 25+ 48 No education 7,185 No education 34 Primary 8,744 Primary 41 Secondary 10,276 Secondary 49 Tertiary 16,454 Tertiary 84 Rural women 6,907 Rural women 33 Rural men 8,707 Rural men 42 Urban women 11,427 Urban women 57 Urban men 12,844 Urban men 62 0 10,000 20,000 0 20 40 60 80 100 Source: Bhutan Labor Force Survey, 2022. 38 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice Returns to education are very high in Bhutan for both Wage discrepancies among some groups of men and women. In appendix B, figure B.21 uses the workers are not immediately explained by point estimates in table B.4 and reports three find- human capital characteristics, which may ings. First, returns to tertiary education for both men further deepen inequalities in the labor and women are higher than returns for secondary market. and primary education. Second, returns to educa- tion are somewhat higher for educated women than Even conditioning for education, age, occupation, and they are for educated men. For example, tertiary-ed- locality, some industries pay better than others. Figure ucated men receive 46–71 percent more in wages than B.23 in appendix B shows the raw hourly wage gap for male workers with no education, whereas the range is the manufacturing sector, together with the wage gap higher for tertiary-educated women, who are likely to estimated by the OLS regression in tables B.4a and B.4b earn 46–94 percent more in wages than female workers in appendix B. For men, the large pay gaps observed with no education. Finally, controlling for occupations when comparing hourly wages in manufacturing with (specification 4) divides the returns to education by a wages in science, information, finance, real estate, and third, suggesting that the most educated tend to opt education disappear when controlling for education, for the best-paid occupations. age, localization, and occupation. However, a large unexplained gap appears between workers in man- Wage gaps across occupations within the same sector ufacturing and workers in accommodation and food illustrate the importance of other types of skills beyond services (–40 percent), as well as workers in arts or education, age, and location. Figure B.22 in appendix other services (–29 percent). On the other hand, com- B shows that the raw pay differences between services pared with their wages in the manufacturing sector, and sales workers and workers in other occupations men are relatively better paid in the energy and water ranged from 0 to 88 percent for men and –24 percent to sectors (+22 percent) and in the construction sector 77 percent for women, which likely reflects the skill dif- (+20 percent). Women working in the manufacturing ferential and worker self-selection into various occu- sector receive a significantly lower hourly wage than pations. Once conditioned on similar demographic those working in most of the sectors (the wage penalty (age, education, and location) and industry character- ranges from 30 to 54 percent), except for transporta- istics, pay differences between occupations decrease, tion, accommodation and food services, science, and but some gaps remain. For the same age and education, administration, where the hourly wage is statistically managers earn 29–52 percent more and professionals similar. 38–44 percent more than services and sales workers in the same industry. This finding illustrates how other Women are also less likely to receive wages comparable types of skills beyond education matter, giving rise to to those of men with similar characteristics. This gap compensating wage differentials.14 is mostly driven by the differential impact of marriage on men and women. The raw wage gap between men and women of working age is 12 percent (excluding family workers). According to table B.5 in appendix B, women and men have different productive character- istics and are oriented toward different types of jobs. These pay gaps between industries and occupations persist in sign and magnitude after controlling for the sector of employment (public/private). All else 14.  being equal, and notably within a given industry and occupation, women earn, on average, 12 percent less per hour in the private sector than in the public sector. By contrast, men working in the private sector earn 13 percent more per hour than in the public sector. Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 39 Based on the Blinder-Oaxaca decomposition method- the gender-differentiated impact of marriage on labor ology described in box 2.1, the first panel in table B.5 in outcomes. All else being equal, the gender pay gap is appendix B shows that differences in worker charac- 8 percentage points lower for unmarried women and teristics contribute in part to the gender gap in wages. men than married women and men. This is further For example, working women elect to work in slightly evidence of the high constraints and expectations that better-paid locations, which reduces the gender gap in married women and men face in the work and family wages. If these women lived in the same places as men, sphere.15 the wage gap would be higher. The observed differ- ences in education between women and men also help Workers in state-owned enterprises are more likely to mitigate a gender wage gap, which would be higher if receive higher wages than workers with similar char- both had the same education level. Differences in living acteristics in the private sector. According to 2022 data, arrangements (marriage, household composition) the raw hourly wage gap between the public sector partly explain the observed wage gap. Finally, it seems (including SoEs) and private sector is 24 percent in that working women choose to work in the worse-pay- favor of the public sector. Using the same Blinder-Oax- ing industries (setting the industry choice of women aca decomposition methodology, the gap is fully to that of men would lead to a decrease in the gender explained when taking into account gender and age, wage gap of 6.6 percentage points), but in slightly bet- but, even more important, education and occupations. ter-paid occupations than men (setting occupation However, in a comparison of the wages of workers in choice of women to that of men would lead to a 2.4 private companies and those of workers in SoEs (where percentage point increase in the explained gender jobs are possibly the most comparable), the wage gap gap). However, three-quarters of the gross gender is much higher, 38 percent. Half of the gap remains wage gap is unexplained because of the differing unexplained after taking into account age, gender, returns to worker characteristics of men and women. location, and education. The decomposition suggests Most of the unexplained portion is, in fact, driven by that differences across workers in terms of education Box 2.1. Blinder-Oaxaca decomposition methodology The Blinder-Oaxaca decomposition is a statistical econometric method for explaining observed wage differences between two groups (here, between working-age Bhutanese women and men, as well as between workers in state-owned enterprises and private sector workers). The observed average differ- ence in the hourly wage is decomposed into a part that can be explained by differences in character- istics, with the remainder due to differences in returns to characteristics (the unexplained part). For example, Bhutanese women have a significantly lower employment rate than men and earn signifi- cantly less. Do they have less access to employment because they are less educated than men? Or do they have different returns for the same characteristics—for example, does having young children penalize them differently in accessing employment? This finding is consistent with the employment gap between men and women. In 2022, the employment rate of working-age women was 21 points lower 15.  than men’s, a difference that cannot be attributed to any observed difference in characteristics and is essentially driven by the differential employment impact of marriage on women and men. Table B.5 in appendix B shows that most of the employment gap is, in fact, driven by the gender-differentiated impact of marriage on employment, which accounts for two-thirds of the gender employment gap. 40 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice and occupation account for half of the gap, whereas monthly salary of Nu 40,600 of a civil servant at the differences in returns to these characteristics explain same level (Zangpo 2024). the other half of the gap (see appendix C for a detailed discussion). Although informal employment is wide- Recent salary hikes announced for public sector and spread, mostly in manufacturing, con- SoE employees will likely further exacerbate the pay struction, wholesale and retail trade, and gap with private sector employees. In July 2023, a pay accommodation and food services, the infor- revision for civil servants was announced to compen- mality premium applies only to those with sate for the rising cost of living. The announced pay low education.16 increase was between 55 and 74 percent. In October 2023, SoEs also announced salary increases of between Informal employment accounted for a considerable 46 and 72 percent of the minimum basic pay. With the share of total employment in 2017. As shown in figure revision, a fresh graduate in grade 8 working at one of 2.29, 11 percent of all workers are employed without a Bhutan’s SoEs (Druk Holding and Investment and its written contract—that is, about 32 percent of all workers companies, DHI) would receive a gross monthly salary in companies, businesses, and nongovernmental orga- of Nu 46,908, which is Nu 6,308 higher than the gross nizations (NGOs)17 or 67 percent of employees when excluding the public sector and armed forces. Figure 2.29. Distribution of total employment types, by location, 2022 Share of workers (%) 80 77 60 51 40 32 28 23 20 18 14 12 11 11 6 4 6 4 2 0 Total Rural Urban Formal employees in companies, businesses, NGOs Informal employees in companies, businesses, NGOs Self-employed/family workers in companies, businesses, NGOs Public services and army Agrifarming Source: Bhutan Labor Force Survey, 2022. Note: NGOs = nongovernmental organizations. 16. Informal employment is defined as wage employment without a written contract. In the BLFS, this question is posed only to employees in public and private companies, private businesses, or NGOs, and so the share of informal employment is computed either over total employment or over employment in the same type of enterprises. BLFS 2017 was the last BLFS wave to collect data on informal work. The Bhutan 2022 Establishment Survey includes an employee module, and, based on another definition of the share of employees without a provident 17.  (retirement) fund, the incidence of informality is 43 percent. Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 41 The share of informal employment in total employ- contract are less educated—for example, 23 percent ment is much higher in urban areas (23 percent) than do not have a degree, compared with 16 percent of in rural areas (4 percent), but the share of informal formal employees, and only 13 percent have a tertiary employment among wage earners in companies, busi- diploma, compared with 19 percent of formal employ- nesses, and NGOs is somewhat higher in rural loca- ees. Also, wage earners without a written contract are tions (70 percent) than in cities (66 percent). slightly more likely (5 percentage points) to live in the poorest households.18 Informal wage earners and formal wage earners working in the same type of enterprises (that is, com- Informal wage earners and formal wage earners have panies, businesses, and NGOs) are similar. Table B.6 in very similar jobs in terms of hours worked and hourly appendix B shows that much like formal wage earners, wage. As shown in figure 2.30, the average number of 67 percent of informal wage earners are men; 25 weekly hours for formal employees is 49 versus 50 for percent are under age 25; 75 percent live in an urban informal employees, and the average hourly wage is area; and 41 percent operate in the Thimphu region. also very close (55 Nu and 53 Nu an hour in real 2010 The only significant difference between wage earners Nu, respectively). The hours and wage distributions who are formally employed and informally employed are also very similar. is their diploma. Wage earners without a written Figure 2.30. Weekly working hours and real hourly wages of employees in companies, businesses, or nongovernmental organizations, by formality status, 2017 Share of workers (%) Share of workers (%) 60 25 50 20 40 15 30 10 20 5 10 0 0 10 20 30 40 50 60 70 80 90 100 10 30 50 70 90 110 130 150 170 Hours worked per week Wage per hour Formal employees Informal employees Formal employees Informal employees Source: Bhutan Labor Force Survey, 2017. Table B.7 in appendix B presents the determinants of having an unwritten contract, along with the determinants of not having provident funds, which is 18.  how informality can be determined using the 2022 Establishment Survey (ES). In the table, columns (1)–(3) rely on the 2017 BLFS, and column (4) relies on the 2022 ES with similar controls. The determinants of having a written contract are not the same as the determinants of having a provident fund—edu- cation, location, and age matter much more in the access to a provident fund than to a written contract. The results obtained using the 2022 ES and the results obtained with the 2017 BLFS are quite comparable, as evident in columns (3) and (4), notably regarding the negative effect of education on the likelihood of not having a provident fund. The magnitude of the effect of education is much higher when using the BLFS than when using the ES, which can be explained by the fact that ES contains only registered firms, where the least educated have a greater chance of benefiting from a provident fund. 42 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice The informality premium exists only for unedu- that 14 percent of formal employees do not receive any cated workers, who earn 23 percent more when they benefits, and 27 percent of informal employees report do not have a written contract. Column (1) in table having no access to employment benefits. The most B.7 in appendix B confirms that there are no differ- frequent benefit for informal employees is access to a ences in the hourly wage when comparing employees leave system or a weekly rest period, followed by over- from private firms with or without a written con- time payment and access to a provident fund. Access to tract. Column (2) indicates that this finding cannot be compensation for a work-related injury, disability, or explained by a different distribution of informal and death is the benefit for which the gap between employ- formal jobs in location in which informal jobs would ment statuses is the largest (19 percent for infor- be concentrated in urban areas, which also happen mal employees versus 42 percent for formal ones). In to pay more. Column (3) highlights that only the least figure 2.31, overall 28 percent of formal employees educated benefit from an informality premium of 22 have access to all the benefits listed, but none of the percent on average, which allows them to reach almost informal employees do. the same wage rate as formal or informal workers with a primary school degree.19 Column (4) controls for industry and occupation fixed effects to show that the Figure 2.31. Formal and informal jobs and heterogeneous effect of informality on wages for the attached benefits, 2017 uneducated cannot be explained by the fact that infor- Share of employees (%) mal workers would move into industries or occupa- 80 tions that also pay more. 74 73 Taken together, manufacturing, construction, whole- 60 56 56 55 sale and retail trade, and accommodation and food ser- 52 vices account for 55 percent of the unwritten contracts 42 42 in 2017 in Bhutan. As shown in table B.8 in appendix B, 39 40 40 informal workers are 2.5 percentage points more likely to work in wholesale and retail trade or in arts and 27 28 28 other services than formal workers in the same types 19 20 of jobs and are 2 percentage points less likely to work 14 in the education sector. Also, informal contracts are more likely to apply to the unskilled. Professionals and 0 0 technicians are underrepresented and craft workers Informal Formal and plant operators are overrepresented in the pool of None Compensation for workplace injury Gratuity workers without a written contract. Provident fund Overtime payments Weekly rest period Leave system All Finally, written employment contracts tend to come with different types of job benefits, but this relation- ship is far from deterministic. Figure 2.31 indicates Source: Bhutan Labor Force Survey, 2017. The marginal effect of informality on the wage is nonlinear. Derivation of the econometric specification in table B.7 in appendix B indicates that the 19.  marginal effects of being informal for each educational level are as follows: uneducated (ref. category): 23 percent (informality premium); primary: 22–17 percent, ~0 effect (not significant); secondary: 23–20 percent, ~0 effect (not significant); tertiary: 23–16 percent, ~0 effect (not significant). Informality matters only for the uneducated; the premium is offset for the rest. Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 43 Aggregating many of the dimensions asso- workers. Although the gender differences are not too ciated with quality of employment into one wide, male workers enjoyed better job quality than index presents consistent results on the wid- female workers. ening gaps between urban and rural areas and between male and female workers. 2.6 Internal mobility Job quality seems to have increased over time in recent years and stalled in 2022, with urban and male workers exhibiting better job quality than their rural and Internal mobility in the labor market is prev- female counterparts. The job quality index (JQI) is used alent and can support workers in finding to evaluate the job quality associated with the current better-quality employment. week’s principal activity based on the number of hours in the current week from BLFS (see box 2.2). Figure 2.32 Over one-third of the working-age population born suggests that job quality increased from 2018 through in Bhutan has lived at least once in a dzongkhag other 2021 and remained the same in 2022. Workers in urban than their current dzongkhag of residence. This pro- areas appear to have had better job quality than rural portion is highest in urban areas, where more than Figure 2.32. Job quality index, by location and gender, 2018–22 Index 3.5 3.3 3.1 2.9 2.9 3.0 3.0 3.0 2.9 3.0 2.8 2.8 2.8 2.7 2.6 2.7 2.6 2.7 2.7 2.6 2.5 2.5 2.5 2.0 1.5 1.0 0.5 0 National Urban Rural Female Male 2018 2019 2021 2022 Source: World Bank, forthcoming b (based on BLFS). Box 2.2. Job quality index—dimensions and methodology The job quality index (JQI) is defined for the principal activity in the current week. It has the following dimensions: (1) benefit: the Bhutan government offers free health insurance countrywide; (2) satisfac- tion: the worker did not work excessive hours (more than 48 hours per week in the principal activity), does not have a second paid job, and either works full-time (less than 40 hours a week in the principal activity) or works part-time but does not want to work more; (3) stability: the worker has been working in the occupation for more than three full years. Income is compared with the US$3.65 per day inter- national poverty line times the national average dependency ratio. The JQI is calculated as the average sum of the dimensions. 44 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice one of every two working-age individuals has lived place in which they were born—are a minority, albeit at least once in a dzongkhag other than their current overrepresented in migrants currently living in rural dzongkhag of residence. Only 1 percent of individuals areas (23 percent, compared with only 7 percent of the born and currently living in Bhutan have ever lived migrants in urban areas). abroad. Working-age men and women born in Bhutan have the same probability of having experienced inter- Surprisingly, internal migration flows are not driven nal or external migration (see figure B.25 in appendix by urbanization. Figure B.27 in appendix B shows that B). Meanwhile, education increases the probability of 45 percent of individuals who ever migrated across internal and external migration—for example, individ- dzongkhags were previously living in an urban area, uals with no education are two times less likely to have compared with 54 percent who had lived in a rural moved across dzongkhags than individuals with a ter- area. In terms of location choices, urban areas appear tiary diploma. to be more attractive—60 percent of internal migrants end up living in an urban area. The urban–urban flow Of the working-age Bhutanese with a migration back- is almost as important as the rural–urban flow (28 ground, half currently live in a district different from percent versus 32 percent). their district of birth, but similar to the one in which they previously lived. According to figure B.26 in Overall, the main reasons for deciding to move are appendix B, a significant share of migrants live in a dis- work and family, but these numbers mask important trict different than the one in which they were born gender and location differences. Six in 10 migrant men or in which they previously lived, suggesting multiple move for work reasons, whereas almost seven in 10 migrations. This finding is particularly true for work- migrant women change dzongkhag for family reasons ing-age individuals currently living in urban areas (42 (figure 2.33). Family reasons dominate for those cur- percent versus 30 percent for rural dwellers). Return rently residing in rural areas (50 percent), whereas migrants—that is, migrants who currently live in the work reasons dominate for those who have moved to Figure 2.33. Reason for moving reported by Bhutan-born working-age individuals, 2019 a. By location and gender b. By education level Share of migrants (%) Share of migrants (%) 80 80 68 64 59 59 60 60 50 48 45 45 44 42 43 44 40 40 40 40 32 34 28 28 22 23 24 19 21 14 20 17 20 14 10 10 9 9 64 8 5 3 5 56 5 4 6 2 3 0.3 0 0 Total Rural Urban Men Women Total None/NFE Primary Secondary Tertiary Monastic Work-related Family-related Work-related Family-related Education-related Other Education-related Other Source: Bhutan Labor Force Survey, 2019. Source: Bhutan Labor Force Survey, 2019. Note: NFE = nonformal education. Note: NFE = nonformal education. Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 45 urban areas (48 percent). The higher the level of edu- Summary cation, the higher is the probability that a migrant moved for work (64 percent for the most-educated An analysis of 10 waves of the BLFS reveals that since migrants, compared with 23 percent for the least-ed- 2013 the size of the working-age population has ucated migrants). declined, but the demographic dividend has remained stable, and the skill level of working-age men and The current labor market status of migrants suggests women has increased rapidly. The pandemic affected internal mobility can help workers improve their the labor market, most notably driving women’s labor labor market outcomes. Only 1 percent of migrants force participation, and subsequently female unem- who moved for work are currently unemployed, and ployment, to increase in 2020 and 2021. In 2022, the 5 percent are inactive. Half of the migrants who had recovery was uneven as labor force participation rates family reasons or education reasons to migrate are for women dropped to pre-pandemic levels. currently inactive, most likely due to family duties or studies (figure 2.34). Interestingly, moving across Chapter 2 finds that the main challenges facing dzongkhags allows a change in labor market status: of workers in Bhutan are the limited inclusion of women the individuals unemployed in their previous loca- in productive employment; the high unemployment tion, 45 percent are now employees, and 21 percent rates among educated youth in urban areas; the seg- are self-employed or family workers. The same type mentation of the labor market into low-productivity of transitions can be observed when looking at previ- agricultural and public sector employment; and the ously inactive individuals, 29 percent of whom became low quality of employment in the private sector. employees and 21 percent of whom became self-em- ployed or family workers. Figure 2.34. Current labor market status of Bhutan-born working-age migrants, 2019 a. By migration motive b. By labor market status in origin dzongkhag Share of migrants (%) Share of migrants (%) 80 80 80 68 58 60 45 46 60 56 40 53 27 29 50 23 21 18 21 21 20 9 40 36 7 4 2 1 30 29 0 Employee (premigration) (premigration) Unemployed (premigration) Inactive (premigration) Self-employed/family 20 14 14 11 10 5 7 4 2 1 0 Work- Family- Education- Other related related related Employee Self-employed/family Employee Self-employed/family Unemployed Inactive Unemployed Inactive Source: Bhutan Labor Force Survey, 2019. Source: Bhutan Labor Force Survey, 2019. 46 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice Women face challenges in engaging in meaningful and The labor market in Bhutan is mostly dominated by quality employment. The participation of prime-age agricultural employment, followed by public sector women is strongly affected by their marital status, the employment, and it is segmented along the lines presence of young children, as well as the labor market of gender, location, and education. Men and high- participation of other women in their family and com- skilled workers in urban areas dominate public sector munity. In the labor market, women tend to work in employment, while women are more likely to remain low-productivity sectors such as agriculture, manufac- employed in agriculture and work as self-employed turing, or services as self-employed or family workers, or family workers. Structural transformation happens and they have limited access to private employment or slowly, with the share of agriculture remaining high. public sector jobs. Decomposition analysis finds that gender pay and employment gaps in the labor market An assessment of job quality through the lens of hours are driven by the differential impact of marriage on worked, wages, and informality indicates that signif- women and men. This finding is further evidence icant gaps linger in hours worked and wages across of the large constraints and high expectations that industries. Informality is still widespread, especially in married women and men face in the work and family urban areas and industries in which overwork dom- sphere. inates—namely, construction, wholesale and retail trade, transportation, and accommodation and food The alarmingly high unemployment rates among services. Those with high education and skills are more young and educated urban workers remain a chal- likely to have better quality employment through lenge. The reservation wages of the unemployed are better wages. However, hourly wage gaps persist across 60–90 percent below the wages observed for workers industries, even after controlling for returns to edu- of similar age and education. And yet the unemployed cation and experience. Cross-dzongkhag mobility for express a strong preference for public sector jobs and work reasons is widespread (mobility within urban believe that a mismatch between their qualifications areas accounts for a large share), which could improve and the available positions in the private sector is a key the quality of employment opportunities in Bhutan. reason for their unemployment. Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 47 Chapter 3 Firm Dynamics in Bhutan and Its Alignment with Labor Supply Mariana Viollaz and Jumana Alaref Introduction 3.1 Profile of firms in Bhutan Chapter 3 explores the extent to which the challenges The private sector is dominated by low-productiv- facing workers highlighted in chapter 2 stem from ity microenterprises concentrated in a few economic the limited productive opportunities created by the sectors that do not grow and employ mostly low- to private sector. Section 3.1 analyzes the characteristics of semi-skilled workers. registered firms in Bhutan by sector, region, size, and characteristics of their workforces. It also examines Over 95 percent of firms in Bhutan have, on average, patterns in job creation and destruction and explores fewer than five employees, and they capture only 37 whether these patterns are related to sectoral produc- percent of all employment (figure 3.1, panels c and tivity. Section 3.2 takes a look at the key occupation d). Ninety-five percent of firms have fewer than five and education categories expected to be in demand in employees (cottage firms), followed by firms with 5–19 the near future and compares them with the profile of employees (small firms that represent 3 percent of the the labor force discussed in chapter 2 to uncover any total). Cottage firms employ 37 percent of all workers potential skill mismatches in the labor market. This in Bhutan. By comparison, large firms, which make section also discusses the hiring difficulties and supply up only 0.3 percent of firms in the country, employ 32 shortages that firms encounter, which could arise from percent of all workers. The vast majority of firms (97 some skill and spatial mismatches in the labor market. percent) are individual proprietorships. In addition, the section explores the hiring of foreign workers to fill vacant positions and discusses firm Most firms in Bhutan are geographically concentrated and employee access to training support as a way to in three regions and are concentrated in a few eco- improve worker productivity and align labor demand nomic sectors with low labor productivity (figure 3.1, and supply. Section 3.3 then looks into firm business panel b). In 2022, 69 percent of firms were located management practices and perceptions of the difficul- in the regions of Gelephu (31 percent), Thimphu (23 ties firms encounter in expanding their growth. percent), and Punakha (15 percent). The distribution of firms by their main economic activity indicates that 80 percent of them are in the wholesale and retail trade sector (56 percent) and the accommodation and food 48 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice services sector (24 percent), both of which have low Higher-order skills related to the use of digital technol- labor productivity. Indeed, the accommodation and ogies are rarely utilized by firms. In 2022, 89 percent food service sector suffers from negative labor pro- of firms reported using e-payment. As for other digital ductivity. A comparison of the 2018 Economic Census technologies, the percentage of firms that use them is and the 2022 Establishment Survey (ES) indicates that low, ranging from 19 percent for social media to 0.6 between 2018 and 2022 firms in Bhutan became more percent for outsourced data centers. When disaggre- geographically concentrated and diversified in terms gated by economic sector, the data reveal that the use of of economic activity (see figure D.1 and table D.1 in e-payment is high in all of them. Firm websites are used appendix D).20 widely in the education sector (55 percent of firms) and financial services (44 percent). These two sectors stand Firm dynamism in Bhutan is limited. In 2022, 3 percent out as those with the highest usage of social media (89 of firms were new (less than one year of operation), percent of firms in the education sector and 91 percent and 26 percent were long-established (10 or more in financial services). Over 50 percent of firms in the years). Almost 75 percent of firms with fewer than five construction, education, mining, and financial ser- employees have been operating for 1–9 years, whereas vices sectors use accounting systems. 25 percent are long-established firms. This finding suggests there is limited firm dynamism in Bhutan: very small firms remain small, and inefficient ones do Job creation is prevalent in the more produc- not exit the market. tive sectors, which tend to hire more skilled workers. However, these sectors remain very The majority of firms employ low- to semi-skilled small and are unlikely to absorb the rising workers, with up to medium level of education (see number of high-skilled workers on their own. table D.2 in appendix D). An analysis of the individual characteristics of the workforce reveals that in 2022, An analysis of past patterns of job creation and destruc- 61 percent had a primary or secondary education; tion indicates that both declined when comparing the almost 20 percent had no education; and 30 percent of pre- and post-pandemic periods and that, on average, workers had vocational qualifications. Ten percent of job creation was always higher than job destruction workers had less than one year of experience working (see figure 3.2 and table D.3 in appendix D). In 2019, at their firm, suggesting some level of employee churn. the average number of workers hired across all firms Also in 2022, 3 percent of workers were foreign-born, was 1.1, and this number declined to 0.7 in 2020 and and 47 percent were women. Finally, the majority of the 2021. The average number of workers leaving firms fell workforce consisted of regular employees (91 percent) from 0.6 in 2019 to 0.5 in 2020 and 0.4 in 2021. Accord- followed by casual employees (6 percent). ing to the firms, the main reason for worker exit was voluntary resignation (58 percent of firms), followed by layoff due to the pandemic or other reasons (42 percent of firms). In 2018, 62 percent of firms were located in the regions of Thimphu, Gelephu, and Punakha (versus 69 percent in 2022). The wholesale and retail trade 20.  and accommodation and food services sectors accounted for 83 percent of firms in 2018, whereas in 2022 these sectors represented 80 percent. This dif- ference is explained mainly by the increase in the share of manufacturing (5–6 percent), construction (1–3 percent), and administrative support services (1–2 percent). Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 49 Figure 3.1. Profile of firms by economic sector, region, size, and employment share, 2018 and 2022 a. By economic sector b. By region 2.6 23.5 Agriculture, forestry, and fishing Thimphu 0.6 24.8 0.2 13.9 Mining and quarrying 0.4 Phuentsholing 20.1 5.4 Manufacturing 31.2 6.4 Gelephu 22.4 Electricity, gas, steam, and 0.0 air-conditioning 0.2 Samdrup 5.9 Jongkhar 6.7 Water supply, sewerage, and 0.0 waste management 0.1 14.6 Punakha 1.2 14.5 Construction 2.6 10.9 Trashigang 61.6 11.6 Wholesale and retail trade 55.9 2022 2018 0.3 Transportation and storage 0.7 c. By size Accommodation and 21.3 food services 23.9 95.9 Cottage Information and communication 0.4 88.8 0.5 0.1 2.9 Finance and insurance Small 0.2 8.8 0.0 0.9 Real estate 0.1 Medium 1.9 Professional, scientific, and 0.5 technical services 1.0 0.3 Large 0.5 Administrative and 1.3 support services 2.4 2022 2018 0.6 Education 1.0 d. By share of employment and firm size, 2022 0.1 Human health and social work No. of employees (thousands) Share of employment (%) 0.1 40 40 Arts, entertainment, 1.2 30 30 and recreation 1.2 20 20 2.9 10 10 Other services 2.7 0 0 Cottage Small Medium Large 2018 2022 Total employees Share of employment Sources: 2022 Establishment Survey and 2018 Economic Census. Note: Panels a, b, and c are showing percentages. In panels c and d, cottage firms have fewer than five employees; small firms between five and 19 employees; medium firms between 20 and 99 employees; and large firms 100 employees or more. 50 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice When disaggregated by firm size, all firm categories Figure 3.2. Average number of workers hired exhibited positive net job creation—that is, more jobs and exiting firms, 2019–21 were created than abolished (table D.3 in appendix D). The average number of workers hired was greater 1.1 than the average number of workers laid off regardless of firm size. The value of net job creation had a posi- tive relationship with the size of the firm. In addition, firms experiencing positive job creation during the 0.7 0.7 pandemic were more likely to report a positive finan- cial performance.21 0.6 0.5 Disaggregation by region reveals a more heteroge- 0.4 neous trend, with some regions having negative net job creation and others showing positive rates even during the pandemic (see figure 3.3 and table D.3 in appen- dix D). Before the pandemic, the regions of Thimphu and Gelephu were more dynamic in terms of having higher job creation than destruction compared with 2019 2020 2021 2019 2020 2021 other regions. In Thimphu and Gelephu, the average Employees hired Employees exited number of workers hired in 2019 surpassed the average number of workers leaving firms by 0.8 and 0.7, respec- Source: 2022 Establishment Survey. tively (in the other regions, the numbers were between 0.2 and 0.3). In 2020, Thimphu and Gelephu were the only regions with positive net job creation—a distinc- tion that continued in 2021. Figure 3.3. Net job creation, by year and region, 2019–21 0.8 0.7 0.5 0.4 0.3 0.3 0.3 0.3 0.2 0.2 0.2 0.2 0.0 0.0 0.0 0.0 -0.1 -0.1 Thimphu Gelephu Phuentsholing Punakha Samdrup Trashigang Jongkhar 2019 2020 2021 Source: 2022 Establishment Survey. Note: Net job creation is the average number of workers hired by firms minus the average number of workers exiting firms. In 2019, 29 percent of firms were operating at a loss, which increased to 61 percent in 2020 and 2021. On the other hand, the percentage of firms breaking 21.  even or operating at a profit declined over time. In terms of correlation between experiencing a loss and hiring in 2020, the net job creation rate was only 0.08 for firms operating at a loss, 0.224 for firms breaking even, and 1.02 for firms making a profit. Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 51 Analysis by economic sector reveals different patterns creation. However, in 2021 some sectors experienced in some sectors, especially the smaller ones that had negative net job creation, such as agriculture, forestry, positive net job creation even during the pandemic and fishing; electricity, gas, steam, and air-condition- (see figure 3.4 and table D.3 in appendix D). In 2019, all ing; water supply, sewerage, and waste management; sectors hired more workers than the average number and administrative and support services. they laid off. The sectors with higher net job creation were those having a smaller share of firms in the In 2021, the relatively more productive sectors hired market. They include financial and insurance activi- more workers, while the least productive ones hired ties, where the average number of workers hired was fewer (figure 3.5). A comparison of net job creation 5.4 higher than the number of workers leaving firms; rates by economic sector and sectoral labor productiv- construction with a net job creation of 4.7 workers; and ity in 2021—measured as GDP per worker—reveals that electricity, gas, steam, and air-conditioning, where the some of the sectors with the highest job creation rates net job creation reached 4.2 workers, on average. In in 2021 were also among the most productive: financial, 2020, most sectors continued to enjoy positive net job insurance, and real estate activities; transportation, Figure 3.4. Net job creation, by year and economic sector, 2019–21 Agriculture, forestry, and fishing Mining and quarrying Manufacturing Electricity, gas, steam, and air-conditioning Water supply, sewerage, and waste management Construction Wholesale and retail trade Transportation and storage Accommodation and food services Information and communication Finance and insurance Real estate Professional, scientific, and technical services Administrative and support services Education Human health and social work Arts, entertainment, and recreation Other services -3.0 -2.0 -1.0 0.0 1.0 2.0 3.0 4.0 5.0 6.0 No. of jobs created 2019 2020 2021 Source: 2022 Establishment Survey. Note: Net job creation is the average number of workers hired minus the average number of workers exiting firms. 52 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice storage, and communications; and mining and quar- productive sectors, such as agriculture and hotels and rying. On the other hand, the three sectors with low restaurants. net job creation rates exhibited lower levels of labor productivity: agriculture, forestry, and fishing; hotels Fast-growing sectors hire high-skilled workers. Figure and restaurants; and wholesale and retail trade. The D.1, panel a, in appendix D shows that, compared with two exceptions were the electricity and water supply the historically large sectors, small but faster-grow- sector, which exhibited a negative job creation rate in ing sectors hire workers with very different education 2021 despite its above-average labor productivity, and levels, and many of them hire a substantial share of the construction sector, which created the most jobs workers with a tertiary diploma. In 2022, the main pro- despite its lower productivity when compared with viders of high-skilled jobs for university graduates were the other sectors, such as transportation, storage, and the education and public administration and defense communication, and mining and quarrying. sectors (figure D.1, panel b, in appendix D), which is in line with the finding in chapter 2 that the public sector This finding is consistent with how sectoral employ- remains the main employer. More than one in 10 uni- ment grew over a longer time period, from 2013 to versity graduates worked in a small job-creating sector: 2021. Figure 2.26 in chapter 2, which merged sectoral 6 percent in financial, insurance, and real estate activ- employment data from the Bhutan Labor Force Survey ities; 4 percent in information and communication; (BLFS) with recent data from the national accounts and 2 percent in professional, scientific, and techni- to capture labor productivity at the sector level, cal services. However, these sectors remain small and shows that the share of the more productive sectors unable to absorb the increasing supply of high-skilled of total employment grew after 2013. Symmetrically, job-seekers. Financial, insurance, and real estate the sectors that shrank the most are also the least sectors made up 0.2 percent of firms and 2 percent of Figure 3.5. Labor productivity and net job creation rate, 2021 No. of workers 5 4 3 2 1 -1 -2 -3 supply fishing Wholesale and retail trade Private social and recreational services Electricity and water Agriculture, forestry, and Hotels and restaurants Manufacturing Mining and quarrying Education and health Transportation, storage, real estate Construction and communications Finance, insurance, and Workers hired and exiting firms Log(sector productivity/total productivity) Sources: 2022 Establishment Survey (ES) and National Accounts Statistics (NAS) 2022. Note: Labor productivity is measured as the sector-level value added per worker in 2021. Because of the different sectoral aggregations adopted by the NAS and the ES, the figure groups transportation and storage with information and communication, education with health, and art, entertainment, and recreation with other services. Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 53 employment in 2022, and the information and com- and food processing, woodworking, garment, and munication sector made up 0.5 percent of firms and 1 other craft and related trades workers (4 percent). percent of employment. Current trends in education and labor demand prospects suggest that the oversup- When looking at the education level that firms expect ply of high-skilled job-seekers (and the high unem- to demand in the future, mid-level education stands ployment in this group) is likely to grow over time and out. The expected labor demand for workers with a will require the appropriate policies to improve pro- primary or secondary level of education is 42 percent ductivity and job creation in high value-added indus- of the total. The expected demand for workers with no tries in the private sector. education is also high, reaching 27 percent of the total. The remaining share is certificate level (17 percent), diploma (5 percent), bachelor’s degree (8 percent), and 3.2 Labor demand prospects master’s degree and above (0.6 percent). A comparison of the current occupational distribu- Future vacancies in the private sector will be tion of employment and the expected labor demand concentrated mostly in occupation categories distribution by occupation reveals a potential increase requiring low to medium levels of educa- in the shares of total employment of services and sales tion and some specific technical skills, and workers, craft and related trades workers, and elemen- they will not match the profile of the current tary occupations (see figure 3.6 and table D.4 in appen- job-seekers. dix D). Although services and sales workers account for 37 percent of total employment, this occupational About one-third (29 percent) of firms expect to have category accounts for 44 percent of the expected labor new vacancies in the next one or two years, mainly in demand. Within this category, the expected labor services and sales (table 3.1). The distribution of new demand is concentrated in personal services workers vacancies by occupation indicates that these firms will and sales workers. The share of craft and related trades mainly need personal services workers (23 percent workers of total employment is 11 percent, and they of the expected new vacancies) and sales workers (20 account for 18 percent of the expected labor demand. percent). These two occupational categories are fol- Within this category, major increases are expected lowed by building and related trades workers (7 percent in the building and related trades and food process- of the total expected labor demand); food preparation ing, woodworking, and garments. Finally, elementary assistants (5 percent); laborers in mining, construc- occupations account for 6 percent of total employ- tion, manufacturing, and transportation (5 percent); ment and 11 percent of the expected labor demand. 54 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice Table 3.1. Distribution of expected vacancies over the next one or two years, by occupation, 2022 Managers Chief executives, senior officials, and legislators 1.6 Administrative and commercial managers 1.3 Production and specialized services managers 0.3 Hospitality, retail, and other services managers 0.4 Professionals Science and engineering professionals 1.6 Health professionals 0.2 Teaching professionals 0.9 Business and administration professionals 2.3 Information and communication technology professionals 0.4 Legal, social, and cultural professionals 1.1 Technicians and associate Science and engineering associate professionals 2.1 professionals Health associate professionals 0.3 Business and administration associate professionals 0.7 Legal, social, cultural, and related associate professionals 3.2 Information and communications technicians 0.6 Clerical support workers General and keyboard clerks 1.0 Customer services clerks 1.1 Numerical and material recording clerks 4.0 Other clerical support workers 0.0 Services and sales workers Personal services workers 23.3 Sales workers 20.3 Personal care workers 0.0 Protective services workers 0.3 Forestry workers Market-oriented skilled agricultural workers 0.6 Craft and related trades Building and related trades workers (excluding electricians) 6.9 workers Metal, machinery, and related trades workers 3.1 Handicraft and printing workers 0.9 Electrical and electronics trades workers 2.3 Food processing, woodworking, garment, and other craft and related trades workers 4.5 Plant and machine opera- Stationary plant and machine operators 1.0 tors and assemblers Assemblers 0.1 Drivers and mobile plant operators 2.9 Elementary occupations Cleaners and helpers 0.4 Agricultural, forestry, and fishery laborers 0.3 Laborers in mining, construction, manufacturing, and transportation 4.6 Food preparation assistants 5.0 Refuse workers and other elementary workers 0.4 Source: 2022 Establishment Survey. Note: Refuse workers = workers who collect, process, and recycle garbage from buildings, streets, and other public places, according to the ILO’s International Standard Classification of Occupations (ISCO). Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 55 Figure 3.6. Distribution of current employment and expected labor demand, by occupation, 2022 a. Current employment Share of total employment (%) Elementary occupations 6 11 Plant and machine operation and 5 assembly 4 Craft and related trades 11 18 Forestry 0.1 0.6 Services and sales 37 44 Clerical support 8 6 Technician and associate professions 11 7 Professional 12 6 Managerial 9 4 0 10 20 30 40 50 Current composition Expected labor demand b. Expected labor demand, by occupation Services and sales Craft and related trades Elementary 9 Building and related trades 3 1 Personal services Cleaning (excluding electricians) 7 0.4 23 Metal, machinery, and 4 Agriculture, forestry, and 0.5 24 related trades 3 fishery 0.3 Sales 20 Mining, construction, 0.4 3 Handicraft and printing manufacturing and 1 5 transportation 0.1 Personal care 0 Electrical and electronics 1.9 1 Food preparation trades 2 5 3 Food processing, Protective services 2 Refuse work and other 0.4 0.3 woodworking, garment 4 elementary occupations 0.4 and other craft and… Percent Percent Percent Current composition Expected labor demand Source: 2022 Establishment Survey. Note: Refuse work = collection, processing, and recycling of garbage from buildings, streets, and other public places, according to the ILO’s International Standard Classification of Occupations (ISCO). 56 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice Firms expect to demand workers for services and sales percent), and building and related trades (5 percent)— and craft and related trades—the two occupations with see table D.5 in appendix D. higher expected labor demand. These occupations are characterized by low levels of education and specific A comparison of the distribution of the expected labor technical skills at the certificate level (figure 3.7). For demand by education and the distribution of current services and sales, the expected demand for workers job-seekers points to a potential mismatch between with no education or with primary levels is 23 percent labor demand and supply. Figure 3.8 reveals a short- and 18 percent of the total demand, respectively. For age of job-seekers with low levels of education and craft and related trades, the pattern is more accentu- with some specific technical skills (gained through ated. The expected demand for workers with no edu- certificates or diplomas) to fill positions mainly in cation is 38 percent, while those with primary level services and sales and craft and related trades. This are 7 percent of the total. A substantial share of the finding can be attributed in part to the high share of expected labor demand in this occupation (31 percent) the working-age population with low education levels is assigned to the certificate level. who remain outside the labor force, especially women (see figure B.16, panel c, in appendix B). According to In terms of the labor demand prospects of firms, the chapter 2, 39,077 prime-age women with no educa- occupation category expected to grow over the next tion are outside the labor force, compared with only five years is services and sales. Thirty-one percent of 4,076 prime-age men with no education. This finding firms expect to have new occupations in the next five points to the need to improve Bhutan’s activation pol- years. The jobs expected to emerge correspond mainly icies to increase the attractiveness and accessibility to sales occupations (28 percent), personal services (24 of the available low-skilled positions, targeting the Figure 3.7. Expected labor demand for services and sales workers and craft and related trades workers, by education, 2022 Share of total demand (%) 40 38 35 31 30 25 23 23 20 18 16 15 13 10 8 7 7 7 5 4 3 3 0 No education Primary Middle-secondary Higher-secondary Certificate Diploma Bachelor's Services and sales workers Craft and related trades workers Source: 2022 Establishment Survey. Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 57 unemployed and the inactive, especially women in about a third of firms in Bhutan faced hiring difficul- urban areas. A twin challenge is creating attractive ties in 2022 (table 3.2). The average, however, hides het- jobs for the fast-growing cohorts of young university erogeneous patterns by region and economic sector. graduates. Figure 3.8 displays graphically the oversup- More than half of firms in Samdrup Jongkhar (56 ply of secondary- and tertiary-educated job-seekers percent) and Trashigang (55 percent) reported having able to fill the current and expected labor demand for difficulties, together with 38 percent in Punakha, 32 workers who fall in these education categories. percent in Thimphu, 31 percent in Phuentsholing, and 25 percent in Galephu. The agriculture, forestry, and fishing sector (45 percent of firms) and the informa- Firms experience labor shortages and hir- tion and communication sector (42 percent) faced the ing difficulties that negatively affect firm most difficulties. By contrast, only 2 percent of firms performance. in the electricity, gas, steam, and air-conditioning sector and none of the firms in the real estate activities Given the observed skill mismatches in the labor sector reported having difficulties hiring or retaining market, it is not surprising that the 2022 ES found that workers. Figure 3.8. Comparison of the expected labor demand and the current labor force and inactive population, by education, 2022 Percent 70 Undersupply Oversupply 60 50 40 30 Oversupply Important to activate, esp. Undersupply 20 among women 10 0 No education Primary Secondary Certificate Diploma University and above Future demand Working-age population Employed Job-seekers Inactive Sources: 2022 Establishment Survey and Bhutan Labor Force Survey, 2022. 58 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice Table 3.2. Percentage of firms facing hiring An applicant supply shortage is the main hiring diffi- difficulties, 2022 culty reported by firms, followed by a lack of required skills or education level (figure 3.9). In 2022, 41 percent Share of firms experiencing of firms reported having no or few applicants as hiring difficul- the main challenge in hiring. The second challenge, ties (%) faced by 36 percent of firms, was related to the edu- Overall 32.87 cation level and skills of the workforce. Specifically, By region 22 percent of firms found applicants did not have Thimphu 32.3 the required technical skills; 8 percent indicated that Gelephu 24.8 applicants lacked the required education level; and 6 Phuentsholing 31.3 percent reported that applicants lacked the required Punakha 38.5 soft skills. Other difficulties in hiring included appli- Samdrup Jongkhar 55.6 cants expecting wages higher than what firms can offer Trashigang 55.0 (29 percent) and applicants lacking the required work By sector experience (28 percent). Agriculture, forestry, and fishing 44.7 Mining and quarrying 35.8 Figure 3.9. Percentage of firms reporting Manufacturing 34.8 hiring difficulties, by reason, 2022 Electricity, gas, steam, and air-conditioning 2.3 Water supply, sewerage, and waste 9.0 Share of firms (%) management Construction 33.2 There were no or 41 few applicants Wholesale and retail trade 30.7 Transportation and storage 12.9 Applicants expected wages higher than 29 Accommodation and food services 38.0 what we could offer activities Applicants lacked required Information and communication 42.2 28 work experience Financial and insurance activities 35.0 Applicants lacked required Real estate activities 0.0 22 technical skills Professional, scientific, and technical 18.1 Administrative and support services 23.6 Applicants did not like 9 activities location of the job Education 37.2 Applicants lacked required Human health and social work activities 6.2 qualification/education level 8 Arts, entertainment, and recreation 33.3 Applicants lacked required Other service activities 34.0 6 soft skills Source: 2022 Establishment Survey. 0 10 20 30 40 50 Source: 2022 Establishment Survey. Note: Figure shows the percentage of firms responding yes to each type of hiring difficulty. Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 59 As presented in table 3.3, the disaggregation of hiring expecting higher wages is also an important factor difficulties by occupation indicates that a lack of appli- for this occupation. Among craft and related trades cants in services and sales and craft and related trades is workers, no or few applicants is important for all occu- one of the main difficulties. The lack of technical skills pation subcategories. And the lack of technical skills is affects craft and related trades workers more than ser- an important reason for the difficulties facing workers vices and sales workers. For services and sales workers, in the metal, machinery, and related trades and the having no or few applicants is the first or second hiring electrical and electronics trades. difficulty in all occupation subcategories. Applicants Table 3.3. Types of hiring difficulties, by occupation, 2022 There were Applicants Applicants Applicants Applicants Applicants Applicants no or few lacked lacked lacked expected did not like lacked applicants required required required wages location of required (%) qualifica- technical soft skills higher the job (%) work expe- tion/edu- skills (%) (%) than what rience (%) cation level we could (%) offer (%) Managers 27.7 21.5 0.0 0.0 22.7 12.3 15.9 Professionals 9.7 23.5 32.9 2.2 12.3 1.9 17.5 Technicians and associate 35.2 7.1 19.3 4.3 18.8 1.2 14.1 professionals Clerical support workers 16.9 12.5 14.0 9.5 34.0 10.7 2.4 Services and sales workers Personal services workers 25.0 0.9 10.9 4.0 28.3 9.4 21.4 Sales workers 42.8 3.4 5.8 6.0 14.7 8.0 19.3 Personal care workers n.a. n.a. n.a. n.a. n.a. n.a. n.a. Protective services workers 26.6 0.0 26.6 0.0 46.8 0.0 0.0 Forestry workers n.a. n.a. n.a. n.a. n.a. n.a. n.a. Craft and related trades workers Building and related trades 31.1 3.2 20.4 2.1 12.4 4.6 26.1 (excluding electricians) Metal, machinery, and 14.1 2.5 33.6 6.1 20.5 5.7 17.6 related trades Handicraft and printing 24.6 0.0 15.5 0.0 11.1 0.0 48.8 Electrical and electronics 29.9 7.9 32.2 0.9 19.9 2.2 7.0 trades Food processing, wood- 24.1 2.3 15.3 4.4 16.0 10.0 27.9 working, garment, and other crafts Plant and machine operators 38.3 12.3 9.8 0.0 27.7 2.1 9.8 and assemblers Elementary occupations 39.9 5.9 10.2 0.5 24.2 4.6 14.6 Source: 2022 Establishment Survey. n.a. = Not available. 60 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice In 2022, 15 percent of firms reported facing worker Figure 3.10. Percentage of firms facing shortages (figure 3.10). The distribution by occupation worker shortages, by occupation and reveals that these shortages mainly affected services education level, 2022 and sales workers (39 percent) and craft and related trades workers (26 percent). The distinction by edu- a. By occupation cation level shows that firms mainly face shortages of low-educated workers: 49 percent of worker shortages Managers 1 can be attributed to a primary education or less. This Professionals 5 finding is in line with the earlier findings on the labor Technicians and associate demand prospects of firms and the mismatch with the professionals 6 profiles of current job-seekers (see table D.6 in appen- Clerical support workers 3 dix D). Services and sales workers 39 Worker shortages as experienced by firms across Forestry workers 0.3 occupations, education levels, and regions suggest a Craft and related trades workers 26 certain degree of spatial mismatch. From a compari- Plant and machine operators son of the ratio of unemployed to employed workers 3 and assembly across regions and by education level as a measure of Elementary occupations 18 the tightness of the labor market, it becomes appar- ent that the regions reporting hiring difficulties (such as Samdrup Jongkhar and Trashigang) suffer the most b. By education level from supply shortages (figure 3.11). In those regions, the ratio of unemployed to employed workers is near No education 36 zero for the uneducated and graduates with a primary education—the two categories in which supply short- Primary 12 ages and labor demand are concentrated. Chapter 2 Middle-secondary 14 finds that while cross-dzongkhag mobility is common in Bhutan, it is less common for the less educated, which Higher-secondary 11 could be exacerbating spatial mismatches. Certificate 20 Diploma 3 Bachelor's 4 Master's and above 0.3 Source: 2022 Establishment Survey. Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 61 Figure 3.11. Ratio of number of job-seekers to number of workers, by education and region, 2022 a. Thimpu b. Gelephu c. Phuentsholing None/NFE 0.067 None/NFE 0.010 None/NFE 0.010 Primary 0.053 Primary 0.014 Primary 0.025 Secondary 0.181 Secondary Secondary 0.117 0.103 Tertiary 0.170 Tertiary 0.099 Tertiary 0.150 d. Punakha e. Samdrup Jongkhar f. Trashigang None/NFE 0.018 None/NFE 0.009 None/NFE 0.004 Primary 0.040 Primary 0.000 Primary 0.021 Secondary 0.110 Secondary 0.085 Secondary 0.137 Tertiary 0.124 Tertiary 0.019 Tertiary 0.048 Source: Bhutan Labor Force Survey, 2022. Note: NFE = nonformal education. Worker shortages have a negative impact on firm of firms reported facing difficulties with retaining performance (table 3.4). Sixty percent of firms indi- workers. The distribution by occupation indicates that cated that worker shortages have a strong impact on in services and sales, retaining difficulties are more the workload of the existing staff; 47 percent found common (39 percent), followed by craft and related that they have a high impact on the loss of productiv- trades (26 percent). The hiring difficulties and labor ity; and 40 percent pointed to their impact on the loss shortages faced by many Bhutanese firms can likely of profits or sales. Worker shortages also have a high be attributed to a combination of barriers. Many low- impact on a firm’s ability to grow and diversify (40 skilled women in urban areas remain outside the labor percent of firms) and can lead to loss of markets (38 force, mostly due to household and care responsibili- percent), lower-quality output (29 percent), and firm ties. Other reasons preventing workers from moving to closure (25 percent). regions where some of those wage opportunities exist could include mobility, financial, and skill barriers. As Retaining workers is less of a problem for firms than noted earlier, for craft and related trades occupations, hiring (see table D.7 in appendix D). Only 6 percent lack of required technical skills is one of the main 62 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice Table 3.4. Impacts of worker shortages on firm performance, 2022 Impact of worker shortages High Moderate Low (% of firms) (% of firms) (% of firms) Loss of productivity 45.7 44.4 9.9 Loss of markets 37.8 50.6 11.6 Inability to grow and diversify the establishment 39.1 49.4 11.4 Higher workload for existing staff 60.5 33.0 6.5 Firm closure/shutdown 24.7 32.8 42.5 Lower-quality output 28.6 44.4 27.1 Decrease in profit/revenue 40.3 43.8 16.0 Source: 2022 Establishment Survey. reasons for hiring difficulties. In addition, workers Samdrup Jongkhar and Phuentsholing, followed by may lack information on where jobs are. Chapter 4 Phunaka and Galephu, where 3 percent of workers are describes how the absence of an up-to-date labor foreign. The lowest rates are in Thimphu (2 percent) market information system leaves job-seekers with no and Trashigang (1 percent). access to timely information on the occupations most in demand. There is also evidence that some Bhuta- Craft and related trades have a higher presence of nese workers have high reservation wages for work in foreign workers—47 percent (figure 3.12). Most foreign certain occupations, particularly in the construction workers in this occupation category are from India and sector (Kuensel 2021). work in the construction sector. According to a Septem- ber 2023 article in Business Bhutan, this sector alone accounted for a staggering 84.6 percent of approved Firms do not employ a high share of foreign foreign labor permits. The next occupations ranked workers to fill in labor shortages. by share of total foreign workers are technicians and associate professionals (11 percent), professionals (10 Although worker shortages in the domestic labor percent), and elementary occupations (10 percent). market could lead firms to hire low-skilled foreign workers from neighboring countries, only a few firms The main reason cited by 31 percent of firms for hiring in Bhutan hire foreign workers, and the percentage of foreign workers is the lack of skilled domestic workers firms doing so has declined substantially over time. (see table D.8 in appendix D). Because most foreign According to the 2022 ES, 7 percent of firms hired at workers are employed in craft and related trades, least one foreign worker in 2022, a lower share than it is likely that foreign workers are better equipped in 2017 when 9 percent hired workers from abroad, with technical skills. Foreign workers are also hired according to the Economic Census. The percentage of because they have better workmanship (as reported foreign workers in all firms is 3 percent, but this share by 20 percent of firms) and better work attitudes (19 rises to 42 percent when conditioning on firms hiring at percent). Approximately 10 percent of firms reported least one. Some heterogeneities appear across regions. that foreign workers are easier to manage or will accept The percentages of foreign workers is 4 percent in both a lower wage rate. Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 63 The private sector does not have links to Figure 3.12. Distribution of foreign workers, by training institutes to address labor short- occupation, 2022 ages, despite its current training needs and positive perceptions of the value of worker Share of foreign workers (%) training. Managers 7 The majority of firms do not have links to public pro- Professionals 10 viders of training, which highlights the disconnect Technicians and 11 between training programs and labor demand in associate professionals Bhutan. Between 79 and 91 percent of firms indicated Clerical support workers 2 they do not have any connections with public or private Services and colleges in the country, public technical or vocational sales workers 4 institutions, private training institutes, government entities, private sector associations, or external train- Forestry workers 0 ing and education institutes (table 3.5). This lack could Craft and related 47 explain the high percentages of firms planning to trades workers provide in-house training (40 percent) or planning to Plant and machine 9 operators and assembly resort to on-the-job learning (56 percent). Fifty-seven percent of firms have provided in-house training in Elementary occupations 10 the past—a surprising finding given the myriad of tech- 0 20 40 60 nical and vocational education and training (TVET) programs available and administered by the Minis- Source: 2022 Establishment Survey. try of Education and Skills Development (reviewed in chapter 4). It also highlights the need for a demand- driven approach and closer collaboration between TVET institutions and employers to help address the observed skill mismatches in the labor market. Table 3.5. Level of connection with potential partners for providing training, 2022 Public or Public Private Government Private External private col- technical or training sector body sector training and leges in the vocational institutes association/ education country institutes body institutes No link at all 91.2 91.2 91.2 79.4 86.3 90.3 Poor link 2.1 2.0 2.7 8.0 4.0 2.7 Moderate link 1.8 1.5 1.6 7.7 4.2 2.0 Strong link 0.7 0.9 0.5 2.5 1.5 0.8 Not relevant 3.6 3.7 3.5 2.0 3.5 3.7 Plan to establish link in the future 0.5 0.8 0.6 0.4 0.5 0.5 Source: 2022 Establishment Survey. Note: Table indicates the percentage of firms responding yes to each type of training. 64 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice In 2022, about a quarter of firms reported having Table 3.6. Training needs for the next five training needs, mainly at the short-term training level years, 2022 (table 3.6). According to the 2022 ES, 27 percent of the Share of firms surveyed firms indicated having training needs for the with training needs (%) next five years. Among these firms, 69 percent reported Overall 32.87 having short-term training needs, followed by certifi- Level of training cate level (19 percent) and diploma (6 percent). Seven- Master’s 2.1 ty-three percent of the firms considered these training Bachelor’s 4.2 needs a high priority, and 40 percent of them have Diploma 6.4 funding to provide their workforce with the needed Short-term training 68.8 training. Certificate 18.6 Priority level Firms also understand the value of training because High 73.5 past training has successfully improved workers’ abili- Moderate 25.7 ties and skills (see table D.9 in appendix D). Firms indi- Low 0.8 cated that from the training provided the last three Funding availability years they had obtained significant improvements in Firm has funding 39.9 worker confidence (61 percent of firms), work pro- Source: 2022 Establishment Survey. ductivity (61 percent), ability to work independently (59 percent), organization productivity (57 percent), job-specific technical skills (52 percent), leadership Table 3.7. Training critical for the current skills (51 percent), problem-solving skills (48 percent), occupation and funding plans, according to and creative and critical thinking (45 percent). employees, 2022 Share of employees (%) Training needs from employees’ perspectives match Training level those reported by firms (table 3.7). In 2022, 72 percent Master's/PhD 3.8 of workers indicated that short-term training was crit- Diploma 10.8 ical for their current occupation, and 69 percent of Short-term training 71.9 firms said they had training needs at this level. Sev- Online learning 2.1 enteen percent of workers reported that the level of In-house training 9.0 training needed for their occupation was a diploma, Bachelor's 2.3 bachelor’s, master’s, or PhD degree; 9 percent reported Plan to fund in-house training; and 2 percent reported online Self 8.1 training. My organization 40.2 Office project 1.4 Donor scholarship 13.8 No funding 36.5 Source: 2022 Establishment Survey. Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 65 The number of workers trained has declined over time, can work independently; and have adequate work likely because of funding constraints. The average experience. number of workers trained in 2019 was 9.1, but that number declined to 8.9 in 2020 and to 5.6 in 2021.22 Training in digital and information technology (IT) Although 40 percent of workers indicated they plan to should receive priority in view of workers’ percep- fund training through the firm, 36 percent reported tions of their own skills (figure 3.13). Specifically, 82 not having funding. The remaining share corresponds percent of employees indicated their use of advanced to self-funding (8 percent of workers) and office digital tools (such as software development, program- project (1 percent), as shown in table 3.7. ming, data analysis, or cyber security) is poor or very poor. Sixty-one percent reported the same for the use The share of firms hiring TVET or tertiary graduate of data analytic software or tools (such as SPSS, Excel, workers is low. However, among those who hire TVET and STATA), and 43 percent for the use of work-related graduates, their perception of TVET quality is positive, software or tools (such as Word, Power Point, Adobe). which highlights the potential for expanding the col- On the other hand, 53 percent of workers reported that laboration between TVET and the private sector. In their use of basic computer tools (including accessing 2022, only 4 percent of firms hired TVET graduates and the internet, typing, and sending emails) is good or 6 percent hired tertiary graduates. The low percent- very good. The low level of digital and IT skills reported age of firms hiring TVET graduates is likely related to by workers is in line with the low level of education the low number of TVET graduates each year who are of the workforce. The correlation between employ- unable to keep up with current demand (as discussed ees’ perceptions of the use of basic computer tools and in chapter 4). Among the firms hiring these workers, their educational level is 0.60 (that is, workers with most find that TVET and tertiary graduates have skills higher levels of education have better perceptions of relevant to the job, including soft, managerial, liter- their digital skills). The correlation coefficient for the acy, and numeracy skills; are creative and innovative; use of work-related tools is 0.65; data analytic tools, 0.55; and advanced digital tools, 0.41. Figure 3.13. Employees’ perceptions of their digital and information technology skills, 2022 Share of employees (%) 70 65 60 50 45 40 30 31 30 23 20 19 21 20 14 14 16 17 16 16 17 12 10 10 7 6 2 0 Use of basic computer tools Use of work-related Use of data analytic Use of advanced digital tools software/tools software/tools Very poor Poor Average Good Very good Source: 2022 Establishment Survey. Among workers who reported receiving any kind of training since joining their firms, 25 percent attended short-term training in specific fields, 20 per- 22.  cent received in-house training, and 7 percent attended online training. 66 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 3.3 Barriers to firm growth and firm size shows that the share of firms noting their business management practices importance increases as firm size increases (see table D.10 in appendix D). Firms face barriers to growth related to Policies and regulations related to the business climate investment climate factors and labor regu- are more likely to be considered a major constraint lations, although the severity of each factor to business management and growth than labor-re- varies by firm size. lated regulations. In 2022, among the factors consid- ered a major or very severe constraint to firm growth, In 2002, most firms reported that having human business climate factors weighed more heavily than resources and favorable policies and regulations in labor factors. Among business climate factors, access place were important factors for business expansion to markets, finance, and raw materials or goods were or diversification (table 3.8). The fact that 69 percent considered to be major or very severe constraints to of firms considered human resources important for business management (table 3.9). According to the the expansion or diversification of business is in line 2017 Investment Climate Assessment, fewer Bhutanese with the findings that a lack of access to trained labor firms had a loan or a line of credit in 2015 (47 percent) to fill existing vacancies is negatively affecting firm than in 2009 (59 percent). The decline likely stemmed growth. Meanwhile, 92 percent of firms indicated from lack of credit information and from a complex, that having favorable policies and regulations in place unpredictable, and ineffective restructuring and insol- was important for their plans. This finding aligns with vency regime, despite the fact that the government had existing evidence that firms spend considerable sums reduced the amount of collateral required for a loan. on checking the requirements of government regula- Access to markets was also likely hampered by trade tions (World Bank 2015). Although 22 percent of firms and logistical deficiencies and limited foreign invest- considered access to finance important and 16 percent ment (Santini, Tran, and Beath 2017). considered markets important, a disaggregation by Table 3.8. Importance of factors in business expansion or diversification plans, 2022 Least important Most important (%) (%) 1 2 3 4 Human resources 9.8 21.1 53.4 15.7 Finance 36.0 41.5 18.7 3.8 Market 50.9 33.5 12.8 2.9 Favorable policies and regulations in place 3.5 4.1 15.1 77.4 Source: 2022 Establishment Survey. Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 67 Table 3.9. Constraints in the management of firms, 2022 No constraint Minor con- Moderate Major con- Very severe (%) straint (%) constraint (%) straint (%) constraint (%) Business climate factors Internet access and connectivity 7.8 12.1 10.8 11.6 11.7 Customs and trade regulations 9.0 10.5 8.6 7.5 7.1 Business licensing and operations permits 10.0 8.3 6.4 4.8 10.5 Access to finance 8.0 10.6 10.7 14.9 10.3 Access to raw materials/goods 7.4 12.0 11.8 15.1 12.6 Access to markets 6.1 12.2 17.4 19.2 18.7 Policy uncertainty 9.0 9.6 9.4 7.4 11.3 Corruption, crime, theft, and disorder 11.4 5.6 3.3 3.4 2.8 Labor factors Stringent labor law and regulations 10.9 5.8 6.4 3.3 4.9 High worker turnover 10.9 5.3 4.9 7.3 5.1 Overall market wage level 9.6 8.0 10.3 5.5 4.9 Source: 2022 Establishment Survey. Disaggregation by firm size shows that in 2022 more likely to be considered a major constraint by labor-related regulations were a bigger constraint for larger firms. Among smaller firms, access to finance, larger firms, whereas investment climate factors were markets, and raw materials were major constraints. more relevant for smaller and medium-size firms. For larger firms, stringent labor regulations, high Figure 3.14 and table D.11, panel b, in appendix D show worker turnover, and market wage level gained more that investment climate factors were more binding for prominence. smaller firms, whereas labor-related regulations were Figure 3.14. Major constraints to growth, by firm size, 2022 Share of firms (%) 100 80 60 40 20 0 Cottage Small Medium Large Internet access and connectivity Customs and trade regulations Business licensing and operations permits Access to finance Access to raw materials/goods Access to markets Policy uncertainty Corruption, crime, theft, and disorder Stringent labor law and regulation High worker turnover Overall market wage level Source: 2022 Establishment Survey. 68 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice Labor-related regulations are a constraint for larger (World Bank 2020a). Firms in Bhutan are burdened firms because they are more likely than cottage firms to with administrative hurdles, including unclear and comply with them. In 2022, the low level of compliance multiple licensing procedures and lack of informa- with labor regulations could be attributed to the large tion on licensing requirements. For example, upon share of small firms in the economy and the low level opening a business an entrepreneur must undertake of compliance among them (see table 3.10 and table eight procedures at a cost of about 4 percent of the D.12 in appendix D).23 The compliance of firms with country’s income per capita. Upon closing a business, fewer than five employees ranges from 4 to 33 percent, a draft Insolvency Bill must be adopted and imple- whereas the rate among firms with 100 employees or mented. As for labor market efficiency, the Labor and more is over 90 percent. Although no information is Employment Act places few restrictions on the hiring available about the reasons for noncompliance, it is and firing of workers. likely that firms with fewer than five employees cannot afford to comply with regulations because they are too Finally, private sector activity may be hampered by costly. For larger firms, noncompliance can be consid- competition policy that encourages the dominance of ered an evasion problem—that is, firms evade the regu- SoEs in key economic sectors and may be crowding out lations to obtain a monetary benefit. private sector investment. As mentioned in chapter 2, the pay gap between SoEs and the private sector is The overall findings on the investment climate and wide, and a large share of it remains unexplained by labor-related regulations are in line with the exist- workers’ characteristics. ing literature on barriers to doing business in Bhutan Table 3.10. Compliance with labor regulations, by firm size, 2022 no. of employees <5 5–19 20–99 100+ Our establishment has an Internal Service Rule (ISR). 14.7 67.7 89.0 92.7 Our Internal Service Rule (ISR) is endorsed by the Department of Labor. 11.4 52.9 79.9 89.5 We have occupational health and safety in place. 32.7 70.4 86.7 99.3 We have occupational health and safety policy in place. 19.9 58.9 78.5 98.1 We provide basic personal protective equipment (PPE) to our employees. 32.7 69.0 79.5 94.3 We have a provident fund for our employees with a recognized financial institute. 13.1 64.9 94.4 99.0 We provide overtime payment to our employees. 17.3 58.0 59.9 83.7 We provide pay slips/evidence of wages paid to our employees. 29.4 80.1 93.6 100.0 We have a written contract/term of employment for our staff and new recruits. 10.9 52.5 81.3 87.7 We issue an appointment letter at the time of appointment of new recruits. 12.6 55.9 88.7 92.9 We have clear job roles and responsibilities for our staff and new recruits. 29.2 80.7 94.0 100.0 We provide maternity leave. 25.1 72.2 93.9 100.0 We provide paternity leave. 23.5 67.9 91.4 94.0 We have a group insurance scheme (GIS) for our employees. 4.0 28.8 60.7 91.5 We have a sexual harassment policy/grievance system in place. 12.1 47.1 75.9 96.8 Source: 2022 Establishment Survey. The disaggregation by sector shows that the wholesale and retail trade and accommodation and food services sectors—the two most important sectors in 23.  the country—have a compliance rate below that for the entire economy. Chapter 2 describes how low-quality employment prevails in those sectors. Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 69 Although some firms have business manage- and interrelated challenges facing the private sector ment practices in place, such as salary incre- development agenda in Bhutan. The first challenge is ment and employment promotion systems, improving the quality of employment in the private such practices do not fully alleviate their con- sector and creating high value-added jobs for the straints to hiring and accessing trained labor. fast-growing cohorts of young university graduates. The private sector is currently underdeveloped and is In 2022, 51 percent of firms in Bhutan reported having dominated by microenterprises that do not grow and a salary increment system, and 18 percent had an are geographically concentrated in few regions. They employee promotion system (figure 3.15). Correlat- are also not sufficiently diversified in terms of eco- ing these percentages with firms facing hiring difficul- nomic activity, and they mostly belong to the whole- ties, worker shortages, and retention difficulties across sale and retail trade and accommodation and food economic sectors indicates that services sectors. Those dominant economic sectors have, on average, low labor productivity and employ • Having a salary increment system or an employee mostly low- and semi-skilled workers. They also have promotion system does not correlate with hiring weak employment conditions as evidenced by low difficulties. This finding could be related to low compliance with labor regulations and the prevalence entry salaries—that is, hiring difficulties do not of nonwritten contracts as highlighted in chapter 2. decline when providing a salary increment or an Although job creation over the last few years has taken employee promotion system because of a low entry place in sectors that are relatively more productive and salary. are likely to employ more skilled workers, those sectors • When firms have a salary increment or an employee remain very small and unlikely able to absorb the promotion system, they are less likely to experience increasing supply of high-skilled job-seekers in light worker shortages. Once workers are hired, these of the current education and labor demand prospects. systems are attractive for workers who may want to stay at the firm. The second challenge is making the available low- • When firms have a salary increment or an employee skilled positions more attractive and accessible to promotion system, they are more likely to face the unemployed and the inactive, especially women retention difficulties. By providing a salary incre- in urban areas. One reason firms in Bhutan, despite ment system, firms are able to attract the best rising shares of unemployment, were facing difficul- workers (in terms of their skills), and so other firms ties in 2022 in filling positions in services and sales and want them. The relationship could operate in the craft and related trades was few or no applicants. Labor other direction as well, with firms implementing shortages also may have a spatial dimension in which these systems to minimize retention difficulties. regions reporting hiring difficulties are those with very few job-seekers, especially those in the low- and semi-skilled education categories most in demand. Difficulties in accessing labor mainly stem from the Summary high share of low-educated workers (mostly women) outside the labor force. They could, with the appro- Chapter 3, using the 2022 Establishment Survey, sheds priate training and support, fill some of those vacan- light on the extent to which lagging labor market out- cies. A second reason is that firms are unable to meet comes among workers are an outcome of the type of their need for some specific technical skills (through employment opportunities created by the private certificates or diplomas) because they have no links sector in Bhutan. The chapter describes two important to vocational training institutes. Other reasons could 70 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice Figure 3.15. Correlations between salary increment/employee promotion systems and hiring difficulties, worker shortages, and retention challenges across economic sectors, 2022 a. Employee promotion system and b. Employee promotion system and hiring difficulties worker shortages 50 60 50 y = - 0.063x + 26.604 40 R = 0.0205 Worker shortages Hiring difficulties 40 30 30 20 20 10 y = - 0.0001x + 26.171 10 R = 7E - 08 0 0 0 50 100 0 50 100 Employee promotion system Employee promotion system c. Employee promotion system and d. Salary increment system and hiring retention difficulties difficulties 60 y = 0.3239x + 1.4961 50 50 R = 0.6571 40 Retaining difficulties Hiring difficulties 40 30 30 20 20 10 10 y = 0.0503x + 24.503 0 0 R = 0.0096 0 50 100 0 50 100 Employee promotion system Salary increment system e. Salary increment system and worker f. Salary increment system and shortages retetention difficulties 60 60 y = - 0.1653x + 34.091 y = 0.121x + 3.6244 50 R = 0.1213 50 R = 0.2563 Retaining difficulties Worker shortages 40 40 30 30 20 20 10 10 0 0 0 50 100 0 50 100 Salary increment system Salary increment system Source: 2022 Establishment Survey. Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 71 include mobility, financial, skill, or informational The challenges described in this chapter are related barriers that prevent low-skilled workers currently to a myriad of other barriers to firm growth related to engaged in low-productivity livelihoods from moving investment climate factors and labor regulations. The to regions where some of those wage opportunities extent to which some of those barriers are binding exist. Indeed, chapter 2 finds that internal mobility varies by firm size, which may require a tailored in the labor market among low-skilled workers is not approach to supporting private sector development. common, with individuals with no education being Among smaller firms, access to finance, markets, and two times less likely to have moved across dzongkhags raw materials appear to be major constraints. For larger than individuals with a tertiary diploma. Many Bhuta- firms, stringent labor regulations, high worker turn- nese workers are also unwilling to work in many vacant over, and market wage level are more prevalent. The occupations in the construction sector. dominance of SoEs in key strategic economic sectors is also distorting the playing field with the private sector. 72 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice Chapter 4 Bhutan’s Employment Support Programs and Delivery System Esther Bartl Introduction 4.1 Objectives of employment support programs and their Chapter 4 examines the extent to which Bhutan’s rationale for Bhutan employment support programs and delivery systems effectively address some of the challenges facing both workers and firms. As outlined in chapters 2 and 3, There is a strong case for targeted employ- those challenges include low female labor market ment support programs that improve the insertion, limited worker skills and productivity, and skills of vulnerable groups and support both the difficulties firms face in accessing trained labor. workers and firms in Bhutan. Section 4.1 reviews employment support programs— ALMPs are government programs aimed at helping active labor market programs (ALMPs) and the tech- unemployed workers find a job and improving the nical and vocational education and training (TVET) access of underemployed workers to better pro- sector—specifically, their broad objectives and their fessional opportunities (Romero and Kuddo 2019). rationale in Bhutan, building on the constraints identi- Interventions mainly include (1) training and skill fied in chapters 2 and 3. Section 4.2 assesses the extent development programs to enhance job-relevant skills, to which Bhutan’s employment delivery system, con- representing nonformal TVET; (2) job search and sisting of ALMPs, the TVET sector, employment services matching assistance such as job search training, coun- centers (ESCs), and a labor market information system seling, and monitoring to help job-seekers look for (LMIS), could help address the challenges facing the jobs more effectively; and (3) private sector incentive labor market in Bhutan. The chapter concludes with programs that offer the unemployed microfinance a variety of recommendations to strengthen Bhutan’s schemes and wage subsidies to cover the labor costs employment delivery system and programs. borne by employers (Brown and Koettl 2012; Kuddo 2012).24 Usually, effective programs offer opportunities The definitions of AMLPs and the specific policies they entail vary in the literature. For an overview of ALMP classifications in the literature, see, for exam- 24.  ple, Romero and Kuddo (2019). Kluve (2016) refers to a fourth category of ALMPs that focus on the direct creation and provision of public works and other activities that produce public services and goods. The goal is to keep the most disadvantaged individuals in contact with the labor market and avoid loss of human capital during often prolonged periods of unemployment (Kluve 2016). This category is not explicitly outlined here as these programs are less relevant in low- and middle-income countries. Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 73 for lifelong learning and can be coordinated with education. Meanwhile, on the labor supply side only 10 other human capital programs such as early child- percent of job-seekers have no education, and there hood programs and education programs to be espe- is a shortage of job-seekers with certificates. In addi- cially impactful. The objectives of ALMPs are further tion, there are signs of a spatial mismatch in the labor described in box 4.1. market in which job-seekers with the required quali- fications (particularly those with lower levels of edu- Similar to ALMPs, TVET programs are also instrumen- cation) are not located in regions experiencing supply tal in strengthening the employability of job-seekers by shortages. promoting the development of skills demanded by the labor market in Bhutan. However, TVET courses are Many employers still do not sufficiently rely on TVET, generally geared more toward educated workers who although specific technical skills will be of greater have completed high school at a minimum, whereas importance in the future. And yet only a small percent- ALMPs target more vulnerable groups with lower age of individuals in the labor force have TVET qualifi- levels of education. In addition, TVET courses tend to cations. As documented in chapter 3, many firms are be longer in duration and more specialized than ALMP experiencing labor shortages. They also have training training. needs but have no links to potential providers of train- ing. Among the few firms that hire TVET graduates, In Bhutan, a large inactive population—especially firms’ perceptions of the quality of TVET graduates are women— need targeted support programs. As described positive. in chapter 2, the decline of labor force participation has mostly been driven by inactive young women. One Finally, the country’s strong dependence on the agri- in three women is not in education, employment, or cultural sector and its vulnerability to natural disasters training (NEET) because of family duties. During the add to the existing weaknesses in the labor market. COVID-19 pandemic, the rise in unemployment was As shown in chapter 2, the structural transformation mostly concentrated in urban areas, especially among has remained slow. Bhutan’s labor market is domi- women. nated by an agricultural sector that does not guaran- tee livelihoods for many rural households, leaving Employment support programs are needed to help them in poverty. Because of the country’s high risk reduce the extensive skill and spatial mismatches in of facing natural disasters such as flooding and land- the labor market. As illustrated in chapter 3, firms slides, Bhutanese households are vulnerable to shocks are demanding or expect to demand workers with a that put their incomes at risk. Climate models project low to medium level of education and with some spe- a substantial increase in the likelihood of droughts and cific technical skills (through certificates or diplomas) heatwaves in the future (World Bank and ADB 2021). to fill positions mainly in services and sales and craft To effectively support individuals in the job market, and related trades. Seventy percent of the expected ALMPs and the TVET system will need to consider these new vacancies are for workers with up to a secondary vulnerabilities. level of education, and 27 percent are for those with no 74 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice Box 4.1. Objectives of active labor market programs Active labor market programs (ALMPs) support both the demand and supply sides of the labor market. Their interventions target labor demand by promoting self-employment and entrepreneurship. They can boost the labor supply by increasing individuals’ employability through providing skills training and the means to increase one’s success in finding a suitable job. ALMPs typically cover vulnerable groups such as elderly workers, long-term unemployed, women, spe- cial-needs individuals, and youth (Castro et al. 2020). Although these programs, by design, should be suitable for large parts of the population, their main target group is disadvantaged youth such as unem- ployed and out-of-job youth, low-skilled youth, and school dropouts; those with limited access to edu- cation and the formal labor market; and those not in education, employment, or training, or NEETs (Kluve 2016). Labor market training programs are the most widespread ALMPs aimed at increasing the human capital of two groups of beneficiaries, the general population of unemployed and low-income youth (McKenzie 2017). These interventions can include classroom vocational/technical training, on-the-job training, and soft skills training. According to evidence from Latin America and the Caribbean (LAC), training programs of four months or less are significantly less likely to have positive effects. The LAC study also finds that female participants and long-term unemployed tend to benefit more from pro- grams than other groups (Kluve 2016). McKenzie (2017) suggests that in low- and middle-income coun- tries, the strongest positive effects of ALMPs providing vocational training are on employment. A special form of this type of ALMPs is the TVET system that provides youth with vocational skills while they tran- sition from school to work and thus aims to boost their employability. Job search assistance programs seek to increase the efficiency of the job search process and the quality of the resulting job matches. Job search assistance programs can include job search training, counsel- ing, monitoring, and job clubs (Kluve 2016). In randomized control trials in developing countries, no significant effects of matching and search assistance could be found on employment; their advantage is that they are much cheaper than vocational training and wage subsidies (McKenzie 2017). Private sector incentive programs include the provision of microfinance products to the unemployed and underemployed and wage subsidies to employers. A prominent example of these programs is microfinance schemes in which microloans are offered to low-skilled individuals, often in combination with training opportunities to support business development (OECD and EC 2021). In addition, wage subsidies are provided to employers to cover part of their labor costs, which may have a lasting positive impact on employment. For example, evidence suggests that in developing countries wage subsidies may help increase the share of formally employed workers (Aşık et al. 2022). Also, workers hired as a result of wage subsidies may be able to learn on the job and increase their productivity above the min- imum wage (McKenzie 2017). Wage subsidies can also help households smooth temporary shocks and boost temporary employment creation (McKenzie 2017). Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 75 Box 4.1. Continued The program costs of ALMPs vary widely, with some ALMPs more cost-effective than others. Training programs have variable costs, depending on their content, length, and equipment. The cost of ALMPs that provide subsidies depends on the amount of these transfers. A country’s political and economic context and wage levels usually affect these subsidies (Angel-Urdinola and Leon-Solano 2013). Evidence from Romania shows that employment services and public works programs tend to be more expensive than training programs (Rodriguez-Planas and Benus 2010). 4.2 Bhutan’s employment support Bhutan’s employment support programs and delivery programs and delivery system system consist mainly of ALMPs, TVET courses, ESCs, and an LMIS (figure 4.1). A distinction is made in this Bhutan spends more than any other South Asian chapter between ALMPs and TVET programs because country on ALMPs and TVET. As of 2019, 0.1 percent of their profiles of beneficiaries differ, as well as the the country’s gross domestic product (GDP) was spent length and type of intervention. ESCs are mandated to on skills development and training, with only Singa- provide labor intermediation and help achieve quality pore spending more, 0.3 percent. Bhutan’s ALMPs matches between labor demand and labor supply. A offer relatively generous benefits—the average ALMP well-functioning and up-to-date LMIS in which all benefit is equivalent to 26 percent of GDP per capita survey and administrative-related labor market data (ADB 2019). In addition, Bhutan has an established in Bhutan could be integrated is important to support TVET sector through which job-seekers and workers evidence-based policy making around workforce can access a variety of training and skill development training needs and employment promotion services. opportunities. An LMIS could also support policies for reducing skill mismatches in the labor market by enabling a better Figure 4.1. Pillars of the employment delivery system in Bhutan Pillars of Bhutan’s Employment Delivery System ALMPs TVET Employment Labor market system services information centers system Bhutan’s labor market Source: World Bank. Note: ALMPs = active labor market programs; TVET = technical and vocational education and training. 76 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice understanding of the profiles of workers and how they emphasized the importance of structural change and align with the skills and occupations demanded by the the importance of enhancing institutionalized skills private sector. training for the labor market (see box 4.2). In the upcoming 13th Five-Year Plan, MoESD is mandated to increase by 2029 the share of the workforce with certi- The legal mandate for the provision of fications in vocational and technical skills to 80 percent employment support programs is shared by (MoESD 2022). Another key policy is the 2013 National two ministries and is underlined by the 12th Employment Policy, which stresses the importance Five-Year Plan of the Royal Government of of ALMPs in promoting entrepreneurship and the Bhutan (RGoB). employability of vulnerable groups in society and aims to foster stronger collaboration between industry and The Ministry of Education and Skills Development TTIs (National Council of Bhutan 2015). (MoESD) is the primary government agency respon- sible for overseeing training and skills development Beginning in September 2024, the RGoB is planning programs. Since November 2022, the Department of to implement the Gyalsung National Service Program, Workforce Planning and Skills Development (DWPSD) envisioned to be a one-year program mandatory for in MoESD has been carrying out the majority of ALMPs all youth age 18. According to the program’s webpage, and overseeing the public TVET institutions known as it will consist of “three months of basic military train- Technical Training Institutes (TTIs). Before govern- ing followed by National Education, Life Skills and spe- ment restructuring became effective in January 2023, cialized training in various fields ranging from home the Ministry of Labor and Human Resources (MoLHR) construction technologies, computing and entrepre- was the agency in which the majority of ALMPs and the neurship to focused development of skills in agricul- TVET programs were consolidated. ture.” 25 It is unclear what the mandate of either MoICE or MoESD will be in administering the program and The Ministry of Industry, Commerce, and Employ- how the program relates to ongoing training efforts. ment (MoICE) is responsible for administering selected ALMPs and promoting the country’s entrepreneurship agenda. MoICE administers a wage subsidy program ALMPs in Bhutan are small, fragmented, and known as the Youth Engagement and Livelihood not specifically focused on the activation of Program, or YELP, and implements the Startup and women. Cottage and Small Industries (CSI) Development Flag- ship Program, which includes overseeing five business The majority of ALMPs cater to vulnerable youth and incubation centers and providing entrepreneurship provide labor market training that focuses on upskill- training. ESCs under the Department of Employ- ing (table 4.1). The selected currently active ALMPs are ment and Entrepreneurship are mandated to connect YELP, the CSI Development Flagship Program, Special job-seekers with employers and to provide them with Skills Development Program (SSDP), Village Skills referrals and job search assistance. Development Program (VSDP), Critical Skills Train- ing (CST) program, Critical Capability Development Although over the last decade a variety of policies have (CCD), and Skills Development Plan (SDP). The list of been implemented to boost employment and employ- ALMPs in table 4.1 is not comprehensive because of ability of the population, the 12th Five-Year Plan (2018– knowledge gaps.26 23) provided the most decisive guidelines. The plan 25. http://www.gyalsunginfra.bt/?page_id=175. 26.  Summaries of the presented ALMPs appear in appendixes E and F. Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 77 Box 4.2. The 12th Five-Year Plan (2018–23) The 12th Five-Year Plan (FYP) was aimed at promoting economic growth through sector diversification away from agriculture and toward high-productivity sectors (MoLHR 2021b). Bhutan’s economy remains undiversified, with agriculture making up the largest share of the economy (17 percent), followed by electricity (16 percent), and construction (15 percent). As a major strategy to promote the diversification of the economy, the 12th FYP introduced the Startup and Cottage and Small Industries (CSI) Develop- ment Flagship Program to create new CSIs and improve the competitiveness of existing ones. The plan acted as an important guide for policy interventions to promote skills and education as well as the employability of job-seekers. It acknowledged the current shortcomings of the TVET system such as a poor technical and institutional capacity to deliver relevant, quality programs; weak coor- dination among key stakeholders; and a limited intake capacity in TTIs. The plan promoted strength- ening the quality, relevance, and access of vocational training programs to improve the employability of TVET graduates. In addition, the plan acknowledged that the undiversified economy has led to a concentration of jobs in a few sectors: agriculture (51 percent), wholesale and retail trade (9 percent), and public administration (8 percent). It called for mandated government agencies to promote pro- grams to enhance the quality of and access to skill development opportunities and the employability of job-seekers. Finally, the 12th FYP stressed the importance of decentralization—that is, the empowerment of local governments. The RGoB is seeking to achieve greater flexibility and autonomy in making choices and setting priorities in the provision of public goods and service delivery. The 12th FYP emphasized the importance of the variations across Bhutan’s local labor markets and the unique regional employment characteristics. Therefore, a goal is the fair allocation of resources in dzongkhags that have high unem- ployment in agriculture. 78 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice Table 4.1. Overview of selected ALMPs Name Ministry Type Services provided Target groups No. of beneficiaries Youth Engagement MoICE Labor market On-the-job training in Youth 2,545 and Livelihood training; private various vocational fields; (2022–23) Program (YELP) sector employment wage subsidy; various assistance financial support of entrepreneurship Startup and MoICE Labor market Entrepreneurship training, Youth 2,079 (2022– Cottage and Small training; job search business competitions, 23) (training Industries (CSI) assistance; private business incubation centers component) Development sector employment Flagship Program assistance Special Skills MoESD Labor market Vocational training in Vulnerable groups such 938 (2022, Development training trades such as plumbing, as armed forces, monks/ SSDP and Program (SSDP) basic tailoring, and basic nuns, juveniles and delin- VSDP) carpentry quents, prisoners, former gang members Village Skills MoESD Labor market Vocational skills and vil- Rural individuals Development training lage-specific training such Program (VSDP) as tailoring Critical Skills MoESD Labor market Vocational skills training Youth 506 (2022) Training (CST) training such as baking, fash- ion design, and online freelancing; Critical Capability MoESD Labor market Upskilling and reskilling Employees in non–civil 990 (2022) Development (CCD) training training in private sector service sector such as accounting, book- keeping, and e-commerce Skills Development MoESD Labor market Upskilling and reskilling Laid-off employees due 1,881 (2022) Plan (SDP) training courses in welding, fitting, to COVID-19, job-seekers, 3D printing, entrepreneur- overseas returnees ship learning, employer matching Sources: MoESD 2022; YELP: MoLHR 2021a and Bhutan Today 2023; CSI Development Flagship Program: MoICE 2023a. Note: MoESD = Ministry of Education and Skills Development; MoICE = Ministry of Industry, Commerce, and Employment. Although women are the largest group out of the labor safe work spaces. Such programs can play a critical role force, the existing ALMPs cater mostly to unemployed in supporting vulnerable women through job search youth (see table F.1 in appendix F). The YELP, CST, and assistance, mentorship, and affordable childcare. CSI programs focus on the employability and employ- ment of youth. The VSDP caters to the rural popula- All ALMPs are small programs with limited regional cov- tion, and the SSDP focuses on other vulnerable groups. erage. The plethora of programs and their fragmenta- The SDP was established to increase the employabil- tion may be limiting the impact on vulnerable groups. ity of young job-seekers and all those vulnerable indi- In 2022–23, 2,545 individuals went through YELP, the viduals affected by the COVID-19 pandemic. None of largest ALMP in the country. Of the five skill develop- these programs is gender-sensitive—that is, includes ment programs under MoESD, the majority had less a component that specifically targets women by pro- than 1,000 beneficiaries that year. Currently, YELP is viding childcare support, transportation subsidies, or the only ALMP that provides wage subsidies, while all Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 79 other programs provide training opportunities. There same time, that prevents individuals who may not be are no plans to scale up YELP so that more vulnerable the most vulnerable from attending multiple ALMPs. individuals can be supported. In fact, there are reports Meanwhile, short- and long-term tracking of gradu- that the program is currently not operational because ates‘ labor market outcomes has not been undertaken of limited funds. In addition, it is not clear how many in a systematic way, which makes assessing training individuals have been trained in specific locations and quality and relevance to the labor market difficult. To by private or public training providers, and where they address this shortcoming, MoESD, with World Bank have been working since graduation.27 It appears that support, recently launched a tracking system within its most beneficiaries are located in the urban centers existing management information system (MIS) to sys- such as Thimphu and Bumthang, limiting their impact tematically collect employment data on all of its train- on other vulnerable populations in rural areas. In ing graduates within two years of training completion. general, many ALMPs lack sustainable mechanisms for funding, and the plethora of programs for a small Lack of system links across ALMPs mandated by dif- country may be diluting their impact. ferent ministries may be contributing to the limited impact of these programs. It is unclear how MoESD’s ALMPs do not focus sufficiently on improving links ALMPs are aligned with the ALMPs under MoICE. For to employers and being demand-driven. YELP is the example, there is no clear way in which trainees in the only program that provides some links to employers MoESD skills programs interested in entrepreneurship via on-the-job training. In addition, the vocational could be linked to or referred to training opportunities training offered by ALMPs does not appear to be sys- provided by MoICE’s CSI Flagship Program. tematically informed by employer assessments. None of the existing ALMPs offers the soft skills training that could help individuals build and maintain pro- The public TVET system is well established, fessional relationships, better manage their time, but it remains too small to meet labor and solve problems (see table D.3 in appendix D). As demand and is not sufficiently linked to a result, training may not equip individuals with the employers. skills that employers need, significantly reducing the likelihood that graduates have positive labor market Bhutan’s TVET system complements ALMPs by provid- outcomes. Adding to the problem is the lack of trans- ing young job-seekers with training opportunities in a parency about where the training provided by ALMPs variety of vocational fields (box 4.3). The average TVET take place beyond TTIs. student is equally male or female ages 20–24 and has a higher secondary education and a lower socioeco- Meanwhile, ALMPs are not rigorously evaluated. There nomic background (MoESD 2022, 2023a). According to are gaps in the data on registries, graduates’ long-term a 2022 TVET study, 52 percent of individuals with TVET labor market outcomes, and the quality of the support qualifications work as technicians and associate pro- provided by ALMPs. No monitoring and evaluation fessionals. The TVET system is a means to intergener- (M&E) mechanism is in place that could reveal which ational occupational mobility in Bhutan: 48 percent of ALMPs are meeting their goals. And there is no trans- the 2021 TVET graduates are from families that work in parent mechanism to ensure that the most vulnerable the agricultural sector, and 64 percent of their parents individuals are selected for a specific ALMP and, at the or guardians do not have any education (MoESD 2022). 27.  The 2022 Tracer study is an overview of the dzongkhags where the training courses provided by ALMPs have been implemented. 80 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice Box 4.3. The TVET system in Bhutan Since the 1960s, the technical and vocational education and training (TVET) system has been providing youth with training with the goal of building a modern workforce. Major objectives of the TVET sys- tem are to offer courses relevant to the Bhutanese labor market; increase the attendance at vocational training programs; provide guidance and support for vocational training programs; provide Technical Training Institutes (TTIs) with guidance and support on TVET policies, interventions, and strategies; design, develop, and review TVET curricula; develop and review guidelines for Training of Teachers and respective curricula; and provide TVET providers with capacity development programs (MoLHR 2020). The goal, then, is to achieve a globally competitive workforce and create a more cohesive society, secur- ing stable incomes for all citizens (MoESD 2022). Nine vocational training institutes under the direct management of the Ministry of Education and Skills Development (MoESD) provide youth with a variety of training opportunities. Currently, public TTIs are located in Chumey, Khuruthang, Rangjung, Samthang, and Thimphu. In addition, two institutes for Zorig Chusum (IZCs) specialize in training of local arts and craftsmanship: the College of Zorig Chu- sum (CZC) in Trashiyangtse and the National Institute for Zorig Chusum (NIZC) in Thimphu. The Jigme Wangchuck Power Training Institute (JWPTI) in Dekiling, Sarpang, and the Rural Development Training Center (RDTC) in Zhemgang offer additional vocational training (MoESD 2022). In the TTIs, regular technical training lasts from three months to two years, and at the IZCs courses last from one to six years. The entry level required to enroll in vocational courses is grade 10—that is, the final year of the basic educational (middle-secondary) level (MoLHR 2020). In addition to the public TTIs, a growing number of TVET institutes are being set up by private promot- ers. In 2022, 80 percent of all private and public TVET institutions were under private ownership. To register, a private training provider must provide information on physical resources, teaching, learn- ing, and the internal quality management system. Private training providers can complement public TVET institutions because they help meet training needs and can offer more specialized services. The registration process for private TVET providers includes an institutionalized assessment based on points for different components (MoESD 2022). Overall, the quality of TVET programs and students’ in their future jobs. Almost 60 percent of respondents satisfaction have improved in recent years. In the past, were satisfied with the duration of the TVET course the main reasons for student dissatisfaction with the and on-the-job training, while the rest felt it was either TVET system were its limited course selection, the too short or too long (MoESD 2022). inadequate skills offered, and poor teacher qualifica- tions (MoLHR 2020). According to graduate feedback Despite significant improvements in the TVET system, in 2022, on average, 70–80 percent of graduate respon- the number of individuals with TVET qualifications has dents regarded the overall course content and TVET remained low, not meeting the demands of the private teaching as good or very good. Most graduates have sector. In 2022–23, 401 individuals graduated from TTIs found the skills gained through TVET courses useful and IZCs of whom only about one-third were female Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 81 (table 4.2). Between 2018 and 2022, the total number of in 2022, lower than the share in 2021, 3.2 percent TVET graduates declined from 962 to 401, and before (MoESD 2022). In recent years, there has been a wid- 2022 about two-thirds of TVET graduates were female ening gap between the number of TVET graduates and (figures 4.2 and 4.3). Among all employed individuals the growing demand for vocational skills required by in Bhutan, only 2.5 percent had TVET qualifications employers, as noted in chapter 3. Table 4.2. Number of regular graduates in TTIs and IZCs, by year and gender 2018–19 2019–20 2020–21 2021–22 2022–23 Total Institute M F T M F T M F T M F T M F T M F T College of Zorig 68 75 143 12 29 48 36 63 99 16 62 78 125 76 201 257 305 569 Chusum (CZC) Jigme 46 101 147 18 39 57 16 50 66 12 25 37 0 0 0 92 215 307 Wangchuck Power Training Institute (JWPTI) National 44 97 141 2 55 57 50 76 126 39 71 110 143 57 200 278 356 634 Institute for Zorig Chusum (NIZC) TTI-Chumey 27 22 49 40 42 82 0 0 0 0 0 0 0 0 0 67 64 131 TTI-Khuruthang 43 45 88 18 34 52 30 59 89 24 43 67 0 0 0 115 181 296 TTI-Rangjung 56 122 178 20 55 75 23 62 85 34 68 102 0 0 0 133 307 440 TTI-Samthang 19 128 147 17 72 89 9 111 120 8 29 37 0 0 0 53 340 393 TTI-Thimphu 11 58 69 7 27 34 7 49 56 8 50 58 0 0 0 33 184 217 Total 314 648 962 134 353 494 171 470 641 141 348 489 268 133 401 1028 1,952 2,987 Source: MoESD 2023b. Note: No data were available for the Rural Development Training Center (RDTC). F = females; IZCs = institutes for Zorig Chusum; M = males; TTI = Technical Training Institute; T = total. Figure 4.2. Total number of TVET graduates, Figure 4.3. Share of TTI graduates, by gender, by year and gender, 2018–22 2018–22 No. of TVET graduates Share of TTI graduates (%) 1,000 100 800 80 600 60 400 40 200 20 0 0 2018 2019 2022 2021 2022 2018 2019 2020 2021 2022 Female Male Total Female Male Source: MoESD 2023b. Source: MoESD 2023b. Note: The displayed years indicate the respective fiscal year. For example, 2018 refers Note: The displayed years indicate the respective fiscal year. For example, 2018 refers to fiscal 2018/19. TVET = technical and vocational education and training. to fiscal 2018/19. TTI = Technical Training Institute. 82 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice The TVET programs offered have not been demand- employment opportunities in all the places where driven because links with employers have been public TVET institutions are located. Certain areas may missing, leading to high unemployment among TVET also have a higher demand for graduates that TVET graduates. As shown in chapter 3, not many employ- institutions cannot meet. The importance of the geo- ers indicated that they had connections to public graphic proximity of TVET institutions to sectors and TVET institutions. In addition, there is no specific evi- employers has not been assessed. dence that the design and delivery of TVET training are carried out in collaboration with employers. And there Finally, the TVET system is not inclusive enough because has been no discussion of sharing costs with employ- private training providers, and thus their graduates, ers and to what extent they could subsidize vocational are mostly concentrated in Thimphu. Although the training. public TVET institutions are spread across the country with an emphasis on urban areas, the majority of As a result, the unemployment rate among TVET grad- private training providers are registered in Thimphu, uates is relatively high, and those who find jobs do with only a few in the eastern and central parts of the not work in future-oriented sectors. In 2022, only 67 country (MoESD 2022). The establishment of private percent of TVET graduates were able to find employ- training providers in Thimphu seems largely driven by ment within a year of graduation. Among the unem- economic considerations as it has the highest concen- ployed TVET graduates, about 49 percent indicated tration of suitable individuals and potential employ- that they needed further employment facilitation ers. By focusing strongly on the capital city, private support (MoESD 2022). As discussed in chapter 3, the training providers cannot currently target vulnerable highest in-demand sectors in the future will be ser- rural individuals sufficiently and thus cannot contrib- vices and sales, as well as craft and related trades. Yet as ute to enhancing rural economies. of 2022, 52 percent of employed individuals with TVET qualifications worked as technicians and associate professionals, 20 percent as professionals, 9 percent ESCs are not empowered enough to play a in the services and sales sectors, and 7 percent in craft. pivotal role in job placement and matching. It appears that TVET courses equip individuals with skills that may not be in demand in sufficiently high ESCs play an important role in connecting job-seek- numbers in the future (MoESD 2022). ers with employers and facilitating the process of job matching (MoICE 2023b). Currently, the five ESCs Public TVET institutions are not close geographically to across the country provide job-seekers and employers employment opportunities. In 2022–23, the only stu- with various services such as job matching and place- dents enrolled at public TVET institutions graduated ment, information on current job vacancies, and help from the College of Zorig Chusum (CZC) and Jigme with arranging job interviews.28 An increasing number Wangchuck Power Training Institute (JWPTI). No one of job-seekers have been visiting ESCs in past years, has graduated from TTI-Chumey since fiscal 2020/21 with an increase by 18 percent from fiscal 2021/22 to (table 4.2). It is unclear from which regions TVET stu- fiscal 2022/23. dents come and whether they move after gradua- tion. Of those in the 2021 graduate cohort, 44 percent Although more and more job-seekers use ESCs, gaps in were employed in Thimphu, followed by 13 percent in training and staffing remain, severely limiting the ser- Paro (MoESD 2022). Apparently there are insufficient vices offered. Employment officers are well trained in 28.  In addition to the five ESCs, two new labor and employment offices are being set up in the Trongsa and Mongar districts. Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 83 job counseling, but they are not dedicated full-time to Bhutan has a new LMIS, but it is still in its ESCs. Currently, only ESC Thimphu has two perma- early stages and needs further development. nent staff members, whereas all the other ESCs rely on interns, with high turnover. Due to this shortage The four key functions of an advanced LMIS are defined of trained staff, mostly simple tasks such as providing by World Bank (2021) as follows: (1) job matching; (2) basic information and online registration in the job career and skills guidance; (3) government support; vacancy and training portal (Bhutan LMIS-BLMIS29 ) and (4) labor market intelligence. Bhutan has recently can be offered. Only ESC Thimphu is able to provide made advances in developing a LMIS to publish digest- some individual counseling sessions due to its many ible information on the labor market and make it easily years of experience. However, no data are available to accessible to job-seekers, employers, and training ascertain the level and quality of its services. institutes, as well as across ministries. By publishing updated labor market information, an LMIS can serve Data on ESCs are scarce because the performance all labor market stakeholders by helping students, the of ESCs is not monitored. Currently, the only avail- inactive, job-seekers, and other groups make informed able data are the registration of visitors in handwrit- decisions about career choices, as well as increase their ten logbooks. Yet data on the visitors, job-seekers, awareness of the occupations, vacancies, and skills and employers interacting with ESCs and the services available in the labor market. It can also inform cur- offered are largely missing. Without gathering data in riculum design and the provision of training through a systematic way, efforts to build an M&E system that knowledge of the most in-demand occupations. could monitor, evaluate, and improve the perfor- mance of ESCs are difficult. Development of the LMIS is in its early stages. The current LMIS platform is administered by the Data All ESCs except the one in Thimphu are located in gov- Science Project on Labor Market Initiative under the ernment administrative buildings and lack proper supervision of Royal Office for Media.30 However, it is equipment and resources. Apart from ESC Thimphu not clear to what extent data sharing arrangements which has its own dedicated space, all other ESCs are are in place between the Data Science Project and the located in the ministry’s regional offices. Thus ESCs ministries that collect and store administrative labor have to follow the rules of the administration. For market data. It is also not clear whether the mandate example, visitors are obligated to dress formally to of the Data Science Project is to only compile data on visit the centers, which significantly reduces the acces- the labor market from different sources and analyze it sibility of ESCs. In all ESCs, visits are limited to the offi- to produce and disseminate labor market information, cial working hours of the ministry. Computers with or whether it is required to initiate new data collection reliable internet access are scarce, and office space is (special-topic surveys) to fill information gaps. mostly limited to one room with basic furniture. It is, then, difficult to provide more sophisticated services In addition, gaps in the published information remain. such as counseling. For example, the LMIS portal currently does not contain information on earnings and thus returns to education and occupational skills. There is also a lack of clarity on data sources and methodology. In addition, 29. The BLMIS platform allows each worker to create a unique profile and apply to multiple TVET training and job vacancies. It can be accessed at https:// blmis.gov.bt/. 30.  For a more detailed information about the portal, go to https://data.bt/. 84 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice the usefulness of some of the data presented are not whereas men usually work in higher-paying sectors. necessarily obvious to the average intended users of Chapter 2 points out that family responsibilities, espe- information (that is, the general audience, including cially childcare, mainly prevent women from enter- students and job-seekers). ing the labor force and finding well-paying jobs. One solution may be to make the current ALMPs more gen- In the future, the LMIS portal could be linked to the der-sensitive by adding offerings that are specifically BLMIS job matching and training registration portal, tailored to women. In addition, new ALMPs could be so that students, employers, job-seekers, and train- developed that tackle the issue of work-life reconcilia- ing service providers can easily access and understand tion by providing free or affordable childcare services, information about the labor market, thereby allow- as well as job search assistance, mentorship, and coun- ing them to make informed hiring decisions, as well as seling for women. ALMPs that provide transportation decisions about their careers and the use of services. allowances targeting low-income and rural women could help to facilitate their mobility to regions or loca- tions experiencing labor supply shortages. Efficient 4.3 Conclusion: Ways to ALMPs could also increase women’s access to quality strengthen employment support jobs. Evidence from the Jovénes programs in Latin programs in Bhutan and America has demonstrated that combining classroom international best practices and on-the-job training could significantly increase women’s employment chances and boost wages (box 4.4). This chapter describes several ways in which Bhu- tan’s employment delivery systems could be strength- Enhancement of the demand-driven nature of voca- ened to more effectively support all those in the labor tional training could be achieved through robust market. First, the priority must be to tailor programs collaboration with employers, aligning ALMPs with to women to move them into quality employment, sought-after skills, employers’ expectations, and the build stronger links between job-seekers and employ- prevailing market demands. To pave the way for this ers, and boost the relevance of the TVET system for transformation, it is advisable that the relevant min- the labor market. Second, steps should be taken to cut istries engage employers as early as possible during back on the number of ALMPs and consolidate them the program design phase. This proactive partnership in a way that would enable the government to scale with employers would be pivotal in tailoring ALMPs to up the impactful ones. This consolidation would help match precisely industry, sector, and company pre- improve the allocation of spending and provide fiscal requisites. Also, ALMPs executed in cooperation with space to implement initiatives where clear gaps exist. employers could ensure that the offerings effectively Finally, institutional reforms such as strengthening endow students with qualifications indispensable in the coordination between stakeholders and the data the private sector. Heightened employer involvement sharing systems, as well as promoting evidence-based could entail the integration of more on-the-job train- programming, could help increase the efficiency of the ing opportunities and the provision of private sector whole system. employment support, including wage subsidies. This multifaceted approach would not only cultivate a more It is important to develop ALMPs that bring inactive responsive vocational training landscape, but also help women into employment and increase women’s access meet beneficiaries’ aspirations and the demands of the to quality jobs. Currently, women work predominantly job market. in agriculture and are often unpaid or underpaid, Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 85 Box 4.4. Global evidence from Latin America’s Jovénes programs Since 1991, the Jovénes programs, a comprehensive labor market intervention in Latin American coun- tries, have improved the labor market outcomes of their participants. The goal of these programs is to increase skills and improve the employability of poor, disadvantaged, and uneducated youth ages 16–29 (Puerto 2007). A meta-analysis of Jovénes programs in Argentina, Chile, Colombia, the Domin- ican Republic, Panama, and Peru indicates that the programs tend to significantly increase (by 5 to 20 percentage points) an individual’s employment chances, especially women’s. Also, the programs appear to boost earnings, ranging from approximately 10 to 25 percentage points (Ibarraran and Rosas 2009; Kluve 2016). A major strength of the Jovénes programs is the transparent financing and the provision of the training. Governments select training courses competitively through a process in which public and private firms or training institutions can participate, and so the training courses reflect the needs of the employers. Finally, the labor market intervention combines classroom training with subsequent work experience in firms. The training focuses on basic and specific trades, complemented with job search assistance, counseling, and information (Ibarraran and Rosas 2009). Enhancing institutionalized certification of training of graduates’ skills. Also, teacher qualifications should courses and teacher qualifications can strengthen the be regularly assessed and certified so that the teach- quality of the TVET system. Course accreditation was ing meets uniform standards. Bhutan could establish a made mandatory in the 12th Five-Year Plan through governmental body similar to the Korea Skills Quality the Critical Skills Training program. In the future, all Authority (KSQA) described in box 4.5, which certifies TVET courses and training provided through ALMPs vocational training and teacher qualifications. should be certified to increase employers’ recognition 86 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice Box 4.5. The Republic of Korea’s TVET system Since the 1960s, technical and vocational education and training (TVET) programs have contributed to the Republic of Korea’s rapid economic growth and shock resilience. During the 1960s and 1970s, the TVET system provided training for large parts of Korean society to meet the country’s quickly rising labor demands (Lee 2008). In the 1980s, the TVET system was aimed at improving the skill set of work- ers. In the 1990s, the TVET institutions expanded and, together with the Employment Insurance Act, helped to boost the Korean economy after the Asian financial crisis (Ra and Kang 2012). Since then, the goal has been to streamline the division of responsibilities and cooperation among TVET stakeholders (UNESCO-UNEVOC 2018). Various ministries and research institutions have distinct responsibilities in the coordination of the TVET system. The Ministry of Education is responsible for the vocational education taught in schools. The Ministry of Employment and Labor regulates vocational training that is based on programs offered through public or private training institutes and the employment insurance fund (UNESCO-UNEVOC 2018). Founded in 1997, the Korea Research Institute for Vocational Education and Training (KRIVET) conducts research on the development and provision of TVET programs, qualification frameworks, the management of qualifications, the assessment of TVET institutes, and the provision of counseling ser- vices (KRIVET 2022). The TVET system is well-structured within Korea’s educational system, providing many professional opportunities. After six years of primary education and an additional three years of lower-secondary education, individuals can move to three years of vocational and technical training. These graduates are then able to go on to vocational colleges at the postsecondary, nontertiary level (usually two to three years). Graduates of these postsecondary, nontertiary courses are then able to go to universities at the tertiary level. Courses there last from three months to one year for nondegree courses and two years for associate degree and industrial degrees. In addition to formal TVET courses, the Ministry of Employ- ment and Labor and the Ministry of Education offer nonformal TVET programs to promote greater access to various educational opportunities for all citizens and foster lifelong learning (UNESCO-UN- EVOC 2018). Institutionalized quality insurance ensures that the TVET programs meet the needs of Korea’s modern labor market. The Ministry of Employment and Labor and KRIVET have developed and maintained over 331 National Competency Standards (UNESCO-UNEVOC 2018). The Korea Skills Quality Authority (KSQA) is responsible for ensuring the quality of the TVET system (OECD 2020). Moreover, TVET teach- ers must acquire national technical qualifications in their area of expertise and work experience. The educational requirement for teachers and trainers in upper-secondary TVET education is a teacher’s certificate awarded by a college of education, or a master’s degree from a graduate school of education, or completion of a teaching course at a university. Teachers in postsecondary, nontertiary education and tertiary TVET education are required to have a doctoral degree in their specific area (UNESCO-UN- EVOC 2018). Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 87 Private sector assistance programs that provide agri- self-employment in past decades (box 4.6). In Bhutan, cultural and rural workers with microloans pack- MoICE’s CSI flagship program and business incubators aged with entrepreneurial training can help improve have remained focused on industrial development in their productivity, connect them to markets, and urban centers without a special focus on the needs of upgrade their agricultural practices. The agricul- rural populations. The development of entrepreneur- tural sector still dominates Bhutan’s economy, and ship programs that offer vulnerable rural populations structural transformation has been slow. This factor, microloans could boost job creation, which may be together with accelerated climate change leading to especially helpful to accelerate structural transforma- extended droughts that threaten harvests, place agri- tion. However, the provision of microloans should be cultural and rural workers in a vulnerable position in carefully designed to mitigate the challenges faced by which they are barely able to secure sufficient house- previous initiatives. The design process should draw hold income. Best practices from Bangladesh confirm lessons from the National Development Bank, which that the country’s sophisticated microfinance system has reported a high incidence of nonperforming loans has been able to help the rural poor to move into in the agricultural sector (World Bank, forthcoming a). Box 4.6. Bangladesh’s microfinance system Grameen Bank launched the first microfinance operation in Bangladesh in 1976. In the early years, loan products targeted rural nonfarming activities to help the rural poor become productive and acquire assets (Khandker 1998). As of 2018, 750 microfinance institutions (MFIs) were providing about 3.5 million borrowers with microcredit (Microcredit Regulatory Authority 2018). Bangladesh’s common microfinance model is based on the Grameen model, which assumes that group members are best able to assess each other’s creditworthiness (Osmani 2016). Peer pressure then ensures that group members make installments according to the regulations of the MFI (Srinivas 2015). Weekly group meetings pro- mote transparency and guarantee the timely repayment of installments (Haldar and Stiglitz 2016). Access by the poor to microfinance has contributed to poverty reduction in Bangladesh. Microcredit allows the poor to protect, increase, and diversify their sources of income (Littlefield, Murduch, and Hashemi 2003). According to one study, a 10 percent increase in the average microcredit provided to females in a village led to a 0.42 percent increase in household nonland assets and a 0.47 percent increase in household net worth (Khandker and Samad 2014). Microfinance products in Bangladesh have also enabled the poor to establish microenterprises: the interest rates of microcredit are low enough that the poor can repay them and are still able to accumulate capital (Osmani 2016). 88 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice Box 4.7. Green Jobs Green jobs and promotion of the green economy are intended to address climate change and deliver decent work for all. The concept of green jobs refers to the transformation of economies, labor mar- kets, enterprises, and workplaces into a sustainable economy that provides decent work (ILO 2009). Such jobs, ranging from manual to highly skilled, can be found in many sectors of the economy such as transportation, construction, agriculture, recycling, and energy supply. The goal is to reduce the need for energy and raw materials, thereby minimizing pollution, avoiding greenhouse gas emissions, and protecting and restoring biodiversity and ecosystems (CEDEFOP 2009). Over the period 2009–14, the Green Jobs Initiative launched by the International Labour Organization, United Nations Environment Programme, International Organization of Employers, and International Trade Union Confederation was a global effort to provide knowledge, build partnerships, and offer governments policy advice to promote the implementation of national and sectoral policies to create green jobs (ILO 2016). Since then, many low- and middle-income countries have implemented projects to promote green jobs. In the Philippines, the Promotion of Green Economic Development project (2013–16), implemented by the country’s Department of Trade and Industry (DTI) in cooperation with the German Federal Ministry for Economic Cooperation and Development (BMZ), supported micro, small, and medium enterprises (MSMEs) in implementing environmentally friendly, climate-sensitive strategies. The project aimed to (1) provide MSMEs with information and raise their awareness of green economic development; (2) promote business matchmaking and link enterprises with green business services providers; and (3) provide a green policy framework for the DTI so it could mainstream green initiatives in its programs. The result was that more than 300 DTI employees learned to contribute to the transformation to a green economy and green jobs based on their participation in workshops, on-the-job training, and informa- tion trips (GIZ 2016). In South Africa, the project TVET and the Promotion of Innovation for Green Employment (2015–17) supported the construction of power plants generated from renewable sources and the generation of green jobs. The project was implemented by South Africa’s Department of Higher Education and Training and Department of Science and Technology in cooperation with the Germany’s BMZ. It aimed to cooperate with universities, technical and vocational education and training (TVET) colleges, and technology transfer institutions to supply the experts and technologies required to operate a green economy. Hands-on training was provided for water/wastewater and waste disposal management; installing and maintaining solar water heaters and photovoltaic systems; and increasing energy and resource efficiency in production processes. A major project achievement was the integration of the module “Renewable Energy Technologies” into a course available at seven TVET colleges with 500 stu- dents. Teaching and learning materials were developed and used for this course (GIZ 2017). Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 89 Bhutan’s ALMPs should also be linked to other human the specific design features to increase the likelihood of capital programs in the health and education sectors. program success. Designing efficient ALMPs that com- As shown in chapter 2, returns to education are very plement each other and are coherent requires data. As high for both women and men, especially women with discussed in this chapter, the LMIS includes a platform a tertiary education who tend to earn relatively higher where workers can create their profiles and apply for wages. In the future, the DWPSD could effectively TVET training. In the future, the LMIS will need to be coordinate with the Ministry of Health and MoESD to further developed because information gaps remain. expand human development programs to cover vul- It will be important to provide accurate and timely nerabilities along the life cycle. Programs to strengthen data on ALMP performance outcomes, assessing their maternal health and infant health could play an essen- relevance and quality, as well as data on the selection of tial role in strengthening Bhutan’s labor force. In addi- beneficiaries. Integrated data systems can be especially tion, ALMPs that go hand in hand with programs that important for effective crisis responses: scale-up of boost the educational outcomes of primary and sec- existing programs or the creation of new ones requires ondary schoolchildren would be important. a fast, effective response to reach newly vulnerable individuals and households. Strengthening coordination between stakeholders could increase the efficiency of Bhutan’s employment Evidence-based programming could be promoted by delivery systems. Currently, the several ministries introducing a M&E system to track the effectiveness of involved in the provision of ALMPs have overlapping ALMPs to help fill current gaps. In addition to enhanc- mandates. Inter-ministerial coordination, especially ing Bhutan’s LMIS, M&E frameworks could be devel- between MoESD and MoICE, could be strengthened oped and evaluations and robust impact assessments to avoid duplication and inefficiencies. It is recom- conducted for all existing ALMPs. Results-based indica- mended that a steering committee be formed to engage tors related to the quality of the services or whether the stakeholders. Members would represent the responsi- provided services have helped to improve labor market ble ministries, the Kidu Foundation, and potentially outcomes of the beneficiaries should be tracked. An international organizations to reach a consensus on M&E department could be established to take respon- labor market interventions, their goals, beneficiaries, sibility for running large-scale randomized, controlled and outcome tracing. trials to evaluate the effectiveness of ALMPs. To com- plement these trials, Bhutan could introduce an online Strengthening data sharing systems can improve the database devoted to research on global best practices coordination and ability to use ALMPs efficiently. It is by ALMPs. Examples of such databases can be found in critical to choose the right type of ALMP and identify Denmark and the United Kingdom (box 4.8). 90 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice Box 4.8. Documenting global best practices in labor interventions Some countries are capturing and sharing information on labor market policy evaluations. Jobeffekt (Denmark) Jobeffekter.dk is a knowledge bank launched in 2013 by the Danish Agency for Labor Market and Recruitment. It provides a quick, accessible, and up-to-date overview of the labor market policies and programs that have had a positive effect based on more than 500 Danish and international research- based studies. Each uploaded study is assessed by researchers. Publications are available for many target groups, including those who are short- and long-term unemployed. See https://jobeffekt.dk/. What Works Network (United Kingdom) This network is made up of 13 research centers committed to increasing both the supply of and demand for evidence in their policy area, and their output is tailored to the needs of decision-makers. The avail- able evidence contains impact evaluations and academic work. Findings are shared in an accessible way to inform policy decisions. See https://www.gov.uk/guidance/what-works-network. Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 91 Chapter 5 Policy Directions Jumana Alaref This report highlights four broad challenges in Bhu- Policy direction 1: Accelerate tan’s labor market: (1) low female labor force partici- job creation in the private pation, especially among low-skilled women in urban sector by implementing vertical growth policies (sector-specific), areas; (2) the low productivity of agricultural workers, especially women in rural areas; (3) limited labor pro- ductivity and job creation in the private sector in urban as well as horizontal reforms areas, resulting in high unemployment among edu- (sector-neutral). cated youth and in low employment quality outside the public sector; and (4) the difficulties facing private Bhutan’s forthcoming Country Economic Memoran- sector firms in accessing trained labor for low-skill dum (CEM) emphasizes the importance of adopting vacancies, which may be one of the factors contribut- vertical policies that support the growth of specific ing to limited productivity and growth (see table 5.1). sectors in view of limited private sector development outside the agricultural and public sectors. The labor From these challenges emerge several directions for demand prospects indicate that the current sectors orienting public policies and programs to address the most dominant within the private sector will be unable constraints facing workers and employers in Bhutan. to absorb the educated working-age population in The policy directions are summarized in this chapter. urban centers into high-value-added jobs that make delivery systems effectively address some of the chal- use of their skills and aspirations. Selecting sectors to lenges facing both workers and firms. As outlined in support their growth through government subsidies chapters 2 and 3, those challenges include lo and programs requires careful analysis to mitigate risks.31 The CEM emphasizes that countries that imple- ment successful vertical interventions “rely on market signals to hold themselves accountable, sustain com- petitive pressures, and drive innovation in the devel- opment and export of new products.” The following sectors that have created jobs over the last few years, especially for skilled workers, but are currently small could be considered: financial, 31.  insurance, and real estate activities; information, communication, and technology; and professional, scientific, and technical activities. A 2022 United Nations Development Programme (UNDP) study proposed inward and outward-focused industrial propositions for three sectors: agriculture, creative, and digital (UNDP 2022b). This study could form the basis for future work by the Royal Government of Bhutan (RGoB). 92 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice Successful diversification policies also require imple- start-ups; and (3) skill-related policies to strengthen menting horizontal policies that support job cre- firms’ capabilities by focusing on managerial practices ation and growth across all sectors. As highlighted in and soft skills. Private sector and governance reforms this report, most firms in Bhutan remain small and are needed on the investment climate and foreign do not expand over time. Although young, dynamic, direct investment, access to finance, and the efficiency high-growth enterprises (sometimes referred to as of the state-owned enterprise (SoE) sector. “gazelles”) are increasingly recognized in the litera- ture as critical for the creation of better-quality jobs (Farole et al. 2017; ILO 2019), some evidence suggests Policy direction 2: Strengthen that it is difficult to identify firms with a high poten- economic inclusion programs to tial for growth. Policies to support firm growth are support rural workers and help them improve their productivity more likely to succeed if they focus instead on factors related to innovation, agglomeration and network economies, managerial capabilities and workers’ skills, and access to protections against global links, and financial development. These policies climate change vulnerabilities. can, in turn, contribute significantly to increasing the probability of a high-growth episode (Goswami, Med- Rural workers remain locked in low-productivity and vedev, and Olafsen 2019). subsistence employment, with limited access to knowl- edge, green skills, markets, services, and protections To improve innovation in Bhutan, the Ministry of against climate change vulnerabilities. Improving agri- Industry, Commerce, and Employment (MoICE) over- cultural productivity is imperative to accelerating sees incubation and acceleration centers under five structural transformation and to the reallocation of royal colleges. These centers are mandated to support workers toward better-quality jobs in the services and entrepreneurship development as part of an effort to industry sectors (World Bank, forthcoming a). boost economic growth and job creation for educated workers. However, the centers are still in their early There is scope to strengthen the present skills develop- stages and also face funding uncertainties. The quality ment programs under the Ministry of Education and of their incubation and acceleration services could Skills Development (MoESD) to improve rural workers’ be further strengthened to connect entrepreneurs links with climate change adaptation and provide them to the broader entrepreneurial ecosystem in Bhutan. with a targeted package of training. Technical training Improvements could link existing and aspiring entre- can help improve farmers’ productivity and agricul- preneurs to (1) access to financing; (2) market valida- tural climate-friendly practices, as well as climate resil- tion; (3) business plan development; (4) research and ience and mitigation. Training in critical business skills, development support; (5) mentorship; and (6) supply mentoring, and psychosocial support can encourage chains and markets. workers to take risks and diversify their income-gen- erating activities. Training could also focus on the skills Complementary horizontal policies that relate to labor required for green jobs to facilitate workers’ transition market, private sector, and governance reforms are to the green economy. needed. Labor market reforms include: (1) labor reg- ulations that can support workers’ mobility and firms’ System linkages between MoESD or MoICE and the access to labor; (2) a functional labor market infor- Ministry of Agriculture and Livestock (MoAL) could mation system (LMIS) that can identify skills available allow for beneficiary referral and the provision of a in the labor market and support the hiring needs of coordinated package of economic inclusion services. Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 93 Those services could include technology transfer, and (3) reduce program fragmentation and scale up improved inputs, microloans, and off-farm activi- programs proven to be relevant and impactful. ties that connect farmers, especially youth, to urban markets. 1. Improve the design of existing programs to improve their efficiency and effectiveness Policy direction 3: Strengthen employment support programs The TVET sector can be reoriented to improve its con- to support the activation of low- nection with the private sector. This report finds that skilled women in urban areas, vocational training after secondary education (by means of certificates and diplomas) may be a promis- improve the skills of vulnerable ing route to employment given the growing demand workers, and help employers with for those skills and employers’ own training needs. their hiring and training needs. Demand-driven TVET (such as through on-the-job training) remains essential to support employers Although employment support programs are unlikely directly in meeting their training and vacancy needs. to have an impact on their own without addressing Achieving this support requires collaboration between binding constraints on the demand side, they have an the Department of Workforce Planning and Skills Divi- important role to play in improving the productiv- sion (DWPSD) within MoESD and employers in services ity and skills of the Bhutanese workforce and alleviat- and sales and craft and related trades—two occupa- ing the skill mismatches. The institutional and policy tions with the highest expected demand in the future. framework for designing employment programs to address the constraints highlighted in this report sets The role of ESCs in providing job intermediation the foundation to support workers and employers. and matching policies remains essential, but it needs Bhutan spends more than the average South Asian strengthening. With the appropriate resources, capac- country on skills development programs, and it has ity, and investments, ESCs can establish relationships a well-established technical and vocational educa- with local employers, engage in vacancy collection, and tion and training (TVET) sector. In addition, Bhutan directly support vulnerable job-seekers with referrals has a network of employment services centers (ESCs) to the appropriate employment support services and that are tasked with providing labor intermediation with job placement. Further support by ESCs of the and matching services as set forward by the National unemployed and inactive can be provided to increase Employment Policy of 2013. the attractiveness of and access to current low- and semi-skilled vacancies by providing on-the-job assis- However, a majority of the existing programs are frag- tance, counseling, and mobility support to cover job mented, are too small to make a tangible impact, have search costs for vulnerable workers and place them very limited regional coverage, or are inadequately directly in regions and sectors with wage employment designed to address relevant constraints. There is opportunities. This report finds that cross-dzongkhag scope to strengthen existing policies and programs mobility is associated with improved labor market out- through key reform principles: (1) improve the design comes, but low-skilled and vulnerable workers are less of programs to improve efficiency and effectiveness; likely to move compared to the higher-skilled. (2) introduce new design features within existing pro- grams and strengthen system linkages to address gaps; 94 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 2. Introduce new design features to existing to further employment support, such as on-the-job programs and strengthen system linkages to training or apprenticeships. maximize the impact of programs Across employment support programs, strengthening 3. Reduce program fragmentation and scale the inclusion of vulnerable workers with lagging labor up programs proven to be relevant and market outcomes is important. This effort may neces- impactful sitate either introducing new design features within existing programs or strengthening system linkages For its population size, Bhutan has too many ALMPs. across several programs to improve beneficiary refer- Most remain small and are discontinued every few ral and coordination and to maximize the impact of years due to lack of sustainable funding. For example, several programs. For example, women emerge as a most of the skills development programs under MoESD clearly vulnerable group from the analysis. To improve have less than 2,000 beneficiaries. The Youth Engage- their inclusion in TVET and support their overall labor ment and Livelihood Program (YELP) under MoICE—a market insertion, offering them childcare facilities is scheme that combines wage subsidies with on-the- important, as well as other enabling services (such as job training—supported almost 2,500 beneficiaries in transportation support for vulnerable women from fiscal 2022/23 and faces uncertain funding prospects. rural areas and safe training spaces). Across active labor Both MoESD and MoICE could focus on consolidating market programs (ALMPs), childcare subsidies can programs and scaling up a few impactful ones. This be an add-on benefit targeting women from low-in- approach could also improve the allocation of spend- come families. It is also essential that ESC resources are ing and create the fiscal space to fund new design directed toward helping women, especially low-skilled initiatives where gaps clearly exist (such as in the avail- mothers with young children in urban areas, reengage ability of childcare and transportation subsidies for in the labor market and that these resources provide women from low-income families). them with targeted case management, coaching, job search support, and training in relevant fields such as digital and information technology. Policy direction 4: Strengthen monitoring and evaluation (M&E) Many employment support programs can be linked in all employment promotion programs to improve their impact to other human capital programs to address vulnera- bilities along the life cycle of individuals. Low-income young mothers benefiting from employment support on labor market outcomes and under either MoICE or MoESD could be linked to other inform scale-up decisions. human capital programs to strengthen maternal and infant health, most notably under the Ministry of The M&E agenda in Bhutan remains essential to Health’s Maternal Child and Health Program. Employ- inform evidence-based policy making around employ- ment support programs also need to be linked to one ment policies. None of the available employment pro- another to maximize impact. For example, many train- motion programs has been evaluated, and monitoring ees benefiting from skills development under MoESD data on the performance and impact of these pro- could be referred to entrepreneurship skill programs grams are scarce and limited. In addition, Bhutan lacks under MoICE if they are interested in self-employ- the high-quality data from frequent firm-level panel ment. In addition, vulnerable trainees who are not surveys on productivity and occupational demand labor market-ready after training could be linked outlooks that could have an impact on its ability to Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 95 undertake evidence-based policy making to support countries such as Indonesia have achieved, requires firm growth. Real-time labor market data on workers important technical capacity and resources (Granata, and vacancies through online job postings could Posadas, and Testaverde 2022). Therefore, investing in transform skills development programs to improve robust information systems and conducting frequent their connection to labor demand. Although Bhutan firm-level surveys are imperative as Bhutan seeks recently built a single intake portal for workers and to strengthen its own workforce development and firms,32 the use of data to produce labor market intelli- support policies for firms. gence valuable to different users, similar to what other Table 5.1. Mapping of policy directions according to four broad labor market challenges Challenge Underlying constraint Suggested policy direction Low female labor Childcare options to help women strike a balance Strengthen active labor market programs (ALMPs) in force participation, between paid employment and care duties are urban areas in the short term to provide childcare especially among limited. subsidies for low-income families with young moth- low-skilled women ers. A medium-term agenda can focus on improving in urban areas early childhood care and development access and quality in both urban and rural areas. Social norms are associated negatively with wom- Offer as part of existing ALMPs or by employ- en’s labor force participation. ment services centers (ESCs) under the Ministry of Industry, Commerce, and Employment (MoICE) targeted information and awareness campaigns, as well as group training sessions on female agency, empowerment, and the benefits of women’s work. Skill development opportunities may be more Strengthen the design of skills development train- limited for young women than young men because ings and technical and vocational education and their not in education, employment, or training training (TVET) and incorporate gender-sensitive (NEET) shares are higher. policies to engage women. Stronger outreach by ESCs is needed to support low-skilled women in their job search efforts, as well as in case manage- ment and training referrals. Low productivity Many workers, especially women in rural areas, are Provide a targeted, comprehensive package of of agricultural self-employed or family workers, with no prospects economic inclusion services that includes skills workers, especially for improved economic mobility. training in technical areas, climate mitigation, and women in rural Agricultural workers often lack access to training business and risk management practices. System areas to improve their productivity, to input and mar- linkages between multiple ministries can support kets, and to protections against climate change the delivery of a coordinated package of services vulnerabilities. and maximize its impact. 32.  Bhutan Labor Market Information System, https://blmis.gov.bt/. 96 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice Table 5.1. Continued Challenge Underlying constraint Suggested policy direction Limited productiv- Over 95 percent of firms in Bhutan are cottage-size To support private sector development, productivity, ity and job creation firms (with an average of five employees or less). and growth, implement both vertical and horizontal in the private There are very few medium and large firms in size. reforms. They include reforms to improve the quality sector Firms are geographically concentrated and not suffi- of incubation and acceleration services provided by ciently diversified in terms of economic activity. The MoICE and implementing complementary reforms to dominant economic sectors (such as wholesale and labor regulations, investment climate, governance, retail) are, on average, characterized by low labor as well as labor market information systems. productivity. Demand is largest to fill low-skilled positions that do not match the profile of an increasingly educated workforce suffering from unemployment. Firms face multiple binding constraints to growth. Among smaller firms, these constraints are access to finance, markets, and raw materials, and among larger firms, they are stringent labor regulations and high worker turnover. Many workers have low-quality employment in the private sector, with no written contracts and a higher share of overwork. Difficulties facing Many of the available low-skilled positions appear Implement activation policies that stimulate the private sector firms to be either unappealing or inaccessible to those labor supply to address the shortage of low-skilled in accessing trained who are unemployed or inactive (such as women). workers. This policy direction is tied to the earlier labor to fill low- point on addressing constraints to women’s labor skilled vacancies, force participation, because low-skilled women who which may be one could fill some of the existing vacancies are largely of the factors con- inactive. Another option is to hire more foreign tributing to limited workers to fill vacancies. productivity and Firms have no connections to the vocational or Implement demand-driven TVET that can support growth training institutes that could help them access employers directly with their training and vacancy skilled labor and address worker shortages. needs and implement on-the-job-training schemes with a sustainable funding mechanism. Job-seekers with the educational levels most in Implement job intermediation and matching demand are not located in the regions experiencing policies by ESCs that can support better matches labor shortages. between labor supply and demand. Some of those policies could promote internal mobility by provid- ing on-the-job assistance, counseling, and trans- portation allowances to help vulnerable workers relocate to urban centers with wage vacancies. Vulnerable workers include self-employed (own-ac- count) and family workers with low education and those engaged in low-quality livelihoods in rural areas. Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 97 Appendixes Appendix A: Data Sources Bhutan Labor Force Survey (2013–22) This report uses 10 waves of the Bhutan Labor Force Survey (BLFS) for 2013–22. The BLFS provides annual data on the labor market, including the proportion of the economically active and inactive population, the labor force participation rate, and labor force data segregated by sex, age, education, location, occupation, hours of work, etc. The data are used to project the country’s future labor supply and to inform policies and programs related to job creation and poverty reduction. BLFS data are representative at the national level and cover both urban and rural areas across 20 (districts). Because of limited access, several hard-to-reach areas were excluded from the survey. The sample sizes are as follows: 6,000 households (2013–16); 8,010 households (2017); 9,012 households (2018–2020); and 10,130 house- holds (2021 and 2022). The BLFS was led by the Labor Market Information and Research Division of the Department of Employment and Human Resources between 2013 and 2017, after which the mandate for the survey shifted to the National Statis- tics Bureau. A key limitation of BLFS 2018–22 is that many important modules collected in the earlier rounds were removed, such as employment history, benefits attached to a job (to infer the level of informality among wage employees), and history of migration (both domestically and abroad). Bhutan Establishment Survey (2022) In Bhutan, 4,700 establishments responded to the 2022 Establishment Survey (ES). Survey data provide estimates representative of the national level and the six regional levels, as well as by economic sector and establishment size. The sampling frame relies on the employer-employee registration data from the Bhutan job vacancy platform of the Bhutan Labor Market Information System. Sectors covered by the ES do not include public administration and defense, activities of households as employers, and activities of extraterritorial organizations and bodies. Also excluded are establishments not registered under the EER 2020–21, household-based businesses and household-based subsistence growing of crops and rearing of livestock, religious institutions, government agencies, and the armed forces. 98 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice The ES was led by the Labor Market Information and Research Division (LMIRD) in close coordination with the Human Resource Planning and Coordination Division (HRPCD) of the Department of National Human Resource Development (DNHRD). The World Bank provided technical assistance on design, sampling, and implementation. Appendix B: Supplementary Figures and Tables, Chapter 2 Figure B.1. Skill composition of working-age population, overall and by gender, 2013, 2018, 2022 a. Overall b. Females c. Males Percent Percent Percent 2.4 2.81 2.69 0.4 0.5 0.3 100 100 100 4.5 6.6 10.2 4.5 5.3 5.2 6.33 8.22 12.2 8.3 9.9 14.3 80 80 26.8 80 28.89 30.4 31.46 33.2 31.1 34.1 32.5 9.2 35.1 60 60 6.3 60 11.29 8.63 6.6 9.38 13.5 11.1 40 40 40 12.3 59.2 56.2 51.09 48.88 49.6 20 41.63 20 20 42.6 41.2 33.0 0 0 0 2013 2018 2022 2013 2018 2022 2013 2018 2022 None/NFE Primary/ECCD None/NFE Primary/ECCD None/NFE Primary/ECCD Secondary Tertiary Secondary Tertiary Secondary Tertiary Monastic Monastic Monastic Source: Bhutan Labor Force Survey, 2022. Note: ECCD = early childhood care and development; NFE = nonformal education. Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 99 Figure B.2. Labor force participation rates, by gender and age, excluding agricultural output for own consumption, 2021 a. Females Participation rate (%) 100 90 80 70 60 50 40 30 20 10 0 15-19 20-24 25-29 30-34 35-39 40-44 45-49 50-54 55-59 60-64 65+ Age Standard definition Excluding own consumption Excluding own/mainly own consumption b. Males Participation rate (%) 100 90 80 70 60 50 40 30 20 10 0 15-19 20-24 25-29 30-34 35-39 40-44 45-49 50-54 55-59 60-64 65+ Age Standard definition Excluding own consumption Excluding own/mainly own consumption Source: Bhutan Labor Force Survey, 2021. 100 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice Figure B.3. Labor force participation, 2021 a. By gender and location, excluding own consumption/mainly own consumption 44 Rural 58 66 83 Female 60 Urban 67 73 87 60 Rural 67 79 89 Male 96 Urban 94 95 93 0 10 20 30 40 50 60 70 80 90 100 Participation rate (%) None/NFE Primary/ECCD Secondary Tertiary b. By gender and location, excluding own consumption 84 Rural 86 82 84 Female 62 Urban 70 74 87 93 Rural 96 96 92 Male 97 Urban 95 95 93 0 10 20 30 40 50 60 70 80 90 100 Participation rate (%) None/NFE Primary/ECCD Secondary Tertiary c. By gender and location, using standard definition 92 Rural 93 85 84 Female 65 Urban 72 75 87 98 Rural 98 97 92 Male 97 Urban 95 96 93 0 10 20 30 40 50 60 70 80 90 100 Participation rate (%) None/NFE Primary/ECCD Secondary Tertiary Source: Bhutan Labor Force Survey, 2021. Note: ECCD = early childhood care and development; NFE = nonformal education. Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 101 Figure B.4. Extensive and intensive margins of employment during pandemic, by gender, 2019–21 a. Inactivity during pandemic b. Unemployment during pandemic Inactivity rate (%) Unemployment rate (%) 50 10 9 45 8 7 40 6 35 5 4 30 3 25 2 1 20 0 2019 2020 2021 2019 2020 2021 Female Male Female Male c. Hours worked per week during pandemic No. of hours worked (conditional on working) 54 53 52 51 50 49 48 47 46 2019 2020 2021 Female Male Source: Bhutan Labor Force Survey, 2019–21. 102 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice Figure B.5. Part-time work and satisfaction with part-time work during pandemic, by gender, 2019–21 a. Evolution of share of part-time work in b. Evolution of the share of individuals total employment wishing to work longer hours, conditional on working 34 hours or less (substantial part-time) Share of substantial part time in employment (%) Share of substantial part-time workers wishing 25 to work more hours (%) 50 20 40 15 30 10 20 5 10 0 0 2019 2020 2021 2019 2020 2021 Female Male Female Male Source: Bhutan Labor Force Survey, 2019–21. Figure B.6. Evolution of share of family workers and self-employed workers, by gender, in agriculture, 2019–2021 a. Family workers b. Self-employed workers Share of family workers (%) Share of self-employed workers (%) 80 70 70 60 60 50 50 40 40 30 30 20 2019 2020 2021 2019 2020 2021 Female Male Female Male Source: Bhutan Labor Force Survey, 2019–21. Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 103 Figure B.7. Evolution of share of self-employed workers and employers, by gender, in nonagricultural sectors, 2019–21 a. Self-employed workers b. Employers Share of self-employed workers, excluding agriculture (%) Share of employers, excluding agriculture (%) 45 3 35 2 25 1 15 0 2019 2020 2021 2019 2020 2021 Female Male Female Male Source: Bhutan Labor Force Survey, 2019–21. Figure B.8. Evolution of real monthly earnings, by gender and sector, 2019–21 a. Agricultural sector b. Nonagricultural sectors Average monthly earnings, agriculture (Nu) Average monthly earnings, excluding agriculture (Nu) 6,500 14,000 13,000 5,500 12,000 11,000 4,500 10,000 3,500 9,000 2019 2020 2021 2019 2020 2021 Female Male Female Male Source: Bhutan Labor Force Survey, 2019–21. 104 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice Figure B.9. Unemployment rate, by age and education, 2013–22 a. By age b. By education Unemployment rate (%) Unemployment rate (%) 40 20 30 15 20 10 10 5 0 0 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 15-24 25-54 55+ None/NFE Primary/ECCD Secondary Tertiary Source: Bhutan Labor Force Survey, 2019–21. Note: ECCD = early childhood care and development; NFE = nonformal education. Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 105 Figure B.10. Predicted labor force participation rate, by age and gender, along the life cycle and by location and education, holding constant other individual, household, and local characteristics a. Along the life cycle Females Males Predicted participation rate (%) Predicted participation rate (%) 100 100 90 90 80 80 70 70 60 60 50 50 40 40 30 30 20 20 10 10 0 0 15-19 20-24 25-29 30-34 35-39 40-44 45-49 50-54 55-59 60-64 65+ 15-19 20-24 25-29 30-34 35-39 40-44 45-49 50-54 55-59 60-64 65+ Source: Bhutan Labor Force Survey, 2022. Note: The predicted participation rate is the average predicted probability of participation (95 percent confidence interval). For each observation, the predicted probability of participation is computed by setting the age group to a specific value, leaving all the other variables unchanged (using specification 3 in tables B.3a and B.3b). In a second step, the predicted probabilities are averaged. For example, it is expected that 37.8 percent of men participate in the labor market if they are ages 15–19 and have otherwise the same distribu- tion of characteristics observed in the data. b. By location and education (males and females, ages 25–54) Females Males Predicted participation rate (%) Predicted participation rate (%) 90 100 80 70 95 60 90 50 40 85 30 20 80 10 0 75 None/NFE Primary Secondary Tertiary None/NFE Primary Secondary Tertiary Urban Rural Urban Rural Source: Bhutan Labor Force Survey, 2022. Note: The predicted participation rate is the average predicted probability of participation (95 percent confidence interval). For each observation, the predicted probability of par- ticipation is computed by setting the location and education level to a specific value, leaving all other variables unchanged (using specification 4 in tables B.3a and B.b). In a second step, these predicted probabilities are averaged. For example, it is expected that 42 percent of women participate in the labor market if they are uneducated, live in an urban area, and have otherwise the same distribution of characteristics observed in the data (versus 69 percent for otherwise similar uneducated rural women). NFE = nonformal education. 106 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice Figure B.11. Average marginal effect of household demographics on labor force participation rate, by gender, holding other individual and local characteristics constant, 2022 a. Females b. Males In couple In couple HH head/spouse HH head/spouse Children, ages 0-2 Children, ages 0-2 Children, ages 3-5 Children, ages 3-5 Children, ages 6-11 Children, ages 6-11 Children, ages 12-15 Children, ages 12-15 HH members over 65 HH members over 65 HH members with HH members with disability/illness disability/illness Other adults Other adults -0.1 -0.05 0 0.05 0.1 0.15 -0.05 0 0.05 0.1 0.15 0.2 Average marginal effect Average marginal effect Source: Bhutan Labor Force Survey, 2022. Note: The average marginal effect of each variable is also reported in tables B.3a and B.3b, column 3. Because the dependent variable is a dummy variable, the model is estimated with a logistic specification, and the marginal effect of a variable x is not constant with the level of x. Therefore, the marginal effect of each variable is calculated for each observation in the sample using the coefficients from the logit model underlying tables B.3a and B.3b and the data. The average of these marginal effects is then calculated, giving the average marginal effect (95 percent confidence interval). For example, on average, all else being equal, a man who is a member of a couple is 11.2 percentage points more likely than a single man to participate in the labor market. HH = household. Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 107 Figure B.12. Average marginal effect of local norms on labor force participation rate, by gender, 2022 a. Females b. Males Share of working in HH Share of working in HH Share of working women in HH Share of working women HH Local log hourly wage Local log hourly wage Local share in public sector Local share in public sector Local share in agriculture Local share in agriculture Local unemployment rate Local unemployment rate Local inactivity rate Local inactivity rate -2 -1.5 -1 -0.5 0 0.5 1 -2.5 -2 -1.5 -1 -0.5 0 0.5 Average marginal effect Average marginal effect Source: Bhutan Labor Force Survey, 2022. Note: The average marginal effect of each variable is also reported in tables B.3a and B.3b, column 3. Because the dependent variable is a dummy variable, the model is estimated with a logistic specification, and the marginal effect of a variable x is not constant with the level of x. Therefore, the marginal effect of each variable is calculated for each observation in the sample, using the coefficients from the logit model underlying tables B.3a and B.3b and the data. The average of these marginal effects is then calculated, giving the average marginal effect (95 percent confidence interval). HH = household. 108 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice Figure B.13. Rates of individuals not in education, employment, or training (NEET), by gender, age, and location, 2022 a. NEET rates, females, ages 15–24 b. NEET rates, males, ages 15–24 15% 23% 23% 19% 59% 62% NEET Employed Studying/training NEET Employed Studying/training c. NEET rates, female and male rural youth, d. NEET rates, female and male urban youth, ages 15–24 ages 15–24 15% 24% 21% 64% 56% 21% NEET Employed Studying/training NEET Employed Studying/training Source: Bhutan Labor Force Survey, 2022. Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 109 Figure B.14. Status of elderly in the household, by gender, 2022 a. Females b. Males Percent Percent 100 100 80 80 60 60 40 40 20 20 0 0 55-59 60-64 65-69 70-74 75-79 80+ 55-59 60-64 65-69 70-74 75-79 80+ Household member Head/partner with family Household member Head/partner with family Head/partner alone Head/partner alone Source: Bhutan Labor Force Survey, 2022. Figure B.15. Labor force participation rate of the elderly, according to status in the household, by gender and age, 2022 Participation rate (%) 100 89 83 80 73 72 62 63 58 59 60 51 42 41 40 33 27 24 20 14 9 3 5 0 55-64 65-74 75+ 55-64 65-74 75+ Female Male Alone/with partner Household head/partner As family member Source: Bhutan Labor Force Survey, 2022. 110 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice Figure B.16. Distribution of education levels, by current status in labor market, number of job- seekers, number of inactive, and gender, 2022 a. Distribution of education levels Percent 100 90 80 70 60 50 40 30 20 10 0 Employed Unemployed Inactive, ages 25-64 Employed Unemployed Inactive, ages 25-64 Employed Unemployed Inactive, ages 25-64 Female Male Total None/NFE Primary Low secondary Middle secondary High secondary Certificate/diploma Bachelor's and above Monastic b. Number of job-seekers, by education c. Number of inactive (ages 25–64), by education 12,000 40,000 35,000 10,000 30,000 8,000 25,000 6,000 20,000 15,000 4,000 10,000 2,000 5,000 0 0 None/NFE Primary Secondary Tertiary Monastic None/NFE Primary Secondary Tertiary Monastic Female Male Total Female Male Total Source: Bhutan Labor Force Survey, 2022. Note: NFE = nonformal education. Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 111 Figure B.17. Main reason for working more than 48 hours per week, overall and by gender, education, location, and age, 2022 a. By gender b. By education Reason for overwork (%) Reason for overwork (%) 60 80 70 50 60 40 50 40 30 30 20 20 10 10 0 None/NFE Primary Secondary Tertiary Monastic Total 0 Female Male Total More income Requirement of the job Other More income Requirement of the job Other c. By location d. By age Reason for overwork (%) Reason for overwork (%) 70 60 60 50 50 40 40 30 30 20 20 10 10 0 15-19 20-24 25-29 30-34 35-39 40-44 45-49 50-54 55-59 60-64 65+ Total 0 Urban Rural Total More income Requirement of the job Other More income Requirement of the job Other Source: Bhutan Labor Force Survey, 2022. Note: NFE = nonformal education. 112 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice Figure B.18. Prevalence of underemployment, overall and by gender, education, location, and age, 2022 a. By gender b. By education Share of part-time employment (%) Share of part-time employment (%) 9 9 8 8 7 7 6 6 5 5 4 4 3 3 2 2 1 1 0 0 Female Male Total None/NFE Primary Secondary Tertiary Monastic Total 20 hours per week or less 21-34 hours per week 20 hours per week or less 21-34 hours per week c. By location d. By age Share of part-time employment (%) Share of part-time employment (%) 9 14 8 12 7 10 6 5 8 4 6 3 4 2 2 1 0 15-19 20-24 25-29 30-34 35-39 40-44 45-49 50-54 55-59 60-64 65+ Total 0 Urban Rural Total 20 hours per week or less 21-34 hours per week 20 hours per week or less 21-34 hours per week Source: Bhutan Labor Force Survey, 2022. Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 113 Figure B.19. Share of workers in part-time employment wishing to work more hours, overall and by gender, education, age, and location, 2022 a. By gender b. By education Share of part-time workers (%) Share of part-time workers (%) 10 16 14 8 12 10 6 8 6 4 4 2 2 0 None/NFE Primary Secondary Tertiary Monastic Total 0 Female Male Total c. By location d. By age Share of part-time workers (%) Share of part-time workers (%) 14 30 28 26 12 24 22 10 20 18 8 16 14 12 6 10 8 4 6 4 2 2 0 15-19 20-24 25-29 30-34 35-39 40-44 45-49 50-54 55-59 60-64 65+ Total 0 Urban Rural Total Source: Bhutan Labor Force Survey, 2022. Note: NFE = nonformal education. 114 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice Figure B.20. Predicted log hourly wage, by gender and age, 2022 a. Female b. Male Predicted log hourly wage Predicted log hourly wage 4.2 4.2 4 4 3.8 3.8 3.6 3.6 3.4 3.4 3.2 3.2 3 3 15 18 21 24 27 30 33 36 39 42 45 48 51 54 57 60 63 15 18 21 24 27 30 33 36 39 42 45 48 51 54 57 60 63 Age Age Source: Bhutan Labor Force Survey, 2022. Figure B.21. Returns to education, by gender, 2022 a. Female b. Male Primary Primary Secondary Secondary Tertiary Tertiary 0 0.5 1 1.5 0 0.2 0.4 0.6 0.8 Log real hourly wage Log real hourly wage Spec 1 Spec 2 Spec 3 Spec 4 Spec 1 Spec 2 Spec 3 Spec 4 Source: Bhutan Labor Force Survey, 2022. Note: The specifications in these figures are from Table B.4. The dependent variable in the specifications is log real hourly wage (base 2010). Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 115 Figure B.22. Raw and conditional log real hourly wage gap, by gender and occupation, 2022 a. Female b. Male Wage gap Wage gap 0.9 0.9 0.7 0.7 0.5 0.5 0.3 0.3 0.1 0.1 -0.1 -0.1 -0.3 -0.3 -0.5 -0.5 Managers Professionals Technicians Clerical Craft Operators Elementary Managers Professionals Technicians Clerical Craft Operators Elementary Raw gap Conditional gap Raw gap Conditional gap Source: Bhutan Labor Force Survey, 2022. Note: The raw log hourly wage gaps are plotted in blue, the conditional gaps from specification 4 of tables B.4a and B.4b are plotted in orange, together with their 95 percent confi- dence interval. Reference is Services and Sales workers. Figure B.23. Raw and conditional log real hourly wage gap, by gender and industry, 2022 a. Female b. Male Wage gap Wage gap 1 1 0.8 0.8 0.6 0.6 0.4 0.4 0.2 0.2 0 0 -0.2 -0.2 -0.4 -0.4 -0.6 -0.6 -0.8 Energy/water Construction Trade Transportation Hotels/restaurants Administration Information/comms. Finance/real estate Science Public Education Health Arts/other Energy/water Construction Trade Transportation Hotels/restaurants Finance/real estate Science Administration Information/comms. Public Education Health Arts/other Raw gap Conditional gap Raw gap Conditional gap Source: Bhutan Labor Force Survey, 2022. Note: The raw log hourly wage gaps are plotted in grey, the conditional gaps from specification 4 of tables B.4a and B.4b are plotted as red diamonds, together with their 95% confidence interval. Reference is Manufacturing. 116 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice Figure B.24. Share of Bhutan-born working-age individuals, by current location, gender, education, and migration status, 2022 a. By current location and gender b. By education Share of individuals (%) Share of individuals (%) 100 100 80 80 60 60 40 40 20 20 0 0 Total Rural Urban Male Female Total None/NFE Primary Secondary Tertiary Monastic Never moved outside dzongkhag Never moved outside dzongkhag Lived in a different dzongkhag Lived abroad Lived in a different dzongkhag Lived abroad Source: Bhutan Labor Force Survey, 2022. Figure B.25. Share of Bhutan-born working-age individuals by current location, gender, and migration profile, 2022 a. By current location and gender b. By education Share of individuals (%) Share of individuals (%) 60 60 50 50 40 40 30 30 20 20 10 10 0 0 Total Rural Urban Male Female Total None/NFE Primary Secondary Tertiary Monastic One move Multiple moves Return migrant One move Multiple moves Return migrant Source: Bhutan Labor Force Survey, 2022. Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 117 Figure B.26. Distribution of Bhutan-born working-age migrants, by location of origin and destination, 2022 Share of individuals (%) 35 30 25 20 15 10 5 0 Rural to urban Urban to urban Rural to rural Urban to rural Source: Bhutan Labor Force Survey, 2022. Table B.1. Change in female entrants in labor force, 2018–20 Difference for new female entrants Mean for new female entrants in in 2020 with respect to new 2018 and 2019 entrants in 2018–19 Age 27.75 -3.210*** (0.631) (0.743) Education Uneducated 0.201 -0.114*** (0.0303) (0.0347) Primary 0.0703 -0.0497** (0.0129) (0.0164) Secondary 0.407 0.0988* (0.0297) (0.0446) Tertiary 0.314 0.0717 (0.0362) (0.0486) Monastic 0.00716 -0.00641 (0.00425) (0.00432) Urban area 0.606 -0.0186 (0.0917) (0.103) Household head/spouse of the household head 0.411 -0.128** (0.0369) (0.0461) Married 0.474 -0.161*** (0.0332) (0.0436) *p < .05, **p < .01, ***p < .001 118 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice Table B.2. Characteristics of the unemployed in 2019 and their evolution, 2020–21 (2) (1) Pre-pandemic means and Effect of the pandemic proportions Demographic profile of unemployed Average age 0.664 27.33*** (0.626) (0.544) 15–24 -0.00142 0.417*** (0.0284) (0.0217) 25–54 -0.00578 0.579*** (0.0288) (0.0220) 55–64 0.00721 0.00368 (0.00451) (0.00261) 65+ 0 0 () () Gender (1=male) -0.0146 0.435*** (0.0268) (0.0197) Uneducated 0.00830 0.113*** (0.0289) (0.0257) Primary education 0.0176 0.0910*** (0.0288) (0.0261) Secondary education 0.0430 0.412*** (0.0281) (0.0203) Tertiary education -0.0688 0.384*** (0.0530) (0.0500) Location (1=rural) -0.0403 0.405*** (0.126) (0.121) Proportion of household head/partner 0.0613 0.302*** (0.0356) (0.0301) Proportion of married 0.0371 0.397*** (0.0316) (0.0248) Number of children, ages 02 0.0337 0.108*** (0.0201) (0.0130) Number of children ages 2–5 0.0392 0.0871*** (0.0202) (0.0132) Number of children, ages 6–12 0.0464 0.200*** (0.0309) (0.0219) Number of children, ages 13–15 0.0508* 0.131*** (0.0248) (0.0181) Number of adults over 65 -0.0241 0.125*** (0.0262) (0.0222) Number of disabled/ill 0.0330* 0.0557*** (0.0165) (0.00947) Number of other adults 0.0749 1.812*** (0.142) (0.124) Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 119 Table B.2. Continued (2) (1) Pre-pandemic means and Effect of the pandemic proportions Region of the unemployed Bumthang -0.00106 0.0159* (0.00923) (0.00809) Chukha -0.0344 0.133 (0.0781) (0.0764) Dagana 0.00589 0.0232 (0.0152) (0.0121) Gasa -0.000187 0.00214 (0.00204) (0.00185) Haa -0.00391 0.0154 (0.00866) (0.00800) Lhuentse -0.00287 0.00548 (0.00354) (0.00327) Monggar -0.0174 0.0306 (0.0197) (0.0189) Paro 0.0162 0.0973* (0.0510) (0.0469) Pema Gatshel -0.0258 0.0475* (0.0229) (0.0222) Punakha 0.0265* 0.0108 (0.0123) (0.00712) Samdrup Jongkhar -0.00827 0.0310 (0.0239) (0.0232) Samtse -0.0184 0.0529* (0.0274) (0.0254) Sarpang -0.0619 0.120* (0.0611) (0.0598) Thimphu 0.120 0.292 (0.199) (0.195) Trashigang 0.0236* 0.00675 (0.00959) (0.00516) Trashi Yangtse 0.00480 0.00592 (0.00547) (0.00396) Trongsa -0.0143 0.0320 (0.0177) (0.0166) Tsirang -0.0194 0.0285* (0.0145) (0.0141) Wangdue Phodrang 0.0127 0.0386 (0.0309) (0.0279) Zhemgang -0.00202 0.0106 (0.00728) (0.00656) 120 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice Table B.2. Continued (2) (1) Pre-pandemic means and Effect of the pandemic proportions Reason for unemployment Skills mismatch -0.0918* 0.474*** (0.0371) (0.0326) Recent graduation -0.0175 0.214*** (0.0273) (0.0223) Terminated 0.209*** 0.0487*** (0.0189) (0.00712) Resigned -0.0234 0.0970*** (0.0168) (0.0138) Family 0.000268 0.0473** (0.0171) (0.0153) Other -0.0769** 0.119*** (0.0258) (0.0249) Duration of unemployment <1 month 0.0421 0.152*** (0.0294) (0.0231) [1–5 months] 0.0742* 0.249*** (0.0356) (0.0303) [6–11 months] 0.0679** 0.160*** (0.0232) (0.0170) [12–23 months] -0.101*** 0.214*** (0.0209) (0.0173) 2 years + -0.0832** 0.224*** (0.0311) (0.0272) Characteristics of the desired future job Public service -0.0788** 0.527*** (0.0291) (0.0212) Private business/company 0.0657* 0.359*** (0.0310) (0.0222) Public company -0.000705 0.110*** (0.0177) (0.0131) Agriculture 0.0138** 0.00313 (0.00523) (0.00238) Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 121 Table B.2. Continued (2) (1) Pre-pandemic means and Effect of the pandemic proportions Main reason for this choice Salary 0.0718** 0.143*** (0.0275) (0.0213) Working conditions -0.0287 0.195*** (0.0230) (0.0181) Reputation 0.0118 0.0330*** (0.0103) (0.00695) Personal interests -0.0334 0.426*** (0.0465) (0.0413) Job security -0.0216 0.203*** (0.0251) (0.0195) Reasons cited for this choice Concerned with salary 0.0950** 0.202*** (0.0306) (0.0235) Concerned with working conditions 0.0174 0.359*** (0.0350) (0.0273) Concerned with reputation 0.0501* 0.130*** (0.0220) (0.0158) Concerned with personal interest 0.0426 0.588*** (0.0417) (0.0359) Concerned with job security 0.0419 0.315*** (0.0299) (0.0226) Reservation wage -0.506 9948.6*** (417.4) (353.3) Have worked before 0.165*** 0.349*** (0.0408) (0.0359) Employment history Former sector: Public -0.0531 0.245*** (0.0364) (0.0297) Former sector: Private 0.118** 0.666*** (0.0369) (0.0296) Former sector: Agriculture -0.0653* 0.0896*** (0.0256) (0.0243) Former industry: Agriculture/Mining -0.0849** 0.124*** (0.0258) (0.0237) Former industry: Manufacturing -0.0574 0.127*** (0.0300) (0.0269) 122 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice Table B.2. Continued (2) (1) Pre-pandemic means and Effect of the pandemic proportions Former industry: Energy/Water -0.0327 0.0608*** (0.0188) (0.0170) Former industry: Construction -0.0129 0.0993*** (0.0254) (0.0199) Former industry: Trade -0.0293 0.137*** (0.0334) (0.0295) Former industry: Transportation -0.0270 0.0646*** (0.0218) (0.0176) Former industry: Hotels/Restaurants 0.120*** 0.113*** (0.0298) (0.0198) Former industry: Information/Communication -0.0447 0.0613** (0.0240) (0.0231) Former industry: Finance/Real Estate -0.00144 0.0125 (0.0144) (0.0129) Former industry: Science -0.00159 0.0119 (0.00890) (0.00728) Former industry: Administration 0.156*** 0.0196* (0.0227) (0.00783) Former industry: Public -0.0271 0.0619** (0.0217) (0.0192) Former industry: Education -0.00738 0.0684** (0.0265) (0.0233) Former industry: Health -0.00641 0.0137 (0.0103) (0.00934) Former industry: Arts/Other services 0.0561** 0.0252* (0.0193) (0.0124) Former occupation: Managers 0.0142 0.0575*** (0.0218) (0.0170) Former occupation: Professionals -0.00357 0.146** (0.0576) (0.0548) Former occupation: Technicians 0.0220 0.0941** (0.0346) (0.0301) Former occupation: Clerical -0.0315 0.131*** (0.0290) (0.0242) Former occupation: Services/Sales 0.117* 0.179*** (0.0471) (0.0401) Former occupation: Agriculture -0.0637* 0.0977*** (0.0283) (0.0268) Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 123 Table B.2. Continued (2) (1) Pre-pandemic means and Effect of the pandemic proportions Former occupation: Craft -0.0429 0.101*** (0.0305) (0.0281) Former occupation: Operators 0.0154 0.0737* (0.0325) (0.0293) Former occupation: Elementary -0.0271 0.120** (0.0404) (0.0371) *p < .05, **p < .01, ***p < .001 Table B.3a. Determinants of female labor force participation (1) = 1 if participating (2) (3) Own characteristics: 20–24 0.414*** 0.430*** 0.431*** (0.0153) (0.0173) (0.0184) 25–29 0.571*** 0.573*** 0.570*** (0.0160) (0.0150) (0.0154) 30–34 0.577*** 0.576*** 0.573*** (0.0134) (0.0143) (0.0151) 35–39 0.603*** 0.579*** 0.576*** (0.0153) (0.0154) (0.0160) 40–44 0.619*** 0.565*** 0.560*** (0.0204) (0.0216) (0.0218) 45–49 0.591*** 0.513*** 0.512*** (0.0214) (0.0238) (0.0249) 50–54 0.563*** 0.483*** 0.480*** (0.0257) (0.0271) (0.0284) 55–59 0.499*** 0.409*** 0.407*** (0.0276) (0.0300) (0.0307) 60–64 0.444*** 0.355*** 0.348*** (0.0316) (0.0336) (0.0334) 65+ 0.127*** 0.0680** 0.0605** (0.0205) (0.0207) (0.0212) Primary 0.0475** 0.0407** 0.0686*** (0.0148) (0.0145) (0.0157) 124 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice Table B.3a. Continued (1) = 1 if participating (2) (3) Secondary 0.0496*** 0.0419** 0.109*** (0.0147) (0.0140) (0.0227) Tertiary 0.0754*** 0.0492* 0.159*** (0.0209) (0.0208) (0.0374) Monastic -0.273*** -0.264*** -0.251*** (0.0616) (0.0696) (0.0602) Rural 0.132*** 0.128*** 0.115*** (0.0205) (0.0211) (0.0191) Household characteristics: = 1 if in couple -0.0275* -0.0216 (0.0127) (0.0130) = 1 if household head or spouse 0.113*** 0.105*** (0.00919) (0.00885) # of children 0–2 -0.0986*** -0.0958*** (0.00982) (0.00961) # of children 3–5 -0.0585*** -0.0611*** (0.00940) (0.00975) # of children 6–11 -0.0343*** -0.0357*** (0.00671) (0.00652) # of children 12–15 -0.0395*** -0.0421*** (0.00957) (0.00933) # of household members over 65 0.0367*** 0.0337*** (0.00875) (0.00823) # of household members with disability/illness 0.00238 0.000652 (0.0207) (0.0204) # of other adults -0.0144*** -0.0158*** (0.00396) (0.00394) Share working -0.0570*** -0.0751*** (0.0153) (0.0133) Share working women 0.0752*** 0.0634*** (0.0164) (0.0149) Local labor market characteristics: Local inactivity rate -1.194*** (0.169) Local unemployment rate 0.191 (0.303) Local share in agriculture -0.0853 (0.115) Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 125 Table B.3a. Continued (1) = 1 if participating (2) (3) Local share in public sector 0.107 (0.194) Local log hourly wage -0.138*** (0.0397) Region fixed effect Yes Yes Yes No. of observations 15,822 15,822 15,822 Note: Reference category is Ages 15–19, Urban, Not educated, Household member, Single. Local labor market indicators for men and women are computed at the dzongkhag level. In addition, the local hourly wage is calculated on the basis of prime-age men or women with the same level of education as the individual. *p < .05, **p < .01, ***p < .001 Table B.3b. Determinants of male labor force participation (1) = 1 if participating (2) (3) Own characteristics: 20–24 0.537*** 0.431*** 0.436*** (0.0275) (0.0245) (0.0247) 25–29 0.797*** 0.523*** 0.530*** (0.0173) (0.0373) (0.0382) 30–34 0.862*** 0.544*** 0.551*** (0.0154) (0.0364) (0.0369) 35–39 0.868*** 0.537*** 0.544*** (0.0177) (0.0389) (0.0395) 40–44 0.884*** 0.550*** 0.557*** (0.0156) (0.0399) (0.0404) 45–49 0.860*** 0.505*** 0.512*** (0.0157) (0.0370) (0.0375) 50–54 0.825*** 0.430*** 0.436*** (0.0220) (0.0491) (0.0498) 55–59 0.757*** 0.316*** 0.324*** (0.0347) (0.0635) (0.0646) 126 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice Table B.3b. Continued (1) = 1 if participating (2) (3) 60–64 0.618*** 0.177** 0.187** (0.0478) (0.0656) (0.0670) 65+ 0.292*** -0.0429 -0.0370 (0.0378) (0.0520) (0.0529) Primary 0.0721*** 0.0418*** 0.0440*** (0.0109) (0.0108) (0.0110) Secondary 0.00534 -0.0101 -0.00590 (0.0155) (0.0135) (0.0134) Tertiary -0.118*** -0.127*** -0.124*** (0.0121) (0.0113) (0.0230) Monastic -0.0945*** -0.0853*** -0.0891*** (0.0167) (0.0135) (0.0138) Rural 0.0242* 0.0386*** 0.0334*** (0.0112) (0.0107) (0.00992) Household characteristics: = 1 if in couple 0.110*** 0.112*** (0.00675) (0.00686) = 1 if household head or spouse 0.137*** 0.134*** (0.00976) (0.0101) # of children 0–2 0.00825 0.00882 (0.00986) (0.00966) # of children 3–5 -0.0131 -0.0140 (0.00740) (0.00739) # of children 6–11 -0.0136* -0.0142* (0.00564) (0.00562) # of children 12–15 -0.000453 -0.00143 (0.00889) (0.00883) # of household members over 65 -0.00766 -0.00864 (0.00453) (0.00444) # of household members with disability/illness 0.0152 0.0136 (0.00894) (0.00887) # Other adults -0.00816** -0.00776** (0.00257) (0.00257) Share working 0.0352** 0.0294* (0.0119) (0.0123) Share working women -0.00478 -0.00912 (0.0139) (0.0131) Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 127 Table B.3b. Continued (1) = 1 if participating (2) (3) Local labor market characteristics: Local inactivity rate -0.260 (0.148) Local unemployment rate -1.356** (0.409) Local share in agriculture 0.0778 (0.0629) Local share in public sector 0.167 (0.0985) Local log hourly wage 0.000120 (0.0269) Region fixed effect Yes Yes Yes No. of observations 14,535 14,535 14,535 Note: Reference category is Ages 15–19, Urban, Not educated, Household member, Single. Local labor market indicators for men and women are computed at the dzongkhag level. In addition, the local hourly wage is calculated on prime-age men or women with the same level of education as the individual. *p < .05, **p < .01, ***p < .001 Table B.4a. Real log hourly wage of women (1) Log real hourly wage (base=2010) (2) (3) (4) Own characteristics: Age 0.0817*** 0.0659*** 0.0624*** 0.0523*** (0.0168) (0.0176) (0.0181) (0.0155) Age # age -0.000798*** -0.000634** -0.000599** -0.000518** (0.000183) (0.000197) (0.000206) (0.000181) Primary 0.160* 0.161* 0.150* 0.0987 (0.0630) (0.0629) (0.0650) (0.0577) Secondary 0.501*** 0.498*** 0.412*** 0.300*** (0.0376) (0.0364) (0.0308) (0.0325) Tertiary 0.959*** 0.968*** 0.805*** 0.471*** (0.0413) (0.0395) (0.0528) (0.0680) Monastic 0.136 0.133 0.122 0.0535 (0.379) (0.375) (0.374) (0.387) 128 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice Table B.4a. Continued (1) Log real hourly wage (base=2010) (2) (3) (4) Rural -0.0236 -0.0142 -0.0198 -0.00687 (0.0299) (0.0296) (0.0315) (0.0308) Household characteristics: = 1 if in couple 0.0433 0.0393 0.0348 (0.0294) (0.0344) (0.0320) = 1 if household head or spouse 0.158*** 0.133*** 0.0974** (0.0365) (0.0343) (0.0313) # of children 0–2 -0.000184 0.00165 -0.000811 (0.0323) (0.0297) (0.0362) # of children 3–5 0.0424 0.0369 0.0358 (0.0261) (0.0276) (0.0268) # of children 6–11 -0.0210 -0.00822 -0.0121 (0.0223) (0.0249) (0.0213) # of children 12–15 0.0319 0.0386 0.0407 (0.0307) (0.0300) (0.0280) # of household members over 65 0.00986 0.00677 0.00950 (0.0260) (0.0289) (0.0289) # of household members with disability/illness 0.0197 0.00467 0.0259 (0.0573) (0.0577) (0.0599) # Other adults 0.0121 0.0138 0.0107 (0.00907) (0.00926) (0.00941) Industry (ref. Manufacturing): Energy/Water 0.436*** 0.506*** (0.0489) (0.121) Construction 0.268*** 0.391** (0.0627) (0.119) Trade 0.196*** 0.344* (0.0463) (0.171) Transportation 0.310** 0.412 (0.103) (0.215) Hotels/Resturants 0.101 -0.0615 (0.0543) (0.123) Information/Communication 0.308** 0.342* (0.105) (0.167) Finance/Real Estate 0.505*** 0.545** (0.0692) (0.171) Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 129 Table B.4a. Continued (1) Log real hourly wage (base=2010) (2) (3) (4) Science -0.0412 -0.0884 (0.0723) (0.0652) Administration -0.236 -0.242 (0.202) (0.132) Public 0.188*** 0.343* (0.0454) (0.154) Education 0.340*** 0.325 (0.0470) (0.167) Health 0.433*** 0.520*** (0.0474) (0.150) Arts/Other services 0.242*** 0.292* Occupation (ref. Services/Sales): Managers 0.523*** (0.111) Professionals 0.439*** (0.0519) Technicians 0.177** (0.0560) Clerical 0.129* (0.0565) Craft 0.181 (0.163) Operators 0.0754 (0.106) Elementary -0.210*** (0.0429) Constant 1.548*** 1.706*** 1.676*** 1.861*** (0.354) (0.358) (0.368) (0.414) Region fixed effect Yes Yes Yes Yes No. of observations 3078 3078 3078 3078 Note: Reference category is Urban, Not educated, Household member, Single, Working in Manufacturing as Services and Sales Worker. *p < .05, **p < .01, ***p < .001 130 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice Table B.4b. Real log hourly wage of men (1) Log real hourly wage (base=2010) (2) (3) (4) Own characteristics: Age 0.0613*** 0.0466*** 0.0449*** 0.0385*** (0.00827) (0.00769) (0.00924) (0.00946) Age # Age -0.000630*** -0.000491*** -0.000456*** -0.000406*** (0.0000837) (0.0000847) (0.000102) (0.000105) Primary 0.0735* 0.0764* 0.0835** 0.0669* (0.0337) (0.0347) (0.0321) (0.0323) Secondary 0.208*** 0.212*** 0.234*** 0.183*** (0.0266) (0.0270) (0.0275) (0.0277) Tertiary 0.704*** 0.711*** 0.703*** 0.462*** (0.0391) (0.0395) (0.0503) (0.0424) Monastic -0.123 -0.0969 0.0130 -0.0285 (0.0825) (0.0786) (0.0573) (0.0575) Rural -0.0605* -0.0464 -0.0615* -0.0615* (0.0266) (0.0269) (0.0275) (0.0297) Household characteristics: = 1 if in couple 0.129*** 0.115*** 0.114*** (0.0345) (0.0334) (0.0305) =1 if household head or spouse 0.0920 0.0965* 0.0804 (0.0547) (0.0480) (0.0468) # of children 0–2 -0.0192 -0.00748 -0.0178 (0.0208) (0.0193) (0.0187) # of children 3–5 -0.00892 -0.00420 -0.0107 (0.0246) (0.0230) (0.0228) # of children 6–11 0.00387 0.00268 0.000316 (0.0192) (0.0238) (0.0203) # of children 12–15 0.0115 0.0161 0.0159 (0.0204) (0.0246) (0.0256) # of household members over 65 -0.0394 -0.0344 -0.0343 (0.0240) (0.0216) (0.0225) # of household members with disabilities/illness -0.0442 -0.0316 -0.0300 (0.0477) (0.0463) (0.0483) # Other adults 0.0266* 0.0250* 0.0307** (0.0117) (0.0107) (0.00982) Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 131 Table B.4b. Continued (1) Log real hourly wage (base=2010) (2) (3) (4) Industry (ref. Manufacturing): Energy/Water 0.221** 0.223*** (0.0717) (0.0655) Construction 0.169** 0.199*** (0.0626) (0.0592) Trade -0.0663 -0.0640 (0.0545) (0.0547) Transportation 0.0315 0.164* (0.103) (0.0777) Hotels/Restaurants -0.290* -0.390*** (0.118) (0.117) Information /Communication 0.217* 0.126 (0.0920) (0.0803) Finance/Real Estate 0.127 0.0891 (0.0975) (0.102) Science 0.180 0.0953 (0.0963) (0.0788) Administration -0.0217 0.0216 (0.0978) (0.0921) Public -0.0179 -0.0246 (0.0641) (0.0628) Education 0.0691 -0.0460 (0.0649) (0.0641) Health 0.0453 0.0162 (0.0800) (0.0742) Arts/Other services -0.201 -0.290** (0.119) (0.111) Occupation (ref. Services/Sales): Managers 0.291*** (0.0637) Professionals 0.383*** (0.0621) Technicians 0.0796* (0.0361) Clerical 0.0187 (0.0495) Craft 0.0336 (0.0523) 132 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice Table B.4b. Continued (1) Log real hourly wage (base=2010) (2) (3) (4) Operators -0.143** (0.0545) Elementary -0.205*** (0.0391) Constant 2.362*** 2.470*** 2.472*** 2.652*** (0.175) (0.162) (0.174) (0.171) Region fixed effect Yes Yes Yes Yes No. observations 5,620 5,620 5,620 5,620 Note: Reference category is Ages 15–19, Urban, Not educated, Household member, Single, Working in Manufacturing as Services and Sales Worker. *p < .05, **p < .01, ***p < .001 Table B.5. Blinder-Oaxaca decomposition of the gender hourly wage gap (1) (2) (3) (4) (5) (6) Log hourly wage of males, exclud- 3.870*** 3.870*** 3.870*** 3.870*** 3.870*** 3.870*** ing family workers (A) (0.0448) (0.0460) (0.0459) (0.0457) (0.0456) (0.0474) Log hourly wage of females, 3.749*** 3.749*** 3.749*** 3.749*** 3.749*** 3.749*** excluding family workers (B) (0.0604) (0.0587) (0.0616) (0.0617) (0.0623) (0.0624) Raw log hourly wage gap (A – B) 0.121*** 0.121*** 0.121*** 0.121*** 0.121*** 0.121*** (0.0207) (0.0191) (0.0213) (0.0215) (0.0220) (0.0208) Explained 0 -0.0224*** -0.0338** -0.0137 0.0295 0.0342 (0.00427) (0.0121) (0.0131) (0.0198) (0.0185) Unexplained 0.121*** 0.144*** 0.155*** 0.135*** 0.0918*** 0.0871*** (0.0207) (0.0175) (0.0163) (0.0165) (0.0154) (0.0146) a. Explained part of the gap (due to differences in the following characteristics): decomposition Dzongkhags -0.0138*** -0.00824** -0.00873** -0.00798** -0.00735** (0.00367) (0.00283) (0.00280) (0.00278) (0.00271) Area -0.00865** -0.00393* -0.00339 -0.00425* -0.00335 (0.00284) (0.00195) (0.00186) (0.00196) (0.00189) Age 0.0114* 0.00770 0.00903* 0.00450 (0.00462) (0.00399) (0.00357) (0.00330) Education -0.0330** -0.0335** -0.0278** -0.0216** (0.0110) (0.0113) (0.00999) (0.00696) Married 0.0223*** 0.0206*** 0.0184*** (0.00324) (0.00297) (0.00267) Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 133 Table B.5. Continued (1) (2) (3) (4) (5) (6) Household composition 0.00193* 0.00178* 0.00120 (0.000964) (0.000875) (0.000775) Industry 0.0381*** 0.0663*** (0.0113) (0.0131) Occupation -0.0238* (0.0101) b. Unexplained part of the gap (baseline difference and difference in returns to the same characteristics): decomposition Dzongkhags -0.0478* -0.0314** -0.0304* -0.0298* -0.0344** (0.0204) (0.0120) (0.0121) (0.0118) (0.0123) Area 0.0254 0.00810 0.00393 0.00638 0.0156 (0.0203) (0.0184) (0.0187) (0.0186) (0.0183) Age 0.0535 0.0473 0.0351 0.0193 (0.0662) (0.0697) (0.0658) (0.0659) Education 0.0292 0.0238 0.0200 0.0267 (0.0771) (0.0713) (0.0715) (0.0735) Married 0.0668 0.0689* 0.0805** (0.0342) (0.0274) (0.0249) Household composition -0.0150 -0.0151 -0.0242 (0.0135) (0.0138) (0.0136) Industry -0.0308 -0.0635** (0.0199) (0.0201) Occupation 0.0302 (0.0205) Constant 0.121*** 0.166*** 0.0957 0.0385 0.0371 0.0369 (0.0207) (0.0221) (0.102) (0.107) (0.104) (0.103) No. of observations 9,056 9,056 9,056 9,056 9,056 9,056 Note: Because alternative reference groups yield different estimates of the contribution of each factor variable to the unexplained part of the gap, this anal- ysis follows Yun (2005) and measures the true contribution of a factor variable as the average effect for every possible specification changing the reference category (following a normalized regression approach). *p < .05, **p < .01, ***p < .001 134 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice Table B.6. Comparison of the demographic characteristics of informal and formal employees working in companies, businesses, or nongovernmental organizations, 2017 Share in informal contracts Difference with share in formal (nonwritten) contracts Gender Females 0.307 0.0263 (0.0331) (0.0213) Males 0.693 -0.0263 (0.0331) (0.0213) Age group 15–24 0.224 0.00607 (0.0222) (0.0227) 25–54 0.748 0.000137 (0.0204) (0.0218) 55+ 0.0277 -0.00621 (0.00390) (0.00511) Education None/NFE 0.233 -0.0743** (0.0303) (0.0229) Primary/ECCD 0.107 -0.0347** (0.00748) (0.0125) Secondary 0.501 0.0710* (0.0152) (0.0280) Tertiary 0.135 0.0531** (0.0254) (0.0194) Education 1 (0–8 items) 0.206 -0.0517** (0.0167) (0.0196) 2 (9–10 items) 0.209 0.0217 (0.0113) (0.0365) 3 (11-12 items) 0.225 -0.00345 (0.0121) (0.0162) 4 (13–14 items) 0.167 0.0186 (0.0106) (0.0168) 5 (15+ items) 0.193 0.0149 (0.0227) (0.0259) Education Rural 0.265 -0.0332 (0.0998) (0.0308) Urban 0.735 0.0332 (0.0998) (0.0308) Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 135 Table B.6. Continued Share in informal contracts Difference with share in formal (nonwritten) contracts Region Thimphu 0.406 -0.0274 (0.223) (0.0213) Gelephu 0.132 -0.0524 (0.0601) (0.0300) Phuentsholing 0.187 -0.00163 (0.105) (0.0217) Punakha 0.128 0.0956* (0.0603) (0.0454) Samdrup Jongkhar 0.0740 0.00417 (0.0428) (0.0117) Trashigang 0.0728 -0.0184 (0.0334) (0.0154) Note: ECCD = early childhood care and development; NFE = nonformal education. *p < .05, **p < .01, ***p < .001 Table B.7. Determinants of the real log hourly wages for employees for private businesses (1) (2) (3) (4) Informality No written contract -0.0350 -0.0159 0.229* 0.228* (0.0569) (0.0621) (0.108) (0.104) Education (ref.: No education/NFE) Primary 0.297*** 0.255*** (0.0760) (0.0748) Secondary 0.579*** 0.435*** (0.0547) (0.0685) Tertiary 1.192*** 0.827*** (0.0606) (0.0700) Informality and education Primary # No written contract -0.175* -0.169 (0.0886) (0.0865) Secondary # No written contract -0.217** -0.205* (0.0763) (0.0834) Tertiary # No written contract -0.168* -0.158 (0.0818) (0.0974) 136 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice Table B.7. Continued (1) (2) (3) (4) Gender Male 0.178*** 0.145** (0.0507) (0.0457) Informality and gender Male # No written contract -0.0407 -0.0331 (0.0795) (0.0755) Geographic controls No Yes Yes Yes Demographic controls No No Yes Yes Job controls No No No Yes No. of observations 2,636 2,636 2,636 2,636 Note: Geographic controls include region fixed effects and a urban/rural dummy. Demographic controls include age, age-squared, and gender. Job controls include industry and occupation fixed effects. *p < .05, **p < .01, ***p < .001 Table B.8. Comparison of the job characteristics of informal and formal employees working for companies, businesses, or nongovernmental organizations Share in informal contracts Difference with share in formal (nonwritten) contracts Manufacturing 0.200 -0.0167 (0.0551) (0.0243) Energy/Water 0.0676 0.0167 (0.0193) (0.0178) Construction 0.151 0.0205 (0.0223) (0.0309) Trade 0.109 -0.0249* (0.0185) (0.0116) Transportation 0.0484 0.00972 (0.00684) (0.0119) Hotels/Restaurants 0.108 0.0143 (0.0228) (0.0223) Information/Communication 0.0543 0.00179 (0.0193) (0.00897) Finance/Real Estate 0.0648 0.0102 (0.00751) (0.0133) Science 0.00574 -0.000325 (0.00325) (0.000718) Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 137 Table B.8. Continued Share in informal contracts Difference with share in formal (nonwritten) contracts Administration 0.0663 -0.000493 (0.0184) (0.00900) Education 0.0266 0.0215* (0.00619) (0.0109) Arts/Other services 0.0540 -0.0289*** (0.00937) (0.00654) Managers 0.0683 0.00411 (0.0118) (0.0141) Professionals 0.117 0.0315* (0.0189) (0.0137) Technicians 0.122 0.0382* (0.00908) (0.0177) Clerical 0.0917 0.00492 (0.00732) (0.0120) Services/Sales 0.151 0.0299 (0.0160) (0.0226) Craft 0.153 -0.0540*** (0.0141) (0.0163) Operators 0.165 -0.0387* (0.0249) (0.0167) Elementary 0.131 -0.0158 (0.0137) (0.0229) *p < .05, **p < .01, ***p < .001 138 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice Table B.9. Determinants of informality employees in public and private businesses (1) (2) (3) (4) 2022 Establishment Bhutan Labor Force Survey, 2017 Survey = 1 if no written = 1 if no provi- = 1 if no provi- contract dent funds dent funds Demographic characteristics Male -0.00818 0.00594 -0.0165 (0.0193) (0.0309) (0.0325) Female 0.00942 [0.0143] Age 0.00459 -0.0285*** (0.00690) (0.00538) Age-squared -0.0000374 0.000299*** (0.0000878) (0.0000707) Primary 0.00256 -0.0706* -0.0632 -0.0469 (0.0441) (0.0355) (0.0355) [0.0301] Secondary -0.0692* -0.158*** -0.162*** (0.0345) (0.0309) (0.0315) Middle secondary -0.061 [0.0253]** Higher secondary -0.0212 [0.0263] Tertiary -0.105* -0.272*** -0.317*** -0.0815 (0.0403) (0.0662) (0.0410) [0.0265]*** Master’s and above -0.0515 [0.0426] Geographic controls (ref.: Thimphu/rural areas) Gelephu 0.0745 0.0196 0.0535 0.229 (0.0401) (0.0386) (0.0423) [0.0253]*** Phuentsholing -0.0428* -0.263*** -0.300*** -0.001 (0.0207) (0.0416) (0.0366) [0.0207] Punakha -0.156*** -0.123 -0.102 -0.0358 (0.0385) (0.0719) (0.0666) [0.0238] Samdrup Jongkhar -0.0587 -0.233** -0.263** -0.0207 (0.0451) (0.0810) (0.0888) [0.0322] Trashigang 0.0193 -0.212*** -0.178** 0.179 (0.0541) (0.0484) (0.0551) [0.0292]*** Urban (ref.: Rural) -0.0368 -0.0883* (0.0331) (0.0436) Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 139 Table B.9. Continued (1) (2) (3) (4) 2022 Establishment Bhutan Labor Force Survey, 2017 Survey = 1 if no written = 1 if no provi- = 1 if no provi- contract dent funds dent funds Industry (ref.: Agriculture): Manufacturing -0.155* -0.0303 -0.00267 -0.0126 (0.0607) (0.0868) (0.0925) [0.0445] Energy/Water -0.213** -0.157 -0.210* -0.175 (0.0795) (0.0862) (0.0832) [0.0459]*** Water 0.0434 [0.145] Construction -0.212** 0.0931 0.0834 0.162 (0.0652) (0.0856) (0.0870) [0.0500]*** Trade -0.0985 0.198* 0.233** 0.0657 (0.0770) (0.0812) (0.0771) [0.0479] Transportation -0.204** 0.212* 0.165 0.0148 (0.0754) (0.0956) (0.0951) [0.0562] Hotels/Restaurants -0.128 0.129 0.142 0.131 (0.0742) (0.101) (0.0976) [0.0466]*** Information/Comunications -0.157 -0.0245 -0.0726 -0.0574 (0.0799) (0.0849) (0.0815) [0.0512] Finance/Real Estate -0.164* -0.160 -0.219* -0.0603 (0.0798) (0.109) (0.105) [0.0428] Real estate activities 0.338 [0.0526]*** Science -0.107 0.226* 0.147 0.0907 (0.0763) (0.104) (0.0933) [0.0809] Admininistration -0.112 0.284** 0.258** 0.0303 (0.0838) (0.0915) (0.0845) [0.0604] Public -0.246* -0.0574 -0.123 (0.120) (0.122) (0.138) Education -0.240* 0.0342 -0.00872 -0.0949 (0.105) (0.0985) (0.0885) [0.0471]** Health 0.291*** 0.194* 0.220** -0.0624 (0.0678) (0.0841) (0.0815) [0.0981] Arts/Other services -0.0566 0.285** 0.276*** 0.242 (0.0867) (0.0850) (0.0792) [0.0713]*** Other services -0.0491 [0.101] 140 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice Table B.9. Continued (1) (2) (3) (4) 2022 Establishment Bhutan Labor Force Survey, 2017 Survey = 1 if no written = 1 if no provi- = 1 if no provi- contract dent funds dent funds Occupations (ref.: Managers) Professionals -0.0177 -0.0390 (0.0520) (0.0411) Technicians -0.0586 -0.0340 (0.0350) (0.0501) Clerical 0.0114 -0.0698 (0.0530) (0.0513) Services/Sales -0.0434 0.0162 (0.0592) (0.0531) Agriculture -0.226 0.00581 (0.254) (0.234) Craft 0.0417 0.150* (0.0567) (0.0605) Operators 0.0420 -0.0336 (0.0495) (0.0490) Elementary 0.0107 0.0826 (0.0687) (0.0633) Constant 0.834*** 1.347*** 0.747*** 0.714 (0.140) (0.149) (0.0862) [0.0511]*** No. of observations 2,636 2,636 2,636 3,920 *p < .05, **p < .01, ***p < .001 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 141 Table B.10. Blinder-Oaxaca decomposition of the gender employment gap (1) (2) (3) (4) Employment rate of men ages 15+ (A) 0.702*** 0.702*** 0.702*** 0.702*** (0.0113) (0.0112) (0.0110) (0.0113) Employment rate of women ages 15+ (B) 0.492*** 0.492*** 0.492*** 0.492*** (0.0248) (0.0254) (0.0251) (0.0249) Raw employment gap (A –B) 0.210*** 0.210*** 0.210*** 0.210*** (0.0179) (0.0175) (0.0175) (0.0171) Explained 0 -0.00164 -0.00877 -0.00360 (0.00107) (0.00578) (0.00552) Unexplained 0.210*** 0.212*** 0.219*** 0.214*** (0.0179) (0.0169) (0.0152) (0.0148) a. Explained part of the gap (due to differences in the following characteristics): decomposition Dzongkhags -0.00177* -0.00225* -0.00233* (0.000891) (0.00104) (0.00106) Area 0.000123 0.000183 0.000180 (0.000334) (0.000495) (0.000488) Age -0.00414 -0.00446 (0.00296) (0.00263) Education -0.00256 -0.00124 (0.00435) (0.00424) Married 0.00263** (0.000895) Household composition 0.00163*** (0.000356) b. Unexplained part of the gap (baseline difference and difference in returns to the same characteristics): decomposition Dzongkhags 0.0242* 0.0300** 0.0310** (0.00958) (0.00940) (0.00956) Area 0.0416** 0.0443** 0.0339* (0.0151) (0.0162) (0.0132) Age 0.0296** 0.0204* (0.00978) (0.0100) Education -0.0290* -0.0291* (0.0138) (0.0136) Married 0.151*** (0.0118) Household composition 0.00289 (0.00502) Constant 0.210*** 0.146*** 0.144*** 0.00342 (0.0179) (0.00825) (0.0203) (0.0251) No. of observations 30,357 30,357 30,357 30,357 Note: Because alternative reference groups yield different estimates of the contribution of each factor variable to the unexplained part of the gap, this anal- ysis follows Yun (2005) and measures the true contribution of a factor variable as the average effect of every possible specification changing the reference category (following a normalized regression approach). *p < .05, **p < .01, ***p < .001 142 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice Appendix C: Public-Private Wage Differential in Bhutan Definitions This analysis uses the 2022 Bhutan Labor Force Survey. It includes workers ages 15 and over in both the public and private sectors, and it excludes workers in agricultural occupations as well as family workers (mainly women, for which wages are unavailable). The public sector includes civil servants and the armed forces. The private sector includes workers in private com- panies and in private businesses. State-owned enterprises (SoEs) include SoEs and government firms. Figures Wages are nominal monthly wages in 2022 for the primary occupation (corresponding to the sector). They are truncated at 1 percent (top only). Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 143 Hourly wages are nominal monthly wages in 2022 divided by four times the number of weekly hours worked in 2022. They are truncated at 1 percent (top only). Main findings Wage gap between the public sector (excluding SoEs) and the private sector The raw hourly wage gap between the public sector (excluding SoEs) and the private sector is 24 percent in favor of the public sector. This gap is explained when taking into account gender and age, but, even more important, education and occupations. This means that wages are higher in the public sector because the sector has relatively more men (who are better paid than women), more experienced workers, and a higher proportion of skilled workers and highly paid occupations than the private sector. The gap at the baseline remains positive, but it is offset in total by the better returns to being a man and being experienced (age) in the private sector than in the public sector (table C.1). Wage gap between the public sector (including SoEs) and the private sector Unsurprisingly, the results are not very different because employment in SoEs and government firms constitutes 20 percent of total public employment. The average raw gap is now 26 percent because wages in SoEs are higher on average. In terms of the decomposition, the same conclusions prevail (table C.2). Wage gap between SoEs and the private sector This decomposition compares workers in private companies with workers in public companies (SoE + gov. firms), with the idea that this is where jobs are most comparable. The sample size is much smaller—N = 1,291 versus N = 8,987 when looking at the wage differential between the public (including SoEs) and private sectors. The hourly wage gap is very high, and 38 percent and 50 percent of it remain unexplained after taking into account age, gender, location, education, and location. The decomposition suggests that differences across workers in terms of education and occupation account for half of the gap, whereas differences in returns to these characteristics explain the other half of the gap (table C.3). 144 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice Table C.1. Blinder-Oaxaca decomposition of wage gap between public sector (excluding SoEs) and private sector Log hourly wage of workers in the public sector, 3.946*** 3.946*** 3.946*** 3.946*** excluding SoEs (A), N = 3,342 (0.0299) (0.0300) (0.0322) (0.0349) Log hourly wage of workers in the private sector (B), 3.708*** 3.708*** 3.708*** 3.708*** N = 4,789 (0.0534) (0.0533) (0.0547) (0.0544) Raw log hourly wage gap 0.238*** 0.238*** 0.238*** 0.238*** (0.0309) (0.0313) (0.0317) (0.0299) Explained -0.000580 0.161*** 0.208*** (0.00646) (0.0195) (0.0320) Unexplained 0.238*** 0.239*** 0.0770** 0.0305 (0.0309) (0.0312) (0.0254) (0.0284) a. Explained part of the gap (due to differences in the following characteristics) Dzongkhags -0.00314 -0.00184 -0.00277 (0.00641) (0.00586) (0.00582) Area 0.00256 0.00143 0.00116 (0.00293) (0.00175) (0.00147) Sex (1=male) 0.0107*** 0.0113*** (0.00308) (0.00324) Age 0.0146** 0.0133*** (0.00455) (0.00374) Education 0.136*** 0.0915*** (0.0189) (0.0152) Occupation 0.0931*** (0.0184) b. Unexplained part of the gap (baseline difference and differences in returns to the same characteristics) Dzongkhags -0.0456 -0.0186 -0.0120 (0.0337) (0.0337) (0.0286) Area 0.0165 0.0184 0.0208 (0.0240) (0.0208) (0.0210) Sex (1=male) -0.0909*** -0.140*** (0.0225) (0.0208) Age -0.0778 -0.129** (0.0457) (0.0445) Education 0.0186 0.0397 (0.0350) (0.0346) Occupation -0.00866 (0.0143) Constant 0.238*** 0.268*** 0.227*** 0.260*** (0.0309) (0.0300) (0.0622) (0.0733) No. of observations 8,131 8,131 8,131 8,131 Note: Real log hourly wage in 2010 Nu.; winsorized at 1 percent (top only). *p < .05, **p < .01, ***p < .001 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 145 Table C.2. Blinder-Oaxaca decomposition of wage gap between public sector (including SoEs) and private sector Log hourly wage of workers in the public sector, 3.970*** 3.970*** 3.970*** 3.970*** including SoEs (A), N = 4,198 (0.0398) (0.0387) (0.0409) (0.0439) Log hourly wage of workers in the private sector (B), 3.708*** 3.708*** 3.708*** 3.708*** N = 4,789 (0.0534) (0.0533) (0.0547) (0.0544) Raw log hourly wage gap 0.262*** 0.262*** 0.262*** 0.262*** (0.0209) (0.0226) (0.0222) (0.0210) Explained 0.00454 0.170*** 0.204*** (0.00598) (0.0173) (0.0278) Unexplained 0.262*** 0.258*** 0.0922*** 0.0588* (0.0209) (0.0228) (0.0224) (0.0271) a. Explained part of the gap (due to differences in the following characteristics) Dzongkhags -0.00110 -0.000989 -0.00143 (0.00585) (0.00534) (0.00535) Area 0.00565 0.00313 0.00257 (0.00320) (0.00213) (0.00192) Sex (1=male) 0.00891* 0.00946* (0.00358) (0.00376) Age 0.0169*** 0.0153*** (0.00472) (0.00378) Education 0.142*** 0.0965*** (0.0159) (0.0149) Occupation 0.0813*** (0.0136) b. Unexplained part of the gap (baseline difference and differences in returns to the same characteristics) Dzongkhags -0.0399 -0.0155 -0.0103 (0.0256) (0.0299) (0.0270) Area 0.0217 0.0209 0.0252 (0.0227) (0.0198) (0.0199) Sex (1=male) -0.0954*** -0.140*** (0.0242) (0.0222) Age -0.0756 -0.119** (0.0431) (0.0414) Education 0.0309 0.0535 (0.0353) (0.0340) Occupation -0.0165 (0.0125) Constant 0.262*** 0.276*** 0.227*** 0.266*** (0.0209) (0.0287) (0.0576) (0.0723) No. of observations 8,987 8,987 8,987 8,987 Note: Real log hourly wage in 2010 Nu; winsorized at 1% (top only). *p < .05, **p < .01, ***p < .001 146 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice Table C.3. Blinder-Oaxaca decomposition of the wage gap between SoEs and private sector Log hourly wage of workers in SoEs (A), N = 856 4.062*** 4.062*** 4.062*** 4.062*** (0.0799) (0.0790) (0.0799) (0.0817) Log hourly wage of workers in private companies 3.684*** 3.684*** 3.684*** 3.684*** (B), N = 435 (0.0791) (0.0845) (0.0842) (0.0834) Raw log hourly wage gap 0.377*** 0.377*** 0.377*** 0.377*** (0.0337) (0.0408) (0.0365) (0.0362) Explained 0.0815** 0.154*** 0.173*** (0.0309) (0.0256) (0.0325) Unexplained 0.377*** 0.296*** 0.224*** 0.205*** (0.0337) (0.0464) (0.0387) (0.0256) a. Explained part of the gap (due to differences in the following characteristics) Dzongkhags 0.0853** 0.0452* 0.0234 (0.0298) (0.0180) (0.0159) Area -0.00381 -0.00121 -0.000248 (0.0118) (0.00391) (0.00128) Sex (1=male) -0.00640 -0.00607 (0.00403) (0.00356) Age 0.0349** 0.0221** (0.0117) (0.00749) Education 0.0810*** 0.0364** (0.0165) (0.0118) Occupation 0.0972* (0.0390) b. Unexplained part of the gap (baseline difference and differences in returns to the same characteristics) Dzongkhags -0.0466 0.0322 0.0287 (0.0554) (0.0500) (0.0360) Area 0.0425 0.0606 0.0586 (0.0800) (0.0599) (0.0463) Sex (1=male) -0.0579 -0.0261 (0.0347) (0.0451) Age 0.219* 0.152* (0.0858) (0.0610) Education 0.216*** 0.166*** (0.0342) (0.0415) Occupation 0.0684* (0.0281) Constant 0.377*** 0.300*** -0.247** -0.243*** (0.0337) (0.0736) (0.0771) (0.0670) No. of observations 1,291 1,291 1,291 1,291 Note: Real log hourly wage in 2010 Nu; winsorized at 1% (top only). *p < .05, **p < .01, ***p < .001 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 147 Appendix D: Supplementary Figures and Tables, Chapter 3 Figure D.1. Education and sectoral employment, 2022 a. Share of each education level, by sector Agriculture/mining Manufacturing Energy/water Construction Trade Transportation Hotels/restaurants Information/comms. Finance/real estate Science Administration Public Education Health Arts/other 0 10 20 30 40 50 60 70 80 90 100 Percent None/NFE Primary Low secondary Middle secondary High secondary Certificate/diploma Bachelor's and above Monastic b. Share of each sector in employment of high-skilled Agriculture/mining 2.73 Manufacturing 3.1 Energy/water 3.05 Construction 3.7 Trade 10 Transportation 1.47 Hotels/restaurants 4.11 Information/comms. 3.82 Finance/real estate 6.27 Science 2.1 Administration 1.94 Public 22.22 Education 29.98 Health 4.56 Arts/other 0.94 0 5 10 15 20 25 30 35 Sectoral distribution of employment (%) Source: Bhutan Labor Force Survey, 2022. 148 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice Table D.1. Profile of firms in Bhutan, 2018 and 2022 2018 2022 N % N Weighted N % Geographic distribution Thimphu 3,508 24.8 1,295 6,230 23.5 Phuentsholing 2,843 20.1 863 3,701 13.9 Gelephu 3,160 22.4 848 8,275 31.2 Samdrup Jongkhar 940 6.7 322 1,567 5.9 Punakha 2,051 14.5 758 3,878 14.6 Trashigang 1,632 11.6 614 2,890 10.9 Firm age Less than 1 year 403 3.2 119 702 2.7 Between 1 and 9 years 9,244 72.9 3,257 18,935 71.4 10 years or more 3,030 23.9 1,323 6,897 26.0 Economic sector Agriculture, forestry, and fishing 363 2.6 25 152 0.6 Mining and quarrying 33 0.2 31 105 0.4 Manufacturing 753 5.4 565 1,695 6.4 Electricity, gas, steam, and air-conditioning 3 0.0 13 48 0.2 Water supply, sewerage, and waste management 5 0.0 7 38 0.1 Construction 168 1.2 151 690 2.6 Wholesale and retail trade 8,523 61.6 1,856 1,4828 55.9 Transportation and storage 45 0.3 78 177 0.7 Accommodation and food services activities 2,946 21.3 1,359 6,331 23.9 Information and communication 57 0.4 53 132 0.5 Financial and insurance activities 17 0.1 20 43 0.2 Real estate activities 3 0.0 3 13 0.1 Professional, scientific and technical 71 0.5 67 276 1.0 Administrative and support services activities 173 1.3 143 647 2.4 Education 86 0.6 94 275 1.0 Human health and social work activities 19 0.1 15 35 0.1 Arts, entertainment, and recreation 172 1.2 113 327 1.2 Other service activities 402 2.9 107 728 2.7 Economic sector Sole proprietorship 13,681 96.8 4,371 25,742 97.0 Partnership 79 0.6 103 355 1.3 State-owned enterprise (SOE) 28 0.2 17 19 0.1 Private limited company 157 1.1 124 286 1.1 Public limited company 16 0.1 26 32 0.1 Foreign direct investment (FDI) business 31 0.2 18 39 0.2 Civil society organization (CSO)/ NGO 21 0.2 22 44 0.2 Public organization 17 20 0.1 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 149 Table D.1. Continued 2018 2022 N % N Weighted N % Cooperatives and groups 116 0.8 1 2 0.0 Others (specify) 5 0.0 1 2 0.0 Size Cottage 12,553 88.8 4,060 25,448 95.9 Small 1,243 8.8 409 777 2.9 Medium 261 1.9 168 227 0.9 Large 77 0.5 63 88 0.3 Sources: 2022 Establishment Survey and 2018 Economic Census. Table D.2. Characteristics of workers in Bhutanese firms, 2022 N Weighted N % Gender composition Share of female employees 1,994 8,259 47.4 Nationality composition Share of foreign workers 1,994 8,259 2.8 Type of employment Regular employees 1,994 8,259 91.4 Contract employees 1,994 8,259 2.5 Casual employees 1,994 8,259 6.1 Educational level No education 1,476 15,825 18.6 Primary and secondary 4,050 16,948 61.5 Degree and above 1,101 16,948 19.9 Experience at the firm Less than 1 year 450 6,821 10.3 Between 1 and 9 years 2,747 46,720 70.2 10 years or more 709 12,989 19.5 Vocational qualification Share of employees with vocational qualification 6,627 85,122 30.7 Sources: 2022 Establishment Survey and 2018 Economic Census. 150 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice Table D.3. Average number of workers hired and workers who exited firms, by year, region, economic sector, size, and legal status, 2019–21 Net rate: workers hired Workers hired Workers exited – workers exited 2019 2020 2021 2019 2020 2021 2019 2020 2021 Geographic region Thimphu 1.5 0.9 0.8 0.7 0.7 0.5 0.8 0.2 0.3 Gelephu 1.3 1.0 1.0 0.6 0.5 0.6 0.7 0.5 0.4 Phuentsholing 1.0 0.5 0.6 0.7 0.7 0.4 0.3 -0.1 0.2 Punakha 0.6 0.3 0.2 0.4 0.3 0.2 0.2 -0.1 0.0 Samdrup Jongkhar 0.6 0.3 0.6 0.3 0.3 0.3 0.3 0.0 0.3 Trashigang 0.5 0.3 0.2 0.4 0.2 0.2 0.2 0.0 0.0 Economic sector Agriculture, forestry, and fishing 1.2 1.0 0.4 0.7 0.5 1.7 0.5 0.5 -1.3 Mining and quarrying 2.3 2.4 2.3 1.6 1.2 1.3 0.7 1.2 1.0 Manufacturing 2.4 1.3 1.4 1.4 1.3 1.0 1.0 0.0 0.5 Electricity, gas, steam, and air-conditioning 10.5 4.8 4.4 6.4 3.1 6.3 4.2 1.6 -1.9 Water supply, sewerage, and waste management 2.4 1.6 0.0 1.3 0.2 0.1 1.1 1.4 -0.1 Construction 12.0 10.2 10.8 7.2 5.6 6.1 4.7 4.6 4.7 Wholesale and retail trade 0.3 0.2 0.2 0.1 0.1 0.1 0.2 0.1 0.1 Transportation and storage 1.0 0.5 0.4 0.3 0.3 0.3 0.7 0.1 0.1 Accommodation and food services activities 0.8 0.3 0.3 0.4 0.4 0.2 0.5 -0.1 0.0 Information and communication 4.6 3.9 4.5 2.2 1.9 2.1 2.5 2.0 2.4 Financial and insurance activities 12.0 4.6 4.9 6.6 3.7 3.3 5.4 0.9 1.6 Real estate activities 0.0 0.9 1.5 0.0 0.9 0.0 0.0 0.0 1.5 Professional, scientific and technical 1.0 0.4 0.6 0.5 0.3 0.4 0.5 0.1 0.2 Administrative and support services activities 1.7 0.8 0.6 1.0 1.1 0.7 0.7 -0.3 -0.1 Education 3.8 2.2 2.6 1.3 1.0 1.4 2.6 1.2 1.3 Human health and social work activities 1.2 0.4 1.3 0.2 0.3 0.5 1.0 0.2 0.8 Arts, entertainment, and recreation 0.7 0.4 0.3 0.4 0.4 0.1 0.3 -0.1 0.2 Other service activities 0.5 0.2 0.4 0.2 0.2 0.2 0.4 0.1 0.2 Economic sector Cottage 0.5 0.3 0.3 0.2 0.2 0.2 0.3 0.1 0.1 Small 5.4 2.9 3.2 2.3 2.3 1.5 3.1 0.7 1.7 Medium 9.9 6.7 7.2 5.4 6.0 4.9 4.5 0.7 2.4 Large 75.7 67.9 66.7 56.0 51.7 55.2 19.7 16.1 11.5 Legal status Sole proprietorship 0.7 0.4 0.4 0.3 0.3 0.2 0.4 0.1 0.2 Partnership 3.2 1.5 1.6 1.0 1.0 0.9 2.2 0.5 0.7 State-owned enterprise (SOE) 11.8 13.4 11.9 22.2 11.0 7.4 -10.5 2.4 4.5 Private limited company 22.9 20.1 20.8 14.5 14.7 14.8 8.5 5.4 6.0 Public limited company 24.1 15.4 17.8 18.7 16.8 15.9 5.5 -1.4 1.9 Foreign direct investment (FDI) business 21.2 14.7 14.1 14.7 13.6 15.0 6.5 1.1 -0.9 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 151 Table D.3. Continued Net rate: workers hired Workers hired Workers exited – workers exited 2019 2020 2021 2019 2020 2021 2019 2020 2021 Civil society organization (CSO)/NGO 2.0 1.6 3.6 0.4 0.6 2.1 1.6 1.0 1.5 Public organization 13.4 11.5 14.9 7.8 8.2 9.5 5.6 3.3 5.4 Cooperatives and groups . . 2.0 . . 1.0 1.0 Other (specify) 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Source: 2022 Establishment Survey. Table D.4. Distribution of current employment and expected labor demand, by occupation, 2022 Current Expected labor composition demand Managers 8.7 3.6 Professionals 12.4 6.5 Technicians and associate professionals 11.5 6.8 Clerical support workers 8.1 6.1 Services and sales workers 36.6 43.9 Personal services workers 9.48 23.3 Sales workers 24.35 20.3 Personal care workers 0.11 0.0 Protective services workers 2.61 0.3 Forestry workers 0.1 0.6 Craft and related trades workers 11.2 17.7 Building and related trades workers (excluding electricians) 2.68 6.9 Metal, machinery, and related trades workers 3.78 3.1 Handicraft and printing workers 0.39 0.9 Electrical and electronics trades workers 1.9 2.3 Food processing, woodworking, garment, and other craft and related trades workers 2.46 4.5 Plant and machine operators and assemblers 5.5 4.0 Elementary occupations 6.0 10.7 Source: 2022 Establishment Survey. 152 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice Table D.5. Share of firms expecting redundant occupations or emerging occupations in the next five years, by region and economic sector, 2022 Share of firms Share of firms expecting redundant expecting new occu- occupations (%) pations (%) All firms 0.7 30.6 By region Thimphu 1.3 34.0 Gelephu 0.5 42.1 Phuentsholing 0.2 17.8 Punakha 0.4 30.0 Samdrup Jongkhar 1.6 20.4 Trashigang 0.2 12.7 By sector Agriculture, forestry, and fishing 1.0 43.3 Mining and quarrying 3.1 54.1 Manufacturing 0.7 44.7 Electricity, gas, steam, and air-conditioning 30.4 44.3 Water supply, sewerage, and waste management 0.0 42.1 Construction 2.3 60.1 Wholesale and retail trade 0.6 23.5 Transportation and storage 1.4 30.6 Accommodation and food services activities 0.4 33.4 Information and communication 1.8 58.1 Financial and insurance activities 4.7 26.2 Real estate activities 0.0 80.2 Professional, scientific and technical 0.0 46.6 Administrative and support services activities 0.9 42.1 Education 2.3 64.6 Human health and social work activities 4.3 38.3 Arts, entertainment, and recreation 0.0 27.7 Other service activities 0.2 47.7 Source: 2022 Establishment Survey. Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 153 Table D.6. Main reasons for worker shortages, 2022 Services and sales All (%) workers (%) Skills are not available in Bhutan 12.4 6.9 Skills are not available in my business region/dzongkhag 11.4 9.2 High demand for these skills in Bhutan 5.9 5.6 Lack of resources/business not doing well 7.5 8.5 Recruitment restricted by other agency or higher authority 1.1 0.9 Workers demand wages that are too high/wage rate lower than market rate 28.5 33.6 Applicants lack required experience 22.7 21.3 Other 10.5 14.1 Source: 2022 Establishment Survey. Table D.7. Firms facing retainment difficulties, total and by occupation, 2022 Share of firms facing retaining difficulties (%) All firms 5.97 By occupation Managers 1.2 Professionals 3.8 Technicians and associate professionals 7.7 Clerical support workers 1.8 Services and sales workers 39.1 Forestry workers Craft and related trades workers 25.7 Plant and machine operators and assemblers 4.9 Elementary occupations 15.7 Source: 2022 Establishment Survey. Note: Table shows the distribution of the total number of firms facing retainment difficulties by the occupation for which they face difficulties. Table D.8. Reasons for hiring foreign workers, 2022 Lack of skilled Bhutanese workers 31.2 Cheap wage rate 10.5 Easy to manage 11.4 Better work attitude 19.0 Better workmanship (highly skilled) 20.5 Easy access to foreign workers 4.8 Other (specify) 2.6 Source: 2022 Establishment Survey. 154 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice Table D.9. Outcomes obtained from training in the last three years, 2022 No Moderate Significant improvement improvement improvement Confidence 4.5 34.8 60.7 Work productivity 5.1 34.2 60.8 Organization productivity 5.4 37.7 56.9 Ability to work independently 6.3 34.4 59.3 Leadership skills 9.7 39.2 51.2 Creative and critical thinking 9.0 46.3 44.7 Problem-solving skills 8.0 43.8 48.2 Job-specific technical skills 7.9 40.4 51.7 Source: 2022 Establishment Survey. Table D.10. Importance of factors for business expansion or diversification plans, by firm size, 2022 Least important Most important 1 2 3 4 Cottage Human resources 9.3 20.7 54.1 15.9 Finance 36.2 41.9 18.4 3.5 Market 51.4 33.7 12.3 2.5 Favorable policies and regulations in place 3.3 3.8 15.1 77.7 Small Human resources 20.6 31.1 40.2 8.2 Finance 35.0 32.2 24.0 8.8 Market 38.0 27.7 23.7 10.6 Favorable policies and regulations in place 6.6 9.1 12.0 72.4 Medium Human resources 23.0 28.7 35.5 12.8 Finance 29.8 33.2 28.2 8.9 Market 39.6 30.4 20.8 9.2 Favorable policies and regulations in place 8.3 7.4 15.2 69.0 Large Human resources 36.7 36.8 20.1 6.4 Finance 21.2 20.3 32.0 26.5 Market 28.2 22.4 28.4 21.1 Favorable policies and regulations in place 13.8 20.6 19.6 46.0 Source: 2022 Establishment Survey. Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 155 Table D.11. Constraints in the management of firms, by firm size, 2022 a. Distribution of firms by severity level for each factor (%) No Minor Moderate Major Very severe constraint constraint constraint constraint constraint Cottage Business climate factors Internet access and connectivity 58 23 10 7 2 Customs and trade regulations 67 20 8 4 1 Business licensing and operations permits 74 16 6 3 2 Access to finance 59 20 10 9 2 Access to raw materials/goods 55 23 11 9 2 Access to market 45 24 16 12 3 Policy uncertainty 67 18 8 4 2 Corruption, crime, theft, and disorder 84 10 3 2 0 Labor factors Stringent labor law and regulation 81 11 6 2 1 High worker turnover 82 10 4 4 1 Overall market wage level 72 15 9 3 1 Small Business climate factors Internet access and connectivity 40 18 14 12 16 Customs and trade regulations 31 35 19 12 3 Business licensing and operations permits 71 10 14 3 2 Access to finance 50 23 14 10 3 Access to raw materials/goods 33 22 16 13 16 Access to market 38 21 15 17 9 Policy uncertainty 18 38 17 20 7 Corruption, crime, theft, and disorder 46 32 15 4 3 Labor factors Stringent labor law and regulation 48 25 16 8 3 High worker turnover 15 39 13 21 11 Overall market wage level 30 34 21 8 7 Medium Business climate factors Internet access and connectivity 33 22 20 13 11 Customs and trade regulations 41 35 15 8 1 Business licensing and operations permits 64 23 6 5 3 Access to finance 43 23 13 16 6 Access to raw materials/goods 43 20 18 15 4 Access to market 36 21 23 15 5 Policy uncertainty 37 19 19 12 14 Corruption, crime, theft, and disorder 69 17 7 4 2 156 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice Table D.11. Continued a. Distribution of firms by severity level for each factor (%) No Minor Moderate Major Very severe constraint constraint constraint constraint constraint Labor factors Stringent labor law and regulation 45 26 12 11 6 High worker turnover 38 27 14 16 5 Overall market wage level 46 18 16 14 5 Large Business climate factors Internet access and connectivity 40 18 14 12 16 Customs and trade regulations 31 35 19 12 3 Business licensing and operations permits 71 10 14 3 2 Access to finance 50 23 14 10 3 Access to raw materials/goods 33 22 16 13 16 Access to market 38 21 15 17 9 Policy uncertainty 18 38 17 20 7 Corruption, crime, theft, and disorder 46 32 15 4 3 Labor factors Stringent labor law and regulation 55 21 12 10 2 High worker turnover 39 21 17 17 6 Overall market wage level 40 26 18 11 4 b. Distribution of factors for each specific constraint level (%) Cottage Business climate factors Internet access and connectivity 8 12 11 12 11 Customs and trade regulations 9 10 9 7 8 Business licensing and operations permits 10 8 6 5 11 Access to finance 8 11 11 15 11 Access to raw materials/goods 7 12 12 15 13 Access to market 6 12 18 20 20 Policy uncertainty 9 10 9 7 11 Corruption, crime, theft, and disorder 11 5 3 3 3 Labor factors Stringent labor law and regulation 11 6 6 3 5 High worker turnover 11 5 4 7 4 Overall market wage level 10 8 10 5 4 Small Business climate factors Internet access and connectivity 9 6 8 9 20 Customs and trade regulations 7 12 11 10 4 Business licensing and operations permits 17 3 8 2 3 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 157 Table D.11. Continued b. Distribution of factors for each specific constraint level (%) No Minor Moderate Major Very severe constraint constraint constraint constraint constraint Access to finance 12 8 8 8 3 Access to raw materials/goods 8 7 9 10 20 Access to market 9 7 9 13 11 Policy uncertainty 4 13 10 16 8 Labor factors Stringent labor law and regulation 11 9 9 6 3 High worker turnover 4 13 8 17 14 Overall market wage level 7 11 12 6 9 Medium Business climate factors Internet access and connectivity 7 9 12 10 19 Customs and trade regulations 8 14 9 6 2 Business licensing and operations permits 13 9 4 4 4 Access to finance 9 9 8 12 9 Access to raw materials/goods 9 8 11 12 6 Access to market 7 8 14 11 9 Policy uncertainty 7 8 11 9 22 Corruption, crime, theft, and disorder 14 7 4 3 4 Labor factors Stringent labor law and regulation 9 11 7 8 9 High worker turnover 8 11 9 12 8 Overall market wage level 9 7 10 11 9 Large Business climate factors Internet access and connectivity 3 2 3 3 9 Customs and trade regulations 2 4 3 3 2 Business licensing and operations permits 5 1 3 1 1 Access to finance 3 3 3 2 2 Access to raw materials/goods 2 3 3 3 9 Access to market 3 3 3 4 5 Policy uncertainty 1 5 3 5 4 Corruption, crime, theft, and disorder 3 4 3 1 2 Labor factors Stringent labor law and regulation 32 23 20 20 11 High worker turnover 23 23 28 35 32 Overall market wage level 24 29 29 24 22 Source: 2022 Establishment Survey. 158 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice Table D.12. Compliance with labor regulation, by economic sector (percent), 2022 We provide Our estab- Our Internal We have We have basic personal lishment has Service occupational occupational protective an Internal Rule (ISR) is health and health and equipment Service Rule endorsed by safety in safety policy (PPE) to our (ISR). the DoL. place. in place. employees. Agriculture, forestry, and 14.4 12.6 31.1 27.7 31.1 fishing Mining and quarrying 32.9 32.0 43.3 41.4 52.2 Manufacturing 32.3 23.3 45.1 31.4 49.2 Electricity, gas, steam, and 100.0 100.0 72.7 72.7 93.9 air-conditioning Water supply, sewerage, amd 8.0 46.1 57.9 36.5 36.5 waste management Construction 22.2 21.0 45.1 35.3 70.2 Wholesale and retail trade 13.7 9.6 32.4 21.7 27.4 Transportation and storage 18.4 10.8 14.2 9.4 10.7 Accommodation and food 15.6 16.4 33.7 18.8 39.1 services activities Information and 60.0 40.2 67.9 48.2 59.3 communication Financial and insurance 100.0 94.7 100.0 94.7 56.8 activities Real estate activities 100.0 0.0 0.0 0.0 0.0 Professional, scientific and 39.9 27.8 35.5 32.0 31.2 technical Administrative and support 72.8 53.3 73.6 52.6 55.3 services activities Education 71.3 47.1 75.6 62.8 35.6 Human health and social work 69.7 65.9 72.9 77.2 47.3 activities Arts, entertainment, and 30.1 19.6 52.1 24.3 37.6 recreation Other service activities 18.1 8.6 32.3 14.5 37.0 Source: 2022 Establishment Survey. Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 159 Table D.12. Continued We have a We have We issue provident We provide a written an appoint- fund for our pay slips/ contract/term ment letter employees We provide evidence of employ- at the time with a recog- overtime pay- of wages ment for our of appoint- nized finan- ment to our paid to our staff and new ment of new cial institute. employees. employees. recruits. recruits. Agriculture, forestry, and 15.5 15.4 30.9 10.8 19.8 fishing Mining and quarrying 34.3 42.3 47.4 43.3 48.0 Manufacturing 24.5 40.1 42.9 21.3 23.7 Electricity, gas, steam, and 90.0 35.9 90.0 100.0 93.9 air-conditioning Water supply, sewerage, amd 8.0 14.3 51.5 40.5 40.5 waste management Construction 50.2 67.2 77.5 32.8 29.0 Wholesale and retail trade 14.7 17.0 27.8 10.5 11.5 Transportation and storage 37.2 12.4 22.3 10.9 13.6 Accommodation and food 8.9 11.2 24.8 9.4 11.3 services activities Information and 60.3 55.9 77.4 57.3 63.5 communication Financial and insurance 100.0 72.8 100.0 100.0 100.0 activities Real estate activities 0.0 0.0 0.0 0.0 0.0 Professional, scientific and 36.5 23.1 57.9 32.4 44.2 technical Administrative and support 62.5 31.5 80.0 44.3 53.7 services activities Education 56.4 32.0 82.9 58.5 71.1 Human health and social work 69.7 33.8 92.5 56.0 64.0 activities Arts, entertainment, and 19.5 20.7 35.1 21.5 12.7 recreation Other service activities 15.5 13.0 23.4 13.5 21.8 Source: 2022 Establishment Survey. 160 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice Table D.12. Continued We have clear We have job roles and Group We have a responsibil- Insurance sexual harass- ities for our We provide We provide Scheme ment policy/ staff and new maternity paternity (GIS) for our grievance sys- recruits leave. leave. employees. tem in place. Agriculture, forestry, and 29.4 24.2 29.3 5.3 5.9 fishing Mining and quarrying 50.3 57.4 46.6 21.6 34.2 Manufacturing 44.7 37.8 35.3 11.5 20.2 Electricity, gas, steam, and 93.9 93.9 93.9 84.1 91.8 air-conditioning Water supply, sewerage, amd 12.3 4.0 8.3 8.0 18.3 waste management Construction 58.9 36.8 59.8 13.3 22.1 Wholesale and retail trade 25.5 23.7 22.2 4.8 11.9 Transportation and storage 31.0 26.1 19.5 2.8 12.7 Accommodation and food 29.4 28.3 22.2 3.3 14.4 services activities Information and 76.6 72.7 76.7 23.6 40.5 communication Financial and insurance 100.0 100.0 100.0 94.1 86.7 activities Real estate activities 0.0 0.0 0.0 0.0 0.0 Professional, scientific and 56.6 32.0 37.0 13.6 21.8 technical Administrative and support 79.8 65.5 66.6 40.6 42.9 services activities Education 85.4 74.7 63.4 23.8 61.8 Human health and social work 92.5 69.7 64.0 49.6 65.2 activities Arts, entertainment, and 33.2 27.8 31.4 13.5 17.3 recreation Other service activities 36.6 44.2 26.4 17.1 12.7 Source: 2022 Establishment Survey. Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 161 Appendix E: Overview of Selected ALMPs Youth Engagement and Livelihood Program (YELP) Launched by the Ministry of Industry, Commerce, and Employment (MoICE) in 2019, YELP helps unemployed youth find regular employment by offering training and work experience. The school-to-work transition program provides job-seekers ages 18–29 with various opportunities to gain on-the-job skills and work experience to enhance their employability across sectors such as construction, agriculture, tourism, hydropower, and hospi- tality. In this scheme, job-seekers are connected with employers to minimize the burden on both the employee and employer, and the ministry provides a monthly allowance of Nu 5,000 as a wage subsidy. The duration of the support ranges from three to 24 months. In addition, in recent years the government has begun to financially support youth who decide to start a business after completing the program (Kuensel 2023). In fiscal 2022/23, 2,545 individuals went through YELP (Bhutan Today 2023). Startup and Cottage and Small Industries (CSI) Development Flagship Program The Startup and Cottage and Small Industries (CSI) Development Flagship Program carried out by MoICE pro- motes entrepreneurship and innovation among youth. To boost the creation of start-ups, the program provides start-up entrepreneurship training and competitions for business ideas for those who would like to become entrepreneurs (MoICE 2023a). As pivotal part of the CSI Flagship Program, five business incubation centers under the Royal University of Bhutan (RUB) Colleges have been established to serve as hubs for entrepreneurship, eco- nomic development, and innovation. The training offered by the five incubation centers consists of mentorship, networking opportunities, and information sessions (RGoB 2022). In addition, “Fablabs” have been established to allow aspiring entrepreneurs to showcase their products and ideas to investors. CSIs are supported by the provi- sion of soft and hard skills training with the goal of enhancing their competitiveness and expanding their busi- ness. The project goal is to increase employment and enhance Bhutan’s overall economic growth (MoICE 2023a). In fiscal 2022/23, 2,079 individuals benefited from the entrepreneur training, and 10 start-up events were hosted (MoICE 2023a). Community Skills Training—SSDP and VSDP Founded in 1996, the Special Skills Development Program (SSDP) provides disadvantaged individuals with skills and vocational training. Although the program was originally geared toward developing the skills of the armed forces in vocational professions, it later expanded to other vulnerable groups such as monks and nuns, juveniles and delinquents, prisoners, and former gang members. The training is offered in coordination with organizations such as the Royal Bhutan Police and the Royal Bhutan army. Launched in 1984, the Village Skills Development Program (VSDP) provides rural communities with skills training and capacity building. The goal is to improve the quality of life in rural communities by promoting skills training, capacity building, reviving and preserving traditional crafts and arts, and discouraging rural to urban migration. 162 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice In particular, the VSDP offers skills training such as haircutting, home appliance repair, and tailoring. Village-spe- cific skills training is offered to villages identified as able to produce certain goods with the goal of creating jobs and establishing a reputation for producing quality goods (MoLHR 2019). In 2022, a total of 24 trainings were imple- mented for 938 beneficiaries of both the SSDP and VSDP. Both programs were implemented across 11 dzongkhags (MoESD 2022). Critical Skills Training (CST) The CST program addresses the short-term needs of the labor market to provide young job-seekers and the unem- ployed in the private sector with employable skills. Training opportunities are offered in food production, tailor- ing, fashion design, hair and beauty, massage therapy, and home appliance repair. Courses range in length from 15 days to 12 months. The program has been implemented in cooperation with registered private and public train- ing providers. It is complemented by an entrepreneurship component to enable self-employment. As of 2022, 506 individuals were trained under the CST, of whom 58 percent were females (MoESD 2022). Critical Capability Development (CCD) The CCD program provides continuous learning opportunities for employees outside the civil service sector. The mandate of this program is to provide lifelong learning through reskilling and upskilling support in the private sector. It offers short-term courses of six months or less and long-term courses of more than six months. The fel- lowships offered are in various sectors such as public management, human resources management, and food science and technology. The training programs have been largely implemented by means of fellowship offers from multilateral and bilateral donors. Examples are the Australian Awards Scholarship offered by the Australian gov- ernment and the Diploma in Hospitality and Tourism offered by the Austrian government. In 2022, 990 individu- als were trained under the CCD, of whom 62 percent were female candidates (MoESD 2022). Skills Development Plan (SDP) In 2021, the Skills Development Plan (SDP) was launched to provide skills training to job-seekers and those affected by the COVID-19 pandemic. Among the services provided were easy and diverse access to skills training; the pro- motion of entrepreneurship; and the linking of individuals to professional opportunities. Eligible individuals are employees laid off during the COVID-19 pandemic, registered job-seekers, and overseas returnees. The 108 courses provided under the SDP are completed in nine sectors such as agriculture, business and services, computing and information technology, construction, and creative arts and design. The courses last from one to seven months (MoLHR 2021c). All courses are complemented by two weeks of entrepreneurship learning to foster self-employ- ment. Course participants receive a monthly stipend of Nu 3,500 (approximately US$45). As of 2022, 1,881 benefi- ciaries had been trained, of whom only 35 percent were female (MoESD 2022). Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 163 Appendix F: Supplementary Tables, Chapter 4 Table F.1. ALMPs and gaps by target Other vulnerable Program Urban/rural Women Youth groups Youth Engagement and Livelihood Program P (YELP) Startup and Cottage and Small Industries (CSI) P Development Flagship Program Special Skills Development Program (SSDP) P Village Skills Development Program (VSDP) P Critical Skills Training (CST) P Critical Capability Development (CCD) P Skills Development Plan (SDP) P Source: Own elaboration. Table F.2. ALMPs and gaps by ALMP classification Private sector Labor market Job search employment Program training assistance assistance Youth Engagement and Livelihood Program (YELP) P P Startup and Cottage and Small Industries (CSI) Development Flagship P P P Program Special Skills Development Program (SSDP) P Village Skills Development Program (VSDP) P Critical Skills Training (CST) P Critical Capability Development (CCD) P Skills Development Plan (SDP) P Source: Own elaboration. 164 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice Table F.3. ALMPs and gaps by type of training On-the-job Program Upskilling Reskilling training Soft skills Youth Engagement and Livelihood Program P P (YELP) Startup and Cottage and Small Industries (CSI) P P P Development Flagship Program Special Skills Development Program (SSDP) P Village Skills Development Program (VSDP) P Critical Skills Training (CST) P Critical Capability Development (CCD) P P Skills Development Plan (SDP) P P Source: Own elaboration. Note: Upskilling helps employees acquire more skills and competencies in their current position. Reskilling equips employees to change jobs across differ- ent sectors or organizations. Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 165 References ADB (Asian Development Bank). 2019. “The Social Protection Indicator for Asia: Assessing Progress.” ADB, Manila. ADB (Asian Development Bank) and World Bank. 2021. “Climate Risk Country Profile”. https://www.adb.org/sites/default/files/publi- cation/722636/climate-risk-country-profile-bhutan.pdf. Angel-Urdinola, D., and R. Leon-Solano. 2013. “A Reform Agenda for Improving the Delivery of ALMPs in the MENA Region.“ IZA Journal of Labor Policy 2 (13): 1–25. Aşık, G., L. Bossavie, J. Kluve, E. N. Özen, M. Nebiler, and A. M. Oviedo. 2022. “The Effects of Subsidizing Social Security Contributions: Job Creation or Informality Reduction?“ Policy Research Working Paper 9904, World Bank, Washington, DC. Bhutan Today. 2023. “YELP at Hold until Next Fiscal Year.” https://www.bhutantoday.bt/ yelp-at-hold-until-next-financial-year/?McasCtx=4&McasTsid=20893. Boyreau, G., and M. Rama. 2015. “Bhutan—Public Finance Reforms Towards Economic Self-Reliance.” Washington, DC: World Bank. Brown, A., and J. Koettl. 2012. “Active Labor Market Programs: How, Why, When, and to What Extent Are They Effective? “Europe and Central Asia Knowledge Brief 58, World Bank, Washington, DC. Business Bhutan. 2023. “Foreign Workers Influx Increase.” September. https://businessbhutan.bt/foreign-workers-influx-increase/ #:~:text=The%20report%20reveals%20that%20the,skilled%20labor%20in%20various%20industries. Castro, L. S., V. S. Villalobos, S. Giannozzi, M. C. Dedios, and K. Lindert. 2020. “Provision of Social and Labor Services.” In Sourcebook on the Foundations of Social Protection Delivery Systems, edited by K. Lindert, T. G. Karippacheril, I. R. Caillava, and K. N. Chavez, 247–80. Washington, DC: World Bank. CEDEFOP (European Centre for the Development of Vocational Training). 2009. “Future Skill Needs for the Green Economy.” Research paper, Publications Office of the European Union, Luxemburg. Farole, T., Y. Cho, L. Bossavie, and R. Aterido. 2017. “Bangladesh Jobs Diagnostic.” World Bank, Washington, DC. https://openknowl- edge.worldbank.org/entities/publication/adf7a9e6-b94f-5a08-9278-c9597a655159. GIZ (Deutsche Gesellschaft für Internationale Zusammenarbeit). 2016. “Green Economic Development.” https://www.giz.de/en/ worldwide/18237.html. GIZ (Deutsche Gesellschaft für Internationale Zusammenarbeit). 2017. “TVET and the Promotion of Innovation for Green Employment.” https://www.giz.de/en/worldwide/17848.html. Goswami, A., D. Medvedev, and E. Olafsen. 2019. “High-Growth Firms: Facts, Fiction, and Policy Options for Emerging Economies.” World Bank, Washington, DC. Granata, J., J. Posadas, and M. Testaverde. 2022. “Indonesia’s Online Vacancy Outlook: From Online Job Postings to Labor Market Intelligence 2020.” World Bank, Washington, DC. https://documents1.worldbank.org/curated/en/936031636696719107/pdf/ Indonesia-s-Online-Vacancy-Outlook-From-Online-Job-Postings-to-Labor-Market-Intelligence-2020.pdf. Haldar, A., and J. Stiglitz. 2016. “Group Lending, Joint Liability, and Social Capital: Insights from the Indian Microfinance Crisis.” Politics and Society 44 (4): 459–97. Ibarraran, P., and D. Rosas. 2009. “Evaluating the Impact of Job Training Programs in Latin America: Evidence from IDB-funded Operations.“ Journal of Development Effectiveness 1: 195–216. ILO (International Labour Organization). 2009. “The Green Jobs Program of the ILO.“ ILO, Geneva. 166 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice ILO (International Labour Organization). 2016. “The ILO’s Green Jobs Programme.“ ILO, Geneva. https://www.ilo.org/global/topics/ green-jobs/WCMS_213842/lang--en/index.htm. ILO (International Labour Organization). 2019. “Small Matters: Global Evidence on the Contribution to Employment by the Self- Employed. Microenterprises, and SMEs.” https://www.ilo.org/wcmsp5/groups/public/---dgreports/---dcomm/---publ/docu- ments/publication/wcms_723282.pdf. Khandker, S. 1998. Fighting Poverty with Microcredit: Experience in Bangladesh. Oxford, UK: Oxford University Press. Khandker, S., and H. Samad. 2014. “Dynamic Effects of Microcredit in Bangladesh.” Policy Research Working Paper 6821, World Bank, Washington, DC. Kluve, J. 2016. “A Review of the Effectiveness of Active Labour Market Programmes with a Focus on Latin America and the Caribbean.“ Research Department Working Paper No. 9, International Labour Office, Geneva. Kose, M. A., and F. Ohnsorge, eds. 2023. “Falling Long-Term Growth Prospects: Trends, Expectations, and Policies.” World Bank, Washington, DC. https://www.worldbank.org/en/research/publication/long-term-growth-prospects. KRIVET (Korea Research Institute for Vocational Education and Training). 2022. “History.“ https://www.krivet.re.kr/eng/eu/ea/ euGDADs.jsp. Kuddo, A. 2012. “Public Employment Services and Activation Policies.“ Social Protection and Labor Discussion Paper No. 1215, World Bank, Washington, DC. Kuensel. 2021. “Number and Professionalism: Shortages Facing Bhutan’s Construction Sector.“ https://kuenselonline.com/num- ber-and-professionalism-shortages-facing-bhutans-construction-sector/#:~:text=Construction%20companies%20do%20not%20 prefer,a%20construction%20company%20in%20Thimphu. Kuensel. 2023. “Migration of Bhutanese.“ https://kuenselonline.com/migration-of-bhutanese/?fbclid=IwAR3UaHmd-jFKHzumQ7G_ PZxxemGA6nRYVQRR2NBEFeqyTCE-IPcQg-1nPIw. Lee, C. J. 2008. “Education in the Republic of Korea: Approaches, Achievements, and Current Challenges.“ In An African Exploration of the East Asian Education Experience, edited by B. Frederiksen and J. P. Tan. 155–217. Washington, DC: World Bank. Littlefield, E., J. Murduch, and S. Hashemi. 2003. “Is Microfinance an Effective Strategy to Reach the Millennium Development Goals?” Focus Note Series No. 24, CGAP-Consultative Group to Assist the Poor, Washington, DC. McKenzie, D. 2017. “How Effective Are Active Labor Market Policies in Developing Countries? A Critical Review of Recent Evidence.” World Bank Research Observer 32: 127–54. MoESD (Ministry of Education and Skills Development, Royal Government of Bhutan). 2022. “TVET Statistics of Bhutan.” Department of Workforce Planning and Skills Development, Thimphu. MoESD (Ministry of Education and Skills Development, Royal Government of Bhutan). 2023a. “Summary Report. Tracer Survey Results for Skills and Engagement Programs. An Assessment of Program Delivery and Employment Outcome for Skills and Engagement Programs Implemented between 2020 and 2022.” MoESD, Thimphu. MoESD (Ministry of Education and Skills Development, Royal Government of Bhutan). 2023b. “12th FYP HRD Masterplan Implementation Review and HRD Tracer Report.” MoESD, Thimphu. MoICE (Ministry of Industry, Commerce, and Employment, Royal Government of Bhutan). 2023a. “Startup Flagship Report for World Bank (2019–2023).” Department of Employment and Entrepreneurship, Thimphu. MoICE (Ministry of Industry, Commerce, and Employment, Royal Government of Bhutan). 2023b. “Strategy to Strengthen Employment Service Centers in Bhutan.” Unpublished draft strategy, MoICE, Thimphu. MoLHR (Ministry of Labour and Human Resources, Royal Government of Bhutan). 2020. “TVET Graduate Study Bhutan: Multi- Cohort Online Tracer Survey Report.“ Department of Technical Education, MoLHR, Thimphu. MoLHR (Ministry of Labour and Human Resources, Royal Government of Bhutan). 2021a. “Guideline for Youth Engagement and Livelihood Program.” MoLHR, Thimphu. Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 167 MoLHR (Ministry of Labour and Human Resources, Royal Government of Bhutan). 2021b. “Midterm Review of 12th FYP HRD Masterplan for the Economic Sectors (2018–23).“ MoLHR, Thimphu. National Council of Bhutan. 2015. “Report on the Review of National Employment Policy, Program and Strategies.” Social and Cultural Affairs Committee, Thimphu. OECD (Organisation for Economic Co-operation and Development). 2020. “Enhancing Training Opportunities in SMEs in Korea.“ OECD, Paris. OECD (Organisation for Economic Co-operation and Development) and EC (European Commission). 2021. “The Missing Entrepreneurs 2021: Policies for Inclusive Entrepreneurs and Self-Employment.“ OECD, Paris. Osmani, S. 2016. “Models of Microcredit Delivery and Social Norm.” Working Paper No. 50, Institute for Inclusive Finance and Development (InM), Dhaka, Bangladesh. Puerto, O. 2007. “Interventions to Support Young Workers in Latin America and the Caribbean: Regional Report for the Youth Employment Inventory.” Unpublished manuscript, World Bank, Washington, DC. Ra, Y., and H. S. Kang. 2012. “Vocational Training System for a Skilled Workforce.“ Korea Research Institute for Vocational Education and Training (KRIVET), Seoul. Rodriguez-Planas, N., and J. Benus. 2010. “Evaluating Active Labor Market Programs in Romania.“ Empirical Economics 38 (1): 65–84. Romero, J. M., and A. Kuddo. 2019. “Moving Forward with ALMPs: Active Labor Policy and the Changing Nature of Labor Markets.“ Social Protection and Jobs Discussion Paper No. 1936, World Bank, Washington, DC. Santini, M., T. T. Tran, and A. Beath. 2017. “Investment Climate Assessment of Bhutan: Removing Constraints to Private Sector Development to Enable the Creation of More and Better Jobs.” World Bank, Washington, DC. Srinivas, H. 2015. “Socio-economic Dimensions of Women Entrepreneurs through SHGs: An Empirical Study.” International Journal of Entrepreneurship and Business Development Perspectives, 4(2), 1688-93. UNDESA (United Nations Department of Economic and Social Affairs). 2022. “World Population Prospects.” https://population. un.org/wpp/. UNDP (United Nations Development Programme). 2022a. “Addressing the Youth Unemployment Challenge in Bhutan through a Systemic Portfolio-Based Approach.” https://www.undp.org/bhutan/publications/addressing-youth-unemployment-challenge- bhutan-through-systemic-portfolio-approach-0. UNDP (United Nations Development Programme). 2022b. “Bhutan Workforce Futures: Steppingstones to Industrial Strategic Propositions for the Agriculture, Creative, and Digital Sectors.” https://www.undp.org/bhutan/publications/ bhutan-workforce-futures-stepping-stones-industrial-strategic-propositions-agriculture-creative-and-digital-sectors. UNESCO (United Nations Educational, Scientific and Cultural Organization)-UNEVOC International Centre for Technical and Vocational Education and Training. 2018. “TVET Country Profile. Republic of Korea.“ https://unesdoc.unesco.org/ark:/48223/ pf0000373097. WFP (World Food Programme). 2022. “WFP Strategic Plan 2022–2025.” https://www.wfp.org/publications/ wfp-strategic-plan-2022-25. World Bank. 2015. “Bhutan Enterprise Survey: Country Highlights.” https://www.enterprisesurveys.org/content/dam/enterprisesur- veys/documents/country-highlights/Bhutan-2015.pdf. World Bank. 2020a. “Doing Business in Bhutan Reform Memorandum.” https://documents1.worldbank.org/curated/ en/851061593610680639/pdf/Doing-Business-in-Bhutan-Reform-Memorandum.pdf. World Bank. 2020b. “Human Capital Index Report.” Human Capital Project, World Bank, Washington, DC. https://www.worldbank. org/en/publication/human-capital. World Bank. 2021. “Towards a World-Class Labor Market Information System for Indonesia: An Assessment of the System Managed by the Indonesian Ministry of Manpower.” World Bank, Washington, DC. World Bank. Forthcoming a. Country Economic Memorandum: Bhutan. Washington, DC: World Bank. 168 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice World Bank. Forthcoming b. Bhutan Poverty and Equity Assessment. Washington, DC: World Bank. World Bank. Forthcoming c. Public Expenditure Review: Bhutan. Washington, DC: World Bank. World Bank and MoLHR (Ministry of Labor and Human Resources, Royal Government of Bhutan). 2016. Bhutan’s Labor Market: Toward Gainful Quality Employment for All. Washington, DC: World Bank. https://openknowledge.worldbank.org/entities/publication/ e5734b3d-af13-530e-a89d-77838d95398d. Zangpo, T. 2024. “DHI and Its Companies Become More Attractive.” Kuensel, January 8. https://kuenselonline.com/ dhi-and-its-companies-become-more-attractive/. Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice 169 170 Bhutan Labor Market Assessment Report: Social Protection and Jobs Global Practice