81332 A WORLD BANK STUDY Financial Management Information Systems and Open Budget Data DO GOVERNMENTS REPORT ON WHERE THE MONEY GOES? Cem Dener and Saw Young Min Financial Management Information Systems and Open Budget Data A WO R L D BA N K S T U DY Financial Management Information Systems and Open Budget Data Do Governments Report on Where the Money Goes? Washington, D.C. © 2013 The International Bank for Reconstruction and Development/The World Bank Washington DC 20433 1818 H Street NW,   Telephone: 202-473-1000; Internet: www.worldbank.org Some rights reserved 1 2 3 4 16 15 14 13 World Bank Studies are published to communicate the results of the Bank’s work to the development com- munity with the least possible delay. This volume is a product of the staff of the International Bank for Reconstruction and Development/The World Bank. The findings, interpretations, and conclusions expressed in this volume do not necessarily reflect the views of the Executive Directors of the World Bank or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of the World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Nothing herein shall constitute or be considered to be a limitation upon or waiver of the privileges and immunities of the World Bank, all of which are specifically reserved. Rights and Permissions This work is available under the Creative Commons Attribution 3.0 Unported license (CC BY 3.0) http:// creativecommons.org/licenses/by/3.0. 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: Dener, Cem, and Saw Young Min. 2013. Financial Management Information Systems and Open Budget Data: Do Governments Report on Where the Money Goes? Washington, DC. World Bank. 10.1596/978-1-4648-0083-2. License: Creative Commons Attribution CC BY 3.0 Translations—If you create a translation of this work, please add the following disclaimer along with the attribution: This translation was not created by the World Bank and should not be considered an official World Bank translation. The World Bank shall not be liable for any content or error in this translation. All queries on rights and licenses should be addressed to the Office of the Publisher, The World Bank, 1818 H Street NW, Washington, DC 20433, USA; fax: 202-522-2625; e-mail: pubrights@worldbank.org. ISBN (paper): 978-1-4648-0083-2 ISBN (electronic): 978-1-4648-0084-9 DOI: 10.1596/978-1-4648-0083-2 Cover image: The map (FMIS World Map) was reproduced by the World Bank Map Design Unit. Library of Congress Cataloging-in-Publication Data Financial management information systems and open budget data : do governments report on where the money goes?. pages cm Includes bibliographical references and index. ISBN 978-1-4648-0083-2 (alk. paper) — ISBN 978-1-4648-0084-9 (ebook) 1. Finance, Public—Management—Data processing. 2. Management information systems. 3. Transparency in government. I. World Bank. HJ9745.F56 2013 352.40285—dc23 2013031240 Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 Contents Prefacexi Acknowledgmentsxiii About the Authors xv Abbreviationsxvii Executive Summary xix Chapter 1 Introduction 1 Definitions3 Budget Transparency Instruments: Overview 6 Notes8 Chapter 2 Methodology9 Definition of Indicators 9 Data Collection 12 Data Analysis 14 Validation of Observations 14 Results Reporting 15 Preparation of Guidelines 15 Notes15 Chapter 3 Data17 Status of Government Practices in Publishing Open Budget Data 17 Existence of Dedicated Websites for Publishing Public Finance Data 19 Source and Reliability of Budget Data 22 Scope and Presentation Quality of Public Finance Information25 Contents and Regularity of PF Information 29 Informative Indicators 34 Characteristics of Underlying FMIS Solutions 50 Comparison of Findings with PEFA Indicators 53 Financial Management Information Systems and Open Budget Data  v http://dx.doi.org/10.1596/978-1-4648-0083-2 vi Contents Comparison with Open Budget Index 54 Comparison with Other Dimensions 56 Notes62 Chapter 4 Good Practices63 Timely Publication of PF Data through Dedicated Websites 65 Visibility of FMIS Solutions on the Web 75 Access to PF Information through Dynamic Query Options 85 Publishing Rich Set of Open Budget Data 92 Reliability of PF Information Published on the Web 98 Quality of Presentation and Interactivity 102 Effective Use of Open Budget Data 110 FMIS World Map 118 Note121 Chapter 5 Guidelines for Publishing Open Budget Data123 Availability of Timely and Comprehensive Budget Information124 Disclosure of Details About Underlying Information Systems124 Availability of User-defined (Dynamic) Query and Reporting Capabilities 125 Publication of Reliable and Interlinked Open Budget Data 126 Authentication of the Sources of Public Finance Data 127 Improving the Quality of Presentation 127 Promoting the Effective Use of Open Budget Data 128 Notes129 Chapter 6 Conclusions131 What You See Is (Not Always) What You Get 131 Findings131 Concluding Remarks 136 Appendix A Explanation for Indicators/Questions and   Response Options137 Appendix B Description of the FMIS & OBD Data Set147 Appendix C Overview of Fiscal Transparency Instruments151 Introduction and Objectives 151 Surveys and Indices 151 Standards and Norms 156 Fiscal Transparency Initiatives 157 Appendix D Feedback Providers163 Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 Contents vii Bibliography165 Open Data References 167 Box 1.1 Trust in Government 5 Figures 1.1 Potential effects of publishing reliable open budget data   from FMIS 2 1.2 Core FMIS functions and interfaces with other PFM systems 4 1.3 Key components of integrated FMIS solutions combining   OLTP and OLAP 5 2.1 Composition of the FMIS & OBD data set 13 3.1 Current status of government practices in publishing   OBD from FMIS 19 3.2 Income level and regional distributions for Indicator 1 20 3.3 Income level and regional distributions for Indicator 2 21 3.4 Income level and regional distributions for Indicator 3 23 3.5 Income level and regional distributions for Indicator 4 23 3.6 Income level and regional distributions for Indicator 5 24 3.7 Income level and regional distributions for Indicator 6 25 3.8 Income level and regional distributions for Indicator 7 26 3.9 Income level and regional distributions for Indicator 8 27 3.10 Income level and regional distributions for Indicator 9 28 3.11 Income level and regional distributions for Indicator 10 29 3.12 Income level and regional distributions for Indicators 11 and 12 30 3.13 Income level and regional distributions for Indicators 13 and 14 31 3.14 Income level and regional distributions for Indicators 15 and 16 32 3.15 Income level and regional distributions for Indicators 17 and 18 33 3.16 Income level and regional distributions for Indicators 19 and 20 34 3.17 Income level and regional distributions for Indicator 21 35 3.18 Income level and regional distributions for Indicator 22 35 3.19 Income level and regional distributions for Indicator 23 36 3.20 Income level and regional distributions for Indicator 24 37 3.21 Income level and regional distributions for Indicator 25 38 3.22 Income level and regional distributions for Indicator 26 38 3.23 Income level and regional distributions for Indicator 27 39 3.24 Income level and regional distributions for Indicator 28 40 3.25 Income level and regional distributions for Indicator 29 41 3.26 Income level and regional distributions for Indicator 30 42 3.27 Income level and regional distributions for Indicator 31 42 3.28 Income level and regional distributions for Indicator 32 43 Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 viii Contents 3.29 Income level and regional distributions for Indicator 33 44 3.30 Income level and regional distributions for Indicator 34 45 3.31 Income level and regional distributions for Indicator 35 46 3.32 Income level and regional distributions for Indicator 36 47 3.33 Income level and regional distributions for Indicator 37 47 3.34 Income level and regional distributions for Indicator 38 48 3.35 Income level and regional distributions for Indicator 39 49 3.36 PFM topology supported by FMIS solutions 50 3.37 Functionality of FMIS solutions 51 3.38 Operational status of FMIS solutions 51 3.39 Scope of FMIS solutions 52 3.40 FMIS application software solutions 52 3.41 FMIS technology architecture 53 3.42 Comparison of PEFA scores with FMIS & OBD scores   (65 economies) 54 3.43 Comparison of PEFA scores with FMIS & OBD scores   (121 economies) 55 3.44 Comparison of OBI scores with FMIS & OBD scores   (100 economies) 56 5.1 Publishing timely and reliable open budget data from   FMIS solutions 128 6.1 Source and scope of public finance data 134 B.1 FMIS & Open Budget Data (OBD) data set 147 Maps 4.1 Basic information and web links displayed on   FMIS World Map 119 4.2 FMIS World Map (as of June 2013) 120 Tables 1.1 A summary of fiscal transparency instruments 7 2.1 Key and informative indicators and points assigned 10 2.2 Definition of FMIS & OBD groups to map the status   of government practices 14 3.1 A summary of current government practices in publishing   open budget data 18 3.2 Indicator 40 > Distribution of native and other   language options 49 3.3 Comparison of FMIS & OBD groups with OBI 2012   scores in 100 countries 56 3.4 Comparison of FMIS & OBD groups with OGP in   60 countries 57 Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 Contents ix 3.5 Comparison of FMIS & OBD groups with MTEF   implementation in 181 economies 57 3.6 Comparison of FMIS & OBD groups with e-Gov’12   rankings and indices 58 3.7 Comparison of FMIS & OBD groups with e-Gov’12   index (Online Services) 59 3.8 Comparison of FMIS & OBD groups with e-Gov’12   index (Telco Infrastructure) 59 3.9 Comparison of FMIS & OBD groups with e-Gov’12   index (Human Capital) 60 3.10 Scope and regional distribution of open source policies 61 3.11 Regional distribution of approved open source initiatives 61 3.12 Distribution of FMIS & OBD groups in 66 economies   with open source policies 61 4.1 Good practices in publishing PF information/open budget data 64 4.2 Historical trends and regularity in publishing PF information 65 A.1 Key indicators 138 A.2 Informative indicators 143 B.1 Basic data 148 B.2 FMIS characteristics 148 B.3 Other indicators 149 B.4 Results 150 C.1 Overview of fiscal transparency instruments 159 D.1 Feedback providers 163 Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 Preface Financial Management Information Systems and Open Budget Data: Do Governments Report on Where the Money Goes? is a World Bank Study, initiated in 2012 after an extended stocktaking exercise, to explore the effects of financial management information systems (FMIS) on publishing reliable open budget data, and to identify potential improvements in budget transparency. A rich data set was cre- ated by visiting the government public finance (PF) websites in 198 economies, and collecting evidence on the use of 176 FMIS platforms in publishing open budget data. This study is not intended to develop another index or ranking on budget transparency. The scope is limited to the budget data disclosed by the govern- ments on the web for the details of budget revenues and expenditures, as well as the results achieved. Other important aspects of fiscal discipline and transpar- ency, related to a wide range of extrabudgetary funds, assets, contingent liabilities, and quasi-fiscal operations were not possible to detect through such an external review. A number of key indicators linked with disclosing budget data were defined and measured using a simple scoring scheme. The main findings are explained in several categories to highlight the impor- tant aspects of publishing reliable and meaningful open budget data, and present some of the good practices. Guidelines for publishing reliable open data from FMIS solutions are presented to share options for improving budget transparen- cy. Finally, web links to relevant PF sites and FMIS platforms are presented through the FMIS World Map. Target Audience World Bank teams, government officials, and other specialists involved in FMIS and Open Budget Data projects. Objective Governments’ disclosure of PF information from reliable FMIS databases can improve transparency and accountability, if the data posted on their websites are accurate, easily accessible, and meaningful to citizens. This study reports on the Financial Management Information Systems and Open Budget Data   xi http://dx.doi.org/10.1596/978-1-4648-0083-2 xii Preface availability, source, reliability, and integrity of the budget data published from FMIS, identifies good practices, and provides guidelines on publishing reliable open budget data to assist in exploring the effects of FMIS on budget transpar- ency. Activities Oct 2012 Initiation of activity (P143587—Effects of FMIS on Budget Transparency). Nov 2012 Concept review (approved on November 28, 2012). Jan 2013 Data collection for a new data set on FMIS and Open Budget Data (scanning the PF-related websites in 198 economies) was completed (as an extended stocktaking exercise, building on an earlier database developed in mid-2012). Jan 2013 FMIS World Map beta version was updated (the first release posted in June 2010) to present the web links to 176 FMIS platforms from 198 economies on Google Maps. Apr 2013 Initial findings and data set were shared with government officials and task teams/managers for possible improvements in the data set and results. Jun 2013 Decision meeting (June 19, 2013). Jul 2013 Final report was delivered as a World Bank Study. Key Resources Hyperlinks (Uniform Resource Locators—URLs) to related web sources are shown as underlined text. • The FMIS and Open Budget Data—data set (July 2013 version). Available from the FMIS Community of Practice website (https://eteam.worldbank. org/FMIS). • Cem Dener, Joanna A. Watkins, and William L. Dorotinsky, Financial Manage- ment Information Systems: 25 Years of World Bank Experience on What Works and What Doesn’t, World Bank Study, April 2011. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 Acknowledgments This paper was prepared by the Governance and Public Sector Management Practice (PRMPS) of the World Bank’s Poverty Reduction and Economic Management (PREM) Network. The principal authors of this study, Cem Dener and Saw Young (Sandy) Min (PRMPS), would like to acknowledge the World Bank staff involved in the prepa- ration of this document: Birgül Meta for her valuable contributions during the preparation of the new data set for assessing the impact of FMIS on budget trans- parency; Angela Lisulo for her support during the early stages of data collection; and Jeffrey N. Lecksell, Cartographer, the World Bank Map Design Unit, for reproducing the map on the cover page. Special thanks to the PEFA Secretariat for their valuable support with the comparative analysis of our findings with the complete set of PEFA indicators, and to Joanna A. Watkins for her useful comments on the definitions and key findings. We also benefited from the practical suggestions and guidance of Jim Brumby, Young Kyu Kang, Adrian Fozzard, and Nick Manning in developing this study, as well as the comments of Vivek Srivastava, Verena Maria Fritz, Anupama Dokeniya, Maritza A. Rodriguez, Gert Van Der Linde, Leif Jensen, Paolo de Renzio, and Amparo Ballivian on the final report. Finally, we would like to thank Stephen Knack and Ivor Beazley (concept note stage), as well as Nicola J. Smithers, Vivek Ramkumar (OBI/IBP), Xavier Rame (IMF), and Marijn Verhoeven, for their invaluable peer review comments. The short-term consultancy services related to this study were funded from the Korean Trust Fund. Financial Management Information Systems and Open Budget Data   xiii http://dx.doi.org/10.1596/978-1-4648-0083-2 About the Authors Cem Dener is a senior public sector specialist within the Governance and Public Sector Management prac- tice of the Poverty Reduction and Economic Management network. Since 2004, he has been work- ing on the improvement of public financial manage- ment practices, and the development of Financial Management Information System (FMIS) solutions in all regions. He is one of the authors of the World Bank’s FMIS Study (April 2011). He is also involved in the organization of regional knowledge sharing and learning events and the formation of communities of practice. He received BSME from METU, Ankara, Turkey (1982), MSc from Cranfield Institute of Technology, Bedford, United Kingdom (1985), and PhD from Vrije Universiteit Brussel, Belgium (1992), working on modeling complex systems. Saw Young (Sandy) Min is a Junior Professional Officer with the Governance and Public Sector Management practice of the Poverty Reduction and Economic Management network. She is focusing on fiscal/budget transparency and public financial man- agement and is a core member of the Global Initiative for Fiscal Transparency. She co-authored a case study on the public participation in the budget process in the Republic of Korea and contributed to the publication, Beyond the Annual Budget: Global Experience with Medium Term Expenditure Frameworks. She holds an MPA from Cornell University and is currently pursuing a PhD in Public Finance at George Mason University. Financial Management Information Systems and Open Budget Data   xv http://dx.doi.org/10.1596/978-1-4648-0083-2 Abbreviations AFR Africa Region APEC Asia-Pacific Economic Co-operation BC Budget classification CoA Chart of accounts COFOG Classification of the Functions of Government (UN) COTS Commercial off-the-shelf software CSV  Comma-Separated Values (file storing tabular data in plain-text form) DOC Microsoft Word document format EAP East Asia and Pacific Region ECA Europe and Central Asia Region EGDI e-Government Development Index e-Gov Electronic Government (e-Government) FMIS Financial management information system(s) FT Fiscal transparency GFSM Government Finance Statistics Manual 2001 (IMF) GIFT Global Initiative for Fiscal Transparency HIC High-income country IBP International Budget Partnership ICT Information and communication technology IFMIS Integrated financial management information system KML Keyhole Markup Language LCR Latin America and Caribbean Region LDSW Locally developed software LIC Low-income country LMIC Lower-middle-income country MNA Middle East and North Africa Region MoF Ministry of Finance MTBF Medium-term budgeting framework Financial Management Information Systems and Open Budget Data   xvii http://dx.doi.org/10.1596/978-1-4648-0083-2 xviii Abbreviations MTEF Medium-term expenditure framework MTFF Medium-term fiscal framework MTPF Medium-term performance framework OBD Open budget data OBI Open Budget Index OBS Open Budget Survey ODF Open Document Format (an open standard since 2006) OECD Organisation for Economic Co-operation and Development OGP Open Government Partnership OLAP Online analytical processing OLTP Online transaction processing OSS Open source software PDF Portable Document Format (an open standard since 2008) PEFA Public Expenditure and Financial Accountability PF Public finance PFM Public financial management PPT Microsoft PowerPoint presentation file PREM Poverty Reduction and Economic Management Network PRMPS PREM Governance and Public Sector Management Practice RDF Resource Description Framework SAR South Asia Region TS Treasury system UMIC Upper-middle-income country WWW World Wide Web XLS Microsoft Excel file format for storing tabular data XML Extensible Markup Language (an open standard since 1998) Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 Executive Summary Financial Management Information Systems and Open Budget Data Do Governments Report on Where the Money Goes? In recent years, the topics of budget transparency and open data have been increasingly discussed. Most discussants agree that for true transparency, it is important not only that governments publish budget data on websites, but that the data they disclose are meaningful and provide a full picture of their financial activities to the public. Most governments have made substantial investments in capacity building and technology for the development of financial management information systems (FMIS). The question is, how much of the disclosed infor- mation and documents are reliable? What is the scope of disclosed information? Is there any reliable information about important aspects of fiscal discipline and transparency? Civil society groups and international organizations have developed a number of fiscal transparency instruments and guidelines to evaluate the existence, regu- larity, and contents of certain key budget documents published in the public domain and assess whether the information complies with international stan- dards. However, these instruments do not concentrate on the source and reli- ability of published information, or on the integrity of underlying systems and databases from which governments extract data. Guidance on publishing reliable open budget data from underlying FMIS solutions is scarce. This study is the first attempt to explore the effects of FMIS on publishing open budget data, identify potential improvements in budget transparency, and provide some guidance on the effective use of FMIS platforms to publish open budget data. If the public finance information published on government websites is to be reliable, relevant budget data should ideally be obtained from dependable FMIS platforms and should comply with open data standards.1 Financial Management Information Systems and Open Budget Data   xix http://dx.doi.org/10.1596/978-1-4648-0083-2 xx Executive Summary Five key research questions guided this study: 1. What are the important characteristics of current government web publishing platforms designed for the disclosure of budget data? 2. Is there any evidence on the reliability of open budget data published from FMIS? 3. Are there good practices demonstrating the effects of open budget data from FMIS in improving budget transparency? 4. Why is a “single version of the truth” difficult to achieve in the budget do- main? 5. Can there be some guidelines to improve practices in publishing reliable open budget data from FMIS? The conceptual framework used in this study has several important bases: ▶ Several decades of experience in the development of FMIS solutions in all regions. ▶ Evidence that the reliability and accuracy of government budget data depends on the capabilities and integrity of underlying FMIS platforms. ▶ Existence of proven industry standards for publishing open government data. ▶ Growing demand from citizens for improved budget transparency, account- ability, and participation. ▶ Widespread use of the Internet and web technologies for transforming the public sector management. The study is designed to draw the attention of governments to possible improvements in the accuracy, timeliness, and reliability of budget reporting, simply by publishing on public f­inance websites open budget data that are drawn from underlying FMIS platforms. Methodology In line with the research questions of this study, 20 key and 20 informative indi- cators were identified to assess the status of government websites for publishing open budget data from FMIS. A rich data set was created by visiting the public finance (PF) websites (mainly those of Finance Ministries or Departments) in 198 economies, and using these indicators to collect evidence on the use of 176 FMIS solutions in publishing open budget data. For the purposes of this study, the team focused only on websites that present information on the governments’ budget operations. A score was assigned for each of the key indicators. Using the total scores from the 20 key indicators, the researchers categorized the websites in four groups according to the good practices observed: A=Highly visible; B=Visible; C=Limited visibility; D=Minimal visibility. A survey form was Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 Executive Summary xxi used to share the initial findings with relevant government officials for their vali- dation and feedback on possible improvements (through e-mail exchanges). The results were presented through a comparative analysis of regional and income- level patterns, and through correlation with relevant budget transparency indices. The team identified the government PF websites that displayed good practices and innovative solutions, and created an interactive geospatial map to share important results broadly in a user-friendly format. Guidelines for publishing open budget data from FMIS were also developed to help governments and prac- titioners improve their websites and open data practices. Main Findings Despite the widespread availability of 176 FMIS platforms used by 198 govern- ments around the world, good practices in presenting open budget data from reliable FMIS solutions are highly visible in only 24 countries (12 %). The average score for the performance of 198 governments in publishing open budget data from FMIS is 45.1 out of 100, based on the 20 key indicators. About 93 websites (presenting extensive or significant information) appear to be bene- fiting from underlying information systems while publishing PF data, but most of these do not yet provide open data. Overall, there are only 48 countries (24 %) where civil society and citizens have the opportunity to benefit from PF information published on the web (Citizens Budget and transparency portals) to monitor the budget and hold their governments accountable. In many countries, external audit organizations do not appear to be using the FMIS platforms effectively for monitoring the govern- ment’s financial activities or auditing the budget results. Governments in high- and middle-income economies publish budget data dynamically in various formats, mainly from centralized systems, while many lower-income economies tend to publish static budget data, mostly through docu- ments posted on PF websites. By Region, ECA, LCR, SAR, and EAP appear to perform better in terms of posting budget reports from databases, while MNA and AFR tend to publish static budget reports through PDF files. A large number of governments in AFR do not have PF publication websites. To verify whether the findings of the study are consistent with key observa- tions from other fiscal transparency indices, the distribution of FMIS & OBD scores was compared with such fiscal transparency instruments as Public Expenditure and Financial Accountability (PEFA), Open Budget Index (OBI), and UN e-Government Development Rankings. It was found that the patterns are largely similar, and the FMIS & OBD scores correlate positively with the PEFA indicators and OBI scores. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 xxii Executive Summary The researchers identified 100 cases from various government websites in 53 countries (from all Regions and income levels) to highlight some of the good practices in different areas of publishing open budget data from FMIS. Guidelines Drawing on the observations of this study, the lessons learned from good practice cases, and experience in the development of FMIS solutions and open budget data portals, the team developed a set of guiding principles for government offi- cials, citizens and civil society groups, and oversight agencies to use in improving government practices for publishing open budget data through FMIS platforms. ▶ Availability of timely and comprehensive budget information There should be dedicated government PF websites that provide timely and regular information on budget plans and execution results. The completeness of published PF information (including off-budget fiscal and quasi-fiscal operations, as well as assets and contingent liabilities), and the presentation of budget execu- tion performance through time-series data, are very important. ▶ Disclosure of details about underlying information systems Government PF websites should present the key features of underlying infor- mation systems, promoting the use of interoperability standards and digital signa- ture, and disclosing data protection and information security policies to build confidence in underlying information systems and relevant ICT practices. ▶ Availability of user-defined (dynamic) query and reporting capabilities Government PF websites should have capabilities for interactive multidimen- sional data analysis with flexible and user-friendly dynamic query and reporting options, and the consistency of historical data should be ensured. ▶ Publishing reliable and interlinked open budget data Publishing open budget data (free, online, editable) from FMIS or data ware- house solutions often requires a change in the culture of organizations. Governments can benefit from the various guidelines on publishing “linked open data” to maximize the benefits. The use of open budget data also creates oppor- tunities to add value to public information. ▶ Authentication of the sources of public finance data Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 Executive Summary xxiii Inclusion of the system name and a date/time stamp on published reports is one of the key indicators for the reliability and integrity of underlying informa- tion systems. Appropriate safeguards should be implemented to protect data from unauthorized modification and access, and oversight mechanisms should be in place to ensure the reliability and integrity of systems, the security of opera- tions, and the effectiveness of IT governance and oversight functions. ▶ Improving the quality of presentation Interactive data visualization options, graphical user interfaces, feedback mechanisms, advanced search/reporting options, innovative tools (searchable interactive maps of PF information), broadened access to PF data through mobile applications, and the provision of daily updates on key performance indicators all substantially improve the quality of presentation in PF websites. ▶ Promoting the effective use of open budget data Publishing meaningful open data on budget revenues, spending, and other financial activities is crucial for any government to explain how the public money has been spent. The Citizens Budget is an important instrument to achieve this. Open budget data portals can also be used to support the participatory budgeting process, in which people in a locality or community can jointly decide on priori- ties in the government’s budget and monitor implementation. The results of participatory budgeting, gender focus, or citizen-led expenditure monitoring should be visible in PF websites. Finally, the oversight agencies should benefit from PF web platforms and underlying systems as much as possible, for effective monitoring and assessment of the government’s financial activities. Conclusions So, can we see where the money goes? The study shows that only a small group of countries provide good access to reliable open budget data from underlying FMIS solutions. Many governments publish substantial information on their PF websites, but the contents are (not always) meaningful to provide adequate answers to the question, “Where does the money go?” Therefore, the main conclu- sion of this study is that when it comes to government PF websites, what you see Selected cases demonstrate that even in difficult settings, innovative solu- tions to publish open budget data and improve budget transparency can be developed rapidly, with a modest investment, if there is commitment from the government and strong interest from the public. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 xxiv Executive Summary is (not always) what you get. Many governments need to make additional efforts that will build confidence in the budget data they disclose. As citizens and civil society increasingly demand access to open data about all financial activities, gov- ernments around the world are trying to respond to this democratic pressure. The outputs of this study are expected to provide a comprehensive view of the status of government practices for publishing budget data around the world, and to promote debates around the improvement of PF web publishing platforms to support transparency, accountability, and participation by disclosing reliable information about all financial activities. Future research could explore such important aspects as capturing and posting additional data on other financial activities, and learning more about user perceptions. Note 1. Public finance information covers all public sector revenues, expenditures, assets and liabilities. Budget data includes mainly general government revenues and expenditures. Dependable FMIS platforms are subject to regular IT audits to ensure the reliability and integrity of systems, the security of operations, and the effectiveness of IT governance and oversight functions. Open data are accessible to the public (online) in editable (machine-readable) and reusable format, without any restriction (free/legally open). Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 CHAPTER 1 Introduction Governments around the world are at various stages of implementing public financial management (PFM) reforms, designed to improve the strategic alloca- tion of resources (to promote growth and reduce poverty), operational efficiency (to minimize waste and align spending with revenues), and fiscal discipline (to improve the credibility of the budget). As part of this effort, most have made substantial investments in capacity building and technology for the development of financial management information systems (FMIS). Within the last decade, the use of FMIS has become a critical part of improv- ing budget transparency. Disclosure of public finance (PF) information to citi- zens through FMIS platforms can improve transparency, if the published budget data are accurate, easily accessible, and meaningful. Fiscal transparency in turn can improve trust in government, if the public interprets the motives for pub- lishing the open budget data positively and the transparency is maintained for long periods. However, designing robust FMIS solutions to capture all financial activities and publish open budget data, and measuring the effects of FMIS on budget transparency, continue to be major challenges. There is no widely accepted framework for assessing the quality of web pub- lishing or the performance of the information systems that are used as a basis for recording and reporting open budget data. Core PFM diagnostic studies1 typi- cally analyze government resource allocations within and among sectors, and assess the equity, efficiency, and effectiveness of those allocations in the context of the country’s macroeconomic framework and sector priorities. However, the indicators defined for these assessments are inadequate to check such aspects as the source and reliability of information. If the PF information published on government websites is to be reliable, relevant budget data should ideally be obtained from dependable FMIS platforms and should comply with open data standards. Financial Management Information Systems and Open Budget Data  1 http://dx.doi.org/10.1596/978-1-4648-0083-2 2 Introduction While most governments have FMIS that are capable of providing useful bud- get data, it is not clear to what extent the budget data are published on dedicated websites, with dynamic links to reliable systems for consistent and timely disclo- sure of information in easy-to-understand and machine-readable formats. Also, the scope, completeness, and quality of the PF information vary considerably, and there seems to be no widely used guidance on improving the reliability and reus- ability of budget data. This study is designed to shed light on these less known areas, and to report the findings, together with a geospatial mapping of the results, to explore the effects of FMIS on publishing open data. Although it is well known that the use of open budget data can contribute to improving accountability and transparency, reducing corruption, and enhancing citizens’ trust in government (see Figure 1.1), it is beyond the scope of this study to look at these aspects. This exercise addresses several key questions: 1. What are the important characteristics of current government web publishing platforms designed for the disclosure of budget data? 2. Is there any evidence on the reliability of open budget data published from FMIS? 3. Are there good practices demonstrating the effects of open budget data from FMIS in improving budget transparency? 4. Why is a “single version of the truth” difficult to achieve in the budget domain? 5. Can there be some guidelines to improve practices in publishing reliable open budget data from FMIS? This study is divided into six chapters. Chapter 1 covers the rationale and aims of the study, and definitions used, along with a summary of relevant budget trans- parency indices and standards. Chapter 2 explains the methodology used to iden- tify the important aspects of disclosing open budget data, as well as the simple Figure 1.1  Potential effects of publishing reliable open budget data from FMIS the he study Scope of th d FMIS: Publish open budget Potential improvements in Building trust in government’s • Record all data through FMIS: budget transparency by posting: financial activities by: revenue and • online • reliable • sustaining transparency expenditure • editable/reusable • timely • demonstrating meaningful transactions • searchable • accurate results daily • free/legally open • meaningful • publishing open budget data public finance information from dependable FMIS Source: World Bank data. Note: FMIS = financial management information system. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 Introduction 3 scoring scheme used to categorize current government practices in publishing open budget data from FMIS. Chapter 3 presents the data collected from 198 economies and describes general patterns observed in publishing budget data. Chapter 4 high- lights some of the good practices in publishing reliable and meaningful budget data through FMIS, and provides a geospatial mapping of the results. Chapter 5 presents the guidelines suggested for possible improvements in government practices to publish open budget data through FMIS platforms. Chapter 6 summarizes the key findings and presents the conclusions of this study. Appendixes A–C amplify the information in the study, and Appendix D lists the feedback providers for the study. Bibliography and open data references are presented at the end of the study. Definitions Broadly speaking, financial management information systems are the automation solutions that enable governments to plan, execute, and monitor the budget. Modern FMIS platforms help governments comply with financial regulations and reporting standards, and support decentralized budget operations through ­ centralized web-based information and communication technology (ICT) solu- tions. FMIS platforms also facilitate the disclosure of PF information to citizens to improve budget transparency, government accountability, and participation. Figure 1.2 illustrates the core FMIS functions and their interrelationships. Whenever FMIS and other PFM information systems (for example, e-Procure- ment, payroll, debt management) are linked with a central data warehouse (DW) to record and report all daily financial transactions, offering reliable consolidated results for budget analysis, decision support, performance monitoring, and web publishing, these platforms can be referred to as integrated FMIS (or IFMIS). IFMIS solutions are rare in practice, and to avoid unrealistic expectations, the term should not be used as a synonym for core FMIS functionality. Next-generation IFMIS solutions combine PFM operational systems for online transaction processing (OLTP) with powerful DW capabilities for multidimen- sional online analytical processing (OLAP) to assist in effective forecasting, plan- ning, performance monitoring, and decision support (see Figure 1.3). Innovative IFMIS solutions also allow more detailed analysis by providing dynamic query options to a large number of users, both internal (public organizations) and external (citizens, nongovernmental organizations, businesses), and they support the publication of open budget data. For the purposes of this study, open budget data (OBD) is defined as the gov- ernment budget data that are made accessible to the public (online) in editable (machine-readable) and reusable format, without any restriction (free/legally open). Requirements to protect the confidentiality of personal or classified infor- mation should be considered while posting open budget data. Public finance (PF) information includes the budget data plus other compo- nents of the government’s financial activities (for example, extrabudgetary funds, tax expenditures, quasi-fiscal activities, fixed assets, contingent liabilities). This study is designed to capture evidence on the disclosure of budget data only. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 4 Introduction Figure 1.2  Core FMIS functions and interfaces with other PFM systems Policy development Budget and review  preparation B FMIS:  F = B + T (+ O)  Public investments Audit and  Mgmt of budget evaluation authorizations T Core treasury  system DW Fiscal reports &  Commitment budget review FMIS DB of funds  O Web portal / DW General ledger  Procurement  Publish + monitor accounting Daily & purchasing  operations Debt and aid  Payments and Tax and  O management Treasury Receipts Mgmt customs single Asset /  account inventory Market operations Cash HR mgmt/payroll +  / securities  management pensions Source: World Bank data. Note: For the purposes of this report, FMIS (F) is defined narrowly to include mainly core budget preparation (B) and treasury/budget execution (T) systems, complemented by other (O) modules in some cases. Arrows are used to indicate the linkages between core modules (blue), the interfaces with other systems (gray), as well as the links with policy development and review processes (red). The core FMIS functions and their contributions to PFM practices are more fully explained in Dener, Watkins, and Dorotinsky, 2011. FMIS DB = financial management information system database; DW = data warehouse; HR = human resources. Fiscal transparency (FT) is defined as the ready availability of meaningful infor- mation on fiscal policy and achievements to the public. Budget transparency refers to the full disclosure of budget data on government revenues, allocations, and expenditures (ideally for the whole public sector). Trust in government is defined as the public’s overall assessment of the govern- ment’s current entitlement to enforce its policy decisions, laws, and regulations based on past performance and the view of how the government and its institu- tions are likely to act in the future (Box 1.1). An IFMIS can be considered as a complex system—that is, a network of het- erogeneous components that interact nonlinearly, to give rise to emergent behav- ior; emergence is the way complex systems and patterns arise out of a multiplic- ity of relatively simple interactions. Such information systems exhibit “organized complexity,” and the main challenge is to integrate a limited number of inter- linked PFM functions through a centralized web-based platform that supports countrywide decentralized operations, and provides innovative tools for decision support, performance monitoring, and web publishing. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 Introduction 5 Figure 1.3  Key components of integrated FMIS solutions combining OLTP and OLAP Integrated financial management information system Operational systems Public sector institutions Web portal - intranet Plan M&E Execute FMIS PFM  database Audit Account OLTP E Report FMIS : collaboration and coordination  T Interfaces OLAP L DW : monthly updates and dynamic query options  BI BI Web portal − internet BI Open budget Data DM data warehouse DM • Consolidation DM • Slice and dice  • Drill‐down  Multi-dimensional analytical queries Citizens, NGOs, businesses  Images: jscreationzs/FreeDigitalPhotos.net Source: World Bank data. Note: OLTP = online transaction processing; OLAP = online analytical processing; FMIS = Financial Management Information System; PFM = Public Financial Management; DW = Data Warehouse; NGO = Nongovernmental organization; ETL = extract, transform, load; BI = business intelligence; and DM = data mining. Box 1.1  Trust in Government Trust in government is a multidimensional concept in which citizens expect the government and public officials to be responsive, honest, and competent, “even in the absence of constant scrutiny.”a Trends in trust have been closely examined by recent studies, often with a sense of alarm regarding a potential long-term decline of confidence in governments. However, little work has been done to understand and empirically measure the drivers of trust in public institutions. As Manning, Shepherd, and Guerrero (2010) highlighted in a working paper, while the con- cept of trust in government is clearly important, there are some major definitional problems and associated questions about the strength of any metrics that can be used to capture it. Terms such as performance, trust, legitimacy, and trustworthiness are often used interchange- ably, or at least with meanings that are very specific to the particular situation. Various mea- sures of trust in government that result from surveys are often unclear about the unit of analysis (what is being trusted?) and whether respondents understood trust or confidence in the same way as the interviewers. The assumption made in the working paper is that views of government performance draw on assessments of the past; views of the trustworthiness of public institutions require an estimate about the future; and views on trust and legitimacy box continues next page Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 6 Introduction Box 1.1  Trust in Government (continued)    draw on both of these and are about a current assessment of government—against general and specific criteria respectively. Considering these definitions, this study assesses current government web publishing practices related to the disclosure of PF information for improving budget transparency and accountability. The use of FMIS solutions to publish reliable and meaningful open budget data was analyzed from different perspectives using relevant indicators, and substantial evi- dence was collected about the linkages between published data and underlying information systems, as well as the platforms designed to improve budget transparency. The study does not assume that FMIS has a direct influence on trust in government. The assumption is that if governments meet the minimum requirements for publishing reliable, timely, accurate, and meaningful open budget data from FMIS, and for promoting citizen participation, these good practices may help improve budget transparency and accountabil- ity, and thus contribute to building trust in governments’ financial activities. a. Miller and Listhaug 1990, cited in Manning, Shepherd, and Guerrero 2010, 358. Open source is an approach to the design, development, and free distribution of software, offering practical accessibility to source code. In the context of FMIS modernization/development and open source solutions, innovations refer to improvements on existing ways of doing things, whereas inventions change the way things are done. Open data/content licensing schemes and open data cata- logues, which can be used for publishing open data and knowledge, are visible in many government websites.2 Budget Transparency Instruments: Overview In recent years, the discussion of data openness has been gaining momentum on the global stage. For example, a number of countries and organizations are engaged in discussions of the challenge of collecting and disclosing timely and reliable information about budget operations, extrabudgetary funds, and quasi- fiscal activities.3 The International Budget Partnership (IBP) is one of the largest forums for these discussions. The World Bank has adopted a proactive and committed stance in the discus- sion of data openness by participating in the process, as well as advocating for it in collaboration with other development partners. In April 2010, the Bank made its development data available for download free of charge.4 The Open Development Technology Alliance5 (also known as the ICT Knowledge Platform) was created to enhance accountability and improve the delivery and quality of public services through technology-enabled citizen engagement (for example, using mobile phones, interactive mapping, and social media). The discussions on open government data, and more specifically on budget transparency, are of particular interest to the Governance and Public Sector Management Practice (PRMPS) of the World Bank’s Poverty Reduction and Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 Introduction 7 Economic Management (PREM) Network. The World Bank is one of the inter- national financial institutions taking the lead in the Global Initiative for Fiscal Transparency (GIFT), an initiative that promotes budget transparency, public participation, and accountability globally.6 BOOST is another useful tool developed by the World Bank for transforming detailed government expenditure data from FMIS databases into an easy-to-under- stand data set (XLS) for detailed analysis through pivot tables and geomapping tools. Expenditure data can be combined with information on public institutions, service delivery, and households to facilitate rigorous expenditure analysis. A number of fiscal transparency instruments have been developed within the last decade (Appendix C provides an overview). These instruments can be grouped in three categories according to their functions (Table 1.1). Most of these instruments are designed to evaluate the existence, regularity, and contents of certain key budget documents published in the public domain, as well as the mechanisms for public access. However, they do not examine the source, reliability, quality, and readability of the PF information that governments publish governments on the web—a major gap. Another important challenge is the difficulty in comparing open budget data published by governments in different formats. In February 2012, the IBP pub- lished the Open Budget Survey 2012,7 which revealed that most governments do not meet basic standards of transparency and accountability with their national budgets. The signing of an Open Data Charter by G-8 leaders on June 18, 2013, is an important development to promote transparency, innovation, and accountability. The charter sets out five strategic principles that all G8 members have endorsed: an expectation that governments will publish data openly by default, along with Table 1.1  A summary of fiscal transparency instruments # Fiscal Transparency Instruments # of countries Since Last update A. Surveys and indices  1 Open Budget Index (OBI) 100 2006 2012  2 PEFA PFM Assessment 121 (public: 65) 2005 —  3 IMF Fiscal Transparency ROSC 93 1999 —  4 Global Integrity Index 119 2004 —  5 Right to Information Index 93 2011 2011  6 UN e-Government Survey and Rankings 193 2003 2012 B. Standards and norms  7 IMF Code of Good Practices on FT n.a. 1998 2007  8 OECD Best Practices for Budget Transparency n.a. 2002 — C. Fiscal transparency initiatives  9 Global Initiative for Fiscal Transparency (GIFT) n.a. 2011 — 10 Open Government Partnership (OGP) 55 (+5) 2011 — Source: World Bank data. Note: Data retrieved in June 2013. PEFA = Public Expenditure and Financial Accountability; PFM = Public Financial Management; IMF = International Monetary Fund; FT = Fiscal Transparency; OECD = Organisation for Economic Co-operation and Development; — = not available; n.a. = not applicable. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 8 Introduction principles to increase the quality, quantity, and reuse of the data that are released. G8 members also identified 14 high-value areas (including “Finance and Contracts”) in which they will release data. Each member of the G8 is expected to publish an open data action plan by October 2013, showing how it will make more data available, in line with the charter and its principles. Additionally, the Lough Erne Declaration from the G8 Summit sets out agreed principles for the future, of which one is related to the open data: “Governments should publish information on laws, budgets, spending, national statistics, elections and govern- ment contracts in a way that is easy to read and reuse, so that citizens can hold them to account.” These developments in the area of open data highlight the importance of ensuring the reliability of published data and the integrity of the underlying systems. Good practice cases have shown that the success of open government data projects relies heavily on strong political commitment, skills development, tech- nology platforms, and resources, as well as on demand from citizens and civil society. Although there have been encouraging developments and some successes in this area, there are still challenges that need to be overcome if the public is to reap the full benefits of open data. Notes 1. Public Expenditure Reviews; Public Expenditure and Financial Accountability (PEFA) Assessments. 2. The Open Definition sets out principles to define “openness” in relation to data and content. Creative Commons licenses enable the sharing and use of creativity and knowledge through free legal tools. Data Catalogs website presents a comprehensive list of open data catalogs in the world. 3. http://www.data.gov/opendatasites. 4. The World Bank Open Data portal: http://data.worldbank.org 5. Open Development Technology Alliance is a joint initiative anchored by the World Bank Institute and the ICT Sector Unit and supported by other Bank Networks/ Regions. 6. IMF’s Fiscal Transparency website includes relevant links: http://www.imf.org/exter- nal/np/fad/trans. 7. Open Budget Survey 2012, www.openbudgetindex.org. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 CHAPTER 2 Methodology I II III IV V VI Definition of Data Data Validation of Results Preparation indicators collection analysis observations reporting of guidelines To measure and analyze the effects of financial management information systems (FMIS) on publishing reliable open budget data, the team used a six-part approach: Definition of Indicators The first step was to identify the key indicators (questions) for collecting data about the characteristics of current web publishing platforms, and the metrics (points) for measuring government practices for publishing open budget data from FMIS. The indicators were defined in two categories for assessing the effects of FMIS on publishing open budget data: • Key indicators: 20 indicators were derived from 10 factual questions about the source, scope, and reliability of open budget data. • Informative indicators: 20 indicators were derived from 10 questions provid- ing useful information about other important features. Each indicator is linked with one of the subcomponents of the questions listed below. The details of all questions, subcomponents, and indicators are provided in Appendix A. The questions corresponding to each indicator are also shown in Table 2.1 (for Example, I-1 is linked with question Q1.3). Financial Management Information Systems and Open Budget Data  9 http://dx.doi.org/10.1596/978-1-4648-0083-2 10 Methodology Table 2.1  Key and informative indicators and points assigned Ind Q Key indicators Points Existence of dedicated website for publishing PF data I-1 Q1.3 Is there a dedicated website for publishing PF information? 0/2 I-2 Q2.1 Is there a website/document about the FMIS platform? 0/2 Source and reliability of open budget data I-3 Q3.1 What is the source of PF data? 0/3 I-4 Q3.3 Presence of open budget data (online, editable/reusable, free) 0/1 I-5 Q3.5 Is system name visible in dynamic/static reports? 0/1 I-6 Q3.6 Is system time stamp visible in dynamic/static reports? 0/1 Scope and presentation of PF information I-7 Q4.1 Quality: What is the quality of PF data presentation? 0/2 I-8 Q4.2 Content: Is there a sufficient level of detail? 0/1 I-9 Q4.3 Are the budget results presented easy to understand (Citizens Budget)? 0/2 I-10 Q5.1 Are the BC/CoA details published? 0/1 Contents and regularity of PF information I-11 Q6.1 Is the approved annual budget published? 0/1 I-12 Q6.2 If yes: Regularity of publishing annual budget plans 0/1 I-13 Q7.1 Are MTEF documents published? 0/1 I-14 Q7.2 If yes: Regularity of publishing MTEF plans 0/1 I-15 Q8.1 Are public investment plans published? 0/1 I-16 Q8.2 If yes: Regularity of publishing investment plans 0/1 I-17 Q9.1 Are budget execution reports published? 0/1 I-18 Q9.3 If yes: Regularity of publishing budget execution reports 0/1 I-19 Q10.1 Is the external audit of central government budget operations published? 0/1 I-20 Q10.2 If yes: Regularity of publishing external audit reports 0/1 Inf Q Informative indicators Points I-21 Q11.1 Public expenditures > Consolidated budget reports published? 0/1 I-22 Q11.2 Public expenditures > Sector analysis published? 0/1 I-23 Q11.3 Public expenditures > Regional analysis published? 0/1 I-24 Q11.4 Public expenditures > Gender analysis published? 0/1 I-25 Q11.5 Public expenditures > Analysis of spending for children & youth published? 0/1 I-26 Q11.6 Public expenditures > Debt data published? 0/1 I-27 Q11.7 Public expenditures > Foreign aid/grants published? 0/1 I-28 Q11.8 Public expenditures > Fiscal data on state/local governments published? 0/1 I-29 Q11.9 Public expenditures > Financial statements published? 0/1 I-30 Q11.10 Public expenditures > Public procurement and contracts published? 0/1 I-31 Q12.1 Is there an open government/open budget website? 0/1 I-32 Q13.1 Does FMIS support the PFM needs of state/local governments? 0/2 I-33 Q14.1 Is there a harmonized public accounting system for all budget levels? 0/1 I-34 Q15.1 Are PF data published on the Statistics website? Or another website? 0/3 I-35 Q16.1 Is access to information explained? 0/1 I-36 Q17.1 Are PFM roles and responsibilities clearly explained? 0/1 I-37 Q18.1 Compliance with specific international reporting standards? 0/1 table continues next page Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 Methodology 11 Table 2.1  Key and informative indicators and points assigned (continued) Ind Q Key indicators Points I-38 Q19.1 Web statistics (reports on website traffic)? 0/1 I-39 Q19.2 Which platforms are available for providing feedback? 0/3 I-40 Q20.1 What languages are used to publish PF information? — Source: World Bank data. Note: FMIS = Financial Management Information System; PF = Public Finance; BC = Budget Classification; CoA= Chart of Accounts;  MTEF =­   Medium-Term Expenditure Framework; PFM = Public Finance Management; — = not available. Questions for Key Indicators Q1. Does the Finance Ministry/Department have a website or portal that is dedicated to publishing public finance (PF) information? Q2. Is there a website or document describing the web-based FMIS platform? Q3. What is the source of the PF information that is published on the web? Q4. Is the PF information meaningful to citizens or budget entities? Q5. Is the data structure or full listing of budget classification (BC)/chart of accounts (CoA) published? Q6. Are documents associated with annual budget plans published? Q7. Are documents associated with medium-term expenditure framework (MTEF)1 published? Q8. Are documents associated with public investment/capital budget plans2 published? Q9. Are documents associated with budget execution published? Q10. Are documents associated with the external audit of central government budget operations published? Questions for Informative Indicators Q11. What is the level of detail of the public expenditure/revenue information published online (plans versus actuals, sectoral or regional details, and so on)? Q12. Is there a dedicated website for open government/open budget initiatives? Q13. Is there a web-based application supporting the Public Finance Manage- ment (PFM) needs of state/provincial governments or municipalities as a part of the FMIS? Q14. Is there a harmonized public accounting system supporting all budget levels (unified budget classification and/or chart of accounts)? Q15. Are there duplicate government budget reporting websites other than that of the Finance Ministry/Department (for example, Office of Statistics)? Q16. Is there a web page explaining the policy/regulations for access to PF in- formation, web publishing standards, or frequency of PF reporting? Q17. Is there a web page with links to regulations3 for clarifying the PFM roles and responsibilities? Q18. Are published PF data compliant with the IMF Government Finance Sta- tistics (GFS)and/or UN Classification of the Functions of Government (COFOG) standards?4 Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 12 Methodology Q19. Is there a web page for receiving feedback on PF information/user satis- faction, or for presenting web statistics? Q20. What languages are used to publish the PF information online for exter- nal viewers? The metrics (points) linked with the 20 key indicators were used to measure the current status of government practices for publishing PF data (Table 2.1). Points were linked with the response options (ranging from 0 to 3) for each indicator. The sum of all points (Σ Point) was normalized to obtain a total score (from 0 to 100), reflecting the strength of each country’s platforms for publish- ing open budget data from FMIS or other sources: Σ1 20 Point 1 × 100 = Score Σ 20 Max Point Most of the key indicators (31 of 40) were measured using a simple point scheme (0 or 1) to ensure the collection of consistent responses by all reviewers and avoid ambiguities in interpreting related evidence. Informative indicators were included to complement the data set by providing additional feedback on ease of access to information, compliance with some of the widely used PF reporting standards, and presentation options. The total score was used to identify the group of each country, for a compara- tive analysis of current government practices and the clarification of regional and income level patterns, as explained under the “data analysis” section. Data Collection To explore the effects of FMIS on budget transparency, the team5 created a compre- hensive data set (worksheet with linked data) by reviewing government PF websites (usually maintained by the Ministry of Finance or other relevant public entities) and collecting evidence (web links/URLs and relevant documents/reposts) about the source and reliability of PF information published through FMIS or other platforms. Responses to all questions were obtained directly from the websites by three reviewers. Almost all indicators/questions are designed to measure important characteristics of web publishing platforms (source, reliability, dynamic query options, and so on), and the patterns for publishing PF data (scope, frequency, regularity, and so on), using evidence visible on the websites. The FMIS and Open Budget Data (FMIS & OBD) data set contains five com- ponents for capturing a number of additional fields related to basic data, FMIS solutions, and other relevant indicators, in addition to the key and informative indicators (scores and evidence). Additional information collected about the FMIS solutions (scope, years in operation, type of application software, technol- ogy architecture, and so on) and relevant web links are included to provide more comprehensive feedback on existing systems. The composition of the data set is summarized in Figure 2.1. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 Methodology 13 Figure 2.1  Composition of the FMIS & OBD data set • Name of economy + income level Basic data • Population + GNI + GNI per capita Key • 20 key indicators derived from 10 questions indicators • Points to measure key indicators Informative • 20 informative indicators derived from 10 questions indicators • Points to measure informative indicators • System name (original + abbreviated) FMIS • Topology + operational status + functional scope characteristics • Type of ASW + technology architecture • Selected PEFA indicators Other • OBI scores + e-Gov ranking + MTEF status indicators • Open source policies and regulations + other aspects • FMIS & ODB practice groups Results • Total Points + scores + groups Source: World Bank data. Note: PEFA = Public Expenditure and Financial Accountability; MTEF = Medium-term expenditure framework; OBI = Open budget Index; FMIS = Financial management information system. All key and informative indicators are explained in Appendix A, and all other data set components are described in Appendix B. Most of the questions are designed to measure observable facts about the contents of relevant PF publication websites, and the reviewers used a well- coordinated approach to check the existence of various features consistently. Only Q4 partially reflects the opinions of the reviewers, based on specific criteria for “quality” and “content.” These questions were included as a measure of the initial perception of first-time visitors to the PF websites about the presentation of budget data and the level of detail visible at first sight. Thus, questions Q4.1 (Quality) and Q4.2 (Content) reflect the assessments of external reviewers of whatever is visible on the home page of the main PF publication sites, and Q4.3 highlights the scope of publications related with the Citizens Budget. Obviously, because these responses are based on the initial perceptions of three reviewers, there is a risk of missing website features that may not be so obvious at first glance. However, since the reviewers visited all PF-related websites from the 198 economies at least twice, they believe that the collected evidence and scores will provide useful feedback on the status of relevant govern- ment practices. To overcome the difficulties of screening a large number of web pages prepared in national languages, the reviewers used the Google Chrome web browser and available machine translation options while visiting government PF websites. In Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 14 Methodology most cases the quality of translation was good enough to understand the contents of the website, check the source and reliability of published PF information, and view the contents of reports/data with sufficient detail. Nevertheless, the team may have missed some important details because machine translation options were not avail- able for several languages. In such cases, the reviewers tried to reach relevant country officials and project teams involved in PFM reforms, to clarify various aspects of PF web publishing sites and their linkage with the underlying systems. Eventually, most of the important PF related websites were screened with a sufficient level of under- standing to allow the team to draw some conclusions and identify patterns. Data Analysis After calculating total scores (from 0 to 100) that reflect the status of government practices for publishing open budget data from FMIS or other sources, the team mapped the sites into four groups (A to D) to distinguish good country practices and highlight innovative solutions for publishing open budget data from FMIS (see Table 2.2). A detailed analysis of the data collected from 198 economies, and of general patterns observed in publishing budget data is presented in chapter 3. Validation of Observations The findings of this stocktaking exercise were shared with relevant government officials to check the evidence collected, reflect other perspectives, and improve the accuracy of observations. Country-specific survey forms were automatically created from the data set to share the initial findings, and relevant government officials were invited to provide feedback. Responses received from 43 govern- ments were used to improve the data set and collected evidence. To identify and promote exemplary country systems and good practices support- ing budget transparency, the team also selectively contacted particular countries to learn more about government practices with regard to publishing budget data on PF websites, and to clarify the reliability and integrity of underlying FMIS databases. Table 2.2  Definition of FMIS & OBD groups to map the status of government practices Group Score Current status Description of relevant government practices A 75–100 Highly visible Extensive information; FMIS is used to publish timely open data; easy to navigate; dynamic query options. B 50–74 Visible Significant information; FMIS rarely used to publish open data; static web pages updated regularly. C 25–49 Limited visibility Some information; ongoing activities to improve the web content or publish open budget data on the web. D 0–24 Minimal visibility Minimum or no information; no Finance Ministry/Department website or budget data. Source: World Bank data. Note: FMIS = Financial Management Information System. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 Methodology 15 Results Reporting The team then recorded their observations and findings about each key indicator (see chapter 3). To verify whether the findings of the study are consistent with key observations from other fiscal transparency indices, the distribution of FMIS & OBD scores was compared with such fiscal transparency instruments as Public Expenditure and Financial Accountability (PEFA), Open Budget Index (OBI), and UN e-Government Development Rankings. Finally, the team developed an overview of some of the good practices in publishing reliable and meaningful budget data, and a geospatial mapping of the results (FMIS World Map) on Google Maps (see chapter 4). The FMIS World Map will be updated annually to ensure the visibility of the findings and provide easy and open public access to good practices. Preparation of Guidelines As the last step in this study, the team prepared some guidelines to highlight the important aspects for improving the reliability and integrity of PF information sources and the presentation and quality of budget data published. These guide- lines are intended to help governments improve how they publish PF informa- tion, and to encourage those that show little or no visibility of PF information on the web by laying out some of the achievable good practices. Chapter 5 presents these guidelines. Notes 1. The MTEF consists of three stages: (1) medium-term fiscal framework (MTFF); (2) medium-term budget framework (MTBF); and (3) medium-term performance frame- work (MTPF). 2. This indicator is used to highlight the publication of investment plans published separately (not embedded in the annual budget or the national development plan/ strategy). 3. Online availability of regulations, organic budget law, procurement law,and so on. 4. IMF GFS: International Monetary Fund Government Finance Statistics. UN COFOG: United Nations Classification of the Functions of Government. 5. The three website reviewers were Ms. Saw Young (Sandy) Min, Ms. Birgül Meta, and Mr. Cem Dener. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 CHAPTER 3 Data This chapter presents the descriptive data analysis based on a rich data set created by visiting the government websites in 198 economies, and collecting evidence on the use of 176 financial management information system (FMIS) platforms in publishing budget data. Status of Government Practices in Publishing Open Budget Data The current status of government practices in publishing open budget data is presented in four groups, derived from the scores calculated through 20 key indi- cators (Table 3.1). Open budget data from reliable FMIS solutions are highly visible in only about 24 economies out of 198 reviewed (12 %). In most cases, the lack of timely and meaningful budget data may be an indi- cation of ineffective budget monitoring or greater opportunities for the misuse of funds. Substantial improvements in budget transparency could be achieved simply by publishing reliable open budget data from FMIS or other databases on existing government websites, if there is political will and commitment. Web publishing practices vary significantly among the different Regions of the World Bank and among countries of different income levels (see Figure 3.1). Among the 198 public finance (PF) websites assessed, the average score assigned by the team was 45.1 out of 100. About 69 PF websites (35 %) provide signifi- cant budget information, but only a small portion of this information qualifies as open budget data from FMIS. Forty-five governments (23 %) provide minimal or no budget information on the web; and 60 governments (30 %) provide some information, mostly from archived documents without enough evidence on the use of FMIS databases as the source of the PF data. Financial Management Information Systems and Open Budget Data   17 http://dx.doi.org/10.1596/978-1-4648-0083-2 18 Data Table 3.1  A summary of current government practices in publishing open budget data Group    Good practices in disclosing OBD from FMIS Economies % E Regions %R A Highly visible  (extensive information) 24 12 % 16 10 % Argentina; Australia; Brazil; Colombia; Ecuador; El Salvador; Germany; Guatemala; India; Ireland; Rep. of Korea; Mexico; The Netherlands; New Zealand; Nicaragua; Paraguay; Peru; Russian Federation; Singapore; Slovenia; Spain; Turkey; United Kingdom; United States B Visible  (significant information) 69 35 % 57 34 % Afghanistan; Albania; Armenia; Austria; Bahrain; Bangladesh; Belgium; Bhutan; Bolivia; Bosnia and Herze- govina; Bulgaria; Canada; Cape Verde; Chile; China; Croatia; Czech Republic; Denmark; Dominican Republic; Estonia; Finland; France; Gabon; Georgia; Ghana; Honduras; Hong Kong SAR; Iceland; Indonesia; Italy; Japan; Jordan; Kenya; Kyrgyz Republic; Latvia; Lebanon; Lithuania; Macedonia; FYR; Madagascar; Malaysia; Malta; Mauritius; Moldova; Morocco; Namibia; Nepal; Norway; Pakistan; Philippines; Poland; Portugal; Romania; Serbia; Slovak Republic; Solomon Islands; South Africa; Sri Lanka; Sweden; Switzerland; Tanzania; Thailand; Timor-Leste; Tonga; Uganda; Ukraine; Uruguay; Venezuela; RB; Vietnam; Zambia C Limited visibility  (some information) 60 30 % 56 33 % Algeria; Andorra; Angola; Antigua and Barbuda; Azerbaijan; Bahamas; The; Belarus; Botswana; Burkina Faso; Cambodia; Taiwan, China; Costa Rica; Côte d’Ivoire; Cyprus; Djibouti; Dominica; Egypt; E ­ thiopia; Fiji; Gambia; Greece; Grenada; Guinea-Bissau; Guyana; Haiti; Hungary; Iraq; Israel; Jamaica; K ­ azakhstan; Kosovo; Lao PDR; Lesotho; Liberia; Luxembourg; Macao SAR; Malawi; Maldives; Mauritania; M ­ icronesia; Mongolia; M­ ozambique; Nigeria; Oman; Panama; Papua New Guinea; Rwanda; Samoa; Saudi A ­ rabia; Senegal; Sierra Leone; St. Lucia; ­ mirates; West Bank and Gaza; Yemen; Swaziland; Tajikistan; Trinidad and Tobago; Tunisia; United Arab E ­Zimbabwe D Minimal visibility  (minimal or no information) 45 23 % 39 23 % Barbados; Belize; Benin; Brunei Darussalam; Burundi; Cameroon; Central African Republic; Chad; Comoros; Congo; Congo; Dem. Rep.; Cuba; Equatorial Guinea; Eritrea; Guinea; Iran; Islamic Rep.,; Kiribati; Korea Dem. Rep.; Kuwait; Libya; Liechtenstein; Mali; Marshall Islands; Monaco; Yugoslavia; former; Myanmar; Nauru; Niger; Palau; Qatar; San Marino; São Tomé and Principe; Seychelles; Somalia; South Sudan; St. Kitts and Nevis; St. Vincent and the Grenadines; Sudan; Suriname; Syrian Arab Republic; Togo; Turkmenistan; Tuvalu; Uzbekistan; Vanuatu   Totals > 198 168 Source: World Bank data. Note: The 198 “economies” include all 188 of the World Bank member countries, plus some of the large economies (from European Union [EU], Organisation for Economic Co-operation and Development [OECD], and Asia-Pacific Economic Co-operation [APEC] members). The “Regions” include 168 World Bank member countries that are currently receiving advisory and financial support to implement public sector management reforms. Income level. Although high-income countries (HICs) tend to publish bud- get data regularly (32 out of 54, or 59 %, are in group A or B), most of these economies do not present evidence about the source of open budget data on their websites. Most upper-middle-income (UMICs) and lower-middle-income countries (LMICs) follow similar patterns. Most of the low-income countries (LICs) have limited or no visibility in terms of publishing budget data (29 out of 38, or 76 %, are in group C or D). Of the 35 fragile states, 29 (83 %) are in group C or D, with little or no visibility on the web in terms of publishing PF information. Regional distribution. Among the Regions, the Europe and Central Asia (ECA), Latin America and the Caribbean (LCR), East Asia and the Pacific (EAP), and South Asia (SAR) countries show a large number of good practices in publishing extensive/substantial information (45–50 % of countries in these Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 Data 19 Figure 3.1  Current status of government practices in publishing OBD from FMIS a. Income level distribution b. Regional distribution 25 20 20 15 Economies Economies 15 10 10 5 5 0 0 A B C D A B C D Grade Grade LIC LMIC UMIC HIC AFR EAP ECA LCR MNA SAR Source: World Bank data. Note: LIC = Low-Income Country; LMIC = Lower-Middle-Income Country; UMIC = Upper-Middle-Income Country; HIC = High-Income Country. AFR = Africa Region, EAP = East Asia and Pacific Region; ECA = Europe and Central Asia Region; LCR = Latin America and Caribbean Region; MNA = Middle East and North Africa Region; SAR = South Asia Region. regions are in group A or B). There are only a few good practice examples in the Africa (AFR) and Middle East and North Africa (MNA) Regions (under 25 % of countries). Most AFR and MNA countries have little or no visibility on the web in terms of publishing budget data, and a large proportion are in group C or D. The EU member states (23 out of 27) and OECD members (31 out of 34) present extensive or significant budget data on the web, but again only a few pres- ent evidence on the source and reliability of open budget data (6 in EU, 11 in OECD). Similarly, APEC member economies (18 out of 21 in group A or B) have well-designed publishing sites, but only 7 publish open budget data from FMIS. The remainder of this chapter provides detailed findings from the data analy- sis, broken down by country income level, and region, in the following order: • Existence of dedicated websites for publishing PF data [ I-1, I-2 ] • Source and reliability of budget data [ I-3 to I-6 ] • Scope and presentation quality of PF information [ I-7 to I-10 ] • Contents and regularity of key PF information [ I-11 to I-20 ] • Informative indicators [ I-21 to I-40 ] • Characteristics of underlying FMIS solutions • Comparison of findings with Public Expenditure and Financial Accountability (PEFA) indicators • Comparison with Open Budget Index • Comparison with other dimensions. Existence of Dedicated Websites for Publishing Public Finance Data There are two indicators (I-1 and I-2) under this category, and the responses to relevant questions (derived from Q1 and Q2) are summarized below. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 20 Data I-1 Is there a dedicated public finance publication website? Points Responses Economies %E Regions %R 2 There is a dedicated PF website, and links to 125 63.1 102 60.7 budget-related publications/reports are clearly visible from the home page. 1 Budget data links are not clearly visible from 41 20.7 37 22.0 the home page or posted on separate sites (without a link with home page). 0 There is no dedicated website for publishing 32 16.2 29 17.3 PF information. Taking into account country differences in organizational structures and web practice, the team screened the relevant websites in the following order: • The Finance Ministry/Department (MoF) websites were screened first, to see if all questions can be answered from one dedicated source. • In addition to the MoF, Statistics, and Central Bank websites, other ministry/ agency web publishing platforms were also visited to capture remaining infor- mation not visible in the MoF websites (for example, investment plans, audit reports, procurement). Most of the governments (166 out of 198, or 83.8 %) have dedicated websites to publish PF data, and for 125 of these (63.1 %), a link to budget data is clearly visible from their home pages. Of the 32 governments (16.2 %) that have no PF website, most are LICs or LMICs (Figure 3.2). Most of the fragile states have dedicated websites (23 out of 35) but the scope of published PF information is limited. The pattern of Regional distribution is similar. Most of the countries (139 out of 168, or 82.7 %) have dedicated sites to publish PF information, and 102 Figure 3.2  Income level and regional distributions for Indicator 1 a. Income level distribution b. Regional distribution 50 30 40 25 Economies Economies 20 30 15 20 10 10 5 0 0 0 1 2 0 1 2 Point Point LIC LMIC UMIC HIC AFR EAP ECA LCR MNA SAR Source: World Bank data. Note: LIC = Low-Income Country; LMIC = Lower-Middle-Income Country; UMIC = Upper-Middle-Income Country; HIC = High-Income Country. AFR = Africa Region, EAP = East Asia and Pacific Region; ECA = Europe and Central Asia Region; LCR = Latin America and Caribbean Region; MNA = Middle East and North Africa Region; SAR = South Asia Region. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 Data 21 governments (60.7 %) provide easy access to relevant publications. Of the 29 (17.3 %) governments with no PF publication website, most are in AFR. All EU member states (27), OECD members (34), and APEC member economies (21) have dedicated websites to publish extensive or significant PF information. I-2 Is there a website/document about the FMIS solution? Points Responses Economies %E Regions %R 2 There is a specific website presenting the character- 92 46.5 74 44.1 istics of FMIS solutions, or the current status of FMIS implementation. 1 There is only published reference document(s) 83 41.9 77 45.8 about FMIS implementation. 0 There is no website or document about the FMIS 23 11.6 17 10.1 solution. Nearly half of the governments (92 out of 198, or 46.5 %) have websites that provide useful information about the status of the FMIS and describe the function- ality and technology architecture. However, the remaining 106 countries have little or no information about their FMIS: 83 economies publish some reports partially describing FMIS functionality and scope, but there is no information about the FMIS in 23 economies. This pattern characterizes all income levels (see Figure 3.3). Among the World Bank Regions, 74 countries have dedicated FMIS-related websites. LCR countries are the most informative, in terms of explaining their FMIS through comprehensive information presented in dedicated websites. AFR, ECA, EAP, and SAR follow, with a relatively large group of countries presenting useful information about their FMIS platforms. Most AFR and MNA countries have no information about their FMIS platforms on the web. Only 8 out of 35 fragile states (23 %) have dedicated FMIS websites. Figure 3.3  Income level and regional distributions for Indicator 2 a. Income level distribution b. Regional distribution 35 25 30 20 25 Economies Economies 20 15 15 10 10 5 5 0 0 0 1 2 0 1 2 Point Point LIC LMIC UMIC HIC AFR EAP ECA LCR MNA SAR Source: World Bank data. Note: LIC = Low-Income Country; LMIC = Lower-Middle-Income Country; UMIC = Upper-Middle-Income Country; HIC = High-Income Country. AFR = Africa Region, EAP = East Asia and Pacific Region; ECA = Europe and Central Asia Region; LCR = Latin America and Carribean Region; MNA = Middle East and North Africa Region; SAR = South Asia Region. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 22 Data Most of the EU (27) and OECD (34) members, and all APEC economies (21), have dedicated websites or documents describing their FMIS platforms. Seven HICs have no website for presenting budget data or FMIS platforms, and 11 UMICs/LMICs have little or no visibility on the web. Source and Reliability of Budget Data The results obtained from the four indicators in this category (I-3 to I-6) revealed that only a small group of countries present evidence about the source and reli- ability of the PF data published on their websites. Most of the information is not linked with the 176 FMIS platforms used by the 198 economies, and there is little focus on publishing open budget data (online, free, editable/reusable) from FMIS. I-3 What is the source of PF data? Points Responses Economies %E Regions %R 3 Dynamic website (linked with FMIS databases); interactive query 12 6.1 8 4.8 ­ options for reports (for example, CSV, XLS, ODF, XML, PDF). Argentina; Brazil; Colombia; Finland; Germany; Republic of Korea; Mexico; Peru; Russian Federation; Turkey; United Kingdom; ­ United States 2 Dynamic website (some linked with databases) to present data from a 22 11.1 17 10.1 predefined list of publications (mainly PDF, XLS). ­ eorgia; Bolivia; Chile; China; Denmark; Ecuador; El Salvador; Estonia; G India; Japan; Kyrgyz Republic; Malta; Netherlands; Nicaragua; ­ Paraguay; Portugal; Singapore; Spain; Sweden; Taiwan, China; ­Timor-Leste; Ukraine 1 Static website (not linked to databases) to publish data from unidenti- 132 66.7 114 67.9 fied sources (mainly in PDF format). 0 There is no published PF information. 32 16.1 29 17.2 Of the 34 governments that have dynamic websites (user-defined reports generated online), 12 (mostly HICs and UMICs) provide access to rich set of information through interactive queries, mostly linked with FMIS databases. A large number of countries (132 or 66.7 %; mostly HICs and MICs) maintain static websites presenting various documents from unidentified sources (Figure 3.4). LCR leads in terms of good practices (11 countries with 2 or 3 points), and ECA and EAP follow (6 countries each with 2 or 3 points). Other Regions pres- ent PF data mainly through static websites. AFR has the lowest level of visibility in presenting the source of PF data. Ten EU member states and 15 OECD members have dynamic websites, but most of the developed countries maintain static websites. Similarly, only 10 APEC economies have dynamic PF web publishing sites. Most of the fragile states (21 out of 35) have static websites. Timor-Leste and the Kyrgyz Republic are the only LICs with dynamic websites. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 Data 23 Figure 3.4  Income level and regional distributions for Indicator 3 a. Income level distribution b. Regional distribution 50 35 30 40 25 Economies Economies 30 20 20 15 10 10 5 0 0 0 1 2 3 0 1 2 3 Point Point LIC LMIC UMIC HIC AFR EAP ECA LCR MNA SAR Source: World Bank data. Note: LIC = Low-Income Country; LMIC = Lower-Middle-Income Country; UMIC = Upper-Middle-Income Country; HIC = High-Income Country. AFR = Africa Region, EAP = East Asia and Pacific Region; ECA = Europe and Central Asia Region; LCR = Latin America and Carribean Region; MNA = Middle East and North Africa Region; SAR = South Asia Region. I-4 Presence of open budget data (online, editable/reusable, free) Points Responses Economies %E Regions %R 1 Yes 52 26.3 38 22.6 0 No 146 73.7 130 77.4 Although many governments have FMIS databases and open government initia- tives, disclosure of open budget data is not a common practice. To check the extent to which open data are published, the team screened the formats of “public expen- diture reports,” together with other open data portals publishing budget reports. Open budget data are visible in 52 economies, but are linked with FMIS data- bases in only about half (based on the evidence about dynamic websites from I-3). Several LICs and LMICs (13 out of 51) publish open data (Figure 3.5). Figure 3.5  Income level and regional distributions for Indicator 4 a. Income level distribution b. Regional distribution 50 50 40 40 Economies Economies 30 30 20 20 10 10 0 0 0 1 0 1 Point Point LIC LMIC UMIC HIC AFR EAP ECA LCR MNA SAR Source: World Bank data. Note: LIC = Low-Income Country; LMIC = Lower-Middle-Income Country; UMIC = Upper-Middle-Income Country; HIC = High-Income Country. AFR = Africa Region, EAP = East Asia and Pacific Region; ECA = Europe and Central Asia Region; LCR = Latin America and Carribean Region; MNA = Middle East and North Africa Region; SAR = South Asia Region. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 24 Data LCR and ECA countries lead in publishing open budget data (25 combined), and countries in AFR post the least open budget data. Only 3 of the 35 fragile states can produce some reports using open budget data. Open budget data are visible in less than half of the developed countries (12 EU member states, 19 OECD members, and 10 APEC economies publish open budget data). I-5 Is system name visible in dynamic/static reports? Points Responses Economies %E Regions %R 1 Yes 18 9.1 16 9.5 0 No 180 90.9 152 90.5 It is rare to see the name of the underlying FMIS printed as a part of most frequently published “budget execution reports” in static or dynamic websites. Most countries present PF information through PDF files without any indication about the source of data. Only 18 countries include the name of FMIS solution (4 HICs, 8 UMICs, 5 LMICs, 1 LICs) as the source of published information (Figure 3.6). Among the Regions, relevant good practices are visible in nine LCR countries. Only one fragile state includes the system name in some budget execution reports. Most of the EU, OECD, and APEC economies do not publish the system name as a part of their regular budget reports (visible only in the reports published by three EU member states, four OECD members, and three APEC economies). I-6 Is system time stamp visible in dynamic/static reports? Points Responses Economies %E Regions %R 1 Yes 28 14.1 24 14.3 0 No 170 85.9 144 85.7 Figure 3.6  Income level and regional distributions for Indicator 5 a. Income level distribution b. Regional distribution 60 50 50 40 Economies Economies 40 30 30 20 20 10 10 0 0 0 1 0 1 Point Point LIC LMIC UMIC HIC AFR EAP ECA LCR MNA SAR Source: World Bank data. Note: LIC = Low-Income Country; LMIC = Lower-Middle-Income Country; UMIC = Upper-Middle-Income Country; HIC = High-Income Country. AFR = Africa Region, EAP = East Asia and Pacific Region; ECA = Europe and Central Asia Region; LCR = Latin America and Carribean Region; MNA = Middle East and North Africa Region; SAR = South Asia Region. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 Data 25 Figure 3.7  Income level and regional distributions for Indicator 6 a. Income level distribution b. Regional distribution 50 50 40 40 Economies Economies 30 30 20 20 10 10 0 0 0 1 0 1 Point Point LIC LMIC UMIC HIC AFR EAP ECA LCR MNA SAR Source: World Bank data. Note: LIC = Low-Income Country; LMIC = Lower-Middle-Income Country; UMIC = Upper-Middle-Income Country; HIC = High-Income Country. AFR = Africa Region, EAP = East Asia and Pacific Region; ECA = Europe and Central Asia Region; LCR = Latin America and Carribean Region; MNA = Middle East and North Africa Region; SAR = South Asia Region. The patterns for this indicator are similar to those for system name. Most economies do not print the system date/time from underlying FMIS solutions in frequently published “budget execution reports” or other documents visible on the websites. The 28 countries that do consist of 8 HICs, 10 UMICs, 9 LMICs, and 1 LIC (Figure 3.7). Again, there are 11 good practice cases in LCR countries, which have a better focus on budget transparency and accountability and a longer history of working with FMIS. Only one fragile state includes the system time in some budget execution reports. In most of the EU, OECD, and APEC economies the system time stamp is not printed as a part of frequent budget reports (visible only in five EU member states, five OECD members, and eight APEC economies). Scope and Presentation Quality of Public Finance Information The scope and presentation quality of PF data published in government websites were analyzed using four indicators (I-7 to I-10) derived from two questions (Q4 and Q5). In this section, it is important to note that the assessment of indicators I-7 and I-8 partially depends on the judgment of reviewers, since it is difficult to quantify quality and scope. These questions were included to gauge the percep- tion of first-time visitors about the presentation of budget data and the level of detail visible at first sight. The other indicators are more specific and were included to complement scope questions. I-7 Quality: What is the quality of PF data presentation? Points Responses Economies %E Regions %R 2 Good quality (presented reports are informative and easy to access 69 34.8 50 29.8 and read). 1 Partially acceptable (some of the published PF info is useful). 97 49.0 89 53.0 0 informative). Below desired level (most of the published PF info is not ­ 32 16.2 29 17.2 Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 26 Data Figure 3.8  Income level and regional distributions for Indicator 7 a. Income level distribution b. Regional distribution 35 30 30 25 25 Economies Economies 20 20 15 15 10 10 5 5 0 0 0 1 2 0 1 2 Point Point LIC LMIC UMIC HIC AFR EAP ECA LCR MNA SAR Source: World Bank data. Note: LIC = Low-Income Country; LMIC = Lower-Middle-Income Country; UMIC = Upper-Middle-Income Country; HIC = High-Income Country. AFR = Africa Region, EAP = East Asia and Pacific Region; ECA = Europe and Central Asia Region; LCR = Latin America and Carribean Region; MNA = Middle East and North Africa Region; SAR = South Asia Region. Of the 69 governments that provide comprehensive information about bud- get performance in easy-to-understand formats, 30 are HICs, 16 UMICs, 17 LMICs, and 6 LICs. However, the quality of reporting in a large portion of PF websites (49 %) is only partially acceptable. most of the reports are not very informative, or are too detailed, without clear instructions on how to interpret results. There are 32 websites with little or no attention to the quality of the presentation (Figure 3.8). In the Regions, a relatively small number of PF publishing sites (29.8 %) are well maintained, with regular updates on budget results. LCR stands out with the largest number of good-quality websites (15), followed by ECA (10), EAP (9), and AFR (7). About half of the Regional PF websites present some useful data, but may benefit from substantial improvements; and only 6 fragile states present informative and detailed reports. Most of the EU, OPEC, and APEC members maintain good-quality websites to publish timely budget results. I-8 Content: Is there sufficient level of detail? Points Responses Economies %E Regions %R 1 Yes 148 74.7 % 123 73.2 % 0 No 50 25.3 % 45 26.8 % This indicator serves as a rapid assessment of the extent of information pub- lished on websites. It is not designed to drill down into published reports and analyze the contents, since other indices (for example, open budget index [OBI], Report on Observance of Standards and Codes [ROSC]) review the contents of key publications. This indicator should be used with caution, since there is always a possibility of missing some of the features that may not be obvious at first glance. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 Data 27 Figure 3.9  Income level and regional distributions for Indicator 8 a. Income level distribution b. Regional distribution 50 30 40 25 Economies Economies 20 30 15 20 10 10 5 0 0 0 1 0 1 Point Point LIC LMIC UMIC HIC AFR EAP ECA LCR MNA SAR Source: World Bank data. Note: LIC = Low-Income Country; LMIC = Lower-Middle-Income Country; UMIC = Upper-Middle-Income Country; HIC = High-Income Country. AFR = Africa Region, EAP = East Asia and Pacific Region; ECA = Europe and Central Asia Region; LCR = Latin America and Carribean Region; MNA = Middle East and North Africa Region; SAR = South Asia Region. A large number of economies provide substantial information on various sub- jects, but 50 provide minimal or no information about the contents of publica- tions. The level of comprehensiveness increases from LICs to HICs. Almost all of the EU, OECD, and APEC member economies have comprehensive websites providing useful details about the budget performance. LCR has the highest number of comprehensive websites (28 out of 123), fol- lowed by ECA (27), EAP (21), AFR (27), MNA (13), and SAR (7). Many AFR countries (44 %) have inadequate web contents for sharing budget execution results, as do several EAP and MNA countries. More than half of the fragile states have little or no detail about the budget results on their websites (Figure 3.9). I-9 Are presented budget results easy to understand (Citizens Budget)? Points Responses Economies %E Regions %R 2 Yes (comprehensive information in meaningful 15 7.6 12 7.1 format for the citizens. highly interactive) 1 Yes (basic information about the budget cycle 33 16.7 26 15.5 and some results in meaningful format) 0 No (minimal or no information about Citizens 150 75.7 130 77.4 Budget) This indicator measures whether meaningful budget data are accessible online to citizens in a simple and easy-to-understand format. In many countries, such documents/web contents are referred to as the Citizens Budget, and are consid- ered as an important indicator of fiscal transparency. Obviously, the Citizens Budget should be credible and timely, and the source of information should be reliable. This study simply aims to locate such documents or websites for com- parative analysis, and to expose the contents; it is beyond the scope of this study to analyze such documents, and the impact of citizen feedback on budget plan- ning, in detail. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 28 Data Figure 3.10  Income level and regional distributions for Indicator 9 a. Income level distribution b. Regional distribution 50 50 40 40 Economies Economies 30 30 20 20 10 10 0 0 0 1 2 0 1 2 Point Point LIC LMIC UMIC HIC AFR EAP ECA LCR MNA SAR Source: World Bank data. Note: LIC = Low-Income Country; LMIC = Lower-Middle-Income Country; UMIC = Upper-Middle-Income Country; HIC = High-Income Country. AFR = Africa Region, EAP = East Asia and Pacific Region; ECA = Europe and Central Asia Region; LCR = Latin America and Carribean Region; MNA = Middle East and North Africa Region; SAR = South Asia Region. Only about 15 of the 198 governments provide interactive platforms for citi- zens’ access to meaningful budget data and feedback provision through dedi- cated Citizens Budget websites. Another 33 present useful information for the citizens on some of the important aspects of budget spending; however, most governments do not provide meaningful information to their citizens on budget results. These observations are valid for all income levels (see Figure 3.10). Of developed countries, only a relatively small number have interactive Citizens Budget websites—about 35–40 % of EU, OECD, and APEC members. There are about 12 Citizens Budget websites with an easy-to-understand graphical user interface, mainly in LCR and ECA. Other Regions have about 26 useful Citizens Budget websites presenting mainly static information (PDF files describing the budget performance in easy-to-understand format). However, most countries in all Regions, and particularly in AFR, do not have Citizens Budget websites. Only 3 of the 35 fragile states have useful Citizens Budget websites providing limited information. I-10 Are the details of budget classification/chart of accounts published? Points Responses Economies %E Regions %R 1 Yes 93 47.0 75 44.6 0 No 105 53.0 93 55.4 A budget classification (BC) with a sufficiently detailed segment structure is very important for the production of comprehensive and reliable PF data sets. A unified chart of accounts (CoA) is essential for consistently recording transac- tions and balances (flows and stocks) in the general ledger. BC and CoA are interrelated: it is advisable to design the BC as a subset of the CoA to ensure correspondence between entries recorded in each of these classifications and to Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 Data 29 Figure 3.11  Income level and regional distributions for Indicator 10 a. Income level distribution b. Regional distribution 35 35 30 30 25 25 Economies Economies 20 20 15 15 10 10 5 5 0 0 0 1 0 1 Point Point LIC LMIC UMIC HIC AFR EAP ECA LCR MNA SAR Source: World Bank data. Note: LIC = Low-Income Country; LMIC = Lower-Middle-Income Country; UMIC = Upper-Middle-Income Country; HIC = High-Income Country. AFR = Africa Region, EAP = East Asia and Pacific Region; ECA = Europe and Central Asia Region; LCR = Latin America and Carribean Region; MNA = Middle East and North Africa Region; SAR = South Asia Region. properly link budget and treasury accounts. Many countries use the same number of digits and subsegments for CoA and the Economic segment. This indicator was designed to check the level of detail presented for BC and CoA on PF websites, to clarify the key parameters used in recording and reporting PF data. Of the 198 governments, 93 provide the details of BC/CoA on their websites. About half of the HICs and MICs (50.6 %) publish the data structures (or some- times a full listing) of their BC/CoA (Figure 3.11), as do many EU members (21 out of 27), OPEC countries (27 out of 34), and APEC economies (14 out of 21). More than half of the countries in LCR, ECA, and SAR publish the BC/CoA details, but only 34 % of the countries in AFR, EAP, and MNA do so. Of the 35 fragile states, 10 publish the BC/CoA listings on their websites. These results may indicate that many economies do not pay enough attention to the design (or optimization) of their BC/CoA data structures to be able to capture maximum budget information with minimum number of digits, and to store a large number of transactions historically and retrieve data rapidly. Contents and Regularity of PF Information The 10 indicators (I-11 to I-20) in this category, derived from 5 questions (Q6 to Q10), measure the existence of key budget documents and the regularity of publication in selected categories. The frequency and regularity of key budget execution publications (planned/actual figures for all revenues and expenditures) were captured in I-17 and I-18 (Q9). Additionally, several other indicators (I-21 to I-30) were used to collect information about specific sections of public expen- diture reports published on the web. Some of these details (for example, sectoral analysis or subnational-level spending) are not published at the same frequency as the major spending reports. It was beyond the scope of this study to analyze or verify in depth the details of PF data in various reports. However, relevant web links are included to provide quick access to specific details for further analysis. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 30 Data I-11 Approved annual budget published? I-12 If yes: Regularity of publishing annual budget plans Points Responses Economies %E Regularity Regions %R Regularity 1 Yes 153 77.3 120 128 76.2 98 0 No 45 22.7 — 40 23.8 — Most of the HICs and MICs publish their approved annual budgets on the PF websites. About 82 % of the HICs and UMICs follow good practices in terms of the regularity of reporting—that is, publishing without interruption, at least within the last five years (levels are indicated by a line on each bar in Figure 3.12)—and present their approved budgets on the web before the upcom- ing budget year. Some 120 economies (78.4 %) publish the approved budget regularly, sometimes after the start of the budget year. Almost all EU, OECD, and APEC governments publish their approved annual budgets regularly. Most of the countries in the Regions (except AFR) publish their approved annual budget plans before the relevant budget year, and about 98 (76 %) of them do so regularly (except in AFR, where only 52 % publish these plans regu- larly). Of the 35 fragile states, 20 publish their approved budgets, and 13 of them do so regularly. I-13 Medium-Term Expenditure Framework documents published? I-14 If yes: Regularity of publishing MTEF plans Points Responses Economies %E Regularity Regions %R Regularity 1 Yes 103 52.0 64 81 48.2 44 0 No 95 48.0 — 87 51.8 — Figure 3.12  Income level and regional distributions for Indicators 11 and 12 a. Income level distribution b. Regional distribution 50 35 30 40 25 Economies Economies 30 20 R R 20 e 15 e g 10 g 10 5 0 0 0 1 0 1 Point Point LIC LMIC UMIC HIC AFR EAP ECA LCR MNA SAR Source: World Bank data. Note: LIC = Low-Income Country; LMIC = Lower-Middle-Income Country; UMIC = Upper-Middle-Income Country; HIC = High-Income Country. AFR = Africa Region, EAP = East Asia and Pacific Region; ECA = Europe and Central Asia Region; LCR = Latin America and Carribean Region; MNA = Middle East and North Africa Region; SAR = South Asia Region. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 Data 31 Figure 3.13  Income level and regional distributions for Indicators 13 and 14 a. Income level distribution b. Regional distribution 40 30 25 30 Economies Economies 20 20 R 15 e R g 10 10 e 5 g 0 0 0 1 0 1 Point Point LIC LMIC UMIC HIC AFR EAP ECA LCR MNA SAR Source: World Bank data. Note: LIC = Low-Income Country; LMIC = Lower-Middle-Income Country; UMIC = Upper-Middle-Income Country; HIC = High-Income Country. AFR = Africa Region, EAP = East Asia and Pacific Region; ECA = Europe and Central Asia Region; LCR = Latin America and Carribean Region; MNA = Middle East and North Africa Region; SAR = South Asia Region. About half of the HICs and MICs publish their multiyear plans, or medium- term expenditure framework (MTEF), and 64 regularly update these plans (revising them every year, at least within the last five years). In LMICs and LICs, the regularity of publishing MTEF information is relatively lower (Figure 3.13). About 80 % of the EU, OECD, and APEC economies publish and regularly update their MTEF plans. About 44 ECA, LCR, and SAR countries publish MTEF data regularly, but regular publication of MTEF data is much less common in AFR, MNA, and EAP countries. Of the 35 fragile states, 10 present some reports on MTEF plans, but most do not do so regularly. I-15 Public investment plans published? I-16 If yes: Regularity of publishing public investment plans Points Responses Economies %E Regularity Regions %R Regularity 1 Yes 44 22.2 32 38 22.6 27 0 No 154 77.8 — 130 77.4 — Only 44 governments publish public investment plans (Figure 3.14). However, in many cases (for example, Ukraine) the public investments are included as a part of the approved annual budgets and listed in MTEF docu- ments; hence this indicator may not accurately capture all public investment plans that are made available. Among these 44 economies, about 60 % of HICs and MICs publish investment plans regularly, as do similar proportions of EU, OECD, and APEC economies. A relatively small number of Regional governments (mainly in LCR and ECA) publish multiyear investment plans regularly, and most of the Regional countries do not publish their investment plans separately. A large number of Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 32 Data Figure 3.14  Income level and regional distributions for Indicators 15 and 16 a. Income level distribution b. Regional distribution 50 40 40 30 Economies Economies 30 20 20 10 10 R R 0 0 0 1 0 1 Point Point LIC LMIC UMIC HIC AFR EAP ECA LCR MNA SAR Source: World Bank data. Note: LIC = Low-Income Country; LMIC = Lower-Middle-Income Country; UMIC = Upper-Middle-Income Country; HIC = High-Income Country. AFR = Africa Region, EAP = East Asia and Pacific Region; ECA = Europe and Central Asia Region; LCR = Latin America and Carribean Region; MNA = Middle East and North Africa Region; SAR = South Asia Region. countries (except in AFR) present their investments within the approved annual budget plans; the actual level of publishing investment plans is estimated as 50 % in all Regions. Investment plans are published in a few of the fragile states (4 out of 35). I-17 Budget execution results published? I-18 If yes: Regularity of publishing budget execution results Points Responses Economies %E Regularity Regions %R Regularity 1 Yes 147 74.2 117 124 73.8 96 0 No 51 25.8 — 44 26.2 — Most of the economies (largely HICs and MICs) publish budget execution results at different intervals (Figure 3.15)—monthly in 50 of the 147, quarterly in 18, and annually in the remaining 78. Among these countries, about 80 % publish these results regularly. It appears that only a small group of countries publish budget performance with monthly updates, and a large portion present the results annually. There seems to be room for many countries to make sub- stantial improvements by using their existing FMIS to publish more frequently (monthly). A large number of the regional governments publish their budget execution results, most regularly. Many countries publish the budget performance reports (comparing plans versus actuals) annually. A small number of governments in LCR, ECA, EAP, and SAR benefit from their FMIS capabilities to publish the status of budget execution monthly. However, most countries in AFR, MNA, and other Regions publish annual results after the completion of the budget year, without any benefit for performance monitoring during the execution period. Therefore, there is a substantial opportunity cost in not disclosing the budget Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 Data 33 Figure 3.15  Income level and regional distributions for Indicators 17 and 18 a. Income level distribution b. Regional distribution 50 30 40 25 Economies Economies 20 30 R R e 15 20 e g 10 g 10 5 0 0 0 1 0 1 Point Point LIC LMIC UMIC HIC AFR EAP ECA LCR MNA SAR Source: World Bank data. Note: LIC = Low-Income Country; LMIC = Lower-Middle-Income Country; UMIC = Upper-Middle-Income Country; HIC = High-Income Country. AFR = Africa Region, EAP = East Asia and Pacific Region; ECA = Europe and Central Asia Region; LCR = Latin America and Carribean Region; MNA = Middle East and North Africa Region; SAR = South Asia Region. reports frequently (monthly). About 50 % of the fragile states follow a similar pattern: 14 of the 35 countries publish this information, mostly annually. Budget execution results are presented through regular updates in more than 80 % of the EU, OECD, and APEC economies. However, monthly publication of budget performance based on FMIS databases is rare (as was already explained under Indicators I-3 and I-4). I-19 External audit of central government budget operations published? I-20 If yes: Regularity of publishing external audit reports Points Responses Economies %E Regularity Regions %R Regularity 1 Yes 76 38.4 61 59 35.1 44 0 No 122 61.6 — 109 64.9 — This indicator checks whether budget-related audit reports are published on external audit organization websites. In most of the economies, publication of external audit reports is mandated by law. However, such reports are usually submitted to parliaments to comply with the legislation, and it is rare to see them published on the web to inform citizens. Publication of such reports on PF web- sites could improve budget transparency and accountability in many economies. Hence, there seem to be opportunities to improve current practices. This study found that 76 governments publish some reports about budget execution performance on external audit websites (usually within 6–12 months after the closing of the budget year), and 61 of them (81 %, mainly HICs and MICs) present such assessments regularly (Figure 3.16). Only 6 of the 38 LIC countries publish external audit reports regularly, and under half of the EU, OECD, and APEC governments do so. In summary, most countries do not pay enough attention to presenting external audit reports regularly on the web. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 34 Data Figure 3.16  Income level and regional distributions for Indicators 19 and 20 a. Income level distribution b. Regional distribution 40 40 30 30 Economies Economies 20 20 R 10 e 10 g R 0 0 0 1 0 1 Point Point LIC LMIC UMIC HIC AFR EAP ECA LCR MNA SAR Source: World bank data. Note: LIC = Low-Income Country; LMIC = Lower-Middle-Income Country; UMIC = Upper-Middle-Income Country; HIC = High-Income Country. AFR = Africa Region, EAP = East Asia and Pacific Region; ECA = Europe and Central Asia Region; LCR = Latin America and Carribean Region; MNA = Middle East and North Africa Region; SAR = South Asia Region. Among the Regions, a number of countries in LCR, ECA, EAP, and SAR publish external audit reports related to budget execution, but only a few do so within six months after the closing of the budget year. In fragile states only 6 out of 35 countries publish some audit reports, and only 4 of these do so regularly. Informative Indicators There are 20 informative indicators (I-21 to I-40) derived from 10 questions (Q11 to Q20) to provide additional feedback about other important features of PF publication websites. Indicators I-21 to I-30 are designed to clarify the contents of budget execution reports by providing additional feedback on sev- eral specific report sections that are visible on the PF websites, for further analysis. I-21 Public expenditures > Consolidated budget results published? Points Responses Economies %E Regions %R 1 Yes 140 70.7 116 69.0 0 No 58 29.3 52 31.0 Most of the governments (mainly HICs and MICs) publish consolidated bud- get results at different intervals (Figure 3.17). Almost all EU, OECD, and APEC members present consolidate budget results more frequently (monthly/ quarterly). Of the 168 Regional countries, 116 (most in LCR, ECA, SAR, and EAP) publish consolidated results during the budget execution cycle. Also, 16 of the 35 fragile states present the consolidated budget results with annual updates. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 Data 35 Figure 3.17  Income level and regional distributions for Indicator 21 a. Income level distribution b. Regional distribution 50 35 30 40 25 Economies Economies 30 20 20 15 10 10 5 0 0 0 1 0 1 Point Point LIC LMIC UMIC HIC AFR EAP ECA LCR MNA SAR Source: World bank data Note: LIC = Low-Income Country; LMIC = Lower-Middle-Income Country; UMIC = Upper-Middle-Income Country; HIC = High-Income Country. AFR = Africa Region, EAP = East Asia and Pacific Region; ECA = Europe and Central Asia Region; LCR = Latin America and Carribean Region; MNA = Middle East and North Africa Region; SAR = South Asia Region. Figure 3.18  Income level and regional distributions for Indicator 22 a. Income level distribution b. Regional distribution 50 35 30 40 25 Economies Economies 30 20 20 15 10 10 5 0 0 0 1 0 1 Point Point LIC LMIC UMIC HIC AFR EAP ECA LCR MNA SAR Source: World Bank data. Note: LIC = Low-Income Country; LMIC = Lower-Middle-Income Country; UMIC = Upper-Middle-Income Country; HIC = High-Income Country. AFR = Africa Region, EAP = East Asia and Pacific Region; ECA = Europe and Central Asia Region; LCR = Latin America and Carribean Region; MNA = Middle East and North Africa Region; SAR = South Asia Region. I-22 Public expenditures > Sector analysis published? Points Responses Economies %E Regions %R 1 Yes 145 73.2 119 70.8 0 No 53 26.8 49 29.2 Sector analysis is important for monitoring the effectiveness of spending in priority areas. FMIS can be linked with other databases to present comprehensive information for better forecasting and decision support, by adding specific dimensions to budget data. It appears that such analytical processing platforms are rare in the public sector. Most of the governments (mainly HICs and MICs) publish an analysis of spending by sectors (Figure 3.18). Comprehensive sector analysis is available in a large number of the budget execution reports published by the EU, OECD, and Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 36 Data APEC economies, as well. However, many of these reports are published annu- ally, as a part of budget reviews, and in only a few cases is sectoral analysis pro- vided during the execution period. In the Regions 119 countries publish sector analysis. LCR, ECA, SAR, and AFR countries lead in publishing sector analysis (mainly focused on the educa- tion, health, energy, and transport sectors). Finally, 19 of the 35 fragile states publish sector analysis as part of their annual budget performance assessments. I-23 Public expenditures > Regional analysis published? Points Responses Economies %E Regions %R 1 Yes 58 29.3 47 28.0 0 No 138 69.7 119 70.8 — n/a 2 1.0 2 1.2 Information about the regional analysis of spending is important for the assessment of budget performance at state/province/district levels. When linked with FMIS databases, regional spending patterns can be monitored dynamically (through monthly updates) to improve public finance management (PFM) prac- tices. Few published budget execution reports focus on such aspects. Some of the HICs and MICs publish a regional analysis of spending and pres- ent the results through geomapping as well. However, there is little attention to regional analysis in LICs (Figure 3.19). About half of the EU, OECD, and APEC governments regularly publish regional spending analysis. Of the 47 Regional countries that publish regional analysis, 41 are in ECA, LCR, EAP, and AFR; countries in the other Regions present little information about regional spending patterns on the web. Only 7 of the 35 fragile states pres- ent some reports on regional spending. Figure 3.19  Income level and regional distributions for Indicator 23 a. Income level distribution b. Regional distribution 40 50 30 40 Economies Economies 30 20 20 10 10 0 0 0 1 0 1 Point Point LIC LMIC UMIC HIC AFR EAP ECA LCR MNA SAR Source: World Bank data. Note: LIC = Low-Income Country; LMIC = Lower-Middle-Income Country; UMIC = Upper-Middle-Income Country; HIC = High-Income Country. AFR = Africa Region, EAP = East Asia and Pacific Region; ECA = Europe and Central Asia Region; LCR = Latin America and Carribean Region; MNA = Middle East and North Africa Region; SAR = South Asia Region. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 Data 37 Figure 3.20  Income level and regional distributions for Indicator 24 a. Income level distribution b. Regional distribution 60 50 50 40 Economies Economies 40 30 30 20 20 10 10 0 0 0 1 0 1 Point Point LIC LMIC UMIC HIC AFR EAP ECA LCR MNA SAR Source: World Bank data. Note: LIC = Low-Income Country; LMIC = Lower-Middle-Income Country; UMIC = Upper-Middle-Income Country; HIC = High-Income Country. AFR = Africa Region, EAP = East Asia and Pacific Region; ECA = Europe and Central Asia Region; LCR = Latin America and Carribean Region; MNA = Middle East and North Africa Region; SAR = South Asia Region. I-24 Public expenditures > Gender analysis published? Points Responses Economies %E Regions %R 1 Yes 29 14.6 22 13.1 0 No 169 85.4 146 86.9 Gender analysis is included in expenditure reports published by 29 govern- ments (Figure 3.20), but most of the annual budgets do not include a specific “gender budget” section. Similarly, only some EU, OECD, and APEC governments publish reports with a gender focus. Except for 18 LCR, AFR, and EAP countries, most Regional countries do not present gender analysis in their budget execution reports (published mostly annu- ally). Only 5 fragile states publish some information related to gender analysis. I-25 Public expenditures > Analysis of spending for children and youth published? Points Responses Economies %E Regions %R 1 Yes 24 12.1 % 19 11.3 % 0 No 174 87.9 % 149 88.7 % Analysis of public spending on children and youth is included in the reports published by 24 governments (Figure 3.21), but most of the annual budgets do not include a specific section dedicated to children. Similarly, only a few EU, OECD, and APEC economies publish reports with a focus on children. Of the 19 Regional countries that publish specific expenditures for various needs of children and youth, 14 are in AFR, LCR, and EAP; most Regional coun- tries do not publish such information. Only 3 of the 35 fragile states present some data on spending for children/youth. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 38 Data Figure 3.21  Income level and regional distributions for Indicator 25 a. Income level distribution b. Regional distribution 60 50 50 40 Economies Economies 40 30 30 20 20 10 10 0 0 0 1 0 1 Point Point LIC LMIC UMIC HIC AFR EAP ECA LCR MNA SAR Source: World Bank data. Note: LIC = Low-Income Country; LMIC = Lower-Middle-Income Country; UMIC = Upper-Middle-Income Country; HIC = High-Income Country. AFR = Africa Region, EAP = East Asia and Pacific Region; ECA = Europe and Central Asia Region; LCR = Latin America and Carribean Region; MNA = Middle East and North Africa Region; SAR = South Asia Region. Figure 3.22  Income level and regional distributions for Indicator 26 a. Income level distribution b. Regional distribution 50 35 30 40 25 Economies Economies 30 20 20 15 10 10 5 0 0 0 1 0 1 Point Point LIC LMIC UMIC HIC AFR EAP ECA LCR MNA SAR Source: World Bank data. Note: LIC = Low-Income Country; LMIC = Lower-Middle-Income Country; UMIC = Upper-Middle-Income Country; HIC = High-Income Country. AFR = Africa Region, EAP = East Asia and Pacific Region; ECA = Europe and Central Asia Region; LCR = Latin America and Carribean Region; MNA = Middle East and North Africa Region; SAR = South Asia Region. I-26 Public expenditures > Debt data published? Points Responses Economies %E Regions %R 1 Yes 145 73.2 120 71.4 0 No 53 26.8 48 28.6 Because of the reporting requirements of lenders, most LICs and MICs tend to publish some data on their foreign debt, but information about domestic or sovereign debt is less visible. A large number of governments publish information about their domestic and/or foreign debt (Figure 3.22). Of these, HICs and MICs (122) publish comprehensive information, usually annually. Also, about 60 % of the LICs publish debt data on their websites. Almost all EU, OECD, and APEC economies publish their domestic/foreign debt data on the PF websites. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 Data 39 Most of the Regional countries publish significant information about their debt. A large number of AFR, LCR, and ECA countries publish debt data, mostly annually; and 20 of the 35 fragile states also publish data about their foreign/ domestic debt. I-27 Public expenditures > Foreign aid/grants published? Points Responses Economies %E Regions %R 1 Yes 123 62.1 105 62.5 0 No 73 36.9 62 36.9 — n/a 2 1.0 1 0.6 Development partners and donors providing foreign aid/grants require regular (quarterly/annual) reporting on the use of funds, mainly from the LICs and LMICs. Donors also have obligations to present information about the funds they provide to various countries for budget support or PFM reforms (invest- ments, advisory support, and capacity building). Despite these requirements, most of the economies do not publish comprehensive information about the use of aid/grant funds on their websites. Foreign aid/grant details are included in the public expenditure reports of 123 of the 198 economies (Figure 3.23). Half of the LICs (19 out of 38) publish foreign aid data on their websites. A large number of EU, OECD, and APEC economies (around 75 %) publish on the web the aid/grants they provide to developing countries. In all Regions, 105 countries publish significant aid/grant data, and about half of the fragile states publish information about the aid/grants they receive. Figure 3.23  Income level and regional distributions for Indicator 27 a. Income level distribution b. Regional distribution 40 30 25 30 Economies Economies 20 20 15 10 10 5 0 0 0 1 0 1 Point Point LIC LMIC UMIC HIC AFR EAP ECA LCR MNA SAR Source: World Bank data. Note: LIC = Low-Income Country; LMIC = Lower-Middle-Income Country; UMIC = Upper-Middle-Income Country; HIC = High-Income Country. AFR = Africa Region, EAP = East Asia and Pacific Region; ECA = Europe and Central Asia Region; LCR = Latin America and Carribean Region; MNA = Middle East and North Africa Region; SAR = South Asia Region. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 40 Data I-28 Public expenditures > Fiscal data on state/local governments published? Points Responses Economies %E Regions %R 1 Yes 75 37.9 59 35.1 0 No 121 61.1 107 63.7 — n/a 2 1.0 2 1.2 Regular web publication of PF data on state/local level government spending as a part of budget expenditure reports is not a common practice (Figure 3.24). Of the 198 economies, 75 (mainly HICs and MICs) publish significant informa- tion (quarterly/annually) about the distribution of revenues and expenditures, with some focus on sectoral spending. However, a large number of economies do not include such details in their web publications. Of EU, OECD, and APEC economies, about 70 % publish such data. Among regional countries, 59 (35.1 %) publish data about subnational spend- ing. About half of the LCR and ECA countries provide significant data (some with geomapping). AFR has the lowest level of visibility in this regard. Six of the 35 fragile states publish some information about subnational spending. I-29 Public expenditures > Financial statements published? Points Responses Economies %E Regions %R 1 Yes 106 53.5 85 50.6 0 No 92 46.5 83 49.4 Government financial statements (income statement, balance sheet, and cash flow statement) include the flows and stocks associated with budget operations. The IMF’s GFSM 2001 and the International Public Sector Accounting Standards (IPSAS) call for the inclusion of detailed balance sheets reporting the value of Figure 3.24  Income level and regional distributions for Indicator 28 a. Income level distribution b. Regional distribution 35 40 30 25 30 Economies Economies 20 20 15 10 10 5 0 0 0 1 0 1 Point Point LIC LMIC UMIC HIC AFR EAP ECA LCR MNA SAR Source: World Bank data. Note: LIC = Low-Income Country; LMIC = Lower-Middle-Income Country; UMIC = Upper-Middle-Income Country; HIC = High-Income Country. AFR = Africa Region, EAP = East Asia and Pacific Region; ECA = Europe and Central Asia Region; LCR = Latin America and Carribean Region; MNA = Middle East and North Africa Region; SAR = South Asia Region. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 Data 41 Figure 3.25  Income level and regional distributions for Indicator 29 a. Income level distribution b. Regional distribution 40 35 30 30 25 Economies Economies 20 20 15 10 10 5 0 0 0 1 0 1 Point Point LIC LMIC UMIC HIC AFR EAP ECA LCR MNA SAR Source: World Bank data. Note: LIC = Low-Income Country; LMIC = Lower-Middle-Income Country; UMIC = Upper-Middle-Income Country; HIC = High-Income Country. AFR = Africa Region, EAP = East Asia and Pacific Region; ECA = Europe and Central Asia Region; LCR = Latin America and Carribean Region; MNA = Middle East and North Africa Region; SAR = South Asia Region. financial and nonfinancial assets and liabilities. However, in many countries the data provided in such financial statements are much less than required. Most of the HICs publish their financial statements regularly, but many MICs and LICs do not. Overall, 106 of the 198 governments publish their financial statements on the web (Figure 3.25). Most of the EU, OECD, and APEC econo- mies (85 %) publish significant information regularly. Considering all Regions, 85 of the 168 countries present audited financial statements on the web. Larger economies in LCR, ECA, AFR, SAR, and EAP publish significant information related to financial statements. However, many AFR and MNA governments provide little or no information in this regard. Among 35 fragile states, 11 provide some information on their financial statements. I-30 Public expenditures > Public procurement and contracts published? Points Responses Economies %E Regions %R 1 Yes 53 26.8 % 41 24.4 % 0 No 145 73.2 % 127 75.6 % Only a small number of governments (most HICs) follow good practices by publishing comprehensive information about public tenders and execution of contracts signed on dedicated websites/portals (Figure 3.26). Even among EU, OECD, and APEC economies, fewer than 50 % publish significant information on tenders and contracts signed. Except for several countries in LCR, ECA, and EAP, most Regional govern- ments do not provide comprehensive information about procurement activities. Among fragile states, 5 out of 35 provide some information about tenders. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 42 Data Figure 3.26  Income level and regional distributions for Indicator 30 a. Income level distribution b. Regional distribution 50 50 40 40 Economies Economies 30 30 20 20 10 10 0 0 0 1 0 1 Point Point LIC LMIC UMIC HIC AFR EAP ECA LCR MNA SAR Source: World bank data Note: LIC = Low-Income Country; LMIC = Lower-Middle-Income Country; UMIC = Upper-Middle-Income Country; HIC = High-Income Country. AFR = Africa Region, EAP = East Asia and Pacific Region; ECA = Europe and Central Asia Region; LCR = Latin America and Carribean Region; MNA = Middle East and North Africa Region; SAR = South Asia Region. Figure 3.27 Income level and regional distributions for Indicator 31 a. Income level distribution b. Regional distribution 50 50 40 40 Economies Economies 30 30 20 20 10 10 0 0 0 1 0 1 Point Point LIC LMIC UMIC HIC AFR EAP ECA LCR MNA SAR Source: World Bank data. Note: LIC = Low-Income Country; LMIC = Lower-Middle-Income Country; UMIC = Upper-Middle-Income Country; HIC = High-Income Country. AFR = Africa Region, EAP = East Asia and Pacific Region; ECA = Europe and Central Asia Region; LCR = Latin America and Carribean Region; MNA = Middle East and North Africa Region; SAR = South Asia Region. I-31 Is there an open government/open data website? Points Responses Economies %E Regions %R 1 Yes 61 30.8 41 24.4 0 No 137 69.2 127 75.6 This indicator is designed to capture the web links to country-specific open government/open data initiatives launched as a part of the Open Government Partnership (OGP) or separately. Such websites are visible in 61 economies (Figure 3.27), and interest in launching new open government/open budget websites seems to have grown since 2011. A large number of HICs and UMICs (49 out of 106) have open data websites with significant data on various aspects (e-Services, access to information, and so Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 Data 43 on). Only 12 of 92 LICs and LMICs have launched such portals. More than 80 % of EU, OECD, and APEC economies have open government web portals. Among Regional countries, 24 of the 62 in ECA and LCR maintain useful open government web portals; a total of 41 open government/open data portals are visible in the Regions. Only 1 of the 35 fragile states provides useful informa- tion through open government websites. I-32 Does the FMIS support the PFM needs of state/local governments? Points Responses Economies %E Regions %R 2 Yes (centralized FMIS solution supports the decentralized 16 8.1 15 8.9 SNG automation, data collection, and reporting needs) 1 Yes (FMIS solution provides data collection and consolida- 10 5.0 10 6.0 tion capabilities for the SNGs) 0 No 170 85.9 141 83.9 — n.a 2 1.0 2 1.2 This indicator is designed to capture evidence about the support provided by centralized web-based FMIS applications to decentralized budget operations at subnational levels. Only a small number of governments (26, or 13.1 %) provide FMIS-based solutions to support state/local budget users (Figure 3.28). In most of the economies, state- and local-level budget users rely on their own automation solutions for their PFM needs or operate manually (85.9 %). Of the 16 economies that have countrywide FMIS supporting subnational levels, 14 are HICs and MICs. However, the functional scope of such FMIS support is limited to the submission of payment requests, production of budget reports, and con- solidation of results. Among EU, OECD, and APEC economies, only some sup- port such subnational operations, mainly because of the decentralized nature of PFM practices. Figure 3.28 Income level and regional distributions for Indicator 32 a. Income level distribution b. Regional distribution 50 50 40 40 Economies Economies 30 30 20 20 10 10 0 0 0 1 2 0 1 2 Point Point LIC LMIC UMIC HIC AFR EAP ECA LCR MNA SAR Source: World Bank data. Note: LIC = Low-Income Country; LMIC = Lower-Middle-Income Country; UMIC = Upper-Middle-Income Country; HIC = High-Income Country. AFR = Africa Region, EAP = East Asia and Pacific Region; ECA = Europe and Central Asia Region; LCR = Latin America and Carribean Region; MNA = Middle East and North Africa Region; SAR = South Asia Region. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 44 Data Among the Regions, LCR countries are responsible for over half of the good practices (8 out of 15). Among the fragile states, only Afghanistan runs a central- ized FMIS solution providing some support for local budget users as well. I-33 Is there a harmonized public accounting system for all budget levels? Points Responses Economies %E Regions %R 1 Yes 65 32.8 55 32.7 0 No 131 66.2 111 66.1 — n.a. 2 1.0 2 1.2 One of the main objectives in many FMIS applications is the production of timely and comprehensive budget reports based on a unified CoA supporting all budget levels. This indicator is designed to locate the websites that provide evidence on a unified CoA and harmonized accounting practices. About 65 governments (mainly HICs and MICs) present documents about the unified CoA and harmo- nized accounting in central and state/local budget levels (Figure 3.29). About 65 % of the EU, OECD, and APEC economies follow good practices, benefiting from decentralized solutions and consolidating the results through a centralized FMIS. Among the Regional countries, 55 (mainly from ECA, LCR, and AFR) have harmonized accounting practices and the unified CoA to improve the consolida- tion and monitoring of budget results. Among the fragile states, 6 benefit from harmonized accounting practices. In summary, a large number of countries con- tinue to run daily PFM operations without a unified CoA and harmonized accounting practices. Centralized FMIS solutions and changes in legal/opera- tional frameworks could bring substantial benefits in terms of operational effi- ciency and cost savings, especially in LICs and LMICs. Figure 3.29  Income level and regional distributions for Indicator 33 a. Income level distribution b. Regional distribution 40 40 30 30 Economies Economies 20 20 10 10 0 0 0 1 0 1 Point Point LIC LMIC UMIC HIC AFR EAP ECA LCR MNA SAR Source: World Bank data. Note: LIC = Low-Income Country; LMIC = Lower-Middle-Income Country; UMIC = Upper-Middle-Income Country; HIC = High-Income Country. AFR = Africa Region, EAP = East Asia and Pacific Region; ECA = Europe and Central Asia Region; LCR = Latin America and Carribean Region; MNA = Middle East and North Africa Region; SAR = South Asia Region. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 Data 45 I-34 Are PF data published on the Statistics website? Or another website? Points Responses Economies %E Regions %R 3 Both Statistics and other websites publish PF data. 23 11.6 16 9.5 2 Other website publishes PF data (no Statistics website publishes 36 18.2 31 18.5 PF data). 1 Statistics website publishes PF data. 39 19.7 30 17.9 0 Statistics website with no PF data. 100 50.5 91 54.1 This indicator is developed to capture other websites (Statistics, Central Bank, or other) on which PF information may be published. In about half of the coun- tries, Statistics Organization websites provide useful information about national accounts (public+private sector results) without a separate section on public finance. A number of countries (39) present additional PF data on their Statistics websites, in addition to their main budget publication websites. Additional bud- get reports are visible on other government websites in 36 economies where Statistics websites do not contain PF data. Finally, 23 countries provide compre- hensive PF data on both Statistics and other websites. The EU, OECD, and APEC economies display similar patterns. Regional trends are very similar to income level distributions (Figure 3.30). Of the 35 fragile states, 3 provide PF data on other public websites. I-35 Access to information explained? Points Responses Economies %E Regions %R 1 Yes 53 26.8 34 20.2 0 No 145 73.2 134 79.8 Figure 3.30  Income level and regional distributions for Indicator 34 a. Income level distribution b. Regional distribution 50 35 30 40 25 Economies Economies 30 20 20 15 10 10 5 0 0 0 1 2 3 0 1 2 3 Point Point LIC LMIC UMIC HIC AFR EAP ECA LCR MNA SAR Source: World Bank data. Note: LIC = Low-Income Country; LMIC = Lower-Middle-Income Country; UMIC = Upper-Middle-Income Country; HIC = High-Income Country. AFR = Africa Region, EAP = East Asia and Pacific Region; ECA = Europe and Central Asia Region; LCR = Latin America and Carribean Region; MNA = Middle East and North Africa Region; SAR = South Asia Region. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 46 Data Figure 3.31  Income level and regional distributions for Indicator 35 a. Income level distribution b. Regional distribution 60 50 50 40 Economies Economies 40 30 30 20 20 10 10 0 0 0 1 0 1 Point Point LIC LMIC UMIC HIC AFR EAP ECA LCR MNA SAR Source: World Bank data. Note: LIC = Low-Income Country; LMIC = Lower-Middle-Income Country; UMIC = Upper-Middle-Income Country; HIC = High-Income Country. AFR = Africa Region, EAP = East Asia and Pacific Region; ECA = Europe and Central Asia Region; LCR = Latin America and Carribean Region; MNA = Middle East and North Africa Region; SAR = South Asia Region. Most countries provide little explanation of their policies on access to public information. Among the 198 countries screened, only 53 economies (mainly HICs and UMICs) have dedicated websites presenting the details of their access to information policy or documents (Figure 3.31). Practices in publishing rele- vant information appear to be better in the EU (65 %), OECD (70 %), and APEC (52 %) economies. Among Regional countries, only 34 publish relevant information. Some of the LCR, ECA, and EAP countries (31 of the 34) follow good practices by providing various options for access to information. No fragile state has a relevant website. I-36 Are PFM roles and responsibilities clearly explained? Points Responses Economies %E Regions %R 1 Yes 143 72.2 % 117 69.6 % 0 No 55 27.8 % 51 30.4 % Most of the websites provide useful information about roles and responsibili- ties in the PFM domain (usually under the home page of PF organizations). About 143 economies explain the PFM roles and organizational structure on their dedicated PF publication websites (Figure 3.32). All EU, OECD, and APEC economies have comprehensive information in this regard, as well. In the Regions, 117 countries (69.6 %) provide useful information about the organizational structure and PFM roles. The visibility of such clarifications is less in AFR (about 52 % of countries). Among the 35 fragile states, 16 present useful information about PFM roles. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 Data 47 Figure 3.32  Income level and regional distributions for Indicator 36 a. Income level distribution b. Regional distribution 50 30 40 25 Economies Economies 20 30 15 20 10 10 5 0 0 0 1 0 1 Point Point LIC LMIC UMIC HIC AFR EAP ECA LCR MNA SAR Source: World Bank data. Note: LIC = Low-Income Country; LMIC = Lower-Middle-Income Country; UMIC = Upper-Middle-Income Country; HIC = High-Income Country. AFR = Africa Region, EAP = East Asia and Pacific Region; ECA = Europe and Central Asia Region; LCR = Latin America and Carribean Region; MNA = Middle East and North Africa Region; SAR = South Asia Region. I-37 Compliance with specific international reporting standards? Points Responses Economies %E Regions %R 2 IMF GFS reports are published (including UN COFOG-based functional classification of expenditures). 54 27.3 44 26.2 1 Expenditure reports according to UN COFOG functional classification. 9 4.5 7 4.2 0 Budget reports compliant with national standards only. 135 68.2 117 69.6 This indicator monitors the publication of specific reports (IMF’s Government Finance Statistics and UN’s COFOG) on PF websites. Most of the governments publish budget reports in line with country-specific reporting standards (Figure 3.33). Internationally accepted reports are published on the dedicated PF web- sites of 54 economies, of which 36 are HICs and UMICs. There is 40 % GFS and COFOG compliance among EU, OECD, and APEC economies. Figure 3.33  Income level and regional distributions for Indicator 37 a. Income level distribution b. Regional distribution 50 50 40 40 Economies Economies 30 30 20 20 10 10 0 0 0 1 2 0 1 2 Point Point LIC LMIC UMIC HIC AFR EAP ECA LCR MNA SAR Source: World Bank data. Note: LIC = Low-Income Country; LMIC = Lower-Middle-Income Country; UMIC = Upper-Middle-Income Country; HIC = High-Income Country. AFR = Africa Region, EAP = East Asia and Pacific Region; ECA = Europe and Central Asia Region; LCR = Latin America and Carribean Region; MNA = Middle East and North Africa Region; SAR = South Asia Region. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 48 Data Of the 44 Regional countries that are GFS- and/or COFOG-compliant, 16 are in ECA. Of the 35 fragile states, only 6 publish GFS- and/or COFOG-compliant reports on the web. I-38 Web statistics (reports on website traffic)? Points Responses Economies %E Regions %R 1 Yes 29 14.6 23 13.7 0 No 169 85.4 145 86.3 There are a few good practices in providing feedback on website traffic (num- ber of visitors, most frequently visited web pages, and so on). It is difficult to locate such tools, since most of them are managed internally by public entities and the statistics are not shared. However, several countries disclose such statis- tics (Figure 3.34). I-39 Which platforms are available for feedback provision? Points Responses Economies %E Regions %R 3 A number of feedback options are visible (telephone/chat/fax/ 100 50.5 81 48.2 mail/e-mail/forms/web statistics). 2 Interactive feedback options are visible (telephone/chat/fax/mail). 28 14.2 23 13.7 1 Static feedback options are available (e-mail/feedback forms/web 45 22.7 41 24.4 stats). 0 Not visible or inadequate. 25 12.6 23 13.7 A large number of governments (mainly HICs and MICs) provide several options for providing feedback (Figure 3.35). Most of the ECA, LCR, EAP, and SAR countries follow good practices. Figure 3.34  Income level and regional distributions for Indicator 38 a. Income level distribution b. Regional distribution 50 50 40 40 Economies Economies 30 30 20 20 10 10 0 0 0 1 0 1 Point Point LIC LMIC UMIC HIC AFR EAP ECA LCR MNA SAR Source: World Bank data. Note: LIC = Low-Income Country; LMIC = Lower-Middle-Income Country; UMIC = Upper-Middle-Income Country; HIC = High-Income Country. AFR = Africa Region, EAP = East Asia and Pacific Region; ECA = Europe and Central Asia Region; LCR = Latin America and Carribean Region; MNA = Middle East and North Africa Region; SAR = South Asia Region. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 Data 49 Figure 3.35  Income level and regional distributions for Indicator 39 a. Income level distribution b. Regional distribution 40 25 20 30 Economies Economies 15 20 10 10 5 0 0 0 1 2 3 0 1 2 3 Point Point LIC LMIC UMIC HIC AFR EAP ECA LCR MNA SAR Source: World Bank data. Note: LIC = Low-Income Country; LMIC = Lower-Middle-Income Country; UMIC = Upper-Middle-Income Country; HIC = High-Income Country. AFR = Africa Region, EAP = East Asia and Pacific Region; ECA = Europe and Central Asia Region; LCR = Latin America and Carribean Region; MNA = Middle East and North Africa Region; SAR = South Asia Region. I-40 What languages are used to publish PF information? A majority of the PF web publishing sites are easy to navigate because of the availability of native and other language options. Using online machine transla- tion tools (for example, Google Translate), it was possible to obtain responses to most of the questions while screening the websites. Table 3.2 lists the various language options available in PF websites. Most of the PF publication websites are accessible in English (54 native, 101 other). Although 109 economies present their PF information in at least one other language, most of the second language options (103) provide substantially reduced information, and key budget reports are presented only in the native language (except in Canada, Malta, and Poland, where publications in another language are comprehensive). Some countries publish their results in a third language (28 out of 198), with reduced scope. Table 3.2  Indicator 40 > Distribution of native and other language options Language Code Native Second Third All English eng 54 82 19 155 French fre 23 8 — 31 Spanish spa 21 — — 21 Arabic ara 18 3 — 21 Russian rus 3 6 1 10 Portuguese por 7 1 1 9 Chinese chi 4 — — 4 German ger 4 — 1 5 Dutch dut 3 — — 3 Other — 61 9 6 76 Totals 198 109 28 335 Source: World Bank data. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 50 Data Characteristics of Underlying FMIS Solutions The FMIS & OBD data set includes the important characteristics of 176 FMIS platforms that are visible on the web (see Appendix B.2 for the description of related fields). An existing FMIS database that includes the details of 109 FMIS projects funded by the World Bank Group (71 completed + 30 active + 8 pipe- line; updated every six months) from 60 countries was used to extract and reuse relevant fields consistently. The team visited the other country websites that were not included in this database to collect additional data and capture the status of all 176 FMIS. The results of this expanded FMIS survey are presented in several groups below to clarify the key features of FMIS platforms that can be used as the basis for open budget data in many countries. This data set is also the basis for the FMIS World Map. In addition to the web links provided for access to the descriptions of the 176 FMIS, the new data set includes the full name and abbreviation of all operational systems. There seems to be no naming convention that is shared by a number of countries to define different types of FMIS based on their scope, functionality, technology architecture, or application software platforms. Many countries prefer to give their systems unique names, and several regional patterns have emerged over the years, mainly in LCR, ECA, and AFR. The name Integrated FMIS (IFMIS) or its different language versions is popular in many countries, despite the fact that the term refers mainly to core FMIS solutions, and does not include a data ware- house for full integration with other PFM systems to publish reliable open budget data. As was explained in chapter 1, IFMIS solutions are rare in practice, and using the term as a synonym for core FMIS functionality may be misleading. Regarding the PFM topology, 166 of the 198 economies (83.8 %) operate a centralized FMIS solution to support decentralized operations at various bud- get levels (Figure 3.36). The remaining 32 economies (16.2 %) have distributed FMIS solutions that are managed by executive line ministries, and a centralized system is used to consolidate data from distributed databases. According to the description of the systems in PF websites, most of the FMIS solutions that are operational in 120 countries (68.2 %) support the budget preparation, execution, accounting, and financial reporting functions (B+T) as core capabilities (Figure 3.37), and some of them have additional modules Figure 3.36 PFM topology supported by FMIS solutions 166 Topology 32 0 50 100 150 200 Economies Centralized Decentralized Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 Data 51 Figure 3.37 Functionality of FMIS solutions Functionality 120 56 0 50 100 150 Economies FMIS (B+T) Treasury (T) Source: World Bank data. Note: PFM = Public Finance Management; FMIS = Financial Management Information System. Figure 3.38 Operational status of FMIS solutions a. FMIS operational status b. FMIS operational since 114 75 9 49 48 37 5 15 0 50 100 150 0 20 40 60 80 Economies Economies Operational Pilot In progress Pipeline >2010 2005–10 2000–05 <2000 Source: World Bank data. Note: FMIS = Financial Management Information System. (human resources management information system/payroll, procurement, asset management, and so on). The other 56 solutions (31.8 %) provide core budget execution, accounting, and reporting capabilities only, and are usually referred to as Treasury Systems (T). The team used existing websites and the reference documents posted on the web to check the operational status of FMIS platforms (Figure 3.38). The initial findings were later verified by some of the government officials, and the data set was updated accordingly. The results indicate that 114 of the 176 FMIS solutions (64.8 %) are used to support PFM functions on a daily basis, and there are 9 pilot implementations that are partially operational. It appears that 48 economies are in the middle of modernization projects to improve their FMIS, and there are 5 pipeline projects to implement new solutions. Most of the existing systems (123 of the 176, or 70 %) have been operational since 2005, and they benefit from relatively new web-based platforms that are designed to improve operational efficiency and provide timely decision support. In 106 economies (60 %), FMIS support both central- and local-level budget operations (C+L) of executive ministries (mainly the Finance Ministry or Department) through web-based systems. The FMIS platforms in the 70 Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 52 Data Figure 3.39 Scope of FMIS solutions 106 Scope 70 0 50 100 150 Economies Central+local Central Source: World Bank data. Note: FMIS = Financial Management Information System. Figure 3.40 FMIS application software solutions 97 66 ASW 13 0 50 100 150 Economies COTS LDSW Unknown remaining countries are designed to support mainly central budget operations (C) for the line ministries, and various public entities at state/local budget levels run distributed systems for their needs (Figure 3.39). In addition, a large group of economies (97 of 176, or 55.1 %) have developed FMIS solutions based on commercial off-the-shelf (COTS) packages, customized for public sector needs (Figure 3.40). About 66 economies (37.5 %) have invest- ed considerable time and effort to develop and maintain locally developed soft- ware (LDSW) solutions, mainly to have full access to the source code and data- bases and reduce the costs of licensing and support. The type of application software is unknown in 13 economies (7.4 %), since most of these countries either do not publish such details on the web or operate manually. Because of the advances in technology and the rapid expansion of the World Wide Web infrastructure, most of the FMIS developed in the last decade (149 of 176, or 84.7 %) are web-based solutions. The other 27 solutions (15.3 %) are usually legacy systems based on distributed architecture, and they run on a client- server platform (Figure 3.41). Finally, the data set also includes information about the type of FMIS project initiated or completed in the 176 economies to install ICT solutions and provide advisory support for capacity building and change management. The systems used in 148 economies (84 %) were designed and implemented as a new turnkey solution (first time or replacing a previous system). About 23 systems (13.1 %) were developed as an improved or expanded version of existing FMIS solutions. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 Data 53 Figure 3.41 FMIS technology architecture 149 Tech arch 27 0 50 100 150 200 Economies Web-based Client-server Source: World bank data Note: FMIS = Financial Management Information System; COTS = Commercial off the shelf software; LDSW = Locally Developed Software. The remaining systems were implemented mainly in fragile states during emer- gency technical assistance projects by installing one of the COTS solutions for rapid deployment. The websites did not yield detailed information about the duration and cost of implementation for most of the FMIS solutions. However, the World Bank’s FMIS Database provides reliable information about the cost/duration and other important aspects (scope, functional modules, procurement approach, lessons learned, and so on) of about 109 projects funded by the World Bank in 60 coun- tries (updated twice a year, last update in July 2013; see https://eteam.world- bank.org/FMIS for details). Comparison of Findings with PEFA Indicators The team used some of the Public Expenditure and Financial Accountability (PEFA) PFM Performance Measurement Framework indicators that are related to fiscal transparency in a comparative analysis with the findings of this study (see Appendix B for details). As of January 2013, the PEFA assessments were available for 121 economies, but only 65 countries published their PEFA results on the web; the disclosure of the PEFA results for 32 countries was not permitted by the governments, and these economies were excluded from this analysis. Although draft PEFA reports were available for the remaining 24 economies, their disclosure status was unknown. The FMIS & OBD scores were compared with the PEFA indicators using the publicly available assessments for 65 economies. For this comparison, the PEFA indicator scores were converted to numbers from 1 to 4 (A=4; D=1; other scores—B+, B, C+, C, and D+—are equally distributed with increments of 0.5). Two scatter diagrams were prepared to plot the correlation of 10 selected indicators, as well as the full set of PEFA indicators using average scores (NR/ NA/NU scores, not rated, were excluded while calculating the averages: • The average of PEFA scores for 10 selected indicators related to fiscal trans- parency (Avg 10) was compared with the FMIS & OBD scores. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 54 Data Figure 3.42  Comparison of PEFA scores with FMIS & OBD scores (65 economies) a. PEFA (avg 10) vs. FMIS & OBD scores b. PEFA (avg) vs. FMIS & OBD scores 4 4 y = 0.0133x + 1.8325 y = 0.0114x + 1.9298 R2 = 0.3143 R2 = 0.2801 3 3 PEFA PEFA 2 2 1 1 0 20 40 60 80 100 0 20 40 60 80 100 FMIS & OBD (65 economies) FMIS & OBD (65 economies) Source: World Bank data. Note: Avg 10 means the average of 10 PEFA indicators selected for a comparison with the FMIS & OBD Scores. Avg means the average of all 31 PEFA indicators. NR/NA/NU scores are excluded in average calculations. PEFA = Public Expenditure and Financial Accountability; FMIS = Financial Management Information System; OBD = Open Budget Data. • The average of all 31 PEFA scores (Avg) was compared with the FMIS & OBD scores. Both diagrams demonstrate that there is a positive correlation between the FMIS & OBD scores and the PEFA scores in 65 economies, and the countries with higher PEFA scores tend to perform better in terms of publishing PF data and readiness to disclose open budget data from FMIS solutions (Figure 3.42). The trend observed for 10 selected PEFA indicators was very similar to the pat- tern for the whole set of PEFA indicators. This indicates that the FMIS & OBD scores capture consistent patterns for 65 economies compared to their perfor- mance in PEFA assessments. As the next step, the FMIS & OBD scores were also compared with the full set of PEFA indicators available for 121 countries (Public+Final+Draft assessments) for a broader analysis. Individual country scores were not dis- closed for final and draft reports. However, the average scores were used to prepare two additional scatter diagrams to plot the correlation of 10 selected indicators, as well as the full set of PEFA indicators, for all 121 economies (Figure 3.43). These diagrams also indicate a positive correlation between the FMIS & OBD scores and the PEFA scores. The similar trends indicate that the results obtained from FMIS & OBD key indicators are largely consistent with the patterns observed through PEFA indicators. Comparison with Open Budget Index The 2012 Open Budget Survey (OBS) measures the state of budget transpar- ency, participation, and oversight in 100 countries (see Appendix B for details). The OBS consists of 125 questions and is completed by independent researchers Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 Data 55 Figure 3.43  Comparison of PEFA scores with FMIS & OBD scores (121 economies) a. PEFA (avg 10) vs. FMIS & OBD scores b. PEFA (avg) vs. FMIS & OBD scores 4 4 y = 0.0122x + 1.8166 y = 0.0107x + 1.9064 R2 = 0.2805 R2 = 0.2795 3 3 PEFA PEFA 2 2 1 1 0 20 40 60 80 100 0 20 40 60 80 100 FMIS & OBD (121 economies) FMIS & OBD (121 economies) Source: World Bank data. Note: Avg 10 means the average of 10 PEFA indicators selected for a comparison with the FMIS & OBD Scores. Avg means the average of all 31 PEFA indicators. NR/NA/NU scores are excluded in average calculations. PEFA = Public Expenditure and Financial Accountability; FMIS = Financial Management Information System; OBD = Open Budget Data. in the countries assessed. The Open Budget Index (OBI) is calculated as a simple average of the responses for 95 questions that are related to budget transparency. The OBI assigns each country a score1 that can range from 0 to 100. The OBS is useful in clarifying and comparing the contents of key budget documents in selected countries. However, it contains no questions about the source of infor- mation in published budget documents and the reliability/integrity of the under- lying databases. FMIS & OBD indicators provide additional information on these less known aspects, as well as the visibility of open budget data, to complement the OBS. The scatter diagram of the FMIS & OBD and OBI scores in 100 economies (Figure 3.44) reveals a positive correlation between these scores. Table 3.3 summarizes the distribution of FMIS & OBD practice groups in 100 countries (with OBI scores) to clarify the patterns observed. Most of the findings of the present study are consistent with the OBI scores, when the current status of government practices to publish open budget data (FMIS & OBD practice groups) is compared with the public availability and comprehensiveness of key budget documents (92 out of 100 countries, high- lighted in shaded area). In only a few exceptional cases (outside the shaded area) do the economies perform better in terms of readiness to publish open budget data compared to their OBI score (for example, Bolivia and Ecuador within group A, and OBI scores 12 and 31, respectively). This may be due to the fact that, although they had a well-designed website and clear linkage between pub- lished open budget data and the underlying FMIS, the contents and types of budget documents they published were not adequate to get higher scores in OBI. Also, there are a few countries with no visibility on the web in terms of publish- ing budget data, but their OBI scores are above 40 (for example, Mali within group D, and OBI score 43). Most probably this was due to the fact that the OBI assessment is based on the review of submitted budget documents that are not disclosed on the web. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 56 Data Figure 3.44  Comparison of OBI scores with FMIS & OBD scores (100 economies) 100 y = 0.761x – 1.4941 R2 = 0.5089 80 60 OBI 40 20 0 0 20 40 60 80 100 FMIS & OBD (100 economies) Source: World Bank data. Note: FMIS = Financial Management Information System; OBI = Open Budget Index; OBD = Open Budget Data. Table 3.3  Comparison of FMIS & OBD groups with OBI 2012 scores in 100 countries OBI 2012 A B C D Total 0–20 1 4 13 8 26 21–40 1 12 1 1 15 41–60 7 23 5 1 36 61–80 10 7 — — 17 81–100 4 2 — — 6 Total 23 48 19 10 100 Source: World Bank data. Note: OBI = Open Budget Index. Ten countries have high OBI scores (above 60) and follow good practices (Group A) in publishing open budget data from FMIS: Brazil, Germany, Rep. of Korea, Mexico, New Zealand, Russian Federation, Spain, Sweden, the United Kingdom, the United States). In general, economies with higher OBI scores tend to perform much better in publishing open budget data from reliable FMIS data- bases, and those with lower OBI scores perform less well. Thus, it appears that the indicators selected for this study consistently capture the main characteristics of PF websites that are relevant to the OBS questions. Comparison with Other Dimensions As a final check on the validity of the findings, the team compared the FMIS & OBD scores with other relevant initiatives and dimensions—the Open Government Partnership, MTEF implementation, UN e-Government Development rankings, and Open Source Policies. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 Data 57 Table 3.4  Comparison of FMIS & OBD groups with OGP in 60 countries OGP A B C D Total 0 Developing commitments > 2 — 3 — 5 1 Commitments delivered > 14 33 7 1 55 Total 16 33 10 1 60 Source: World Bank data. Note: OGP = Open Government Partnership; — = not available. Table 3.5  Comparison of FMIS & OBD groups with MTEF implementation in 181 economies MTEF (2010) A B C D Total 0  No MTEF 4 10 20 15 49 1 MTFF 9 36 18 8 71 2 MTBF 7 22 10 3 42 3 MTPF 7 12 — — 19 — Unknown — 1 6 10 17 Total 27 81 54 36 198 Source: World Bank data. Note: MTEF = Medium-Term Expenditure framework; MTFF = Medium-Term Fiscal Framework; MTBF = Medium-Term Budgeting Framework; MTPF = Medium-Term Performance Framework; — = not available. Open Government Partnership Open Government Partnership (OGP) was launched on September 20, 2011. Since then, 55 members joined the group by endorsing the declaration (as of June 2013); and 5 other countries are expected to officially endorse the declaration in 2013, and are in the process of developing commitments. Table 3.4 presents a comparison of the FMIS & OBD groups with the OGP participation status. Most of the governments with endorsed declarations (47 out of 55) have high FMIS & OBD scores (Group A or B), and 11 governments have relatively low scores (Group C and D). It appears that most of the economies involved in OGP are getting ready for publishing open budget data from FMIS solutions to improve their PF web publishing practices. Fourteen countries have endorsed the OGP declarations and already following good practices in publishing open budget data from FMIS solutions (Argentina, Brazil, Colombia, El Salvador, Guatemala, Republic of Korea, Mexico, Netherlands, Paraguay, Peru, Spain, Turkey, United Kingdom, and the United States). Two countries in Group A are in the process of developing commitments (Australia and Ireland). Medium-Term Expenditure Framework Implementation Drawing from the data set in a report published by the World Bank2 on the adop- tion of the medium-term expenditure framework (MTEF) in 181 economies (1990–2010), the study team compared the findings of this study with the cur- rent status of MTEF implementation (see Table 3.5). In the present study, MTEF is viewed as a sequence of three increasingly demanding stages: medium-term fiscal framework (MTFF), medium-term Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 58 Data budgetary framework (MTBF), and medium-term performance framework (MTPF). Of the 19 countries implementing MTPF with an emphasis on the measurement and evaluation of budget performance, all consistently have high FMIS & OBD scores. A large number of governments (113) are included in the MTBF and MTFF, and many of them also have high FMIS & OBD scores. Even of the 49 economies that have no MTEF implementation, 4 have relatively high FMIS & OBD scores (Bolivia, Ecuador, El Salvador, and Guatemala in Group A). This indicates that there is a parallel between the implementation of MTEF at various stages and the publication of open budget data from reliable FMIS data- bases. However, some of the governments with no MTEF focus can still publish substantial budget data to present the performance of their budget implementa- tion (plans vs. actuals; results) and provide useful feedback for the evaluation of their budget. UN e-Government Development Rankings 2012 The overall conclusion of the UN’s e-Government 2012 survey is that “govern- ments must increasingly begin to rethink in terms of e-government—and e-gover- nance—placing greater emphasis on institutional linkages between and among the tiered government structures in a bid to create synergy for inclusive sustainable devel- opment, by widening the scope of e-Gov for a transformative role towards cohesive, coordinated, and integrated processes and institutions.” In the PFM domain, such a transformation is possible through reliable FMIS solutions that provide timely information to a large number of budget users. Therefore, the e-Gov develop- ment ranking is another relevant measure of readiness to publish open budget data from FMIS, in terms of online services, ICT infrastructure, and human capital. Table 3.6 presents the distribution of FMIS & OBD groups in 190 economies included in the UN 2012 e-Government Development rankings. In terms of both e-Gov’12 ranking and index values, the FMIS & OBD scores follow consistent patterns in most of these economies About 90 % of the 40 countries with the top e-Government rankings are in Group A or B regarding their current practices in publishing open budget data. The FMIS & OBD groups shift toward C and D as the e-Gov rankings decline. A small number of economies are exceptions to this trend (15 out of 190, or Table 3.6  Comparison of FMIS & OBD groups with e-Gov’12 rankings and indices eGov rank A B C D Total eGov index A B C D Total 1–40 13 23 2 2 40 0.00–0.20 — 3 3 4 10 41–80 7 18 12 3 40 0.21–0.40 2 16 23 17 58 81–120 5 18 10 7 40 0.41–0.60 7 29 17 7 60 121–160 2 13 15 10 40 0.61–0.80 9 22 8 4 43 161–190 — 6 13 11 30 0.81–1.00 9 8 1 1 19 Total 27 78 52 33 190 Total 27 78 52 33 190 Source: World Bank data. Note: — = not available. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 Data 59 8 %); however, most of the economies display a consistent pattern in FMIS & OBD groupings and e-Gov rankings. As regards the e-Gov indices, more than 90 % of the countries in each index interval display consistent patterns, and only 11 out of 190 economies are outside the shaded area representing the trend. EGDI. The e-Government Development Index (EGDI) is a composite indica- tor measuring governments’ willingness and capacity to use ICT in the delivery of public services. The EGDI is a weighted average of three normalized scores on the most important dimensions: • Scope and quality of online services • Development status of telecommunication infrastructure • Inherent human capital. The four stages of online service development are defined as (a) emerging, (b) enhanced, (c) transactional, and (d) connected (with increasing level of sophisti- cation). Only about 23 % of the economies (43 out of 190) have EGDI scores above 0.60 for connected or transactional online services, with relatively high FMIS & OBD groups. A vast majority of the countries provide emerging or enhanced services through their websites, and this is consistent with the findings of this study (Table 3.7). Telco infrastructure index. The telecommunication (Telco) infrastructure index is an arithmetic average of five indicators per 100 inhabitants: (a) estimated Internet users, (b) number of main fixed telephone lines, (c) number of mobile subscribers, (d) number of fixed internet subscriptions, and (e) number of fixed broadband facilities. The International Telecommunication Union is the primary source of data in each case. Table 3.8 presents the comparison of FMIS & OBD groups with the Telco infrastructure index. Table 3.7  Comparison of FMIS & OBD groups with e-Gov’12 index (Online Services) eGov’12 (OS) A B C D Total 0.00–0.20 — 4 10 19 33 0.21–0.40 1 20 29 10 60 0.41–0.60 9 32 10 3 54 0.61–0.80 9 12 3 1 25 0.81–1.00 8 10 — — 18 Total 27 78 52 33 190 Source: World Bank data. Note: — = not available. Table 3.8  Comparison of FMIS & OBD groups with e-Gov’12 index (Telco Infrastructure) eGov’12 (TI) A B C D Total 0.00–0.20 4 24 29 22 79 0.21–0.40 8 23 9 3 43 0.41–0.60 3 14 9 5 31 0.61–0.80 9 13 4 1 27 0.81–1.00 3 4 1 2 10 Total 27 78 52 33 190 Source: World Bank data. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 60 Data Only about 37 out of 190 (19 %) of the economies have EGDI scores above 0.60 for the telecommunication infrastructure index. However, most of the countries have inadequate Internet connectivity and broadband access, and this is another important aspect to consider in analyzing the relatively lower FMIS & OBD scores in these economies. Human capital index. The human capital index is a weighted average of two indicators: (a) adult literacy rate, and (b) the combined primary, secondary, and tertiary gross enrollment ratio; a two-thirds weight is assigned to adult literacy rate, and one-third to the gross enrollment ratio. The distribution of FMIS & OBD groups with respect to the EGDI index for human capital is presented in Table 3.9. It appears that most of the economies (142 out of 190, or 75 %) score above 0.60, and have relatively high FMIS & OBD groups. Most of the countries are scattered along the trend line indicated by the shaded area in the table. In summary, the findings of this study are largely consistent with the EGDI indices. The human capital index is high in most of the countries, but the online service and Telco infrastructure indices are high only in about 20 % of the econo- mies where the FMIS & OBD groups are also high, indicating a much better level of readiness to publish open budget data, in line with the EGDI indices. Open Source Policies The seventh update to the CSIS Open Source Policy Survey3 was published in 2010. The survey tracks government policies on the use of open source software (OSS). The survey is divided into four categories: (a) research and development, (b) mandates (where the use of OSS is required), (c) preferences (where the use of OSS is given preference, but not mandated), and (d) advisory (where the use of OSS is permitted). The survey also looks at whether an initiative was made at the national, regional, or local level, and whether it was accepted, under consid- eration, or rejected. The study has found a total of 362 open source policy initiatives (Table 3.10), of which 227 are at the national level (66 economies), 117 are at the state or local level (22 economies), and 18 at the international level (EU, OECD, UN). About 69 % of these open source policies have been approved (Table 3.11). The survey results show a greater tendency for the approval of open source research and development initiatives relative to mandatory, preference, or advisory policies. Table 3.9  Comparison of FMIS & OBD groups with e-Gov’12 index (Human Capital) eGov’12 (HC) A B C D Total 0.00–0.20 — 1 — 3 4 0.21–0.40 — 3 7 3 13 0.41–0.60 1 13 11 6 31 0.61–0.80 7 20 15 10 52 0.81–1.00 19 41 19 11 90 Total 27 78 52 33 190 Source: World Bank data. Note: — = not available. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 Data 61 Table 3.10  Scope and regional distribution of open source policies Scope of policy R&D Advisory Preference Mandatory Total National 69 66 60 32 227 State/Local 25 19 52 21 117 International 5 12 1 — 18 Total 99 97 113 53 362 Region Approved Proposed Failed Total Europe 126 35 10 171 Asia 59 20 2 81 Latin America 31 15 11 57 North America 16 11 10 37 Africa 8 1 0 9 Middle East 5 2 0 7 Total 245 84 33 362 Source: World Bank data. Note: — = not available. Table 3.11  Regional distribution of approved open source initiatives Region R&D Advisory Preference Mandatory Total Europe 45 37 36 8 126 Asia 19 16 22 2 59 Latin America 8 6 12 5 31 North America 5 8 2 1 16 Africa 3 1 4 0 8 Middle East 1 2 2 0 5 Total 81 70 78 16 245 Source: World Bank data. Table 3.12  Distribution of FMIS & OBD groups in 66 economies with open source policies OSS Policies/FMIS & OBD Groups > A B C D National-level: 66 economies > 22 38 4 2 State/local-level: 22 economies > 10 12 — — Source: World Bank data. Note: OSS = Open Source Software; FMIS = Financial Management Information System; OBD = Open Budget Data; — = not available. Table 3.12 shows the distribution of FMIS & OBD groups for national- and state/local-level open source policies. This indicates that there is a positive cor- relation between the FMIS & OBD groups and the use of open source policies in the public sector, which in turn promotes the publication of open budget data. At the national level, 60 out of 66 governments (91 %) follow good practices in this regard. For the state/local level, all of the 22 economies with open source policies have high FMIS & OBD scores. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 62 Data With these and earlier comparative analyses, it can be concluded that the indicators defined for this study produce consistent results when compared to other relevant indicators of data transparency. Notes 1. OBI: 0–20: Scant or no info  21–40: Minimal  41–60: Some  61–80: Significant  81–100: Extensive. 2. “Beyond the annual budget: Global experience with medium-term expenditure frameworks,” World Bank, September 2012. doi:10.1596/978-0-8213-9625-4 License: Creative Commons Attribution CC BY 3.0. 3. CSIS Open Source Policy Survey, 2010. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 CHAPTER 4 Good Practices The findings of this study indicate that, although 198 governments around the world use 176 financial management information systems (FMIS) plat- forms, good practices in presenting open budget data from reliable FMIS solutions are highly visible in only about 24 countries (12%). For this study, a good practice can be defined simply as publishing extensive, reli- able, and timely public finance (PF) information, drawn from an FMIS, on easy- to-navigate government websites with dynamic query and reporting options. The purpose of identifying good practices is to allow users to learn from others per- forming well in different areas of publishing PF data, and to share relevant knowledge. This chapter provides an overview of selected cases from all FMIS & open budget data (OBD) practice groups to share information about good practices in effectively addressing the challenges linked with open budget data and transpar- ency. The last section of this chapter describes the FMIS World Map, a geomap- ping application developed on Google Maps to improve the visibility of the findings of this study. In selecting the good practice cases, the following criteria were applied: • Timely publication (I-1): A tradition of regularly publishing consistent PF information through dedicated websites • Visibility of FMIS (I-2): Comprehensive information about the underlying FMIS solution or data warehouse (DW) used for publishing PF information • Dynamic query options (I-3): Access to information for all revenues, alloca- tions, and expenditures through user-defined (dynamic) queries on FMIS databases or DW • Open budget data (I-4): Presenting a rich set of open budget data published from FMIS. Financial Management Information Systems and Open Budget Data   63 http://dx.doi.org/10.1596/978-1-4648-0083-2 64 Good Practices Some cases demonstrate that even in difficult settings innovative solutions to improve budget transparency can be developed rapidly with a modest investment, if there is political will and commitment from the government. • Reliability of PF data (I-5 and I-6): Visibility of system name/time stamp on published reports • Presentation quality (I-7 I-8): Presence of interactive and user-friendly graphi- cal interfaces to display budget data and provide adequate search/download options • Effective use of open data (I-9): Meaningful open data on fiscal policy, budget performance, and achievements available to the public (Citizens Budget). Selected cases from all Regions are presented for each of the good practice categories listed, along with images from PF publication websites and relevant uniform resource locators (URLs) (100 cases from 53 government websites). Important characteristics of highly visible integrated FMIS solutions, including the Digital Budget and Accounting System (or DBAS)1 and other sustained/ emerging platforms (for example, Brazil, the Russian Federation), are also pre- sented. Table 4.1 shows the distribution of good practice cases according to income levels, geographical locations, and FMIS & OBD groups. Table 4.1  Good practices in publishing PF information/open budget data Income Level # Government websites selected to highlight some of the good practices High income 20 Australia; Austria; Taiwan, China; Denmark; Estonia; Finland; France; Germany; Ireland; Japan; Republic of Korea; Netherlands; New Zealand; Norway; Singapore; Slovenia; Spain; Sweden; United Kingdom; United States Upper-middle 17 Argentina; Brazil; Chile; China; Colombia; Dominican Rep.; Ecuador; Jordan; Malaysia; income Mauritius; Mexico; Peru; Russian Fed.; Thailand; Turkey; Uruguay; Venezuela; RB Lower-middle 12 Bolivia; Guatemala; El Salvador; India; Indonesia; Morocco; Nicaragua; Pakistan; income Paraguay; Philippines; Vietnam; Zambia Low income 4 Madagascar; The Gambia; Timor-Leste; West Bank and Gaza Region # Government websites selected to highlight some of the good practices Africa 4 Madagascar; Mauritius; The Gambia; Zambia East Asia and Pacific 13 Australia, China; Taiwan, China; Indonesia; Japan; Malaysia; New Zealand; Republic of Korea; Philippines; Singapore; Thailand; Timor-Leste; Vietnam Europe and Central 15 Austria; Denmark; Estonia; Finland; France; Germany; Ireland; Netherlands; Norway; Asia Russian Federation; Slovenia; Spain; Sweden; Turkey; United Kingdom North and South 16 Argentina; Bolivia; Brazil; Chile; Colombia; Dominican Republic; Ecuador; El Salvador; America Guatemala; Mexico; Nicaragua; Paraguay; Peru; Uruguay; United States; Venezuela; RB Middle East and 3 Jordan; Morocco; West Bank and Gaza North Africa South Asia 2 India; Pakistan table continues next page Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 Good Practices 65 Table 4.1  Good practices in publishing PF information/open budget data (continued) FMIS & OBD Group # Government websites selected to highlight some of the good practices A 24 Argentina; Australia; Brazil; Colombia; Ecuador; El Salvador; Germany; Guatemala; India; Ireland; Republic of Korea; Mexico; Netherlands; New Zealand; Nicaragua; Paraguay; Peru; Russian Federation; Singapore; Slovenia; Spain; Turkey; United Kingdom; United States B 26 Austria; Bolivia; Chile; China; Denmark; Dominican Republic; Estonia; Finland; France; Indonesia; Japan; Jordan; Madagascar; Malaysia; Mauritius; Morocco; Norway; Pakistan; ­ Philippines; Sweden; Thailand; Timor-Leste; Uruguay; Vietnam; Venezuela; RB; Zambia C 3 Taiwan, China; The Gambia; West Bank and Gaza Source: World Bank data. Note: FMIS = Financial Management Information System; OBD = Open Budget Data. The following sections describe the examples. Relevant web links are available in the FMIS & OBD data set posted on the FMIS CoP website: https://eteam. worldbank.org/FMIS. Timely Publication of PF Data through Dedicated Websites A large number of PF websites provide historical information about the approved annual budget plans (153 out of 198, or 77 %) and budget execution results (147 out of 198, or 74 %). However, only about 60 % of these economies follow good practices and publish regularly: their performance in publishing the medium-term expenditure framework (MTEF) and investment plans, as well as the external audit of the budget, is well below the expected levels (see Table 4.2). Although most of these PF websites have been created within the last decade, some of them provide a great amount of historical PF information. Brazil stands out as the best performer in this category, making available annual budget plans and budget execution results since 1980. Several other countries in Latin America and Caribbean Region (LCR) also present historical PF data since the 1980s (Argentina, Chile), but most of the governments publish data covering the last decade. Government websites described in this section illustrate some of the good practices observed during this study, along with ongoing improvements linked with open budget data initiatives. Table 4.2  Historical trends and regularity in publishing PF information Since Budget plans MTEF Investment plans Budget execution External audit # Econ Regular # Econ Regular # Econ Regular # Econ Regular # Econ Regular 1980 3 3 0 0 0 0 2 1 0 0 1990 29 29 12 12 6 6 27 27 13 13 2000 104 80 58 44 30 22 103 81 54 45 2010 17 8 33 8 8 4 15 8 8 2 Totals 153 120 103 64 44 32 147 117 75 60 % 77% 61% 52% 32% 22% 16% 74% 59% 38% 30% Source: World Bank data. Note: MTEF = Medium-Term Expenditure Framework. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 66 Good Practices Africa Mauritius The Ministry of Finance and Economic Development/Ministère des Finances et du Développement Économique website (http://www.mof.mu) presents key information about the organization, current and past national budget, legislation, and public debt. The budget section includes program-based budget estimates and actuals, as well as the public sector investment program online system, pro- viding comprehensive information. Documents can be downloaded in PDF, DOC, or XLS formats. East Asia and Pacific Australia The Central Budget website is the entry point for Australian Government budget material (http://www.budget.gov.au). A rich set of budget information is avail- able: budget speech, strategy, details of revenues and expenditures, appropria- tions, portfolio statements, execution performance, midyear economic and fiscal outlook, and final budget outcome. The past budgets section presents similar details for budget performance starting from 1996. Most of the reports are avail- able in PDF format. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 Good Practices 67 Japan The Ministry of Finance (http://www.mof.go.jp) website presents key PF infor- mation in open data format. Policy documents, net total budget and balance sheet, budget execution performance, and treasury operations are published in PDF, XML, or XLS formats. Some of the government financial statistics are also available from the Statistics (e-Stat) website in open data formats (PDF, CSV, XLS, DB). The Ministry home page also includes web links to relevant govern- ment institutions, daily news, and feedback options for citizens and businesses. Republic of Korea The Ministry of Strategy and Finance website (http://www.mosf.go.kr) provides extensive information about economic policy, performance indicators, publica- tions, budget execution performance, and e-applications. Other important fea- tures are web links to the Digital Budget and Accounting System (DBAS/ dBrain); Treasury bonds; online bidding; statistical information; educational material for children, youth, and researchers; access to information policy; and submission of complaints. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 68 Good Practices New Zealand The New Zealand Treasury/Kaitohutohu Kaupapa Rawa website (http://www. treasury.govt.nz) provides access to extensive information about government finances, budget performance, economy, state sector, and financial statistics. The home page is easy to navigate and includes links to important PF websites. A Treasury Twitter feed, quick links, and topics of current interest sections provide useful feedback for citizens and businesses. A mobile budget site and fiscal time series (1972–2012) present a rich set of open budget data and meaningful infor- mation on public finances. Singapore The Ministry of Finance (http://www.mof.gov.sg) home page is easy to navigate and provides extensive information about the budget process. Budget archives present the details of policies, public spending, and results since 1996. Individuals, civil soci- ety, and businesses can post their views and suggestions through a public consulta- tion section. Mobile applications can be downloaded from the website to learn more about the details of budget plans and execution performance. A budget quiz and download pages provide additional information in an easy-to-understand format. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 Good Practices 69 Europe and Central Asia France The Ministry of Economy and Finance/Ministère de l’Économie et des Finances website (http://www.economie.gouv.fr) provides direct access to regularly updated PF information. Public consultation and communication options include links to social media and WebTV. The budget and public finance website includes Performance Forum, where detailed information on budget performance and results is posted regularly in meaningful formats. The General Directorate of Treasury/Direction Générale du Trésor (http://www.tresor.economie.gouv.fr) website presents comprehensive data on the treasury and debt management. Germany The Ministry of Finance/Bundesministerium der Finanzen website (http://www. bundesfinanzministerium.de) provides, and frequently updates, extensive infor- mation on the federal budget. A services website provides access to key units, data, and reports and strategy documents. The home page has a number of options for communication with citizens and businesses (video, graphics, audio, and pictures) to convey key messages and share progress in ongoing reform activities. Budget execution performance is reported monthly. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 70 Good Practices Russian Federation The Ministry of Finance home page (http://www.minfin.ru) provides access to comprehensive information about the budget, pension, and local government reform activities, reserve and national wealth funds, budget execution perfor- mance, public debt, and audit results. Links to federal agency websites allow users to view and download a rich set of open budget data. The multiyear budget framework is presented, along with updates on the regulations related to spe- cific activities and the description of all MoF information systems. The new budget portal (pilot) is impressive (www.budget.gov.ru). Sweden The Ministry of Finance (MoF) website presents (http://www.sweden.gov.se) the responsibilities, key program activities, news, and publications of the Ministry. Detailed information about the national economy and budget is posted under the “policy areas” section of the website. The MoF web page also includes the financial stability framework, latest legislation, links to relevant agencies, and several interaction options for citizens and businesses. The latest PF forecasts, as well as a follow-up of budget policy objectives and estimate for expenditure ceil- ing, are also presented through relevant web pages and documents. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 Good Practices 71 North and South America Argentina The Ministry of Economy and Public Finance/Ministerio de Economía y Finanzas Públicas (MECON) (http://www.mecon.gov.ar) is the main source of PF infor- mation in Argentina. The website is easy to navigate and provides comprehensive information about the organization, relevant legislation, programs, budget per- formance, and such other important aspects as file queries, documentation and information center, innovation, and tenders. The InfoLEG (Información Legislativa y Documental) section provides online access to all relevant legal documents. Brazil The Brazilian Ministry of Finance/Ministério da Fazenda (http://www.fazenda. gov.br) has one of the most comprehensive federal government websites, provid- ing a rich set of information on all aspects, as well as links to a large number of internal and external PF sites. The National Treasury/Tesouro Nacional has a contemporary website (https://www.tesouro.fazenda.gov.br) publishing detailed budget execution reports from the new FMIS, SIAFI. To improve the efficiency of public spending, a data warehouse extracts data from SIAPE, SIAFI, and SIGPLAN to generate information for decision support and performance moni- toring. Another important source of information is the Federal Budget portal, main- tained by the Ministry of Planning and Budget Management/Ministério do Planejamento, Orçamento e Gestão (http://www.planejamento.gov.br). The portal provides access to federal budget and investment plans, Program for Accelerated Growth/Programa de Aceleração do Crescimento, and multiyear plans. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 72 Good Practices Chile The Ministry of Finance/Ministerio de Hacienda (http://www.hacienda.cl) web- site presents comprehensive information about PF benefits, procedures, and programs, as well as fiscal transparency activities. There is a dedicated web page for citizen participation in policy formulation and execution. The Documents section includes web links to all historical and current data about the budget plans and execution performance obtained from Sistema de Información para la Gestión Financiera del Estado (SIGFE) databases. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 Good Practices 73 Colombia The Ministry of Finance and Public Credit/Ministerio de Hacienda y Crédito Público (http://www.minhacienda.gov.co) maintains an informative and easy-to- navigate website to present substantial information about public finances. Links to SIIF Nation and a Transparency Portal provide access to detailed information about central and local budget activities, including the performance of budget execution, revenue collection and spending details, progress in investments, and the details of all major contracts signed. Ecuador The Ministry of Finance/Ministerio de Finanzas (http://www.finanzas.gob.ec) home page presents comprehensive information about the status of PF and his- torical trends in an easy-to-understand format. Links to SIGEF (Sistema Integrado de Gestion Financiera) and the Government Results/Gobierno por Resultados portal provide access to detailed information on sectoral and depart- mental spending and revenues. The Ministry website includes a large amount of multimedia content (YouTube, presentations, SIGEF e-Learning platform) to inform citizens and civil society with clear and easy-to-understand messages. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 74 Good Practices Mexico The Ministry of Finance and Public Credit/Secretaría de Hacienda y Crédito Publico website (http://www.shcp.gob.mx) presents extensive information about PF, budget revenues and expenditures, treasury operations, regulations, and bud- get transparency. A number of feedback mechanisms are provided for citizens and businesses through a transparency portal and access to information (InfoMex) sites. Information on federal government contracts, reports, wages, regulations, subsidies, services, concessions, and permits issued are also provided. Dynamic query options are available to present reports from PF databases. Middle East and North Africa Morocco The Ministry of Economy and Finance/Le Ministère de l’Economie et des Finances portal (http://www.finances.gov.ma) provides access to key PF informa- tion and timely updates on state budget execution, regulations, and public debt. A gender-responsive budget section and links to PF agencies also present open budget data. The General Treasury/Trésorerie Generale website presents the details of public spending and taxes, along with e-services and forms. Historical data are available from the databases and documents section. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 Good Practices 75 South Asia India The Ministry of Finance regularly presents comprehensive data (http://www. finmin.nic.in) on the Indian state budget, PF statistics, public debt, tenders, and monthly economic reports. The government accounting and reporting functions are supported through the e-Lekha solution, developed by the National Informatics Center for the office of the Controller General of Accounts. Most of the budget execution reports are produced and posted from e-Lekha. Reports posted on the MoF website can be viewed or downloaded in PDF. Visibility of FMIS Solutions on the Web This section highlights some of the websites that present the key features of the FMIS that is used in publishing PF information. Africa Madagascar The Ministry of Finance and Budget (MoFB) maintains a dedicated website for introducing the Integrated Public Financial Management System/Système Intégré de Gestion des Finances Publiques (SIGFIP). The website describes the functional modules and users of the system, provides web links to other Public Finance Management (PFM) systems, and explains the history of the system’s development and the units responsible for managing the system. The MoFB maintains the SIGFIP platform to facilitate data exchange among a number of fragmented information systems. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 76 Good Practices East Asia and Pacific Australia The Central Budget Management System website presents the details of Australia’s new integrated FMIS development process. The background of the project and current status are explained in detail, as are the expected benefits and the changes introduced through the new system. The website provides timely feedback to all stakeholders about the stages of system implementation, and supports the change management activities. Presentations and various feedback mechanisms are also available to improve communication and ­ coordination­. Indonesia The Ministry of Finance’s dedicated website (http://www.depkeu.go.id) presents the main features and development phases of the SPAN (State Treasury and Budgetary System/Sistem Perbendaharaan dan Anggaran Negara) solution, Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 Good Practices 77 which is expected to be operational in 2014. The SPAN website, built on a con- tent management platform, provides useful updates on massive change manage- ment activities, as well as blog posts, news, and announcements of upcoming events. When SPAN is operational, it will be the source of PF information. Republic of Korea The Digital Budget and Accounting System (DBAS) website (https://www.dig- italbrain.gov.kr) provides access to extensive PF information and open budget data. DBAS or dBrain system modules and the history of system development are presented in detail. Information about the PF policy, medium-term fiscal plan, budget system, government securities, cash management, procurement/ tenders, and financial statistics is presented in a user-friendly format and updated regularly. Citizens/businesses can submit questions and download reports in various formats. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 78 Good Practices Malaysia The Government Financial and Management Accounting System (GFMAS) is the accounting system used by the Accountant General Department since 2006. The GFMAS website presents the capabilities of this integrated system (financial planning, budget control, and government accounting). Salary management, gov- ernment loans, investments, and the preparation of public accounts are also covered. Online statistics system and data-mart sections provide useful ­ information about the status of budget execution, as well as historical PF data. Europe and Central Asia France CHORUS is an integrated FMIS solution (based on SAP) that supports financial, budgetary, and accounting management at central and decentralized levels. The CHORUS website presents detailed information about the history of development and system functionality. The system provides timely and reliable PF information to all departments and programs. Another important source of PF information is the Open Government Data portal (http://www.data.gouv.fr). Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 Good Practices 79 Russian Federation The Ministry of Finance uses a dedicated website to regularly present develop- ments related to the design and implementation of a new integrated FMIS solu- tion (e-Budget). The legal and regulatory framework, status of implementation, work plans, system functionality, and working group activities are published, together with updates on the development of a single budget portal, state and municipal payment system, and other components. The new e-Budget system is expected to be operational in 2016. Currently, most of the open budget data is published from Federal Treasury Automation System databases. The United Kingdom The Online System for Central Accounting and Reporting (OSCAR) website pro- vides access to extensive open budget data extracted from a public spending data- base. This data set makes public spending data more directly accessible and can be used with standard spreadsheet software. OSCAR is designed as a user-friendly system to provide the Treasury with key management information and data for public reporting. The system appears to be fully operational, and data covering the first six months of 2012–13 (April-September) have recently been published. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 80 Good Practices North and South America Argentina The Sistema Integrado de Información Financiera (SIDIF) web page provides detailed information about Argentina’s FMIS. The system’s conceptual model, functionality, scope, history of development and upgrades, training manuals, and interfaces with other systems are presented clearly. The website also has options for requesting information, following SIDIF events, and participating in relevant courses. The SIDIF website is updated regularly with information about develop- ments and good practices. SIDIF is the source of the PF information that is pub- lished on the MECON website. Bolivia The Ministry of Economy and Public Finance/Ministerio de Economía y Finanzas Públicas (http://www.economiayfinanzas.gob.bo) home page is easy to navigate and provides access to personnel data, contracts, publications, legal framework, and links to other units. Detailed budget information since 2000 is published from SIGMA (Sistema Integrado de Gestión y Modernización Administrativa). In addition to improving transparency in resource management, SIGMA provides timely and reliable information, interacts with the planning system and public investments, and provides feedback to the results tracking system. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 Good Practices 81 Brazil SIAFI is the Integrated System of Financial Administration of Brazil’s Federal Government (Sistema Integrado de Administração Financeira). Operational since 1987 and thus one of the oldest FMIS solutions around, SIAFI was developed to support budget execution, accounting, and reporting. The new SIAFI is a web- based application (operational since January 2012) to which all SIAFI modules will gradually migrate. It is compliant with the Federal Government’s interoper- ability standards (e-ping). The new SIAFI includes the Accounts Payable and Receivable module. SIAFI websites present the system functionality, reports, and dynamic query options. The SIOP (Sistema de Presupuesto y Planeamiento) portal provides access to the Federal Government’s new budget preparation system (which is based on open source software and freely available to Brazilian states. Monthly, quar- terly, or annual budget reports can be downloaded for current and previous years. Chile Sistema de Información para la Gestión Financiera del Estado (SIGFE) has been operational in Chile since 2007. The new version is a hybrid solution combining budget preparation (based on COTS) and execution locally developed software (LDSW) modules to support key functions, including the daily monitoring of all revenues and expenditures, cash management, the monitoring of multiyear com- mitments and budget programs, and the management of documents and work- flow (supporting electronic documents). Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 82 Good Practices Nicaragua Sistema Integrado de Gestión Financiera (SIGFA), operational since 2002, was upgraded to a web-enabled system in 2009 to support budget preparation, exe- cution, accounting, and reporting. The Ministry of Finance is in the process of selecting the new web-based FMIS solution to support PFM reforms and new requirements. The SIGFA website presents useful information about the PFM reforms (since 1996), and about the e-SIGFA platform, which integrates SIGFA with other PFM systems. Venezuela The Integrated System for Administration and Control of Public Finances/ Sistema Integrado de Gestión y Control de las Finanzas Públicas (SIGECOF) supports budget preparation, execution, and reporting functions. The system is linked with the Debt Management System and Public Investment Management to provide useful information about budget execution performance through the Ministry of Planning and Finance website. The history of system development, legal basis, and digital library are accessible from the SIGECOF website. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 Good Practices 83 Middle East and North Africa Jordan The Ministry of Finance (http://www.mof.gov.jo) initiated the development of the Government Financial Management Information System (GFMIS, http:// www.gfmis.gov.jo) in 2005, to support budget preparation, execution, account- ing, and reporting. GFMIS development was completed in 2012, and the system is expected to be operational in 2013. The GFMIS website presents the history of system development and related publications and change management activ- ities. GFMIS is expected to be the source of PF information for publishing online. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 84 Good Practices South Asia India The Controller General of Accounts (CGA) maintains e-Lekha as a management accounting system for all levels. The system is mainly used for payments and pub- lic account transactions, centralized maintenance of the CoA, and cash manage- ment. The Central Plan Scheme Monitoring System (CPSMS) was introduced in 2011 to monitor all funds releases and payments countrywide (gradually expand- ing to all states by 2018), and included interfaces with Core Banking Solutions for payments. The CGA is now moving toward an integrated FMIS solution (GFMIS) to link budget preparation, execution, accounting, and reporting. Pakistan The Financial Accounting and Budgeting System, developed by Pakistan Audit Department under the PIFRA (Project to Improve Financial Reporting and Auditing), supports budget execution, accounting, and reporting. The PIFRA website provides useful information about the objectives, expected results, func- tional modules, cost, and components of the project. The first phase of the proj- ect was completed in 2004 and the second phase, including the rollout of a payroll module, in 2011. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 Good Practices 85 Access to PF Information through Dynamic Query Options This section describes good practice cases linked with access to PF data through dynamic query options to generate user-defined or predefined reports from FMIS databases. East Asia and Pacific China The Ministry of Finance maintains a comprehensive website for publishing PF data (http://www.mof.gov.cn/) on central and local government activities. Dynamic query options are available for generating reports on central budget revenues and expenditures, monitoring the performance of local budgets, and retrieving historical data on national financial accounts. There is also a section on public participation, as well as online services (to download forms and reports) and updates on ongoing PFM reforms. Timor-Leste The Timor-Leste Budget Transparency Portal provides extensive information on the execution of the national budget. The portal is updated daily from underly- ing FMIS databases, and is accessible to the public, civil society, and develop- ment partners. The portal provides a large number of dynamic query options to monitor the budget (plans, actuals, commitments, obligations), and reports may be downloaded in various formats (PDF, DOC, XLS, XML, and HTML). Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 86 Good Practices Europe and Central Asia Denmark The Ministry of Finance/Finansministeriet website (http://uk.fm.dk) includes a section on publications and a dynamic query window through which users can drill down on public spending details for each institution. The FMIS solution is the source of information for the PF database, and it is possible to select a spe- cific budget program level and entity to produce a listing of relevant spending (plans vs. actuals). The query filter is displayed, together with the level of report detail, at the top of the window. Displayed open budget data can be downloaded in CSV format. Estonia The Ministry of Finance/Rahandusministeerium website (http://www.fin.ee/ riigieelarve) presents the overall status of budget execution through consolidated results for the government sector, which consists of about 3,100 institutions, including nearly 2,700 local governments. The details of budget activities are available from the Statistics Estonia website and databases. Revenue, expenditure and general government debt can be monitored in detail from the Statistics data- base, with a rich set of query and filtering options. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 Good Practices 87 The Netherlands The Ministry of Finance/De Rijksoverheid home page presents, and regularly updates, a comprehensive set of Documents and Publications on budget perfor- mance. User-defined search/query options can be specified to extract relevant publications according to ministry or report types, and there are options to refine the search results. Budget documents, annual plans, guidelines, and other reports can be selected and downloaded as PDF or XLS files from the archives. Reports on the state budget, taxes, and other public finance information can also be obtained from the National Budget web page. Turkey The Ministry of Finance eBudget/eBütçe website provides access to extensive open budget data from PF databases. User-defined or standard reports can be generated online and downloaded in various forms (XLS, PDF). Dynamic query options are provided to specify the reporting period, type of report, and budget institutions, and to select desired budget classification segments, as well as the rows/columns to be displayed. The portal is used for public access as well as secure internal access by government officials. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 88 Good Practices North and South America Argentina The Ministry of Economy and Public Finance (MECON) has a dedicated web- page to present key public finance information, with dynamic query options. Selected open budget data can be displayed on the screen or downloaded (PDF, RTF) for further analysis. The reports indicate the source of information (Business Intelligence database) and include the date/time stamp for the creation of files. The MECON Information web page provides direct access to all important PF information, with clear explanations about the scope and period covered. Bolivia The Ministry of Economy and Public Finance provides dynamic query options for access to a number of budget documents—for example, executive branch financial statements (since 1991) and balance sheets and other reports of finan- cial institutions (since 2003). Another publication website provides access to more than 30 reports related to territorial finance departments since 2000. Open budget data can be displayed on the screen or downloaded in PDF or XLS ­formats. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 Good Practices 89 Brazil The SIAFI portal provides a number of dynamic query options to generate open budget data in HTML or Worksheet format. The month of the last database update is shown at the upper right corner, and the report type, year, and other parameters can be selected from available options. One of the query options for generating reports on personnel expenses or other expenses and capital costs is shown at right. Mexico The Ministry of Finance and Public Credit/Secretaría de Hacienda y Crédito Publico (http://www.shcp.gob.mx) provides access to extensive PF data through dynamic query options. Predefined reports cover the budget performance of all public sector, federal government, and budget organizations and the social secu- rity administration. User-defined reports can be generated by selecting the type of report, period, amount to be displayed, and presentation format. Reports can be displayed online from the PF databases and downloaded in PDF or XLS ­formats. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 90 Good Practices Paraguay The Ministry of Finance/Ministerio de Hacienda website (http://www.hacienda. gov.py) is easy to navigate and provides access to key PF information. An Online Services section includes several query options for access to state records, resource transfer requests, SIARE (the new version of the FMIS), the National System of Public Investments, Taxation, National Register, and Procurement. Query windows provide access to relevant databases, and the reports can be generated online (with download/print options). Peru The Ministry of Economy and Finance/Ministerio de Economía y Finanzas web- site presents comprehensive information about the budget performance through its Economic Transparency Portal/Portal de Transparencia Económica. A large number of dynamic query options are available to generate reports about budget execution, public and private investments, treasury operations, debt, and public accounting from the underlying FMIS. They can be displayed in various formats (XLS, Graphics, PDF). Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 Good Practices 91 United States The US Government launched the Open Government Initiative in 2010 to improve transparency, participation, and collaboration. The Office of Management and Budget launched USAspending.gov as a single searchable website, accessible to the public at no cost, to disclose the details of each federal contract award since 2007 (including subcontracts) and related obligations (not outlays or actual cash disbursements). The Performance.gov portal presents useful informa- tion about the performance of the government in selected areas. The US Department of the Treasury maintains a dedicated website on budget perfor- mance and publishes the Citizens Budget report. Uruguay The Ministry of Economy and Finance (Ministerio de Economía y Finanzas) portal provides access to key PF data through interactive query options. The General Treasury of the Nation (Tesorería General de la Nación) and the General Accounting Office (Contaduría General de Nación) publish relevant details regularly. A budget execution query section can be used to generate reports online from the underlying Integrated Financial Information System (SIIF). The budgets for the current and previous periods can be analyzed in detail. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 92 Good Practices South Asia India The Data & Statistics section of the websites of the Ministry of Finance and the Controller General of Accounts present dynamic query options to create detailed reports on public finances. Selected reports are displayed as a PDF file, with a system name and date/time. Detailed reports on state loan data, expenditure statements, audit reports, national PF summary, central government borrowings, and other areas can be downloaded. Publishing Rich Set of Open Budget Data This section highlights several good practices in publishing a rich set of open budget data. East Asia and Pacific Australia The Publications & Reports section of the Department of Finance and Deregulation website presents the key documents published in the last decade. There are well- established standards for web publishing, content management, open data, informa- tion security, ICT infrastructure, and guide to open source software, among others. Guidance for the Australian Government in publishing Public Sector Information is also available on the Open Government website (http://data.gov.au). Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 Good Practices 93 Republic of Korea The Ministry of Strategy and Finance DBAS website provides access to a rich set of open budget data. Reports can be produced directly from the DBAS data warehouse and displayed online according to selected budget classification seg- ments or other parameters. The download section presents a large number of display options on various operating systems in a range of formats (PDF, DOC, XLS, PPT, and so on) and languages. The budget status, consolidated results ­ (general, central, and local budget levels), and financial indicators are updated daily and presented along with news and economic reports. Taiwan, China The Directorate General of Budget, Accounting and Statistics (DGBAS) website presents extensive PF information obtained from the databases of the Government Budget Accounting system. The central and special budget, as well as city and local government budgets, can be monitored in various formats (XLS, PDF) from the DGBAS website, along with relevant laws and regulations, statistics, and other economic reports. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 94 Good Practices Vietnam The Ministry of Finance website presents (http://www.mof.gov.vn) extensive open budget data about the performance of state and local budget execution from existing information systems. A new Treasury and Budget Management Information System (TABMIS) is expected to be fully operational in 2013 to support the publication of open budget data. All budget reports, investments, and final accounts are regularly updated and displayed online, and displayed results can be downloaded in XLS format. Europe and Central Asia Ireland The Department of Public Expenditure and Reform initiated the Ireland Stat project in 2012 to publish open data about how public money is spent, allocated, and accounted for. The Government Performance Measurement website presents useful and meaningful performance information on three policy areas (Economy, Transport, Environment), and four more areas will be added by 2014 (Health, Education, Public Safety, Government). The pilot project sets out over 480 indi- cators across four categories of information (achievements, actions, costs, com- parisons). Regularly updated indicators can be viewed online and exported in PDF, XLS, or DOC formats. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 Good Practices 95 The Netherlands The open data portal of the Dutch government (https://data.overheid.nl) pro- vides information on public government data and the national Registry Open Data, with references to open data sets in government organizations. One of the open data sets is the State Budget (a summary of expenses, liabilities, and revenue by product of departmental budgets), which can be downloaded in CSV format. Other data sets provide information on education, national roads, legislation, and more. The Ministry of Finance/Ministerie van Financiën website provides access to additional open data on the state budget. Norway The Ministry of Finance/Finansdepartementet maintains a dedicated website for publishing comprehensive information about the performance of the State Budget/Statsbudsjettet (http://www.statsbudsjettet.no). Predefined reports can be displayed online and downloaded in several formats (HTML, PDF, XLS). In addition to detailed information about various aspects of the state budget, policy documents, government priorities, new rules and revised legislation, and press releases are posted regularly. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 96 Good Practices The United Kingdom The Combined Online Information System (COINS) is the database of UK Government expenditures, managed by HM Treasury. All reports are built cen- trally, mainly using SQL and VBA, for output in XLS/CSV (for any period since 2005). COINS has three data streams (forecasts, budget plans/outturns, audited results) and is used for main estimates, national statistics, public expenditure statistical analyses, performance monitoring, fiscal management, ad hoc informa- tion/reports, and public accounts. Data are provided by central government departments, which retain ownership of their data on COINS. North and South America Brazil The Historical Series section and other parts of the National Treasury webpages, and the Federal Budget portal, present a number of reporting options to produce a rich set of open budget data. Most of the reports can be viewed or downloaded in an editable Worksheet format. The Access to Information portal explains the procedures for access to various public sector websites and available open data. The Citizen Information Service/Serviço de Informações ao Cidadão (www.acessoainformacao.gov.br/sistema) provides access to open government data and other information. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 Good Practices 97 Chile The Budget Directorate/Dirección de Presupuestos (DIPRES) Publications web- site provides access to a rich set of open budget data in several formats (PDF, XLS, CSV, XML). Government financial statistics are published regularly, and key PF information on the operations of the central government and public enterprises is presented, together with budget execution updates (monthly/quar- terly) and government debt. Reports are also categorized in terms of the period, institution, and content type. Colombia The Integrated Financial Information System/Sistema Integrado de Información Financiera (SIIF Nation) provides timely and reliable information on consoli- dated results of the general budget, and exercises control over the budget execu- tion of the central government and decentralized units. A territorial assistance section includes an interactive map of all regions for easy access to relevant information. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 98 Good Practices Guatemala The Ministry of Public Finance/Ministerio de Finanzas Públicas (http://www. minfin.gob.gt) website includes several links to important PF data portals on which open budget data from SIAF (Sistema Integrado de Administración Financiera) for the central government and municipalities are available. The Local Government Portal presents the budget performance of municipalities through an interactive map, by posting reports dynamically from the SIAF Muni database. Reliability of PF Information Published on the Web The examples presented in this section highlight websites that include the sys- tem name and a date/time stamp as part of ensuring the reliability of open bud- get data generated from FMIS. Africa The Gambia The Integrated Financial Management Information System (IFMIS) was introduced by the Ministry of Finance and Economic Affairs (MoFEA) in 2007 ­ for budget preparation, execution, accounting, and reporting (including donor- funded projects). Since 2011, the system is used by all central government enti- ties. The IFMIS General Ledger reports include the system name and a date/time stamp. Monthly budget execution results are published on the MoFEA website. System name Date/time stamp Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 Good Practices 99 Zambia The Integrated Financial Management and Information System (IFMIS) is used for the preparation and execution of the state budget. Some of the budget reports include the system name and a date stamp to clarify the source of i ­nformation. System name (IFMIS) Date stamp East Asia and Pacific Republic of Korea The Digital Budget and Accounting System (DBAS) website provides access to extensive open budget data through dynamic query options. Although the DBAS system name and a date/time stamp are not visible on the budget documents posted, a large number of predefined reports are available for providing rapid access to reliable data from DBAS. Thailand The Government Financial Management Information System (GFMIS) website presents a rich set of open budget data on budget execution performance, key indicators, and other important aspects. Most of the reports generated from the system include a footer with the system name and a date/time stamp to clarify the source of information and date of publication. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 100 Good Practices Date/time stamp System name (GFMIS) Europe and Central Asia Slovenia The Ministry of Finance (Ministrstvo za finance) website provides access to key FMIS modules, the MFERAC Uniform Accounting System and SAPPrA web application for Budgeting and Analysis. Bulletins of public finances and other budget reports are generated from these systems, and they include a date/time stamp in the header section and the system name in the footer. System name Date/time stamp North and South America Bolivia The Sistema Integrado de Gestión y Modernización Administrativa (SIGMA) provides reliable information about the budget and performance management. Almost all reports include a header section with the system name indicated on the upper left corner, and a date/time stamp visible on the upper right corner. The date and time of reports from previous budget years reflect the latest pub- lication date and remain the same, demonstrating the consistency of records generated from the system. System name (SIGMA) Date/time stamp from SIGMA Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 Good Practices 101 Brazil The Sistema Integrado de Administração Financeira (SIAFI) is a web-based application providing reliable information from system databases through inter- active query options or archived documents. The system name and a date/time stamp are visible at the top of most budget reports. System name (SIAFI) Date stamp from SIAFI Dominican Republic The Sistema Integrado de Gestión Financiera (SIGEF) provides timely and reliable information on budget execution. The reports published on the Ministry website include a header section with system name and date/time stamp. Other monthly statistics based on the SIGEF database are also available from the Ministry website. System name (SIGEF) Date/time stamp from SIGEF Nicaragua The Ministry of Finance and Public Credit (Ministerio de Hacienda y Crédito Público) is using TRANSMUNI (Sistema de Transferencias Municipales) to man- age transfers to municipalities from the general budget. Almost all reports pub- lished through the TRANSMUNI website include a header section with the system name and a date/time stamp. System name (TRANSMUNI) Date/time stamp Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 102 Good Practices Middle East and North Africa West Bank and Gaza System name Date/time stamp The Financial Management Information System (BISAN) supports budgeting and financial management functions. The reports produced from BISAN about financial operations contain the system name and a date stamp at the bottom, together with additional explanations, to indicate the source of published data. South Asia India E-Lekha supports daily reporting of expenditure in sync with the budget allo- cated to a ministry and its subunits. e-Lekha supports near-real-time reporting, as well as financial monitoring and control. The reports published in the minis- try website include a footer section with the system name and a date/time stamp. System name Date/time stamp Quality of Presentation and Interactivity This section presents some good practice cases that demonstrate the use of inno- vative solutions and highly interactive websites for improving the quality of presentation of open budget data. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 Good Practices 103 East Asia and Pacific Australia The Department of Finance and Deregulation has a well-maintained and infor- mative Procurement website providing comprehensive information about all public tenders and contracts. A number of procurement guidelines are available, including specific documents designed to support ICT procurement activities. Annual procurement plans, notices, and contracts, as well as training/professional development opportunities, are updated regularly. Republic of Korea The Ministry of Strategy and Finance and Digital Budget and Accounting System (DBAS or dBrain) websites present extensive information through well-designed graphical user interfaces. A wide variety of open data display and download options are available, together with feedback mechanisms and social media links. A promotion video and other documents provide information about the dBrain modules, advanced budget performance monitoring functions, development his- tory, lessons learned, user surveys, international cooperation, and more. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 104 Good Practices New Zealand The New Zealand Treasury website provides access to extensive PF information through mobile applications and other interactive tools (Twitter, Facebook, YouTube, and so on). A new budget application (NZ Budget) was launched in December 2012 to present 2012 and 2013 budget data on mobile devices (iPad, iPhone, Android), together with budget-related videos, budget speech, executive summary, and key facts for taxpayers. A tablet version provides additional infor- mation on the Budget Economic and Fiscal Update and the Half-Year update. Philippines The Government of the Philippines initiated the development of the Integrated Financial Management Information System (GIFMIS) to strengthen the public administration and improve PFM performance and service delivery. Key PFM oversight agencies (Commission on Audit, Department of Budget and Management, Department of Finance, and Bureau of the Treasury) are improving their websites and online services to support these activities. PFM-related web- sites were updated substantially in 2012, and several important services (Citizens Portal, Budget 101, My Budget, PFM Portal, e-Payment) are available to citizens. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 Good Practices 105 Singapore The Ministry of Finance maintains a highly interactive and informative budget website for publishing timely open data from the FMIS data warehouse. Each budget year is presented in separate websites with comprehensive information about budget preparation, execution, and results. Budget 2013 mobile applica- tion (iPhone, iPad) allows citizens to access the latest information about the Singapore budget, as well as press releases and announcements, videos, a budget quiz, and an e-mail subscription for the budget speech. Europe and Central Asia Austria The Ministry of Finance (Bundesministerium für Finanzen) provides the Finance Online service through which citizens can file tax returns and request electronic notices, and companies can pay sales tax and income or corporation taxes. The Business Service Portal is another central electronic service platform for elec- tronic billing (mandatory from January 2014) and the registration of employees using digital signatures. Links to performance-based management, Pan European Public Procurement Online, and other services provide an integrated platform for improving the quality of public service delivery. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 106 Good Practices Finland Suomi.fi is the Finnish public administration’s one-stop service for citizens. The portal contains e-services and forms, links to relevant institutions, information packages, legislation, and news from public administration. A service map pro- vides contact and location information for public sector services. The Citizen’s Account service can be used to check the progress of an application or inquiry, send in additional information to support an application, and switch between the services of different organizations without having to register again. Russian Federation The Ministry of Finance is developing e-Budget as an integrated FMIS solution to improve service delivery and transparency, following the rollout of the Federal Treasury Automation System in 2012. As a part of ongoing reforms, a mobile application (Public Services) was developed for smart phones/tablets (on Android, iOS, Windows Phone) for checking tax obligations, applying for or renewing driver’s licenses, paying traffic fines, and other services. The MoF portal also provides information on reaching retirement, migration, obtaining grants and social assistance, and acquiring real estate. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 Good Practices 107 Spain The Ministry of Finance and Public Administration (Ministerio de Hacienda y Administraciones Públicas) maintains a dedicated portal for Autonomous Communities to provide information about local budgets, and access to online services. Citizens and users can log in with digital certificates to use online ser- vices. A list of all electronic services and information systems is available, together with a geomapping tool for displaying the latest data on central and local budgets. Turkey The Public Expenditure and Accounting Information System/Kamu Harcama ve Muhasebe Bilişim Sistemi (KBS) portal was developed by the General Directorate of Public Accounts/Muhasebat Genel Müdürlüğü to provide access to key FMIS functions and online services by more than 200,000 public employ- ees in 60,000 central and local entities. In addition to providing various expendi- ture management modules, the system is used to manage all personnel records and automate payroll calculations/payments. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 108 Good Practices North and South America Brazil The Comptroller General of the Union has an informative website dedicated to raising public awareness and encouraging monitoring of budget spending. The “Get Smart in Public Money/Olho Vivo no Dinheiro Público” program is designed as a learning platform for citizens to monitor the use of public resources. Local leaders, councils, local government officials, teachers, and students are informed about the importance of transparency and accountability in the public sector, and compliance with legal provisions (including a distance education option). Chile The Budget Directorate/Dirección de Presupuestos (DIPRES) maintains an easy- to-use Documents website providing access to open budget data related to pub- lic finance and economic indicators, as well as laws and regulations, presentations, speeches, and statistics. Each PF site has a specific web page on transparency, providing easy access to relevant links on access to information, regulations, pro- curement notices, contracts, and feedback provision mechanisms. Procurement Agency/Dirección Chile Compra also has a very informative website. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 Good Practices 109 Colombia The Fiscal Transparency Portal/Portal de Transparencia Económica presents inter- active query options (http://www.pte.gov.co) to display the budget execution performance in different sectors and provide historical data on budget programs, investments, and the medium-term budget framework. Most of the websites sup- port open budget data in various formats (XLS, PDF, HTML). The portal includes all important web links to PF institutions for presenting additional infor- mation on procurement, debt, assets, and the details of sectoral and regional spending. El Salvador The Ministry of Finance/Ministerio de Hacienda presents substantial information about revenues, expenditures, investments, public debt, and procurement, as well as statistics on public finances, human resources, and foreign trade through the new Fiscal Transparency Portal. The budget monitoring section includes a snap- shot of the execution rate for various institutions (plans vs. actuals) with regular updates. A large number of interactive query options are listed to generate and download desired reports as open data (XLS, PDF). Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 110 Good Practices Mexico The Ministry of Finance and Public Credit/Secretaría de Hacienda y Crédito Publico maintains a highly interactive and easy-to-navigate Budget Transparency/ Transparencia Presupuestaria portal. The portal presents detailed information about public investments, public finance, performance evaluation system, Citizens Budget, and federal agencies. Geomapping of relevant data provides useful feedback on the details of spending (who, why, where, how). The Performance Evaluation System/Sistema de Evaluación del Desempeño is avail- able to track the performance of public policies and budget programs and verify compliance with goals and objectives. There is also a dedicated website for Financial Education/ Educación Financiera to support online learning. Effective Use of Open Budget Data This section highlights government websites that present open budget data effec- tively so that citizens/civil society can readily monitor the performance and transparency of budget spending. East Asia and Pacific Republic of Korea The Ministry of Strategy and Finance DBAS website provides extensive informa- tion about the details of budget spending. The Korean Institute of Public Finance (KIPF) developed the InfoGraphic website to present open budget data in a format that would help taxpayers understand how public money is being spent. The KIPF website presents the results of policy-oriented research on taxation, public budgeting, and state-owned enterprises across various levels of the govern- ment, and it assists the government in formulating and implementing public policies. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 Good Practices 111 Philippines The Public Financial Management reform website presents detailed informa- tion about ongoing activities, including accounting and auditing reforms, GIFMIS implementation, improvements in cash management through central- ized treasury single account operations, and other activities. New features introduced by the Department of Budget Management (DBM) include the online submission of budget proposals. The DBM’s Budget ng Bayan website presents the whole budget process and performance indicators, as well as open budget data. Singapore The Ministry of Finance website presents substantial information about the Singapore Government Budget, with a clear explanation of key budget initiatives to benefit households and businesses. Citizens and civil society can share their feedback, and the MoF responds on the web, summarizing all feedback received and explaining how it was incorporated in the preparation of the new budget. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 112 Good Practices Video highlights and relevant budget documents/speeches provide useful addi- tional feedback on budget planning/execution process. Europe and Central Asia Germany The Ministry of Finance/Bundesministerium der Finanzen provides access to federal budget information through websites, social media, and mobile applica- tions. The interactive website presenting the Federal Budget/Bundeshaushalt (http://www.bundeshaushalt-info.de) revenues and expenditures (according to budget sections, functional category, and the spending groups of all govern- ment institutions) demonstrates good practice. Results are displayed both graphically and in tabulated form, and can be downloaded as open budget data (PDF, XLS). Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 Good Practices 113 Offener Haushalt is a noncommercial project designed to visualize, analyze, and comment on the federal and local budgets, using data available from the Federal Ministry of Finance website and other relevant sources. Norway The Directorate for Financial Management (DFØ) maintains an easy-to-navigate and innovative portal to provide core PFM services for government agencies through shared platforms. DFØ is responsible for managing national accounts and for maintaining the standard CoA, accounting standards, and the govern- ment’s cash. The DFØ manages payroll, accounting, e-Commerce, e-Invoice, and other services, and the operational status of all systems (Agresso, EFB, SAP, Integration Engine) can be monitored online. DFØ also organizes forums for the exchange of experiences and disseminates good practice through professional networks. To provide meaningful information on public finances, the Ministry of Finance has developed a web portal (http://www.ungokonomi.no) that is easy for citizens and civil society organizations to understand and navigate. It provides regular updates on economics and budget issues. Russian Federation The Ministry of Finance maintains a website for regular updates on key perfor- mance indicators about the budget system. Budget data (federal and consolidat- ed), macroeconomic indicators, and government programs are presented through graphs/charts, and can be downloaded in PDF or XLS formats. The Federal Treasury has developed a useful data-mart system to monitor and publish key indicators related to budget execution. Another important web platform is the procurement portal, where all procurement notices, contracts, and suppliers related to federal and local-level tenders are posted and updated regularly. A new Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 114 Good Practices budget portal was launched in early 2013 (www.budget.gov.ru—test version) to publish a rich set of open budget data on federal and regional revenues and expenditures, investments, intergovernmental transfers, procurement activities, and more. There is a discussion forum through which citizens and civil society can provide feedback on possible improvements. MoF > Key Performance Indicators FT > Data-mart Budget portal Turkey The General Directorate of Public Accounts of the Ministry of Finance provides access to extensive PF information based on FMIS (say2000i and KBS) databases. In addition to secure access to KBS, automated payroll calculations (e-Bordro), revolving funds, and statistics, the portal presents service quality standards, regu- lations, and the latest news. Regular updates on local administration budgets and other important indicators are published as graphics and open budget data (XLS, PDF). A civil society website was launched in 2010 (Public Expenditures Monitoring Platform/Kamu Harcamalarını İzleme Platformu) to monitor public expendi- tures and send reports to the parliament about possible improvements in public spending allocations/priorities and policy, based on the budget plans and actual spending data published on the MoF websites. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 Good Practices 115 The United Kingdom “Where Does My Money Go?” was launched in 2007 to promote citizen engage- ment and transparency through analysis and visualization of data about UK public spending. This project, expanded as the “Open Spending” initiative, now includes 212 PF data sets from 53 economies. The graphical user interface explains how tax revenues are divided among the different units, and how much is spent for various functions in total and where. A Country & Regional Analysis section includes links to similar transparency websites. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 116 Good Practices North and South America Argentina The Ministry of Economy and Public Finance maintains a website for citizens and civil society to improve the transparency of public management. Key infor- mation about the execution of the national budget is presented in simple language. Citizens can also access budget data through charts, graphs, and tables ­ for additional details on budget revenues and expenditures, and for comparison with previous periods. Dynamic query options are available to generate selected reports online, with a system date/time stamp and user-friendly presentations. A survey form is also included for receiving feedback from visitors. Dominican Republic The Citizens Portal/Portal del Ciudadano (http://www.portaldelciudadano.gov. do) presents comprehensive information about the execution of the budget based on SIGFE databases and benefiting from business intelligence queries. A number of quick access queries are available: What is spent? Who spends? How is it spent? and Where is it spent? Links to relevant public institutions are includ- ed, along with feedback provision forms. Most of the reports are generated (and can be reproduced) from the SIGEF databases with a clear indication of the system name and a date/time stamp. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 Good Practices 117 Brazil The Brazilian Transparency Portal/Portal da Transparência is one of the fine examples of presenting meaningful PF information to citizens and civil society organizations. It provides a number of dynamic query options on all important aspects, and includes links to other government portals: Federal Budget, National Treasury, Public Accounts, the procurement portal, the Planning and Management Information System, the transparency portal of the Chamber of Deputies, con- stitutional transfers, and more. Transparency Portal Transparency Portal of the Deputies Auditor of Public Accounts Open Government Data Transparency in States and Municipalities Mexico The Budget Transparency/Transparencia Presupuestaria portal of the Ministry of Finance and Public Credit uses the Citizens Budget to share meaningful infor- mation about budget spending, investments, public sector salaries, and other reference documents. Previous Citizens Budget reports are also available (since 2010) in PDF. Citizens Budget 2013 is also available as an iPad application. The Investment section uses interactive maps to present the annual distribution of budget to programs/projects of the federal portfolio by executing unit or by state. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 118 Good Practices FMIS World Map One of the important outputs of this study is the development of the FMIS World Map, which provides access to the web links (URLs) of PF websites in 198 economies, as well as 176 FMIS supporting key PFM operations in these coun- tries. The FMIS World Map is a geospatial mapping application developed on Google Maps; it benefits from free open-source data conversion software (XLS to KML) and a free public mapping option provided by Google. Icons with a letter (A to D) are used as place marks (on capital cities) to reflect the FMIS & OBD group of 198 government websites. When a user selects one of the icons on the map, the following basic information is displayed, along with any Open Government/Open Data web link (Map 4.1). Country Name ... FMIS abbreviation [Functional scope: F/T] Location : Capital city [Income level] System : Full name of the FMIS solution in native language Group : A - D [FMIS & OBD practices] Status : FMIS operational status  [Since: year] Web links  MIS, Finance Ministry/Dept, Central Bank, Statistics, : F and Open Gov/Open Data URLs. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 Good Practices 119 The Beta version of the FMIS World Map was developed in June 2012 and updated several times before the completion of this study in June 2013. The FMIS World Map will be updated annually to improve the visibility of the find- ings, and to provide easy and open public access to good practices (Map 4.2). Map 4.1  Basic information and web links displayed on FMIS World Map Source: World Bank data. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 Map 4.2  FMIS World Map (as of June 2013) 120 Source: FMIS World Map. http://maps.google.com/maps/ms?ie=UTF&msa=0&msid=213542822110887565899.0004c2f44512d9ce6795f Good Practices 121 Note 1. The DBAS/dBrain (IFMIS solution of the Ministry of Strategy and Finance of the Republic of Korea) is the winner (first place in Category 4, EAP region) of the 2012 United Nations Public Service Awards (UNPSA) for promoting the whole-of-govern- ment approach. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 CHAPTER 5 Guidelines for Publishing Open Budget Data Governments’ interest in posting open budget data on their websites has been growing over the last decade, largely because citizens and civil society groups are increasingly demanding information with which to monitor the details of public revenues, expenditures, and debt, as well as the results of government spending and investments. Although some governments have improved the timeliness, quality, and scope of their public finance (PF) reporting, most do not pay enough attention to some of the basic principles for publishing meaningful budget data that can be downloaded, analyzed, and evaluated by citizens and civil society, and for ensuring the source and reliability of the information they publish. Guidance on publishing reliable open budget data from underlying financial management information systems (FMIS) solutions is scarce. These guidelines, prepared to help fill this gap, have three main purposes: •• To assist governments in understanding the key principles for posting reliable open budget data from underlying FMIS solutions when they are developing new PF websites/portals, or improving the contents, functionality, format, or presentation quality of existing websites •• To help civil society organizations learn more about effective monitoring of the budget planning and implementation process and the performance of the government; and to expose them to other country experiences that they can use to improve citizen participation in their own countries •• To assist oversight agencies in strengthening their ability to perform their inter- nal/external oversight role during budget implementation by ensuring that the PF information presented is reliable, and that transaction-based evidence is available from underlying information systems for consistent reporting. The guiding principles for posting reliable open budget data are summarized below: •• Availability of timely and comprehensive budget information •• Disclosure of details about underlying information systems Financial Management Information Systems and Open Budget Data   123 http://dx.doi.org/10.1596/978-1-4648-0083-2 124 Guidelines for Publishing Open Budget Data •• Availability of user-defined (dynamic) query and reporting capabilities •• Publication of reliable and interlinked open budget data •• Authentication of the sources of public finance data •• Improving the quality of presentation •• Promoting the effective use of open budget data. The discussion of each of these principles in this chapter can be supplemented by reference to some of the good practice cases. The FMIS & Open Budget Data (OBD) data set and the FMIS World Map present evidence to assist in the com- parison of similar practices across countries. Availability of Timely and Comprehensive Budget Information The design of official websites dedicated to the publication of PF information should take into account the following factors: •• Existence of dedicated websites for access to PF data. Frequent changes in the uniform resource locator (URL) of PF websites should be avoided. When changes are necessary (for example, because of organizational restructuring, or the merging of two institutions), relevant web links should be maintained for a while with clear explanations, to ensure the continuity of reference URLs. •• Timely and regular publication of budget plans and execution results. A website can be created to present the publication schedule (name, update frequency, publisher, and so on) of key PF documents. Budget plans, execu- tion reports, and other reports should then be posted on the designated web- sites according to announced schedules (for example, weekly or monthly). Governments should also have in place clear procedures for updating and archiving the budget data. •• Completeness of published PF information. Detailed PF information should be posted during the budget year to provide a comprehensive, updated picture of the government’s financial activities, as well as those of state/local governments, when applicable. (See key budget documents suggested in the Open Budget Survey.) The IMF Data Quality Assess- ment Framework for Government Finance Statistics is a useful guideline that comprehensively covers the quality aspects of data collection, pro- cessing, and dissemination.1 •• Presentation of budget execution performance through time-series data. Access to online budget archives presenting the policies, PF information, and ­ major achievements should be provided to facilitate comparative analysis and observation of the variations in transparency and accountability over time. Disclosure of Details About Underlying Information Systems The dedicated websites should explain key aspects of the PFM information sys- tems, along with important functional/technical capabilities, standards, and policies. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 Guidelines for Publishing Open Budget Data 125 •• Presenting the key features of underlying information systems. The website should contain adequate details about the functionality, operational status, technical architecture, scope, number of users, and other important aspects of the underlying information systems (FMIS, data warehouse [DW], or other). •• Promoting the use of interoperability standards and digital signature. The government’s interoperability standards—policies and specifications governing the use of ICT in the government, as well as the rules/formats of interaction among government entities—need to be defined clearly on the dedicated web- site. Public finance management (PFM) operations should use a digital signa- ture as part of ensuring the quality and reliability of the services delivered. •• Disclosure of data protection and information security policies. For the best- managed PFM information systems, relevant publications or FMIS websites should clearly describe the government’s data protection and information ­ security policy and practices. The requirements to protect the confidentiality of personal or classified information should be considered while posting open data. Availability of User-defined (Dynamic) Query and Reporting Capabilities User-defined (dynamic) query tools developed on FMIS databases or DW solu- tions are becoming standard features of integrated PFM systems. •• Capability to analyze multidimensional data interactively. Integrated FMIS solutions should be designed to link operational systems (OLTP—supporting online transaction processing and preserving data integrity, with minimal back-end reporting options) with powerful DW solutions (OLAP—support- ing online high-volume analytical processing and elaborate report generation) to provide interactive query options to analyze multidimensional data from different perspectives. The update frequency of key databases should be dis- played in dynamic query websites to clarify the refresh schedules. •• Flexible and user-friendly dynamic query, reporting, and download options. Interactive query and reporting platforms linked with the underlying FMIS database or DW are ideal for presenting relevant information according to user preferences. Query results can be displayed online in various forms (graphics, charts, tables, text) or downloaded in desired open data formats (for example, XML, CSV, XLS, ODF, DOC) to support internal and external re- porting needs. •• Consistency of historical data. It should be possible to reproduce the gov- ernment financial statistics or other official reports of previous budget years by extracting historical data from integrated FMIS or DW solutions, as a verification of the integrity and reliability of underlying databases. Separate presentation of historical data (frozen) and operational reports (regularly up- dated during the budget year), and proper explanations about the stability and consistency of the data presented, also help in the interpretation of posted results. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 126 Guidelines for Publishing Open Budget Data Publication of Reliable and Interlinked Open Budget Data Publishing open budget data (free, online, editable) from DW solutions linked with FMIS databases is very important to improve the accuracy and reusability of PF data. •• Publishing open budget data on the web requires a cultural change. Posting open budget data requires a change in the mindsets of politicians and govern- ment officials, who must be committed to increasing public confidence by allowing more visibility into operations. This is both an adaptive and a techni- cal challenge for PF officials and ICT specialists, who should manage this change effectively to ensure that their motivations are properly understood and supported. •• Benefiting from the guidelines on publishing open data. A number of guide- lines define the minimum requirements and web publishing standards for open government data—for example, Australia, Brazil, New Zealand, EU Public Sector Information Platform, and The World Bank Open Data. The World Wide Web Consortium has developed specific guidelines to help gov- ernments open and share their data, emphasizing the importance of metadata to clarify the structure of posted data and related standards (for example, Resource Description Framework, or RDF). The World Bank released the Open Government Data Toolkit in 2012, to provide staff and government officials a basic set of resources for initiating and developing an open data program. The toolkit includes five components: (a) Open Data Essentials; (b) Technology Options; (c) Demand and Engagement; (d) Supply and Quality of Data; and (e) Readiness Assessment Tool. Another useful reference is the Open Data Handbook, which presents the legal, social, and technical aspects of open data. These resources can very helpful during the development of open budget data portals. •• Open budget data creates opportunities to add value to public information. As a part of FMIS modernization efforts, some governments are publishing open budget data to provide opportunities for new products and improved service delivery by adding value to PF data, which is difficult and expensive to capture. PF websites should provide multiple access options, including full downloads and Application Programming Interface (API) for developers. •• Paying attention to the legal aspects of open government data. Open data should ideally be license-free (that is, not subject to any copyright, patent, or trademark). However, reasonable privacy, security, and privilege restrictions may be acceptable. The legal aspects of open budget data should be clarified on government websites, either by using existing licensing/legal options (for example, Creative Commons licenses) or by defining country-specific legal requirements. Existing open data licensing options include (a) Public domain (no rights reserved: CCo); (b) Attribution (credit must be given: CC-BY); and (c) Sharealike (data should be shared back: CC-BY-SA). Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 Guidelines for Publishing Open Budget Data 127 Authentication of the Sources of Public Finance Data Visibility of a system name and a date/time stamp on published reports is one of the key indicators for the reliability and integrity of underlying information ­systems. •• Displaying system name and date/time stamp on official reports. The source of information should be clearly visible on all official reports/publications posted on PF data websites. In addition to the system name and the date/time stamp of generated reports, the version of the related FMIS should also be visible. •• Safeguards to protect information from unauthorized modification or access. The integrity of PF information should be ensured by implementing appropri- ate safeguards to protect data from unauthorized modification and access. Also, the availability of information should be secured by preventing the de- nial of authorized access. Privileged access rights should also be monitored by the information and communication technology (ICT) risk and compliance units. Moreover, necessary oversight mechanisms should be in place to ensure the reliability and integrity of databases, the security of operations, and the ef- fectiveness of information technology (IT) governance and oversight functions. Improving the Quality of Presentation Interactive data visualization options, graphical user interfaces, feedback mecha- nisms, and advanced search/reporting options substantially improve the quality of presentation in PF websites. •• Creating searchable interactive maps of PF information. Interactive maps are very useful to present the details of PF information in a user-friendly for- mat (for example, sectors, regions, gender focus, programs/activities). A num- ber of visualization platforms are available (including open source software and free public versions of visualization packages) to rapidly post key PF data on searchable interactive maps (for example, Knoema, Tableau, RapidMiner, Ushahidi). •• Broadening access to PF data through mobile applications. Governments are increasingly sharing financial activities with citizens through mobile de- vices (tablets/smart phones). Some of the advanced applications provide on- line payment and search options as well. •• Providing daily updates on key performance indicators. Advanced applica- tions for data marts (subsets of data stored in a warehouse) can support daily updates on critical PF information. DWs should be updated regularly (often daily) from the operational systems (FMIS databases) to support the timely presentation of selected indicators through data marts (see Figure 5.1). Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 128 Guidelines for Publishing Open Budget Data Figure 5.1  Publishing timely and reliable open budget data from FMIS solutions Budget classification / Unified chart of accounts shared by all budget levels Capture all transactions (OLTP) >>> daily (revenues/expenditures + assets/liabilities + commitments) Consolidate and verify all PF data >>> daily/weekly (accounting & financial reporting) Data Warehouse (OLAP) >>> daily/weekly (plans/results + monitoring + decision support + publish + audit) Publish PF Data (Open Budget Data) >>> weekly/monthly (national + international std. > GFS 2001, COFOG, and so on) Source: World bank data Note: OLTP = Online Transaction Processing; OLAP = Online Analytical Processing; PF = Public Finance; GFS = Government Finance Statistics; COFOG = Classification of the Functions of Government. Promoting the Effective Use of Open Budget Data Publishing meaningful open data on budget revenues, spending, and other finan- cial activities is crucial for any government to explain how public money has been spent. The Citizens Budget is an important instrument to achieve this. •• Measuring the government’s financial performance. Governments should regularly publish on the web the supreme audit institution’s audit reports on the financial performance of the government in the previous budget year. Oversight agencies should have read-only access to FMIS databases, linked with DW solutions, so that they can dynamically monitor and audit the bud- get execution performance. •• Availability of the Citizens Budget and feedback mechanisms. A Citizens Bud- get is a website or document that explains basic budget information using simple and clear language. Using a regularly updated website to present the Citizens Budget is an important indicator of the government’s commitment to improving budget transparency and participation. The International Budget Partnership (IBP) has published a guide on Citizens Budgets (“The Power of Making It Sim- ple,” May 2012), offering useful tips to governments interested in developing a Citizens Budget. The Citizens Budget should be prepared in consultation with citizens about what they would like to know about the budget. Feedback ­provided by the citizens and relevant adjustments should be posted on the Citizens Budget websites to complement the publication of meaningful open budget data. •• Promoting public consultation and participation in the budget process. Par- ticipatory budgeting is a process (different from the Citizens Budget) through which people in a locality or community can participate in budget decision making and affect the government’s budget. The publication of open budget Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 Guidelines for Publishing Open Budget Data 129 data should be complemented by engaging the public in policy making and monitoring. Detailed discussions about public participation/engagement are offered in “The Core Principles for Public Engagement”2 and “Deliberative Public Engagement: Nine Principles.”3 •• Monitoring budget implementation through timely information. Civil society groups play an important role in monitoring and analyzing the government’s budget on the basis of the information available in PF websites, and publish- ing reports in simplified forms to enable citizens to understand what the gov- ernment is doing with their money. As the IBP’s guide on Citizens Budgets explains, in recent years a few governments have taken a similar task upon themselves—for example, El Salvador, Ghana, India, New Zealand, and South Africa. Recently, new budget portals have been launched (for example, Russian Federation) to present substantial information on key performance indicators with daily updates. To improve the monitoring of budget imple- mentation, governments should make meaningful open data available to citi- zens and civil society groups, with timely updates (daily/weekly). •• Demonstrating meaningful results. The implementation status of govern- ment policies and plans should be presented clearly, together with the results of operations, to increase the transparency of government actions. In addition, the results of participatory budgeting, gender focus, or citizen-led expendi- ture monitoring should be published to demonstrate meaningful results. Notes 1. The guideline can be found on http://dsbb.imf.org/images/pdfs/dqrs_gfs.pdf. 2. http://ncdd.org/rc/wp-content/uploads/2010/08/PEPfinal-expanded.pdf. 3. http://www.involve.org.uk/wp-content/uploads/2011/03/Deliberative-public- engagement-nine-principles.pdf. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 CHAPTER 6 Conclusions What You See Is (Not Always) What You Get The average score for the performance of 198 governments in publishing open budget data from financial management information system (FMIS) is 45.1 out of 100, based on 20 key indicators. About 93 websites that present extensive or significant information appear to be benefiting from underlying information systems while publishing public finance (PF) data, but most of them do not yet provide open data. Overall, only 48 countries (24 %) offer PF information well enough that civil society and citizens can rely on it to monitor the budget and hold the governments accountable. Also, in many countries the internal/external over- sight agencies do not appear to be using the FMIS platforms effectively while monitoring the government’s financial activities or auditing the budget results. The findings of this study indicate that developing robust FMIS solutions as the source of reliable open budget data and measuring the effects of FMIS on budget transparency continue to be major challenges in many countries. This chapter summarizes the findings and conclusions of the study. Findings The following research questions were addressed during this exercise: 1.  What are the important characteristics of current government web publishing platforms designed for the disclosure of budget data? By analyzing data collected through 40 indicators, the team found that web publishing practices vary significantly among countries in different regions of the World Bank and at different income levels. Financial Management Information Systems and Open Budget Data   131 http://dx.doi.org/10.1596/978-1-4648-0083-2 132 Conclusions  The status of 198 government web publishing platforms around the world is as follows: •  24 governments (12 %) follow most of the good practices for publishing open budget data from reliable FMIS solutions. •  69 governments (35 %) provide significant budget information, but only a small portion of these publications qualify as open budget data from FMIS. •  60 governments (30 %) provide some information in their PF websites, mostly from archived documents without enough evidence on the use of FMIS databases as the source of published PF data. •  45 governments (23 %) post minimal or no budget information on the web.  Indicator 1: Most of the governments (166 out of 198, or 83.8 %) have dedi- cated websites to publish PF information, and 125 of these (63.1 %) include a link to budget data clearly visible from their home pages.  Indicator 2: About half of the governments (92 out of 198, or 46.5 %) have dedicated websites providing useful information about the functionality and current status of FMIS platforms. Also, 83 governments publish some reports about FMIS functionality/scope, but 23 present no information about FMIS.  Indicator 3: Only 34 governments (17.2 %) have dynamic websites, and 12 of these provide access to extensive information through interactive queries, mostly linked with FMIS. A large number of countries (132 or 66.7 %) maintain static websites presenting various documents from unidentified sources.  Indicator 4: Open budget data are visible in 52 economies (26.3 %), but only about half of these appear to be linked with FMIS databases.  Indicator 5: Only 18 governments (9.1 %) include the name of FMIS solution as the source of published information in budget reports.  Indicator 6: Only 28 governments (14.1 %) present PF data with a system time stamp as an evidence of direct dynamic links to underlying databases in budget reports.  Indicator 7: 69 governments (34.8 %) provide comprehensive information about budget performance in easy-to-understand formats. The quality of re- porting in a large portion of PF websites (49 %) is partially acceptable, but 32 websites (16.2 %) show little or no attention to presentation ­ quality.  Indicator 8: Regarding the general level of detail in budget execution reports, 148 economies (74.7 %) provide substantial data on various aspects, while the remaining 50 provide minimal or no information about the contents of pub- lications.  Indicator 9: Only 15 governments (7.6 %) provide interactive Citizens Bud- get websites through which citizens can gain access to meaningful budget data and provide feedback, and 33 governments (16.7 %) present useful ­ information for the citizens only on budget spending. Most of the governments (75.7 %) do not provide meaningful budget information to ­ their citizens. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 Conclusions 133  Indicator 10: 93 governments (47 %) provide Budget classification (BC)/ Chart of Accounts (CoA) details on their websites.  Indicators 11 and 12: 153 governments (77.3 %) publish their approved an- nual budgets on the PF websites, and 120 of them do so regularly (at least within the last five years).  Indicators 13 and 14: 103 governments (52 %) publish their multiyear plans or MTEF, and 64 of them regularly update their plans (revising them every year, at least within the last five years).  Indicators 15 and 16: Only 44 governments (22.2 %) publish public invest- ment plans separately from their annual or multiyear plans, and 32 do so regularly.  Indicators 17 and 18: 147 governments (74.2 %) publish budget execution results at various intervals, and 117 do so regularly.  Indicators 19 and 20: 75 governments (37.9 %) publish some reports on ex- ternal audit websites about budget execution performance, and 60 of them do so regularly. 2. Is there any evidence on the reliability of open budget data published from FMIS? The evidence from the first 10 key indicators shows that although the 198 economies use 176 FMIS solutions, many governments do not have comprehen- sive web publishing platforms to present reliable open budget data. Good prac- tices for the publication of open budget data from reliable FMIS solutions are highly visible in only about 24 government websites (12 %). 3.  Are there good practices demonstrating how open budget data from FMIS can improve budget transparency? There are some good practice examples of governments that present budget results in a meaningful way so that the citizens can understand where the money goes. All of these economies have dedicated websites presenting the Citizens Budget with regular updates. Some of them also present evidence about public consultations and participatory budgeting. Several publish the results of public consultation and relevant changes in the PF policy and planned investments. Some post the results of participatory budgeting, gender focus, or citizen-led expenditure monitoring, together with interactive monitoring options. Chapter 4 presents selected good practice cases and innovative solutions. 4. Why is a “single version of the truth” difficult to achieve in the budget domain? Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 134 Conclusions In the ICT world, the “single version of the truth” (SVOT) is a technical con- cept describing the data warehousing ideal of having either a single centralized database or at least a distributed synchronized database, that stores all of an organization’s data in a consistent and nonredundant form. Similarly, “single source of the truth” (SSOT) refers to the practice of structuring information models such that every data element is stored exactly once. In the PF/budget domain, these ideals are difficult to achieve, since the public finance manage- ment (PFM) operational systems and data warehouses are so fragmented and widely dispersed that it is very difficult to determine the most accurate version of the PF information at a specific point in time (except frozen historical data stored in data warehouses) (Figure 6.1). Some of the main challenges in capturing reliable open budget data are as follows: • Difficulties in developing unified BC/CoA data structures supporting the needs of all public sector entities • Defining BC/CoA segments in sufficient detail (optimization) to be able to record and report all economic activities with adequate disaggregation (cap- turing program/activity and sector/regional spending) • Capturing all PF transactions daily, and consolidating the results weekly/ monthly for accounting and monitoring needs • Converting country-specific PF data into IMF Government Finance Statistics (GFS), Classification of the Functions of Government (COFOG), or other internationally accepted reporting formats (bridge tables) consistently Figure 6.1  Source and scope of public finance data Public sector* Nonfinancial public sector General government Central government Budgetary central government Centralized Budget classification/chart of accounts database Extra budgetary units MoF > FMIS Social security funds State governments Interfaces Local governments Distributed databases Nonfinancial public corporations Financial public corporations • Financial monetray public corp. (Central Bank) • Financial nonmenatary public corp. * Institutional structure of “public sector,” as defined in the IMF government finance statistics 2001 manual Source: World Bank data. Note: MoF = Ministry of Finance; FMIS = Financial Management Information System. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 Conclusions 135 • Developing the necessary legal and regulatory framework to define the roles and responsibilities in capturing and publishing PF data on the web • Establishing the necessary oversight mechanisms to ensure the integrity and reliability of the FMIS databases used for publishing PF information. For complex systems, Master Data Management (MDM) can provide some practical solutions to improve consistency and control. When new systems are being developed or complete FMIS infrastructure and databases are being mod- ernized, it may be possible to achieve SVOT with careful design and by using some industry standards. If FMIS solutions can be effectively used to capture all transactions daily (at least at the central government level), and if related data warehouses can be updated from operational systems regularly (ideally every day), it may be possi- ble to come close to SVOT through integrated systems. Some of the advanced FMIS solutions have data warehouse components that operate very close to this ideal. 5. Can there be some guidelines to improve the practices in publishing reliable open budget data from FMIS? Using the findings of this study and the experience gained in the development of FMIS solutions funded by the World Bank in 60 countries since the 1980s, the team proposed several guiding principles (chapter 5) that should help govern- ments, citizens/civil society groups, and oversight agencies improve their prac- tices in publishing reliable open budget data from FMIS. The main findings of this study are based on the rapid assessment of 198 PF websites, mostly at the central government level. However, it is important to note that there are a number of PF publication platforms at state, district, or local government levels in more than 30 economies that have fiscal decentral- ization and state-/agency-level FMIS solutions (for example, Brazil, Canada, India, Italy, the United States). At these subnational levels, as well, there are a number of good practices in promoting transparency and participatory budgeting. To verify whether the findings of the study are consistent with key observa- tions from other fiscal transparency indices, the distribution of FMIS & Open budget data (OBD) scores was compared with several relevant fiscal transpar- ency instruments such as Public Expenditure and Financial Accountability (PEFA), Open Budget Index (OBI), and UN e-Government Development Rankings. It was found that the patterns are largely similar, and the FMIS & OBD scores correlate positively with the Public expenditure and financial account- ability (PEFA) indicators and OBI scores. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 136 Conclusions Concluding Remarks So, can we see where the money goes? The study shows that many governments publish substantial budget information on their PF websites, but the contents are (not always) meaningful to answer the question, “Where does the money go?” Only a few countries provide citizens, civil society groups, and oversight agencies with access to reliable open budget data from underlying FMIS solutions. Therefore, the main conclusion of this study is that what you see is (not always) what you get. Additional efforts are needed in many economies for building con- fidence in the budget data disclosed by the governments. However, there is rising demand from citizens and civil society for access to open government data about all financial activities, and many governments around the world are trying to respond to this democratic pressure. Widespread use of the Internet and web technologies is transforming public sector management in many economies. Any country—regardless of income level, geographic location, or infrastructure constraints—can follow good prac- tices in budget transparency if the government is committed to publishing open budget data from reliable FMIS databases (technology or capacity is not the main barrier). Effective budget monitoring mechanisms that benefit from existing FMIS platforms may contribute substantially to the improvement of budget transparency. Selected cases demonstrate that, even in difficult settings, innovative solutions to improve budget transparency can be developed rapidly, with a ­ modest investment, if there is political will and commitment from the ­government. The outputs of this study (key findings and conclusions, updated FMIS & OBD data set, selected good practices, guidelines on publishing open budget data from FMIS, and the FMIS World Map) are expected to provide a comprehensive view of the current status of government practices around the world, and to promote debates about the improvement of government PF web publishing platforms for disclosing reliable and complete information about all financial activities to support transparency, accountability, and participation. This study is not designed to address other challenges related to the collection and disclosure of timely and reliable PF information about off-budget fiscal and quasi-fiscal activities and contingent liabilities; analyze the contents of the pub- lished PF information in detail; or reach citizens and civil society groups around the world to learn more about user perceptions and the level of utilization of open budget data by various stakeholders. These important aspects can be explored in future studies. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 Appendix A Explanation for Indicators/Questions and Response Options To assess the effects of open budget data (published from financial management information systems (FMIS) solutions) on budget transparency, 40 indicators were defined in two categories: 1. Key indicators: 20 indicators from 10 factual questions about the source, scope, and reliability of open budget data. 2. Informative indicators: 20 indicators from 10 questions providing useful infor- mation about other important features. The key indicators are presented below, with a brief explanation about the measurement method, points, and evidence (links/uniform resource locator (URL) of relevant web pages) of observed characteristics. The informative indica- tors and related questions follow the key indicators. Financial Management Information Systems and Open Budget Data   137 http://dx.doi.org/10.1596/978-1-4648-0083-2 Table A.1  Key indicators 138 Questions # Indicators Point Response options Q1. Does the Finance Q1.1 PF home page Uniform Resource Locator (URL) http……….. Ministry/Department have Q1.2 Official name of the Finance Ministry/Department Official name of the Ministry/Dept a website or portal that is Q1.3 1 Is there a dedicated website for ­publishing PF information? 0 No dedicated to publishing pub- 1 Yes (not clearly visible from home page) lic finance (PF) information? 2 Yes (easy access from home page) Q1.4 If Yes to Q1.3 (1 or 2) > PF publication web link/URL http……….. Indicator 1   Dedicated public finance publication website Approach:  The following approach was used to locate the dedicated websites of the public finance organizations and open budget data: a) The Finance Ministry/Department home sites were checked first to see if all questions can be answered from one source. b) In case some of the public finance information was not visible in the MoF websites (for example, investment plans, audit reports, procurement), other websites were visited, depending on the roles/responsibilities of related organizations (for example, Ministry of Economy, Planning). ­ c) In addition to the MoF, Statistics, and Central Bank websites, other ministry/agency web pages (for example, Treasury, Open Budget portal) were visited to ­ capture remaining questions. Evidence:  The URLs of relevant websites (Q1.1 and Q1.4), as well as the official name of the key PF organization (Q1.2). Point: 2= There is a dedicated PF website, and links to budget related information/reports are clearly visible from the home page. 1= PF information links are not clearly visible from the home page, or are posted on separate websites without linkages to the PF home page. 0 = No dedicated website for publishing PF information is visible. Q2. Is there a website or docu- Q2.1 2 Is there a website/document about the FMIS platform? 0 No ment describing the web- 1 Yes (links to relevant documents only) based FMIS platform? 2 Yes (FMIS-related website) Q2.2 If Yes to Q2.1 (1 or 2) > Related web link/URL http……….. Indicator 2   FMIS web page Approach: Public finance organization or other government websites were screened to collect evidence about specific financial management information system (FMIS) web pages/portals, as well as publicly available FMIS-related documents. Evidence: The URL of relevant website (Q2.2). Point: 2= There is a specific website presenting the characteristics of completed FMIS solutions, or the current status of FMIS implementation. 1 = There is only published reference document(s) about FMIS implementation. 0 = No dedicated website for publishing PF information is visible. table continues next page Table A.1  Key indicators (continued) Questions # Indicators Point Response options What is the source of the PF Q3.  Q3.1 3 What is the source of PF data? 0 No published PF data information that is published 1 Static (tables from unidentified sources) on the web? 2 Dynamic (archived docs from systems) 3 Dynamic (data extracted from FMIS DB) Q3.2 If Dynamic (2 or 3) > http……….. Related web link/URL Q3.3 4 Presence of open budget data (online, editable, free) 0/1 No/Yes Q3.4 If Yes to open budget data (1) > Related web link/URL http……….. Q3.5 5 Is system name visible in reports? 0/1 No/Yes Q3.6 6 Is system time stamp visible in reports? 0/1 No/Yes Q3.7 If Yes to Q3.5 or Q3.6 (1) > Related web link/URL http……….. Indicator 3   Source of PF information Approach: The source of PF information was assessed by reviewing the links to related databases (static posting/dynamic query options). Evidence: The URL of relevant dynamic website (Q3.2), since the linkage of PF information with a database is very important. Point: 3= Dynamic website (linked with FMIS/DW databases) with interactive query options to produce reports (for example, PDF, CSV, ODS, XLS, XML). 2= Dynamic website (some linked with databases) to present data from a predefined list of publications (mainly PDF, XLS). 1= Static website (not linked to databases) to publish information from unidentified sources (mainly in PDF format). 0 = There is no published PF information. Indicator 4   Presence of open budget data Approach: The format of published PF data was reviewed to assess compliance with minimum open budget data standards (online, editable, free). Evidence: The URL of relevant website (Q3.4) for open budget data. Point: 0 (No) / 1 (Yes) Indicator 5   System name visible in reports Approach: Check the visibility of the name of FMIS solution used as a basis for PF information in the reports published on the web. Evidence: The URL of relevant website (Q3.7) with a report including the name of FMIS solution used as a basis. Point: 0 (No) / 1 (Yes) Indicator 6   System time stamp visible in reports Approach: Check the visibility of the time stamp (date and time of publication) from the FMIS that is used as a basis for publishing reports. Evidence: The URL of relevant website (Q3.7) with a report including the time stamp from related FMIS solution/database. Point: 0 (No) / 1 (Yes) 139 table continues next page Table A.1  Key indicators (continued) 140 Questions # Indicators Point Response options Q4. Is the PF information mean- Q4.1 7 presentation? Quality: What is the quality of PF data ­ 0 Below desired level (not informative) ingful to citizens or budget 1 Partially acceptable (some useful data) entities? 2 Good quality (informative+easy to read) Q4.2 8 Content: Is there a sufficient level of detail? 0/1 No/Yes Q4.3 9 Citizens Budget: Are the budget results presented easy to 0 No understand? 1 Yes (basic info) 2 Yes (comprehensive info/interactive) Q4.4 If Yes to Q4.3 on CB (1 or 2) > Related web link/URL http……….. Indicator 7   Quality of presentation Evidence:  Perception of the external reviewers about the overall quality of presenting open budget data. This indicator should be used with caution, since there is always a possibility of missing features that may not be so obvious while surfing relevant websites the first time. Point: 2 = Good quality (presented reports are informative and easy to access and read). 1 = Partially acceptable (some of the published PF info is useful). 0 = Below desired level (most of the published PF info is not informative). Indicator 8   Level of detail Evidence:  Perception of the external reviewers about the general level of detail in published reports. This indicator should be used with caution, since there is always a possibility of missing features that may not be so obvious while surfing relevant websites the first time. Point: 0 (No) / 1 (Yes) Indicator 9   Citizens Budget Approach: Citizens Budget websites were screened to identify good practices where budget data are presented in meaningful format. Evidence: The URL of relevant website (Q4.4). Point: 2 = Yes (Comprehensive information in meaningful format for the citizens. Highly interactive.) 1 = Yes (Basic information about the budget cycle and some results in meaningful format). 0 = No Q5. Is the data structure or full Q5.1 10 Are the BC/CoA details published? 0/1 No/Yes listing of budget classifica- Q5.2 If Yes to Q5.1 (1) > Related web link/URL http……….. tion (BC)/chart of accounts (CoA) published? Indicator 10   Budget classification/chart of accounts Approach: The PF websites were screened to spot the BC/CoA details and the revision history, as a basis for published PF data. Evidence: The URL of relevant website (Q5.2). Point: 0 (No) / 1 (Yes) table continues next page Table A.1  Key indicators (continued) Questions # Indicators Point Response options Are documents associated Q6.  Q6.1 11 Is the approved annual budget published? 0/1 No/Yes with annual budget plans Q6.2 12 If Yes to Q6.1 (1): Regularity of publication 0 Not regular published? 1 Published regularly (within last 5 years) Q6.3 Since Year Starting year of publication Indicator 11   Publishing annual budget Evidence: Check the publication of approved annual budget documents on PF websites. Point: 0 (No) / 1 (Yes) Indicator 12   Regularity of publishing annual budget Evidence: Check regular (annual) publication of annual budgets, at least within the last 5 years. Indicate the start year (Q6.3) of publications. Point: 0 (Not regular) / 1 (Published regularly) Are documents associated Q7.  Q7.1 13 Are MTEF documents published? 0/1 No/Yes with medium-term expen- Q7.2 14 If Yes to Q7.1 (1): Regularity of publication 0 Not regular diture framework (MTEF) 1 Published regularly (within last 5-yr) published? Q7.3 Period 3/5 Period (in years) of the MTEF Q7.4 Since Year Starting year of publication Indicator 13   Publishing MTEF Evidence: Check the publication of multiyear expenditure framework documents. Indicate the period (Q7.3) of publications. Point: 0 (No) / 1 (Yes) Indicator 14   Regularity of publishing MTEF Evidence: Check regular (annual) publication of MTEF documents, at least within the last 5 years. Indicate the start year (Q7.4) of publications. Point: 0 (Not regular) / 1 (Published regularly) Are documents associated Q8.  Q8.1 15 Are public investment plans published? 0/1 No/Yes with public investment/capi- Q8.2 16 If Yes to Q8.1 (1): Regularity of publication 0 Not regular tal budget plans published? 1 Published regularly (within last 5-yr) Q8.3 Since Year Starting year of publication Indicator 15   Public investment plans Evidence: Check the publication of public investment plans on PF websites. Point: 0 (No) / 1 (Yes) 141 table continues next page Table A.1  Key indicators (continued) 142 Questions # Indicators Point Response options Indicator 16   Regularity of publishing public investment plans Evidence:  Check regular (annual) publication of public investment plans, at least within the last 5 years. Indicate the start year (Q8.3) of reports. Point: 0 (Not regular) / 1 (Published regularly) Q9. Are documents ­associated Q9.1 17 Are budget execution reports published? 0/1 No/Yes with budget execution Q9.2 ­ ublications If Yes to Q9.1 (1): Frequency of major p 1 Weekly ­published? 2 Monthly 3 Quarterly 4 Annually Q9.3 18 publications If Yes to Q9.1 (1): Regularity of major ­ 0 Not regular 1 Published regularly (within last 5-yr) Q9.4 Since Year Starting year of publication Indicator 17   Budget execution results Approach: The PF websites were screened to spot budget execution results mainly based on economic, administrative, and functional classification. Evidence: Check the publication of budget execution results on PF websites, considering their frequency (Q9.2) and regularity. Point: 0 (No) / 1 (Yes) Indicator 18   Regularity of publishing budget execution results Evidence:  Check regular (annual) publication of budget execution results, at least within the last 5 years. Indicate the start year (Q9.4) of reports. Point: 0 (Not regular) / 1 (Published regularly) Are documents associated Q10.  Q10.1 19 Is the external audit of central government budget opera- 0/1 No/Yes with the external audit of tions published? central government budget Q10.2 20 If Yes to Q10.1 (1): Regularity of publication 0 Not regular operations published? 1 Published regularly (within last 5-yr) Q10.3 Since Year Starting year of publication Indicator 19   External audit of budget operations Evidence: Check the publication of external audit reports about budget operations on PF websites. Indicate the start year (Q10.3) of reports. Point: 0 (No) / 1 (Yes) Indicator 20   Regularity of publishing external audit reports Evidence: Check regular (annual) publication of external audit reports about budget operations, at least within the last 5 years. Point: 0 (Not regular) / 1 (Published regularly) Table A.2 Informative indicators Questions # Indicators Point Response Options What is the level of detail of the p Q11.  ­ ublic Q11.1 21 Consolidated budget results published? 0/1 No/Yes ­expenditure/revenue information Q11.2 22 Sector analysis published? 0/1 No/Yes published online (plans versus actuals, Q11.3 23 Regional analysis published? 0/1 No/Yes sectoral or regional details, and so on)? Q11.4 24 Gender analysis published? 0/1 No/Yes Q11.5 25 Analysis of spending for children & youth 0/1 No/Yes published? Q11.6 26 Debt data published? 0/1 No/Yes Q11.7 27 Foreign aid/grants published? 0/1 No/Yes Q11.8 28 Fiscal data on subnational government/mu- 0/1 No/Yes nicipalities published? Q11.9 29 Financial statements published? 0/1 No/Yes Q11.10 30 Public procurement and contracts published? 0/1 No/Yes Q11.11 Other PF data? 0/1 No/Yes Q11.12 If Yes to Q11.11 (1) > Related web link/URL http……….. Indicators 21 to 30   Public expenditures > Analysis of various dimensions of budget spending Approach: Reports on public spending were scanned further to locate specific information on (i) consolidated budget results; (ii) sectoral spending; (iii) regional spending; (iv) gender budget; (v) children’s budget; (vi) debt data; (vii) aid data; (viii) fiscal data on subnational governments; (ix) financial statements; (x) public procurement and contracts; and (xi) other PF data, together with its URL (Q11.12). Evidence: Check the publication of specific publications on various subcategories of public expenditures. Point: 0 (No) / 1 (Yes) <<< Please note that Other PF Data (Q11.11) is included for additional information only (without a Point). Q12. Is there an open government portal? Q12.1 31 Open government/open budget website? 0/1 No/Yes Q12.2 If Yes to Q12.1 (1) > Related web link/URL http……….. Indicator 31   Open government/open budget website Evidence: Check the presence of open government or open budget websites, and include the URL of relevant website (Q12.2). Point: 0 (No) / 1 (Yes) Q13. Is there a web-based system support- Q13.1 32 Does FMIS support the PFM needs of state/ 0 No ing the SNG PFM needs as a part of FMIS local governments or municipalities? Yes (only data collection/­ 1 solution? consolidation) 2 Yes (support SNG automation needs) Q13.2 If Yes to Q13.1 (1 or 2) > Related web link/URL http……….. 143 table continues next page Table A.2 Informative indicators (continued) 144 Questions # Indicators Point Response Options Indicator 32   Support to subnational government operations from FMIS Evidence: Check the support to subnational government (SNG) operations from centralized FMIS, and indicate the URL of relevant website (Q13.2). Point: 2= Yes (Centralized FMIS solution supports the decentralized SNG automation, data collection, and reporting needs.) 1 = Yes (FMIS solution provides data collection and consolidation capabilities for the SNGs). 0 = No Q14. Is there a harmonized accounting system Q14.1 33 Is there a harmonized public accounting sys- 0/1 No/Yes for all budget levels (unified BC/CoA)? tem for central + state/local governments and municipalities? Q14.2 If Yes to Q14.1 (1) > Related web link/URL http……….. Indicator 33   Harmonized accounting system for SNG Evidence:  Check the presence of harmonized accounting standards for recording and reporting SNG operations, and indicate the URL of relevant website (Q14.2). Point: 0 (No) / 1 (Yes) Are there other government websites Q15.  Q15.1 34 Are PF data published on the Statistics web- 0/3 (please see explanations below) reporting budget results/performance site? Or another website? (for example, Office of Statistics)? Q15.2 If Q15.1 = 0 or 1 > URL of the statistics home http……….. page Q15.3 Official name of statistical organization Official name of statistics office Q15.4 URL of the central bank home page http……….. Q15.5 Official name of central bank Official name of central bank Q15.6 If Q15.1 = 2 or 3 > URL of other major alternative http……….. PF publication website Indicator 34   Other PF publication websites Evidence:  Check the existence of other websites (Statistics, Central Bank or other) publishing PF information, and include the URLs of relevant websites (Q15.2, Q15.4, and Q15.6). The URLs and original names of statistics organization and central bank are included as references. Point: 3 = Both Statistics and other websites publish PF data. 2 = Other website publishing PF data (no Statistics website publishing PF data). 1 = Statistics website publishing PF data. 0 = Statistics website has no PF data. Q16. Is there a website explaining the policy/ Q16.1 35 Access to information explained? 0/1 No/Yes regulations for access to PF information, Q16.2 If Yes to Q16.1 (1) > Related web link/URL http……….. web publishing standards, or frequency of reporting? table continues next page Table A.2 Informative indicators (continued) Questions # Indicators Point Response Options Indicator 35   Access to information Evidence:  Check the presence of dedicated websites providing links or explanations about citizens’ rights for access to PF information, and include the URL of relevant website (Q16.2). Point: 0 (No) / 1 (Yes) Q17. Is there a website with links to regulations Q17.1 36 Are PFM roles/responsibilities clearly 0/1 No/Yes for clarifying PFM roles and responsibilities? ­explained? Q17.2 If Yes to Q17.1 (1) > Related web link/URL http……….. Indicator 36   Regulations on PFM roles and responsibilities Evidence: Check the presence of web pages providing links or explanations about PFM roles and responsibilities, and include the URL of relevant website (Q17.2). Point: 0 (No) / 1 (Yes) Are published PF data compliant with the Q18.  Q18.1 37 Compliance with specific int’l reporting 0/2 (please see explanations below) IMF GFS and/or UN COFOG standards? standards? Q18.2 If Other > Please describe reporting formats Other reporting formats Indicator 37   Reporting standards Evidence: Check the availability of published budget reports in line with IMF GFS or COFOG. Indicate other reporting standards in Q18.2, if any. Point: 2 = IMF GFS reports are published (including UN COFOG-based functional classification of expenditures). 1 = Expenditure reports according to UN COFOG functional classification. 0 = Budget reports compliant with national standards only. Q19. Is there a web page for receiving feedback Q19.1 38 Web statistics (for example, visitors, freq visited 0/1 No/Yes on PF information/user satisfaction, or for web pages) presenting web statistics? Q19.2 39 Which platforms are available for feedback 0/3 (please see explanations below) provision? Q19.3 If Other > Please describe other feedback op- Other feedback mechanisms tions Indicator 38   Web statistics (reports on website traffic) Evidence: Check the existence of web monitoring tools (web statistics and usage reports). Point: 0 (No) / 1 (Yes) table continues next page 145 Table A.2 Informative indicators (continued) 146 Questions # Indicators Point Response Options Indicator 39   Feedback/monitoring Evidence: Check the existence of feedback provision and web traffic monitoring in PF websites. Indicate other feedback mechanisms separately in Q19.3. Point: 3= A number of feedback/monitoring options are visible (telephone/chat/fax/mail/e-mail/feedback forms/web statistics). 2 = Interactive feedback options are visible (telephone/chat/fax/mail). 1 = Static feedback options are available (e-mail/feedback forms/web stats). 0 = Not visible or inadequate. What languages are used to publish PF Q20.  Q20.1 40 What is the native language? Enter language code (ISO 639-2 information online for external viewers? alpha-3) Q20.2 Language #1 option? Indicate language #1 code Q20.3 Language #2 option? Indicate language #2 code Q20.4 Language #3 option? Indicate language #3 code Q20.5 Change of contents for PF data in other lan- 0 Less detail guages 1 Same level of detail in other ­languages Indicator 40   Language Evidence: Indicate the code (ISO 639–2 alpha-3) of publication language(s). Point: Not used. > For Q20.5, Check if the level of detail is different in other languages >>> 0 (Less detail) / 1 (Same level of detail) Appendix B Description of the FMIS & OBD Data Set The data set developed for assessing the effects of financial management infor- mation systems (FMIS) on publishing open budget data is composed of five components (Figure B.1). Key and informative indicators are explained in Appendix A. Other components are described below: Figure B.1  FMIS & Open Budget Data (OBD) data set • Name of economy + income level Basic data • Population + GNI + GNI per capita Key • 20 key indicators derived from 10 questions indicators • Points to measure key indicators Informative • 20 informative indicators derived from 10 questions indicators • Points to measure informative indicators • System name (original + abbreviated) FMIS • Topology + operational status + functional scope characteristics • Type of ASW + technology architecture • Selected PEFA indicators Other • OBI scores + e-Gov ranking + MTEF status indicators • Open source policies and regulations + other aspects • FMIS & ODB practice groups Results • Total Points + scores + groups Source: World Bank data. Note: FMIS = Financial Management Information System; GNI = Gross National Income; PEFA = Public Expenditure and Financial Accountability; OBI = Open Budget Index; MTEF = Medium-Term Expenditure Framework. Financial Management Information Systems and Open Budget Data   147 http://dx.doi.org/10.1596/978-1-4648-0083-2 148 Description of the FMIS & OBD Data Set Table B.1  Basic data Data field Description Values Economy Name of economy + flag Web link to related Wikipedia page Level Income levels according to GNIPC LIC Low income [US$1,025 or less] (2011) LMIC Lower-middle income [US$1,026 to US$4,035] UMIC Upper-middle income [US$4,036 to US$12,475] HIC High income [US$12,476 or more] Population Population (2011) in thousands GNI Gross National Income (2011) in millions of USS (Atlas method) GNIPC GNI per Capita (2011) US$ (Atlas method and PPP) Table B.2  FMIS characteristics Data field Description Values FMIS/TS Abbreviation Abbreviated name of FMIS or Treasury System FMIS Full name Full name of FMIS or TS in original language Topology PFM topology C  Centralized PFM operations D  Decentralized PFM operations Functionality FMIS functional scope B Budget planning/formulation (MTEF, program/ performance based budgeting) T  Treasury (public expenditure management) F  FMIS F = T + B (+O) O  Other FMIS components (procurement, human resources/payroll) Status FMIS implementation status (details 0  T/F was not implemented or not operational of some of the systems are given as 1  T/F is fully/partially operational comments in related cells) 2  T/F is operational for pilot or reduced-scope impl 3  Implementation in progress 4  Pipeline project Op Yr Operational year of FMIS Year in which the FMIS/TS became operational Scope Coverage of budget levels (subnational C  Only central government not included) C+L  Central and Local (district) level budgets Type Type of FMIS project 1 T/F designed+impl as a new turnkey solution  (first time or replacing previous system) 2  Existing T/F improved or expanded 3  New T/F implemented during emergency TA 4  Existing T/F improved or expanded during  emergency TA 5 Improvement or expansion of an existing T/F,  already implemented by the Gov/other donors 0  Not operational ASW Type of application software 0  Not identified yet 1  Locally developed software (LDSW) 2  Commercial off-the-shelf software (COTS) ASW solution ASW vendor/name COTS > Name of ASW package LDSW > Name of database used in developing ASW Arch ASW technology architecture CS  Client-server (distributed system/  decentralized operations) Web  Web-based (centralized system/  decentralized operations) Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 Description of the FMIS & OBD Data Set 149 Table B.3  Other indicators Several indicators related to budget transparency, open data, and e-governance were also included to be able to compare the findings of this study with other assess- ments. Additional columns were added to list the type and total number of open source policies adopted by the national and local governments in 66 economies. Data field Description Values PEFA Yr Year of assessment Year of the last PEFA assessment Stat Disclosure status of PEFA assessment Public Publicly available report www.pefa.org Final Completed, but not publicly disclosed Draft  Draft report completed; status unknown PI-5 [ PEFA ] Classification of the budget [ for details and values of all indicators > www.pefa.org ] PI-6 Comprehensiveness of information included in budget documentation PI-10 Public access to key fiscal information PI-12 Multiyear perspective in fiscal planning, expenditure policy, and budgeting PI-22 Timeliness and regularity of accounts reconciliation PI-23 Availability of information on resources received by service delivery units PI-24 Quality and timeliness of in-year budget reports PI-25 Quality and timeliness of annual financial statements PI-26 Scope, nature, and follow-up of external audit D-2 Financial information provided by donors for budgeting and reporting on project and program aid Avg 10 Average of 10 PEFA indicators selected for a comparison with the FMIS & OBD scores. Avg Average of all 31 PEFA indicators. NR/NA/NU scores are excluded in avg calculations. OBI_12 2012 Open Budget Index 0 - 20  Scant or no information internationalbudget.org 21 - 40  Minimal 41 - 60  Some 61 - 80  Significant 81 - 100  Extensive information OBI_10 2010 Open Budget Index Same as above OGP Status of participation in the Open 0  Developing commitments Government Partnership initiative ­ 1  Commitments delivered MTEF Status of MTEF implementation 0  No MTEF (as of 2010) 1  MTFF (medium-term fiscal framework) 2  MTBF (medium-term budget framework) 3  MTPF (medium-term performance framework) - Unknown eGov12 UN e-Gov Ranking (2012) Ranking among 193 countries included in UNPAN e-Gov Survey the survey, together with the values of indices used to calculate these ratings. eGov e-Gov index 0-1 eOS e-Gov - Online Services 0-1 eTC e-Gov - Telco infrastructure 0-1 eHC e-Gov - Human cap dev 0-1 table continues next page Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 150 Description of the FMIS & OBD Data Set Table B.3  Other indicators (continued) Data field Description Values Region World Bank Region of the economy AFR Africa EAP East Asia and Pacific ECA Europe and Central Asia LCR Latin America and the Caribbean MNA Middle East and North Africa SAR South Asia Zone Zone of various countries according to EU European Community their membership in international EUR Euro zone ­organizations AME America PAC Pacific OTH Other zones OECD OECD membership status OECD Member APEC APEC membership status APEC Member Fragile Fragile state status Fragile Economies Name of economy — OSS Gov National Open Source Policies (in 66 M Mandatory, where the use of OSS is economies) required. CSIS Government Open Source Policies R R&D, where the use of OSS is required. (2010) A Advisory, where the use of OSS is ­permitted. P Preference, where the use of OSS is given preference. OSS Loc States/provinces/cities OSS policies (in 22 Same as above economies) Verif date Date of verification message received from dd-mmm-yy (15-May-13) gov officials Table B.4  Results Data field Description Values Tot Pts Total points (20 key indicators) 0 - 26 Score FMIS & OBD Score 0 - 100 Group FMIS & OBD Practice Group A (Highly visible) B (Visible) C (Limited visibility) D (Minimal visibility) Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 Appendix C Overview of Fiscal Transparency Instruments Introduction and objectives This appendix complements the main report by summarizing the key aspects of various global norms, indices, and initiatives that have emerged in the last decade to promote fiscal transparency (FT). The objective is to clarify the differences between internationally prominent instruments and the methodology followed in this study. Existing FT instruments can be categorized in three parts according to their functions: (a) surveys and indices, (b) standards and norms, and (c) initiatives. A short explanation of each instrument is provided through an extract from relevant websites, and further information can be found through the web links provided. A summary table is also included to compare the important aspects of all instruments (Table C.1). These instruments consider various dimensions in assessing fiscal/budget transparency. The methodology applied in the current study is different than the ones these instruments use. While existing instruments mainly focus on the exis- tence and regularity of certain key budget documents published in the public domain, and on the mechanisms for public access, this study is designed to assess the source, reliability, quality, and readability of published PF information. The purpose of the study is not to develop another FT indicator or standard. Rather, it is a stocktaking exercise, focused on less-known aspects such as the source and reliability of open data, intended to highlight some of the good prac- tices where FMIS is the source of meaningful open budget data. Surveys and Indices 1. Open Budget Index (OBI) http://www.openbudgetindex.org The Open Budget Survey is designed to assist civil society groups and inde- pendent researchers in understanding selected international good practice Financial Management Information Systems and Open Budget Data   151 http://dx.doi.org/10.1596/978-1-4648-0083-2 152 Overview of Fiscal Transparency Instruments benchmarks for budget transparency and accountability, and applying them to the practices they observe in their countries. The first Open Budget Survey was released in 2006, and it has been conducted biennially since then. Currently, the Open Budget Index covers 100 countries (2012). As explained in the 2012 report, the Survey consists of 125 questions and is completed by independent researchers in the countries assessed. Most of these questions (95) deal directly with the public availability and compre- hensiveness of the eight key budget documents that governments should publish at various points of the budget cycle. The remaining 30 questions relate to opportunities for public participation in the budget process, and to the roles played by legislatures and supreme audit institutions in budget formulation and oversight. The Open Budget Index (OBI) is calculated as a simple average of the quantified responses for the 95 survey questions that are related to budget transparency, in order to allow for comparisons across countries and over time. The OBI assigns a score to each country (0–100, in five equal intervals) based on the information it makes available to the public throughout the budget process. The index assesses the availability of eight key budget documents, the quantity of information they provide, and the timeliness of their dissemination to citizens in order to provide reliable infor- mation on each country’s commitment to budget transparency and accountability. 2. PEFA PFM Assessment http://www.pefa.org The PEFA PFM Performance Measurement Framework (introduced in 2005; revised in January 2011) incorporates a public finance management (PFM) performance report and a set of high-level indicators that captures the key aspects that are recognized as being critical for all countries to achieve sound PFM. The indicator identifies comprehensiveness and transparency as one of the critical dimensions of performance of an open and orderly PFM system. It seeks to measure whether the budget and whether the fiscal risk oversight are comprehensive and fiscal and budget information is accessible to the public. The indicators are scored from A to D, with A being the best and D being the lowest score. FT-related indicators are listed below: • PI-5. Classification of the budget A robust classification system allows the tracking of spending on the following dimensions: administrative unit, economic, functional and program. Where stan- dard international classification practices are applied, governments can report expenditure in Government Finance Statistics (GFS) format and track poverty- reducing and other selected groups of expenditure. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 Overview of Fiscal Transparency Instruments 153 • PI-6. Comprehensiveness of information included in budget documentation Annual budget documentation (the annual budget and budget supporting docu- ments), as submitted to the legislature for scrutiny and approval, should allow a complete picture of central government fiscal forecasts, budget proposals, and out-turn of previous years. • PI-10. Public access to key fiscal information Transparency will depend on whether information on fiscal plans, positions, and performance of the government is easily accessible to the general public or at least the relevant interest groups. • PI-12. Multiyear perspective in fiscal planning, expenditure policy and bud- geting Expenditure policy decisions have multiyear implications, and must be aligned with the availability of resources in the medium-term perspective. Countries that have effectively introduced multiannual program budgeting are likely to show good performance on most aspects of this indicator. • PI-22. Timeliness and regularity of accounts reconciliation Reliable reporting of financial information requires constant checking and verifi- cation of the recording practices of accountants—this is an important part of internal control and a foundation for good-quality information for management and for external reports. • PI-23. Availability of information on resources received by service delivery units Front-line units providing services at the community level frequently have problems in obtaining resources that were intended for their use, whether in terms of cash transfers, distribution of materials in kind or provision of centrally recruited and paid personnel. The intended resource provision may not be explicit in budget documenta- tion, but is likely to form part of line ministries’ internal budget estimates preparation. • PI-24. Quality and timeliness of in-year budget reports The ability to “bring in” the budget requires that timely and regular information on actual budget performance be available both to the ministry of finance (and Cabinet), to monitor performance and if necessary to identify new actions to get the budget back on track, and to the ministries, departments, and agencies for managing the affairs for which they are accountable. The indicator focuses on the ability to produce comprehensive reports from the accounting system on all aspects of the budget (flash reports on release of funds to ministries, departments, and agencies are not sufficient). • PI-25. Quality and timeliness of annual financial statements Consolidated year-end financial statements are critical for transparency in the PFM system. To be complete they must be based on details for all ministries, independent departments, and deconcentrated units. In addition, the ability to Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 154 Overview of Fiscal Transparency Instruments prepare year-end financial statements in a timely fashion is a key indicator of how well the accounting system is operating, and of the quality of records maintained. • PI-26. Scope, nature, and follow-up of external audit A high-quality external audit is an essential requirement for creating transpar- ency in the use of public funds. Key elements of the quality of actual external audit are the scope/coverage of the audit, adherence to appropriate auditing standards including independence of the external audit institution (ref. INTOSAI and IFAC/IAASB), focus on significant and systemic PFM issues in reports, and performance of the full range of financial audit such as reliability of financial statements, regularity of transactions, and functioning of internal control and procurement systems. • D-2. Financial information provided by donors for budgeting and reporting on project and program aid Predictability of disbursement of donor support for projects and programs (below referred to only as projects) affects the implementation of specific line items in the budget. 3. IMF Fiscal Transparency Report on Observance of Standards and Codes (ROSC) http://www.imf.org/external/NP/rosc/rosc.aspx The fiscal transparency module of the ROSC documents a country’s current practices, assesses compliance with the Code of Good Practices on Fiscal Transparency, and establishes country-specific priorities for improving fiscal transparency. ROSC summarizes the extent to which countries observe certain internationally recognized standards and codes. The International Monetary Fund (IMF) has recognized 12 areas and associated standards as useful for the operational work of the Fund and the World Bank: accounting; auditing; anti- money-laundering and countering the financing of terrorism; banking supervi- sion; corporate governance; data dissemination; fiscal transparency; insolvency and creditor rights; insurance supervision; monetary and financial policy trans- parency; payments systems; and securities regulation. Reports summarizing countries’ observance of these standards, prepared and published at the request of the member country, are used to help sharpen the institutions’ policy discus- sions with national authorities, and in the private sector (including by rating agencies) for risk assessment. Short updates are produced regularly, and new reports are produced every few years. Between 1999 and March 2013, 93 countries from all Regions and levels of eco- nomic development had posted their fiscal ROSCs on the IMF’s Standards and Codes web page, and 29 countries had undertaken updates or complete assessments. 4. Global Integrity Index http://www.globalintegrity.org/report/methodology The Index, launched in 2004, covers 119 countries. It is based on an Integrity Indicators scorecard that assesses the existence, effectiveness, and citizen access Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 Overview of Fiscal Transparency Instruments 155 to key governance and anticorruption mechanisms through 320 actionable indi- cators. A simple aggregation method is used to produce a country’s aggregate scorecard. The lead researcher for the country assigns original indicator and sub- indicator values. Each indicator score is then averaged within its parent subcat- egory, producing a subcategory score. The subcategory score is in turn averaged with the other subcategory scores into a parent category score. Category scores are averaged to produce an overall country score. The Global Integrity Report groups countries into five performance “tiers” according to their aggregated score: • Very strong (90+) • Strong (80+) • Moderate (70+) • Weak (60+) • Very weak (< 60) Because some aspects of governance and anticorruption mechanisms are harder to measure definitively, some categories require a more complex matrix of subindicator questions than others. The categories are equally valued, even if some categories are derived from a more lengthy series of subindicators/ ques- tions than others. Similarly, the subcategories are equally valued within their parent category. In other words, each score (subindicators, indicator, and so on) is equally weighted with its peers addressing the same subcategory/category. However, indicators from different categories are not necessarily equally weighted. Our approach of using equally valued concepts and adding subordinate elements as needed has produced score weightings that reflect the six main conceptual cat- egories evenly. Although we recognize the rationale for a nonequal weighting system (to give emphasis to issues of greater import), we have yet to develop a compelling defense for valuing certain categories, subcategories, or indicators more importantly than others. 5. Right to Information (RTI) Index http://www.rti-rating.org/index.php The RTI Rating is a system that began in 2011 for assessing the strength of the legal framework for guaranteeing the right to information in a given country. It is limited to measuring the legal framework, and does not measure quality of implementation. Currently, the Index covers 93 countries. In some cases, countries with relatively weak laws may nonetheless be very open, because of positive implementation efforts, while even relatively strong laws cannot ensure openness if they are not implemented properly. Regardless of these outlying cases, over time a strong access to information law can contribute to advancing open- ness and help those using it to defend and promote the right of access to information. At the heart of the methodology for applying the RTI Rating are 61 indicators. For each indicator, countries earn points within a set range of scores (in most Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 156 Overview of Fiscal Transparency Instruments cases 0–2), depending on how well the legal framework delivers the indicator, for a possible total of 150 points. The indicators are drawn from a wide range of international standards on the right to information, as well as comparative study of numerous rights to informa- tion laws from around the world. The indicators are grouped into the following seven main categories: Section Max Points 1. Right of access 6 2. Scope 30 3. Requesting procedures 30 4. Exceptions and refusals 30 5. Appeals 30 6. Sanctions and protections 8 7. Promotional measures 16 Total score 150 6. UN e-Government Survey and Rankings http://www2.unpan.org/egovkb/global_reports/index.htm The Global e-Government Development Survey, first released in 2003, pres- ents a systemic assessment of how governments use information and communi- cations technology (ICT) to provide access and inclusion for all. Each Survey offers insights into the different strategies and common themes in e-government development among and across regions. By studying broad patterns of e-govern- ment use, it identifies countries that have taken a leadership role in promoting e-government development and those where the potential of ICT for develop- ment has not yet been exploited. The Survey aims to inform and improve the understanding of policy makers’ choices in their e-government program undertakings. It is a useful tool for gov- ernment officials, researchers, and the representatives of civil society and the private sector to gain a deeper understanding of the relative position of a country vis-à-vis the rest of the world’s economies. In this way the Survey rankings hope to contribute to the e-government efforts of the member states as they move to provide access for all. It ranks 193 countries. The Survey consists of four parts: information dissemination and outreach, access and usability, service delivery capability, and citizen participation and interconnectedness. Standards and Norms 7. IMF Code of Good Practices on Fiscal Transparency http://www.imf.org/external/np/pp/2007/eng/051507c.pdf The Code identifies a set of principles and practices to help governments pro- vide a clear picture of the structure and finances of government (released in 1998; Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 Overview of Fiscal Transparency Instruments 157 updated in 2007). It underpins the voluntary program of fiscal transparency assessments called ROSC fiscal transparency modules. It identifies four areas: • Clarity of roles and responsibilities • Open budget processes • Public availability of information • Assurances of integrity. 8. OECD Best Practices for Budget Transparency http://www.oecd.org/dataoecd/33/13/1905258.pdf. The Organisation for Economic Co-operation and Development (OECD) Best Practices are a reference tool issued in 2002. They support the full disclosure of all relevant fiscal information in a timely and systematic manner and provide a series of best practices in the areas of principal budget reports, specific disclo- sures, quality, and integrity. The Best Practices are in three parts. Part 1 lists the principal budget reports that governments should produce, and their general content. Part 2 describes specific disclosures to be contained in the reports, including both financial and nonfinancial performance information. Part 3 highlights practices for ensuring the quality and integrity of the reports. The Best Practices are organized around specific reports for presentational reasons only. It is recognized that different countries will have different reporting regimes and may have different areas of emphasis for transparency. The Best Practices are based on different member countries’ experiences in each area. It should be stressed that the Best Practices are not meant to constitute a formal “standard” for budget transparency. Fiscal Transparency Initiatives 9. Global Initiative for Fiscal Transparency (GIFT) http://fiscaltransparency.net/ The Global Initiative for Fiscal Transparency (GIFT) was launched in 2011 as a multistakeholder action network working to advance and institutionalize global norms and significant, continuous improvements on fiscal transparency, partici- pation, and accountability in countries around the world. GIFT aims to achieve this by advancing incentives, norms, technical assistance, and new technologies. GIFT mobilizes a wide range of stakeholders—national authorities, the private sector, civil society, and international organizations—in support of fiscal transpar- ency initiatives at the global and national levels. At the global level, GIFT seeks to strengthen the normative framework for fiscal transparency by harmonizing norms and standards and addressing gaps related to participation, the role of the legislature, and open data. At the national level, GIFT facilitates multistakeholder engagement on fiscal transparency by Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 158 Overview of Fiscal Transparency Instruments sharing international experience and providing support to multistakeholder forums and technical advice in the implementation of their work programs. 10. Open Government Partnership http://www.opengovpartnership.org Open Government Partnership (OGP) is an international organization pro- moting multilateral initiatives and seeking strong commitments from participat- ing government institutions to promote transparency, increase civic participation, fight corruption, and harness new technologies to make government more open, effective, and accountable. This initiative was launched on September 20, 2011, with the endorsement of a declaration by eight countries (Brazil, Indonesia, Mexico, Norway, Philippines, South Africa, United Kingdom, and United States). Since then, 47 additional members have joined the group endorsing the declaration (as of June 2013); 5 other countries are expected to officially endorse the declaration in 2013, and are in the process of developing commitments. The OGP provides an international forum for dialogue and sharing among governments, civil society organizations, and the private sector, all of which con- tribute to a common pursuit of open government. OGP stakeholders include participating governments as well as civil society and private sector entities that support the principles and mission of OGP. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 Table C.1  Overview of fiscal transparency instruments Category Instrument Number of countries Since/updates Stated objective Methodology Surveys Open 94 2006 To evaluate whether govern- http://www.openbudgetindex.org and Budget biennial ments give the public The Open Budget Survey is based on a detailed questionnaire to guide ­indices access to budget informa- civil society researchers from each country through each of the four Index last: 2010 tion and opportunities to stages of the budget process. The questionnaire contains a total of (OBI) next: 2012 participate in the budget 123 questions. The responses to 92 of the questions that evaluate process at the national public access to budget information are averaged to form the Open level. Budget Index. The remaining 31 questions cover topics related to opportunities for public participation in the budget process and the ability of key oversight institutions of government to hold the execu- tive accountable. PEFA 121 2005 By providing a common pool http://www.pefa.org PFM Public: 65 of information for measur- The PEFA framework was created as a high-level analytical instrument that ing and monitoring PFM consists of a set of 31 indicators and a supporting PFM Performance Re- Assessment performance progress, and port, providing an overview of the performance of a country’s PFM sys- a common platform for dia- tem. Drawing on the established international standards and codes and logue about PFM reform, other commonly recognized good practices in PFM, it forms part of the it aims to contribute to the “strengthened approach to supporting PFM reform”, which emphasizes development of effec- country-led reform, donor harmonization and alignment around the tive country-owned PFM country strategy, and a focus on monitoring results. The indicators are systems. scored from A to D, with A being the best and D being the lowest score. IMF 93 1999 ROSCs summarize the extent http://www.imf.org/external/NP/rosc/rosc.aspx Fiscal to which countries observe ROSCs document a country’s current practices, assess compliance with certain internationally recog- the Code of Good Practices on FT, and establish country-specific Transparency nized standards and codes. priorities for improving FT. ROSC Global Integrity 119 2004 The Global Integrity Index http://www.globalintegrity.org/report Index assesses the existence and The Global Integrity Index is generated by aggregating more than effectiveness of, and citizen 300 Integrity Indicators systematically gathered for each country access to, key national- covered. For the 2009 Global Integrity Index, those indicators level anticorruption comprised more than 100,000 peer-reviewed questions and mechanisms used to hold answers scored by in-country experts. Several rounds of review are governments accountable. conducted at the international level to ensure that cross-country comparisons are valid. In addition, all assessments are reviewed by a country-specific, double-blind peer review panel comprising ad- 159 ditional local and international subject matter experts. table continues next page Table C.1 Overview of fiscal transparency instruments (continued) 160 Category Instrument Number of countries Since/updates Stated objective Methodology Right to 93 2011 To assess the strength of the http://www.rti-rating.org/index.php ­Information legal framework for guar- (RTI) Index anteeing the right to infor- There are 61 indicators. For each indicator, countries earn points within a set range of scores (in most cases 0–2), depending on how well mation in a given country. the legal framework delivers the indicator, for a possible total of 150 It is limited to measuring points. The indicators are drawn from a wide range of international the legal framework, and standards on the right to information, as well as a comparative does not measure quality study of numerous rights to information laws from around the of implementation. world. UN 193 2003 To systemically assess how http://www2.unpan.org/egovkb/global_reports/index.htm governments use informa- E-gov Survey biennial tion and communications The Survey consists of four parts: information dissemination and out- and Rank- reach, access and usability, service delivery capability, and citizen since 2004 technology (ICT) to provide ings participation and interconnectedness. Countries are ranked accord- access and inclusion for the last: 2012 ing to the score of the Survey. public. Standards IMF n.a. 1998 To identify a set of principles http://www.imf.org/external/np/pp/2007/eng/051507c.pdf and and practices to help gov- Code of Good Last update: It identifies four areas: clarity of roles and responsibilities, open budget norms ernments provide a clear Practices on processes, public availability of information, and assurances of 2007 picture of the structure and Fiscal Trans- integrity. finances of government. parency Another update, initiated in 2012, is under way. Best Practices for Budget Transparency OECD n.a. 2002 To support the full disclo- http://www.oecd.org/dataoecd/33/13/1905258.pdf sure of all relevant fiscal information in a timely and The Best Practices are in three parts. Part 1 lists the principal bud- get reports that governments should produce and their general systematic manner and content. Part 2 describes specific disclosures to be contained in provide a series of best the reports, including both financial and nonfinancial performance practices in the areas of information. Part 3 highlights practices for ensuring the quality and principal budget reports, integrity of the reports. The Best Practices are based on different specific disclosures, quality, member countries’ experiences in each area; they are not meant to and integrity. constitute a formal “standard” for budget transparency. table continues next page Table C.1 Overview of fiscal transparency instruments (continued) Category Instrument Number of countries Since/updates Stated objective Methodology INITIATIVES Global ­Initiative n.a. 2011 To contribute to significant, http://fiscaltransparency.net for Fiscal continuous improvements GIFT mobilizes a wide range of stakeholders. At the global level, GIFT ­Transparency on fiscal transparency, seeks to strengthen the normative framework for fiscal transpar- (GIFT) public participation, and ency by harmonizing norms and standards and addressing gaps accountability by advanc- related to participation, the role of the legislature, and open data. At ing incentives, norms, the national level, GIFT facilitates the multistakeholder engagement technical assistance, and on fiscal transparency by sharing international experience and new technologies. providing support to multistakeholder forums and technical advice in the implementation of their work programs. Open 55 committed 2011 To promote transparency, http://www.opengovpartnership.org ­Government 5 in progress increase civic participa- Open Government Partnership (OGP) provides an international forum Partnership tion, fight corruption, and for dialogue and sharing among governments, civil society organi- harness new technologies zations, and the private sector, all of which contribute to a common to make government more pursuit of open government. OGP stakeholders include participat- open, effective, and ac- ing governments as well as civil society and private sector entities countable. that support the principles and mission of OGP. Source: World Bank data. Note: IMF = International Monetary Fund; OECD = Organisation for Economic Co-operation and Development; PEFA = Public Expenditure and Financial Accountability; PFM = Public financial manage- ment; n.a. = not applicable. 161 Appendix D Feedback Providers The feedback provided by the following government officials (Table D.1) on the initial findings of this study was very useful to improve the quality of the data set and the evidence collected. The authors are grateful to all government offi- cials (from 43 economies) for their valuable support. The team would also like to express sincere appreciation to the World Bank staff and other financial man- agement information systems (FMIS) Community of Practice members who assisted in the coordination of efforts and collection of responses on the FMIS and Open Budget Data survey. Table D.1  Feedback providers Economy Government officials The World Bank staff Others (FMIS CoP) Albania Odeta Kromici Evis Sulko Argentina Raul Rigo Mamadou Deme, Alejandro Solanot Armenia Grigor Aramyan Davit Melikyan Azerbaijan Nazim Gasimzade Bangladesh Ranjit Chakraborty Dilshad Dossani, Jonas Fallov Barbados Juanita Thorington-Powlett, Nancy Headley Belize Zita Magana Perez Botswana Grace Nkateng Colombia David Morales Jeannette Estupinan Shirley Herreño Oscar E. Escobar Eduardo Rodriguez Croatia Marijana Müller Cyprus Maria Dionysiou El Salvador Lilena Martínez de Soto Finland Marko Oja, Economics Dept. Ghana Sammy Arkhurst Ismaila Ceesay Hong Kong SAR, China Thebe Ng Indonesia MoF officials Hari Purnomo table continues next page Financial Management Information Systems and Open Budget Data   163 http://dx.doi.org/10.1596/978-1-4648-0083-2 164 Feedback Providers Table D.1 Feedback providers (continued) Economy Government officials The World Bank staff Others (FMIS CoP) Ireland Joe Kirwan, John Palmer, Dept of Thomas Ferris Finance Republic of Korea Hyangwoo Jeong, Ji-yeon Kim Kyrgyz Republic Nurida Baizakova Zhanybek Ybraiym Uulu Lao PDR Sifong Oumavong Minh Van Nguyen Latvia Sintija Dadzīte Lithuania Grazina Steponenaite Madagascar MoF officials Anne-Lucie Lefebvre, Haja Andriamarofara Mauritius R. Kalleechurn Myanmar MoF Budget Department Netherlands Rense Posthumus New Zealand Nicola Haslam, Emma Taylor Ian Storkey Nicaragua MoF officials Daniela Felcman, Alberto Leyton Norway Pål Ulla Paraguay María Teresa de Agüero Mamadou Deme Poland Piotr Dragańczuk Iwona Warzecha Russian Federation Renold Rubies Irina Rostovtseva San Marino Roberta Mularoni Singapore Wang Shihui Slovak Republic Katarína Kováčová Slovenia Senka Maver, Vesna Derenčin Solomon Islands Norman Hiropuhi Timothy Bulman South Sudan David Martin Parminder Brar, Adenike Oyeyiola Spain Carmen Castaño Tajikistan MoF officials Hassan Aliev Ukraine Roman Chuprynenko, Konstantin Tetiana Kovalchuk Stanytskyy United States of America Dustin Brown, Regina Kearney Joanna Watkins Vietnam MoF officials Quyen Vu, Khanh Linh Thi Le Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 Bibliography Dener, Cem, Joanna A. Watkins, and William L. Dorotinsky. 2011. Financial Management Information Systems: 25 Years of World Bank Experience on What Works and What Doesn’t. World Bank Study. Washington, DC: World Bank. International Budget Partnership. 2011. A Guide to Transparency in Public Finances: Looking Beyond the Budget. International Budget Partnership. ———. 2012. The Power of Making It Simple: A Government Guide to Developing Citizens Budgets. Includes five briefs that go beyond the eight key budget reports covered in the guide: • Extrabudgetary Funds • Tax Expenditures • Quasi-Fiscal Activities • Contingent Liabilities • Future Liabilities International Monetary Fund. 2012. Data Quality Assessment Framework for Government Finance Statistics. Washington, DC: IMF. ———. 2012. Fiscal Transparency, Accountability, and Risk. Washington, DC: IMF. Manning, Nick, Geoffrey Shepherd, and Alejandro Guerrero. 2010. “Why Trust in Government Matters?” In Results, Performance Budgeting and Trust in Government. Washington, DC: World Bank. National Coalition for Dialogue & Deliberation. 2009. The Core Principles for Public Engagement. National Consumer Council. 2008. Deliberative Public Engagement: Nine Principles. Padgett, John F., and Walter W. Powell. 2012. The Emergence of Organizations and Markets. Princeton: Princeton University Press. (review of the book). Public Expenditure and Financial Accountability (PEFA) Program. Supreme Audit Institutions Performance Measurement Framework (SAI-PMF), Draft Version 2.0, September 2012. World Bank FMIS Database (1984–2013), updated July 2013. Financial Management Information Systems and Open Budget Data   165 http://dx.doi.org/10.1596/978-1-4648-0083-2 Open Data References Creative Commons (CC) Licenses for Open Source: http://creativecommons.org. G8 Open Data Charter; Improving the transparency and accountability of government and its services, 18 June 2013. List of open data catalogs: http://www.datacatalogs.org. Open Data Handbook: http://opendatahandbook.org. Open Knowledge Foundation Open Data Guide: http://opendatacommons.org/guide. Open Knowledge Foundation Guide: http://opengovernmentdata.org. Publishing Open Government Data: http://www.w3.org/TR/gov-data. Putting Government Data Online: http://www.w3.org/DesignIssues/GovData.html. UN DevInfo: http://www.devinfo.org. World Bank Open Data portal: http://data.worldbank.org. World Bank Open Government Data Toolkit (OGD Toolkit), released in Nov 2012: http://data.worldbank.org/open-government-data-toolkit. World Wide Web Foundation, Open Gov Data: http://webfoundation.org/projects/ogd. Financial Management Information Systems and Open Budget Data   167 http://dx.doi.org/10.1596/978-1-4648-0083-2 Environmental Benefits Statement The World Bank is committed to reducing its environmental footprint. In support of this commitment, the Publishing and Knowledge Division leverages electronic publishing options and print-on-demand technology, which is located in regional hubs worldwide. Together, these initiatives enable print runs to be lowered and shipping distances decreased, resulting in reduced paper consumption, chemical use, greenhouse gas emissions, and waste. The Publishing and Knowledge Division follows the recommended standards for paper use set by the Green Press Initiative. Whenever possible, books are printed on 50% to 100% postconsumer recycled paper, and at least 50% of the fiber in our book paper is either unbleached or bleached using Totally Chlorine Free (TCF), Processed Chlorine Free (PCF), or Enhanced Elemental Chlorine Free (EECF) processes. More information about the Bank’s environmental philosophy can be found at http://crinfo.worldbank.org/wbcrinfo/node/4. Financial Management Information Systems and Open Budget Data http://dx.doi.org/10.1596/978-1-4648-0083-2 F inancial Management Information Systems and Open Budget Data: Do Governments Report on Where the Money Goes? is part of the World Bank Studies series. These papers are published to communicate the results of the Bank’s ongoing research and to stimulate public discussion. Financial Management Information Systems and Open Budget Data: Do Governments Report on Where the Money Goes? is a World Bank Study, initiated in 2012 after an extended stocktaking exercise, to explore the effects of Financial Management Information Systems (FMIS) on publishing reliable open budget data and identify potential improvements in budget transparency. A rich data set was created by visiting the government public finance websites in 198 economies, and collecting evidence on the use of 176 FMIS in publishing open budget data. This study is not intended to develop another index or ranking on budget transparency; rather, it examines the budget data disclosed by the governments on the web for the details of budget revenues and expenditures, as well as the results achieved. This is a tour around the world in search of reliable open budget data, intended to share some of the good practices and possible answers to a key question: “Where does the money go?” The study shows that only a small group of governments provide their citizens, civil society groups, and oversight agencies access to reliable, accurate, and meaningful open budget data from underlying FMIS solutions. However, given the rising demand from citizens and civil society for improved and complete open budget data about all financial activities, many governments around the world are trying to respond to this democratic pressure. Several cases demonstrate that even in difficult settings, innovative solutions to improve budget transparency can be developed rapidly, with a modest investment, if there is political will and commitment. To help governments with their endeavor and to encourage those that provide little or no public finance information on the web, the report concludes with several achievable recommen- dations and guidelines on publishing open budget data using existing FMIS solutions. World Bank Studies are available individually or on standing order. This World Bank Studies series is also available online through the World bank e-library (www.worldbank.org/elibrary). ISBN 978-1-4648-0083-2 90000 9 781464 800832 SKU 210083