How to Measure Decentralization: A Summary of the Global Literature Review on Decentralization Measurement © [2024] International Bank for Reconstruction and Development / The World Bank 1818 H Street NW Washington DC 20433 Telephone: 202-473-1000 Internet: www.worldbank.org This work is a product of the staff of The World Bank with external contributions. The findings, interpretations, and conclusions expressed in this work do not necessarily reflect the views of The World Bank, its Board of Executive Directors, or the governments they represent. The World Bank does not guarantee the accuracy, completeness, or currency of the data included in this work and does not assume responsibility for any errors, omissions, or discrepancies in the information, or liability with respect to the use of or failure to use the information, methods, processes, or conclusions set forth. 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 construed or 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 The material in this work is subject to copyright. Because The World Bank encourages dissemination of its knowledge, this work may be reproduced, in whole or in part, for noncommercial purposes as long as full attribution to this work is given. Any queries on rights and licenses, including subsidiary rights, should be addressed to World Bank Publications, The World Bank, 1818 H Street NW, Washington, DC 20433, USA; fax: 202-522-2625; e-mail: pubrights@worldbank.org. Contents Acknowledgements iv 1 Introduction 1 2 Unpacking the Concept of Decentralization 2 3 How to Measure Decentralization 4 4 Key Reflections on Indicators 7 5 Conclusion 9 References 10 Annex 1: The RAI Indicators – Self-rule and Shared Rule 12 Acknowledgements This paper is prepared by Yoshihiro Saito (Public Sector Specialist, ESAC1). The author wishes to thank reviewers Shabih Ali Mohib (Practice manager of the ESAC1), Serdar Yilmaz (Practice Manager of the EGVPF), Marcela Rozo (Senior Public Sector Specialist, ESAC1), Nayan Joshi (Economist, ESAC1) and Ojeswee Pande (Consultant ESAC1) for providing guidance and the feedback on the content. iv 1 Introduction As decentralization is a cross-cutting concept, there are many ways to measure it. This short note aims to unpack the concepts of decentralization by conducting a global literature review. It starts with a review of the traditional ways of approaching decentralization and the history of its development, including arguments on the merits and limitations of each method. While there are some limitations, the author concludes that integrated quantitative and qualitative approaches using Regional Authority Index (RAI) type indicators are the most comprehensive and objective measures to assess decentralization with its quantifiable score-based method, which employs greater objectivity than purely quantitative methods. 1 2 Unpacking the Concept of Decentralization Decentralization is a means of sharing authority and/or resources between at least two tiers of government. As the nature of decentralization is cross-cutting, different scholars view it from different angles. For example, political scientists who draw from the liberal tradition argue that decentralization helps to deepen and consolidate democracy by devolving power to local govern- ments (Diamond and Tsalik 1999). Economists who draw from a market theory of local expendi- ture argue that decentralization helps to improve resource allocation through better knowledge of local preferences and competition among localities (Oates 1972). Other scholars warn against the devolution of power to sub-national officials and show that it can augment distributional conflicts (Treisman 2000), foster sub-national authoritarianism (Williams 2000) and exacerbate patronage (Theriault 2003). 2.1 TYPES OF DECENTRALIZATION Despite these varied approaches to decentralization, one of the most prevalent ways of framing it is to view it as comprising three components: political, administrative, and fiscal decentralization. The definition and scope of these components varies depending on the country, and there is no agreed over-arching definition for each concept. However, according to the World Bank1, these are some theoretical principles that capture the salient characteristics of the three components of decentralization: • Political decentralization transfers policy and legislative power from central governments to autonomous, lower-level assemblies and local councils that have been democratically elected by their constituencies. To be effective, it requires regular elections, clearly defined jurisdictions and powers, and the appropriate legal, political, and functional space. • Administrative decentralization places planning and implementation responsibilities in the hands of locally situated civil servants who are under the jurisdiction of elected local 1 Edited based on the definitions on the World Bank website: https://www.worldbank.org/en/topic​ /communitydrivendevelopment/brief/Decentralization 2 governments. To be effective, it requires an ability to make independent staffing decisions and to negotiate conditions of service (although the center may retain a useful role in training). • Fiscal decentralization accords substantial revenue and expenditure autonomy to local gov- ernments, including the power to levy taxes and user charges. To be effective, it requires linking service delivery responsibilities with revenue generation, increasing revenue autonomy, building capacity to analyze data for budget decisions, and establishing proper fiduciary controls. These forms of decentralization are interlinked and influence one another. Indeed, Boex and Yilmaz (2010) point out that successful decentralization needs comprehensive development in all three dimensions. 2.2 WAYS OF DECENTRALIZING These three types of decentralization take shape in three ways: • Deconcentration is the handing over of some amount of authority or responsibility to lower levels within central government ministries and agencies. • Delegation refers to the transfer of managerial responsibility for specifically defined functions to organizations that are outside the regular bureaucratic structure and that are only indirectly controlled by the central government. • Devolution refers to the transfer of power from a higher level of government to a sub-national local authority (Rondinelli and Nellis 1986). 3 3 How to Measure Decentralization 3.1 LACK OF CONSENSUS ON HOW TO MEASURE DECENTRALIZATION Many scholars have been making efforts to measure each of the three forms of decentralization by setting up indicators, but there remains difficulty in establishing clear substantive indices that measure the degree of decentralization (Dardanelli and Wright 2021). This lack of consensus is caused mainly by the different theoretical and methodological approaches followed by researchers (Harguindéguy et al. 2021). For example, Schneider (2003) attempted to set up indicators to assess the three types of decen- tralization about two decades ago. While he pointed out the importance of in-depth investigation into how the three forms of decentralization are interlinked, he argued that the best indicator of the level of fiscal centralization or decentralization is the share of sub-national expenditure and revenue. In a similar vein, followed by the International Monetary Fund (IMF) in its government finance statistics (2013), the Organisation for Co-operation and Development (OECD) established comparative information on fiscal decentralization (2015). This approach is more comprehensive in terms of assessing fiscal decentralization, because its coverage is broader, encompassing additional angles such as sub-national debt. The database also allows users to assess time trends, as it covers the period from 1965 to 2020 in OECD countries. Dickovick (2014) is among the scholars who have sought to determine sub-national autonomy by employing an objective measure. He ranks the autonomy of sub-national governments (SNGs) in relation to the central government according to three categories: a) high – when SNGs spend more than 20 percent of government spending, b) moderate – 10–15 percent of government spending, and c) low – less than 10 percent of government spending. This method has the advantage of being simple and clear-cut, as it is based on the degree of quantifiable data. 3.2 A COMPREHENSIVE METHOD – REGIONAL AUTHORITY INDEX One of the most comprehensive ways to assess decentralization is through the Regional Authority Index (RAI), which was developed by Hooghe et al. (2016). The RAI assesses the 4 fiscal, administrative, and political authority of regional bodies, mainly by looking at self-rule and shared rule through ten dimensions, including policy scope, law-making authority, and fiscal and borrowing autonomy (see Annex 1 for scoring methodology). The measure is sensitive to differ- ences between individual regions in the same country and changes over time. This index enables assessed countries/regions to produce scores that indicate their authority. The key objectives of the RAI are to: a) provide a reasonably valid measure of sub-national government structure that is sensitive to cross-sectional and temporal variation, and b) deliver a measurement of the sub-national authority (Hooghe et al. 2016). One of the biggest advantages of this index is its wide coverage of periods and countries/regions. As of August 2023, the RAI covers 95 countries and a period of nearly 70 years (1950–2018). While the RAI is a comprehensive and quantifiable score-based method that employs greater objectivity than purely quantitative methods, it has some limitations. First and foremost, it contains no mea- surable local government data relating to a population of less than 150,000. It is assumed that this is mainly due to the fact that municipal-level data are rarely available or often not reliable, because of their data-producing capacity or just because collecting municipal-level data requires extensive human resources and time. Second, the central focus is on authority, yet it would be advanta- geous to also consider the expenditure ratio and the volume/degree of expenditure, as argued by Schneider (2003), as these figures are a result of sub-national authority. 3.3 A COMPLEMENTARY METHOD – LOCAL AUTONOMY INDEX In view of these merits and limitations, and to complement such measures, Ladner et al. (2019), established a method of measuring local government autonomy, called the Local Autonomy Index (LAI), based on weighted indicators for seven dimensions2 that consider political, fiscal, and administrative factors. The LAI uses a similar coding system to the RAI and aims to provide scores across 34 years (1990–2014), covering 57 European counties, enabling users to perceive trends in decentralization. This has an advantage compared to other indicators in terms of a) looking at local government autonomy, and b) considering access, that is, the degree of influence on decision-making at higher levels of government. The latter, in particular, offers insight into the importance of vertical inter-governmental coordination, which is a new angle. While geographic coverage is limited to Europe, it gives insight into measuring decentralization at the local level, which could be replicable in other countries if data permits. 2 The dimensions are political discretion, financial autonomy, non-interference, policy scope, legal autonomy organizational autonomy, and access. 5 3.4 DARDANELLI’S MODEL – 7-POINT CODING SCHEME ON 22 POLICY AREAS Considering the merits and limitations of these different measures, Dardanelli et al. (2019) recently formulated a concept, methodology, and theoretical framework for analyzing dynamic de/centralization in federations, with reference to 6 federal countries over the course of 10 years since the establishment of a federation. These scholars outline a method for measuring the dynamics of federalism according to a 7-point coding scheme across 22 policy areas3 and 5 fiscal categories. This theoretical framework has many advantages. Firstly, it allows comparison between the 6 countries using the same foundation and methodology. Assessment in 22 policy areas also allows for a better understanding of the outcome of decentralization by sector, than assessing political and administrative federalism-related indicators. Although political and administra- tive decentralization form the conditions for the outcome, it is hard to quantify these factors. Therefore, this framework has the advantage of seeing the outcome itself and using that as a means to assess the dynamics of de/centralization. Secondly, in terms of measuring fiscal decentralization, this method looks at fiscal auton- omy, rather than the volume of financial resources. While many other scholars have looked at sub-national expenditure and the ratio of their own sources of revenue to the total revenue, this alternative framework goes one step further in measuring fiscal autonomy. For example, if financial resources are heavily earmarked by the central government, sub-national governments have no authority to determine how to use them. That is not a desired structure for sub-national govern- ments subject to fiscal federalism. While Dardanelli et al.’s (2019) concept has many advantages, there are some limitations. Firstly, the case selections are limited to six; thus, it would be beneficial if further studies were conducted to ascertain overall trends. Secondly, the attribution between the 22 policy areas in terms of de/centralization and political/administrative decentralization is unclear. Thirdly, the methodology used is based on the premise that if one of the indicators suggests de/centralization, then this determines the dynamics of the de/centralization. As decentralization is broad and cross-cut- ting in nature, again, it may not be entirely accurate to determine an overriding result from this methodology. 3 The 22 areas are: agriculture; citizenship and immigration; culture; currency and money supply; defense; economic activity; pre-tertiary education; tertiary education; elections and voting; employment relations; environmental protection; external affairs; finance and securities; health care; language; civil law; criminal law; law enforcement; media; natural resources; social welfare; and transport. 6 4 Key Reflections on Indicators It is evident that indicators to assess decentralization have advanced considerably over the last two decades. Starting from Schneider (2003), who attempted to use simple indicators to inves- tigate the degree of three forms of decentralization, recent efforts to study the three types of decentralization are more mature in terms of depth and breadth. For example, the RAI covers 95 countries over the course of nearly 70 years of trends, using a coding system that makes the methodology quantifiable and objective. The following are some key reflections on the indicators used for assessing the three types of decentralization. 4.1 FISCAL INDICATORS Fiscal indicators are more sophisticated in relation to fiscal decentralization, as compared to administrative and political decentralization. This is because the areas of fiscal data, such as bud- get allocation and public expenditure, are readily available, given that most ways of assessing fiscal decentralization are based on quantifiable data, such as the share of sub-national expenditure and own source revenue. This nature of work allows researchers to establish the most objective means of measuring fiscal decentralization. However, fiscal indicators have some limitations. Firstly, sub-state fiscal data are rarely available for the three tiers of government (municipal, provincial or regional) (Ebel and Yilmaz 2002). Second, as Dardanelli and Wright (2021) mention, many scholars tend to ignore the elements of administrative and political decentralization and over rely on fiscal data, which may result in missing the full picture of decentralization. For example, in Nepal, while provincial and local governments received nearly 36.7 percent of federal money through intergovernmental fiscal transfers, which amounted to 64.8 percent of their revenue in financial year 2021, in the past five years, the largest and proportionally growing fund transfers to sub-national governments have been through conditional grants. These are heavily earmarked and less flexible resources for sub-national governments. Only looking at the sub-national share of expenditure or revenue may ignore these contexts. In other words, policy making and policy implementation do not always appear in statistical tables. This point was affirmed by Stegarescu (2005, 301–333), who stressed that fiscal measures usually tend to overestimate the extent of decentralization. 7 4.2 POLITICAL INDICATORS As opposed to fiscal data, this is the least investigated dimension among the three. Most of the scholars who have looked at political indicators have done so in relation to elections at multiple levels. However, there are ample opportunities to explore further subjects, such as the sub-national party system, the election system itself, inter-governmental political relationships, all the aspects of citizen participation in the tiers of government, and so forth, so as to better assess forms of politi- cal decentralization. 4.3 ADMINISTRATIVE INDICATORS Administrative indicators have substantially advanced over the last two decades, with, for example, RAI and LAI covering extensive areas of administrative decentralization. This development helps balance out an over-reliance on fiscal decentralization data and provides a qualitative dimension, whereas fiscal indicators often rely on quantitative approaches. 4.4 OVERARCHING REFLECTIONS One of the potential areas for further research is the co-relations between one indicator and another. While Falleti (2005) investigated sequential relationships among the three types of decentralization, there is no conceptual theory widely used to look at causal factors at the indi- cator level. Another challenge lies in the definitions of the three types of decentralization itself. Despite ample research, there is still a lack of consensus on the definition/scope of the three types of decentralization, which leads to different scholars using their own conceptualization of decentralization. For example, some scholars put public finance management under administrative decentralization, but others conceptualize it under fiscal decentralization. This hinders systematic comparison among scholars. 8 5 Conclusion While acknowledging these merits and limitations, in my view, integrated quantitative and qualitative approaches using RAI-type indicators are the most comprehensive and objective measures to assess decentralization. This approach is even more robust when it includes citizens’ views on decentralization, as outlined by Ladner et al. (2019). While fiscal and administrative indicators appear to be comprehensive, there is ample room for improvement as regards political decentralization. 9 References Boex, Jamie and Yilmaz, Serdar. 2010. An Analytical Framework for Assessing Decentralized Local Governance and the Local Public Sector. IDG Working Paper No. 2010-06. The Urban Institute Center on International Development and Governance. Dardanelli, P., Kincaid, J., Fenna, A., Kaiser, A., Lecours, A. and Singh, A.K. 2019. “Conceptualizing, Measuring, and Theorizing Dynamic De/Centralization in Federations.” Publius: The Journal of Federalism 49(1) Winter: 1–29. https://doi.org/10.1093/publius/pjy036. Dardanelli, Paolo and Wright, Kieran. 2021. Devometrics: How to Measure Decentralisation? A Review of the Literature. Working Paper. Local Government Information Unit (KAR id:96045). Diamond, L. and Tsalik, S. 1999. “Size and Democracy: The Case for Decentralization.” In: L. Diamond (ed.), Developing Democracy: Toward Consolidation (pp. 117–160). The Johns Hopkins University Press. Dickovick, J. T. 2014. “Foreign Aid and Decentralization: Limitations on Impact in Autonomy and Responsiveness.” Wind Energy 34, 194–206. doi: 10.1002/pad.1691. Ebel, Robert D. and Yilmaz, Serdar. 2002. On the Measurement and Impact of Fiscal Decentralization. Policy Research Working Papers. Economic Policy and Poverty Reduction Division, World Bank Institute. Falleti, Tulia G. 2005. “A Sequential Theory of Decentralization: Latin American Cases in Comparative Perspective.” The American Political Science Review 99(3) (Aug. 2005): 327–346. Harguindéguy, Jean-Baptiste, Cole, Alistair, and Pasquier, Romain. 2021. “The Variety of Decentralization Indexes: A Review of the Literature”. Regional and Federal Studies 2021, 31(2): 185–208. ff10.1080/13597566.2019.156 6126ff. ffhal-02055944f. Hooghe, L., Marks, G., Schakel, A.H., Niedzwiecki, S., Chapman Osterkatz, S. and Shair-Rosenfield, S. 2016. “Transparency in Measurement.” In: Measuring Regional Authority: A Postfunctionalist Theory of Governance, Volume I, Transformations in Governance. Oxford: Oxford Academy. https://doi.org/10.1093​/acprof:oso/978 0198728870.003.0001. IMF. 2013. Government Finance Statistics. International Monetary Fund. https://data.imf.org/?sk=a0867067​ -d23c-4ebc-ad23-d3b015045405. Ladner, A., Keuffer, N., Baldersheim, H., Hlepas, N., Swianiewicz, P., Steyvers, K. and Navarro, C. 2019. “The Local Autonomy Index (LAI)”. In: Patterns of Local Autonomy in Europe. Governance and Public Management. Palgrave Macmillan, Cham. https://link.springer.com/chapter/10.1007/978-3-319-95642-8_9. 10 Oates, W.E. 1972. Fiscal Federalism. New York: Harcourt Brace. OECD. 2015. Fiscal Decentralisation Database. Organisation for Co-operation and Development (OECD). https://www.oecd.org/tax/federalism/fiscal-decentralisation-database/. Rondinelli, D.A. and Nellis, J.R. 1986. “Assessing Decentralization Policies in Developing Countries: The Case for Cautious Optimism.” Development Policy Review 4: 3–23. https://doi.org/10.1111/j.1467-7679.1986.tb00494.x. Schneider, A. 2003. “Decentralization: Conceptualization and Measurement.” Studies in Comprehensive International Development 38, 32–56. https://doi.org/10.1007/BF02686198. Stegarescu, D. 2005. “Public Sector Decentralisation: Measurement Concepts and Recent International Trends.” Fiscal Studies 26: 301–333. https://doi.org/10.1111/j.1475-5890.2005.00014.x Theriault, Sean. 2003. “Patronage, the Pendleton Act, and the Power of the People.” Journal of Politics 65: 50–68. 10.1111/1468-2508.t01-1-00003. Treisman, Daniel. 2000. “Decentralization and Inflation: Commitment, Collective Action, or Continuity.” The American Political Science Review 94(4) (Dec. 2000): 837–857. Williams, E. 2000. “Subnational Politics and Democratization in Mexico.” Wayne A. Cornelius, Todd A. Eisenstadt and Jane Hindley (eds). La Jolla, C.A.: Center for U.S.-Mexican Studies, University of California. World Bank. 2013. Decentralization Brief Webpage. https://www.worldbank.org/en/topic/communitydriven​ development/brief/Decentralization 11 Annex 1: The RAI Indicators – Self-rule and Shared Rule TABLE 1. Self-rule Authority to be Dimension Measured Score Weighted Indicator Institutional Extent to which 0–3 0 No functioning general purpose administration at regional level depth a regional 1 Deconcentrated, general purpose, administration government is 2 Non-deconcentrated, general purpose, administration subject to central autonomous government veto rather than 3 Non-deconcentrated, general purpose, administration not subject to central deconcentrated government veto Policy scope Range of 0–4 0 Very weak authoritative competencies in a), b), c), or d), whereby a) economic policies for policy; b) cultural/educational policy; c) welfare policy; and d) one of the which a regional following: residual powers, police, own institutional set–up, local government government is 1 Authoritative competencies in one of a), b), c), or d) responsible 2 Authoritative competencies in at least two of a), b), c), or d) 3 Authoritative competencies in d) and at least two of a), b), or c) 4 Criteria for 3 plus authority over immigration, citizenship, right of domicile Fiscal Extent to which a 0–4 0 Central government sets the base and rate of all regional taxes autonomy regional 1 Regional government sets the rate of minor taxes government can 2 Regional government sets the base and rate of minor taxes independently tax 3 Regional government sets the rate of at least one major tax: personal income, its population corporate, value added, or sales tax 4 Regional government sets base and rate of at least one major tax Borrowing Extent to which a 0–3 0 Regional government does not borrow (e.g., centrally imposed rules prohibit autonomy regional borrowing) government can 1 Regional government may borrow under prior authorization (ex-ante) by the borrow central government and with one or more of the following centrally imposed restrictions: a) golden rule (e.g., no borrowing to cover current account deficits) b) no foreign borrowing or borrowing from the central bank c) no borrowing above a ceiling d) borrowing is limited to specific purposes 2 Regional government may borrow without prior authorization and under one or more of a), b), c), or d) 3 Regional government may borrow without centrally imposed restrictions Representation Extent to which 0–4 Assembly: a region has an 0 No regional assembly independent 1 Indirectly elected regional assembly legislature and 2 Directly elected assembly executive Executive: 0 Regional executive appointed by central government 1 Dual executive appointed by central government and regional assembly 2 Regional executive appointed by a regional assembly or directly elected 12 TABLE 2. Shared Rule Authority to be Dimension Measured Score Weighted Indicator Law making Extent to 0–2 0.5 Regions are the unit of representation in national legislature which regional 0.5 Regional governments designate representatives in national legislature representatives  egions have majority representation in national legislature based on 0.5 R co-determine regional representation national legislation Executive Extent to which 0–2 0 No routine meetings between central and regional governments to negotiate control a regional policy government 1 Routine meetings between central and regional governments without legally co-determines binding authority national policy in 2 Routine meetings between central and regional governments with legally intergovernmental binding authority meetings Fiscal control Extent to 0–2 0 Neither the regional governments nor their representatives in national which regional legislature are consulted over the distribution of national tax revenue representatives 1 Regional governments or their representatives in a national legislature co-determine negotiate over the distribution of tax revenue, but do not have a veto the distribution 2 Regional governments or their representatives in a national legislature have a of national tax veto over the distribution of tax revenue revenue Borrowing Extent to which 0–2 0 Regional governments are not routinely consulted over borrowing constraints control a regional 1 Regional governments negotiate routinely over borrowing constraints, but do government not have a veto co-determines 2 Regional governments negotiate routinely over borrowing constraints and sub-national and have a veto national borrowing constraints Constitutional Extent to which 0–4 0 The central government or national electorate can unilaterally reform the reform regional constitution representatives 1 A national legislature based on regional representation can propose or co-determine postpone constitutional reform, raise the decision hurdle in the other chamber, constitutional require a second vote in the other chamber, or require a popular referendum change 2 Regional governments or their representatives in a national legislature can propose or postpone constitutional reform, raise the decision hurdle in the other chamber, require a second vote in the other chamber, or require a popular referendum 3 A legislature based on regional representation can veto constitutional change; or constitutional change requires a referendum based on the principle of equal regional representation 4 Regional governments or their representatives in a national legislature can veto constitutional change (Source: Hooghe et al. 2016, pp 28–29) 13