Local Public Finance and Capacity Building in Asia

Issues and Challenges

Report

OECD Fiscal Federalism Studies

22 December 2020

Download PDF

Chapter 4. Subnational capacity building: An international perspective of the Asia Pacific region

Dorothée Allain-Dupré

Isabelle Chatry

Louise Phung

Abstract

This chapter provides an overview of decentralisation trends and challenges in the Asia-Pacific region, and the enabling conditions that support subnational capacity building. The subnational landscape in the Asia-Pacific region is highly diverse when looking at territorial organisation, decentralisation patterns, subnational finances and capacities. Since the 1990s, decentralisation reforms have been carried out in many Asia-Pacific countries, transferring responsibilities and fiscal resources to subnational governments. These reforms translated into high fiscal decentralisation indicators – above world averages – but several bottlenecks to effective multi-level governance remain. Unclear assignments of responsibilities across levels of government and weak fiscal decentralisation are the most prominent challenges. The chapter provides an international perspective on the enabling conditions that support subnational capacity building, a key condition for decentralisation systems to work more effectively.

This chapter was prepared for the 2nd KIPF-OECD-ADB Roundtable of the Network on Fiscal Relations in Asia (RoNFRA). It builds on the OECD work Making Decentralisation Work: a Handbook for Policy-Makers and on data from the OECD/UCLG World Observatory on Subnational Finance and Investment.

Introduction

Decentralisation has increased around the world in recent decades, according to fiscal and institutional indicators (OECD/UCLG, 2019[1]). Decentralisation refers to the transfer of powers and responsibilities from the central government level to elected authorities at the subnational level, having some degree of autonomy. Decentralisation is also about reconfiguring the relationships between the central government and subnational governments towards more co-operation and a more strategic role for national/federal governments. Decentralisation is a multi-dimensional concept, covering three distinct but inter-related dimensions: the fiscal, administrative and political.

In Asia, many countries started to decentralise after the 1990s and decentralisation started to figure among the most prominent public reforms. Like in all regions of the world, there is considerable heterogeneity among countries in Asia Pacific in terms of the level and types of decentralisation. There are also common features, linked to some common characteristics of the region in terms of fast urbanisation, megacities, high population growth and diverse territorial organisation. These dimensions have an impact on the way decentralisation is shaped, with many asymmetric or differentiated systems within countries (OECD/KIPF, 2019[2]).

With such decentralisation trends in Asia and around the world, subnational governments have gained additional responsibilities in key areas linked to transportation, economic development, energy, education, health, social protection, housing, water and sanitation, for example. They play a key role for public service delivery and investment. Therefore, subnational governments need to develop financial, institutional, administrative and strategic capacities to manage these expanding responsibilities. It is thus essential to understand the enabling conditions, incentives and intergovernmental mechanisms that can support subnational capacity building. Building adequate capacities is a gradual process that takes time and requires long-term involvement and commitment from both central and subnational governments (OECD, 2019[3]).

The current COVID-19 crisis might contribute to redefine current multi-level governance arrangements. governments at all levels must act on all fronts simultaneously and in synchrony, requiring high degree of coordination and clear responsibility allocation while adopting a place-based approach to adjust for local/regional specificities. This need – for flexibility and adaptability – may lead governments to reconsider their multi-level governance systems, to revaluate their policy tools, and to reassess their regional development priorities (OECD, 2020[4]).

Decentralisation reforms are and have been implemented for a wide variety of political, historical, and economic reasons, which vary significantly across countries. Several moves towards decentralisation around the world have been mainly motivated by the quest for more local democratic control, as well as by the desire for greater efficiency in public service delivery and accountability for regional and local development policies. Megatrends linked to the information revolution or digitalisation, the globalisation of economic activity and urbanisation also play an important role in the strengthened role of subnational governments.

The outcomes of decentralisation have been debated for decades. According to the OECD (2019[3]), the question is not whether decentralisation is good or bad in itself; rather, it is a question of under which conditions decentralisation can support local democracy, efficient public service delivery and regional development. The outcomes of decentralisation depend on how it is designed and implemented, and several pre-conditions need to be in place for decentralisation to be successful. The OECD has identified ten preconditions for effective design and implementation of decentralisation reforms (OECD, 2019[3]).

This chapter provides a snapshot of subnational governments’ organisation and finance in the Asia-Pacific region, and focuses and enabling conditions to support subnational capacity building. The first section provides an overview of current trends occurring in Asia Pacific, compared to other regions in the world, in terms of territorial organisation and decentralisation. The second section provides some highlights on subnational government finance in Asia Pacific, compared to other regions of the world and OECD countries. The third section addresses the typical challenges in subnational government capacities, and the fourth section focuses on OECD guidelines for to help policy makers make decentralisation systems work more effectively, in particular by developing adequate capacities at subnational level. The chapter uses the new data and information from the 2019 OECD-UCLG World Observatory on Subnational Government Finance and Investment, notably from the sample of 16 countries of the Asia-Pacific region covered in the Observatory (Box 4.1).

Box 4.1. The OECD/UCLG World Observatory on Subnational Finance and Investment

The 2019 update of the World Observatory on Subnational Finance and Investment provides a relatively clear picture of the decentralisation frameworks in the world and in Asia Pacific. It identifies the current context and challenges linked to decentralisation, thanks to thorough qualitative and quantitative data collections that lasted two years. This World Observatory is the largest international database on subnational government structure and finance ever produced. It covers dozens of indicators for more than 120 countries, accounting for 86% of the world population and 89% of the world gross domestic product (GDP). The database is complemented by country profiles that provide quantitative and qualitative information on multi-level governance systems around the world. The synthesis report provides analysis by geographical areas, including Asia-Pacific, country income groups and institutional forms (federal or unitary countries).

In the World Observatory, 16 countries in the Asia-Pacific region are covered, including Australia, Bangladesh, Cambodia, the People’s Republic of China (hereafter “China”), India, Indonesia, Japan, Korea, Malaysia, Mongolia, Nepal, New Zealand, the Philippines, Sri Lanka, Thailand and Viet Nam.

Hereafter, in this chapter, when “Asia-Pacific countries” or the “Asia-Pacific region” or “Asia Pacific” are mentioned, this refers to the 16 countries mentioned above. The year of reference of fiscal indicators is 2016 but can vary country to country according to data availability.

Source: OECD/UCLG (2019[1]). More information can be found on the World Observatory website (including country profiles and data): www.sng-wofi.org.

Current socio-economic and territorial organisation trends in Asia Pacific

Fast urbanisation and diverse levels of development

Although great disparities exist among Asia-Pacific countries in terms of levels of development (see Annex Table 4.A.1), there are also common drivers in terms of urbanisation and economic growth. On average, urbanisation growth is high among Asia-Pacific countries, with an annual rate of 2.1% in 2017, slightly higher than the world average of 1.9%, which is driven by Africa (3.7%), and higher than the Latin America and European averages (1.6% and 0.6%, respectively). This can reflect a catch-up trend as, on average, only 53.9% of the Asia-Pacific population lives in urban areas compared to 71.4% in Europe and 72.1% in Latin America (OECD/UCLG, 2019[1]). This catch-up trend, i.e. relatively low urbanisation rate and high urbanisation growth, is the most acute in Cambodia, Nepal, Bangladesh and Viet Nam. More precisely for ASEAN countries, the population in cities is expected to reach 51% by 2050, i.e. 379 million people, with the largest metropolitan areas grow twice as fast as other urban cities. Comparatively, in the world, city populations have doubled over the last 40 years and will increase from 48% to 55% by 2050.

Among Asia-Pacific countries, the average annual population growth of 1% between 2010 and 2015 was close to the world average of 1.2%.1 By 2030, average annual population growth is forecasted to be slightly below 1.0% in Asia Pacific (United Nations, 2019[5]). While the average annual population growth between 2010 and 2015 is negative in Japan, and is below 0.5% in Korea, Thailand, China and Sri Lanka; it is above the world average in Indonesia, Malaysia, Australia, Cambodia, the Philippines and Mongolia, the average annual population growth is above the world average (Table 4.1).

Economic development and impact of COVID-19 crisis

The average real GDP growth in Asia Pacific is the highest regional average in the world, reaching 5.1% in 2017, compared to 3.9% in the world, 4.2% in Africa, 4.1% in Eurasia, 3.4% in Europe and 3.0% in Latin America. In 2017, in ten Asia-Pacific countries, the real GDP growth was more than 5%, the highest growth rates being 7.5% in Nepal and 7.3% in Bangladesh. In Thailand, Sri Lanka, Korea and New Zealand, the real GDP growth is between 3.0% and 3.9%, while it is 2.3% in Australia and 1.7% in Japan. The COVID-19 crisis has however put a halt to this strong economic growth (see Box 4.2). The Asia-Pacific region concentrates 5% of all cases reported at global level and 19% of deaths as of 2 December 2020. Within Asia-Pacific, India registered 60% of cases and 50% of deaths related to the COVID-19. Most other affected countries are Indonesia, Bangladesh and Philippines (ECDC, 2020[6]). The impact of the COVID-19 health crisis differs markedly not only across countries, but also across regions and municipalities within countries (see Box 4.2).

Box 4.2. COVID-19 crisis impact on economic development and multi-level governance

The World Bank estimates a contraction to -0.9% of economic growth in 2020 for the East Asia and Pacific region1, with long-lasting effects if the COVID-19 economic consequences remain unremedied (World Bank, 2020[7]). Korea has been the least affected country among OECD countries and is expected to register the smallest decline in GDP, just over 1% in 2020, due to effective measures to contain the spread of the COVID-19 in spring. Australia and New Zealand have also so far been less affected by the COVID-19 crisis, compared to other OECD countries. Real GDP is expected to contract by 3.8% and 4.8% respectively in 2020 with a rebound around 3.2% and 2.7% respectively in 2021. In Japan, however, the COVID-19 shock in early 2020 triggered a major recession and real GDP is projected to shrink by around 5.2% this year (OECD, 2020[8]). Within Emerging Asian countries2, the OECD forecasted that Thailand will be the worst affected, with an estimated GDP decline of 6.7% in 2020, and that Cambodia’s GDP growth will also turn negative in 2020. In China and India, economic output is expected to contract for the first time in decades before returning to a positive growth trajectory in 2021. On the contrary, Viet Nam will outperform its ASEAN-5 counterparts, posting a positive GDP growth for 2020 and the strongest growth in 2021 for the region (OECD, 2020[9]).

Territorial impact of the COVID-19

The COVID-19 crisis has a strong territorial dimension. For instance, the health of populations in some regions is more affected than in others. In India for example, the State of Maharashtra registered 19% of Indian confirmed cases as of end of November 2020, followed by Karnataka (9%) and Andhra Pradesh (8%). In the People’s Republic of China (hereafter ‘China’) 73% of confirmed cases were concentrated in Hubei province as of end of November 2020. Tokyo concentrated 31% of national cases by the end of October 2020. Large urban areas have been hard hit, but within them deprived areas are more strongly affected (OECD, 2020[4]).

Subnational governments – regions and municipalities – are at the frontline of the crisis management and recovery, and confronted by COVID-19’s asymmetric health, economic, social and fiscal impact – within countries but also among regions and local areas. The impact of the COVID-19 crisis and related policy responses (e.g. public health measures, lockdowns, emergency economic and social measures) on subnational government finance is significant. This observation is supported by the results from a survey jointly conducted by the OECD and the European Committee of the Regions (CoR) during the summer 2020 with 300 representatives of regional and local governments in 24 countries of the European Union. The results indicate that in the short and medium terms most subnational governments expect the socio-economic crisis linked to COVID‑19 to have a negative impact on their finances, with a dangerous “scissors effect” of rising expenditure and falling revenues (OECD, 2020[4]).

The current crisis requires all levels of government to act in a context of great uncertainty and under heavy economic, fiscal and social pressures. In addition, governments cannot sequence their policy actions as they need to manage the crisis, notably the second wave of pandemic, design exit strategies and support the recovery all at once. The need – for flexibility and adaptability – may lead governments to reconsider their multi-level governance systems, to revaluate their policy tools, and to reassess their regional development priorities (OECD, 2020[4]).

1. The World Bank East Asia and Pacific region encompasses 37 countries and territories: https://data.worldbank.org/country/Z4

2. 12 countries are covered in the Emerging Asia region as defined by the OECD: the members of the Association of Southeast Asian Nations (ASEAN) member countries: Brunei Darussalam, Cambodia, Indonesia, Lao PDR, Malaysia, Myanmar, the Philippines, Singapore, Thailand and Viet Nam; in addition to China and India.

Human development is very heterogeneous in Asia Pacific. Regarding the Human Development Index (HDI), six Asia-Pacific countries have a world rank below 100 and 12 below 50. Cambodia, Nepal, Bangladesh, India, Indonesia, Viet Nam and the Philippines have the lowest ranks in the Asia-Pacific region. On the opposite end, Australia, New Zealand, Japan and Korea rank among the best countries in terms of HDI. Five Asia-Pacific countries have a GDP per capita higher than the world average: the GDP per capita is more than double the world average in Australia, Japan and New Zealand and close to double the world average in Korea, while it is almost one and a half the world average in Malaysia.

	Urban population growth	Urban population	Population growth	GDP per capita	HDI	Real GDP growth
	Annual %	% of total population	Average annual %	USD PPP	World rank	Annual %
Australia	1.7%	85.9%	1.5%	48 460.0	3	2.3%
Bangladesh	3.2%	35.9%	1.2%	3 868.8	136	7.3%
Cambodia	3.3%	23.0%	1.6%	4 009.0	146	6.8%
China	2.7%	58.0%	0.5%	16 806.7	86	6.9%
India	2.4%	33.6%	1.2%	7 059.3	130	6.7%
Indonesia	2.3%	54.7%	1.3%	12 283.6	116	5.1%
Japan	-0.1%	91.5%	-0.1%	43 279.0	19	1.7%
Korea	0.4%	81.5%	0.4%	38 350.3	22	3.1%
Malaysia	2.2%	75.4%	1.4%	29 448.9	57	5.9%
Mongolia	1.7%	68.4%	1.9%	12 918.4	92	5.3%
Nepal	3.2%	19.3%	1.2%	2 696.7	144	7.5%
New Zealand	2.2%	86.5%	1.1%	41 109.0	16	3.0%
Philippines	2.0%	46.7%	1.6%	8 342.8	113	6.7%
Sri Lanka	1.5%	18.4%	0.5%	12 826.6	76	3.3%
Thailand	1.8%	49.2%	0.4%	17 872.2	83	3.9%
Viet Nam	3.0%	35.2%	1.1%	6 775.8	116	6.8%

Notes: GFCF: Gross Fixed Capital Formation. The year of reference is 2017 but can vary country by country according to data availability. More information can be found on the database. The annual average population growth rate is calculated over the period 2010-15.

Source: Authors’ elaboration based on the 2019 World Observatory on Subnational Government Finance and Investment database, found at www.sng-wofi.org.

Trends in decentralisation and territorial organisation

A number of countries have implemented decentralisation reforms in the Asia-Pacific region, mostly since the late 1980s and 1990s, as in the case of Australia, India, Japan and New Zealand. In countries like Korea or the Philippines, decentralisation policies have been conducted over the past three decades and are still high on the agenda (Box 4.3). For Bangladesh, decentralisation gained momentum in the 2000s while in Indonesia it has been more recent, around the 2010s. In many Asia-Pacific countries, decentralisation reforms continue to be on the agenda, but there are some exceptions, such as Thailand, where local elections have been suspended since 2014 (OECD/KIPF, 2019[2]; Smoke, 2019[10]) and will be held for the first time since in December 2020. Elections will be held at several levels – for provincial administration organisations (PAOs) first, then for tambon administration organisations (TAOs), municipalities and special administration areas (Bangkok Metropolitan Administration (BMA) and Pattaya City) (Government of Thailand, 2020[11]).

Box 4.3. Examples of decentralisation reforms in Asia Pacific

Indonesia engaged in a strong decentralisation process, together with the democratic transition in 1998. The Asian economic crisis participated in speeding up this decentralisation process, which led to two “big bang” decentralisation reforms in 2001 and 2005that granted subnational governments greater political autonomy and transferred to them substantial responsibilities, personnel, assets and resources. The laws 22/1999 on local government – also often referred to as the Regional Autonomy Law – and law 25/1999 on revenue sharing between the central and the regional governments were revised in 2004. Law 32/2004 introduced local direct elections and granted additional powers and responsibilities to subnational governments, giving provinces supervisory powers and strengthening their role as representatives of the central government, particularly in the area of planning and budgeting. Law 28/2009 specifies subnational government revenues, local taxation and charges. In September 2014, the Indonesian government enacted a new Local Government Law which aimed to restructure decentralisation to make the public sector more effective. The same year, the “Village Law” recognised the villages as self-governing entities and allocated them with more authority and resources.

In Japan, the decentralisation process was carried out through several steps over a long period, with the support of a Decentralisation Promotion Committee set up in 1995.The first “Decentralisation Promotion Reform” of 1995 led to the adoption of the Omnibus Decentralisation Law in 2000. This law was followed by the Trinity Reform in 2004-06, which laid out the financial component of the decentralisation reform. The “Decentralisation Promotion Reform” of 2006, granted further authority to local governments, rationalised their functions as well as the power of central government on local authorities and consolidated local administrative systems through municipal mergers.

In Korea, decentralisation started in 1987 with the “Decentralisation for Democratisation” and was followed in 1988 by the Local Autonomy Act and the Local Finance Act. Decentralisation gained further momentum in 1991 with the first election of local councillors and in 1995 with the first election of local governments’ chief executives. The “Special Act on Decentralisation”, enacted in 2004, clarified the principles and methods for decentralisation, transferred new functions to local governments and abolished special administrative agencies. It was followed by a fiscal reform in 2005 and then changed in 2008 for the “Special Act on the Promotion of Decentralisation”. The “Special Act on Decentralization and Restructuring of Local Administrative Systems” enacted in 2013 and revised in 2018 sets up the Presidential Committee on Autonomy and Decentralization, which developed a comprehensive the country’s plan for autonomous decentralisation, included a transfer of additional fiscal authority to local governments and included a devolution of centralised administrative powers to local governments effective from 1 January 2021. The goal of the reform is to enable local governments to autonomously establish and enforce polices tailored to their regional needs.

In the Philippines, the 1987 Constitution established the notions of decentralisation, local autonomy and popular participation. In 1991, a Local Government Code was set up to define the legal and regulatory framework of local governments. A “Master Plan for the Sustained Implementation” was developed in three stages of decentralization reform: (a) formal transfer of functions (1992–93) (b) adjustment by local governments (1994–96) (c) institutionalisation of the decentralised system (1997 onwards). The current political debate in the country is heading towards the acknowledgement of a more explicitly federal form of government. In December 2016, the presidency signed an executive order to prompt a consultative committee tasked with reviewing the 1987 Constitution. A draft Constitution was presented to the President in July 2018. It is proposed to establish 17 federated regions and the National Capital Region. The reform has not yet been adopted.

Source: OECD/KIPF (2019[2]); OECD/UCLG (2019[1]); OECD (forthcoming[12]).

According to the Regional Authority Index (RAI) (Hooghe et al., 2016[13]), an institutional indicator that aims to measure the degree of autonomy of regional governments (defined as states, regions, provinces, Länders, for example), 52 countries out of 81 of the world sample became more regionalised since the 1970s. This trend occurred in all regions of the world, including Asia. During the period 1950-2010, according to the RAI, reforms to strengthen the intermediary level of government (regions, states) have intensified since the 1990s in Asia, corresponding to the bulk of decentralisation reforms in the region (Figure 4.1). Australia, Japan, Korea and the Philippines experienced a significant increase in their RAI scores in the 1990s, while in Indonesia,2 regionalisation started in the 2010s.

Figure 4.1. Regionalisation in Europe, America and Asia since 1950

Levels of government

In Asia Pacific3, nine countries have two-tiered subnational governments4 and seven have three-tiered subnational governments5 (see Annex Table 4.A.3). As previously mentioned, the particularity of China is that its constitution states only three tiers of subnational government, while in practice, four layers exist. Usually, a hierarchical link exists between the different tiers. However, in Japan and Korea, where the subnational governments are two-tiered, there is no hierarchical link. Korea is characterised by a complex architecture with a diversity of organisations within each level. In Bangladesh, there is also no hierarchical link between the three tiers in the rural areas and the single tier in urban ones.

Many countries in Asia Pacific are also characterised by a territorial organisation that differentiates between rural and urban areas. This is also the case in many other countries in Europe or Latin America. This differentiation often exists at the municipal level and is often stated in legal texts or the constitution itself. This is the case in Bangladesh, India, Indonesia, Malaysia, Sri Lanka and Viet Nam. In New Zealand, a distinction is made according to the population size, and in Nepal, the distinction is made according to the population size and the subnational government’s resources. In Cambodia, Mongolia and Thailand, a distinction is made between the capital city and the rest.

As mentioned above, in Bangladesh, while the rural areas are organised in non-hierarchical three tiers, the urban areas are single tiers. In India, there is a differentiation between the 262 771 panchayat, the rural local bodies, and the 4 657 municipalities, the urban local bodies. In most Indian states, the panchayat are organised in three levels, the village, the “development block” and the district level, while the urban areas are classified in three categories according to their population size. In Malaysia, there are 12 city councils, 38 municipal councils (urban) and 98 district councils (rural). In Sri Lanka, there are 24 municipal councils, 41 urban councils and 276 rural councils.

In Viet Nam, the differentiation between rural and urban areas is made at the intermediate level of subnational governments. There are 8 978 communes, 1 581 wards and 603 commune-level towns at the municipal level. At the intermediate level, there are 546 rural districts, 49 urban districts, 51 district-level towns and 67 provincial cities, and at the regional level, there are 58 provinces and 5 centrally run cities (Can Tho, Da Nang, Ha Noi, Hai Phong and Ho Chi Minh City). In 2018 in Viet Nam, Ho Chi Minh City reached a population of 8.1 million inhabitants, and Ha Noi is projected to reach 6.4 million inhabitants by 2030 (United Nations, 2018[14]).

The high concentration of megacities in Asia Pacific can partially explain the special status attributed to urban areas and specific cities (Box 4.4). In 2018, out of the 33 megacities of 10 million or more inhabitants, more than half were in Asia (19). Besides, seven of the world’s ten largest urban agglomerations in 2018 were in Asia Pacific: Tokyo (37 million inhabitants), Delhi (29 million), Shanghai (26 million), Mumbai, Beijing, Dhaka and Kinki Major Metropolitan Area (Osaka) (these last four megacities have close to 20 million inhabitants each).

Box 4.4. Megacities in Asia Pacific and special status attributed to urban areas

Because of the large size of Tokyo, the municipal level in Japan is broken down into 1 718 municipalities (shichouson) and 23 special wards within Tokyo.

In Mongolia, Thailand and Cambodia, a special status is attributed to the capital city. In Mongolia, Ulan Bator (Ulaanbaatar) is at the regional level, its nine districts (duureg) are at the intermediate level, and the 152 neighbourhoods (khoroo) are at the local level. Ulan Bator had a population of 1.5 million in 2018, representing 71% of the urban population and 49% of the total population of Mongolia. Similarly, in Thailand, the Metropolitan City of Bangkok is erected at the regional level. The City of Bangkok reached 12.1 million inhabitants in 2018, representing 17% of the country’s population. In Cambodia, Phnom Penh, the capital city, is at the regional level, the khans are districts of Phnom Penh and are at the intermediate level, and the 236 Sangkats are the lowest tier of government in the capital city. In 2018, Phnom Penh had 1.9 million inhabitants, representing 51% of the urban population and 12% of the total population of Cambodia (United Nations, 2019[15]; United Nations, 2018[14]; OECD/UCLG, 2019[1]).

Number and size of municipalities

Although there is diversity across countries, a common trend across Asia-Pacific is that the number of municipalities per country is high, on average 26 366, and over five times the world average of 5 116 municipal-level governments. If this high number of municipal-level governments is mainly driven by India (267 428), then by Indonesia (83 344), the Philippines (42 045) and Viet Nam (11 162), the median number of municipal-level governments remains high (1 730) and is five times higher than the world median (340).

The size of municipalities, and their fragmentation, vary a lot among Asia-Pacific countries. In India, Indonesia, Philippines and Viet Nam, the municipal landscape is very fragmented with an average municipal size of fewer than 5 000 inhabitants and a total number of municipalities ranging from 11 000 in Viet Nam to over 260 000 in India. On the contrary, the municipal landscape is highly concentrated in China, Korea and Malaysia with an average municipal size over 200 000 inhabitants and fewer than 3 000 municipalities (see Figure 4.2). One explanation for the subnational fragmentation at the municipal level, among others, is geography as several Asia-Pacific countries are composed of islands. This specific geography must be taken into account to understand the territorial organisation and decentralisation systems in Asia-Pacific countries.

Figure 4.2. Average number of inhabitants per municipality by country

In 2018, out of the 33 megacities of 10 million or more inhabitants, more than half were in Asia (19). Besides, seven of the world’s ten largest urban agglomerations in 2018 were in Asia Pacific: Tokyo (37 million inhabitants), Delhi (29 million), Shanghai (26 million), Mumbai, Beijing, Dhaka and Kinki Major Metropolitan Area (Osaka) (these last four megacities have close to 20 million inhabitants each).

Subnational government finance in Asia Pacific

Subnational spending and revenue

Compared to other regions in the world, the Asia-Pacific region has – overall – higher indicators of fiscal decentralisation. Of the 16 countries in the Asia-Pacific region covered by the World Observatory on Subnational Government Finance and Investment, the share of subnational public spending represents on average 9.6% of GDP and 35.4% of public spending,6 which is higher than the world averages of 8.6% of GDP and 24.1% of public spending (Figure 4.3).7 The share of public investment undertaken by subnational governments (40%) is also higher than the world average (36%). This public investment represents 1.8% of GDP (compared to 1.3% on average in the world). Subnational revenue represents on average 9.0% of GDP (world: 8.5%) and 36.0% of public revenue (world: 24.4%), 41.4% coming from taxes (world: 33.0%). Subnational debt reaches on average 9.1% of GDP and 12.4% of the total public debt; it is at 69.3%, on average, of financial debt.

Figure 4.3. Subnational government expenditure as a percentage of GDP and total public expenditure by geographic area, 2016

However, large disparities exist among the 16 Asia-Pacific countries for these fiscal indicators (see Annex Table 4.A.2). For instance, half of the countries in the Asia-Pacific sample8 have subnational expenditure that reached more than one-third of public expenditure, and for four of them, subnational expenditure reached more than 50% of public expenditure in 2016 (Figure 4.3)). However, among the other half, subnational expenditure as a percentage of public expenditure is below the world average, reaching from 19.1% in Thailand to below 10% in Malaysia and Cambodia. In terms of tax revenue, there is also a large diversity of profiles (Box 4.1). Subnational tax revenue in China, India and Japan represents between 40% and 60% of public tax revenue, more than twice the world average (14.7%). In contrast, in New Zealand, the Philippines and Sri Lanka, subnational tax revenue as a share of public tax revenue represents less than half the world average.

A large diversity also exists within the Asia-Pacific region regarding subnational revenue. Taxes are the primary sources of revenue for subnational governments in Cambodia, India and New Zealand (over 50%), while grants and subsidies are the main sources of subnational revenues in many Asia-Pacific countries, including Indonesia, Sri Lanka and the Philippines (over 65%) (Box 4.5). Tariffs and fees, and property income, are relatively low compared to world averages, except in New Zealand and Australia (Figure 4.4).

Box 4.5. Subnational revenues in Asia-Pacific countries: A diversity of profiles

In 2016, less than 20% of subnational revenue came from taxes in Indonesia, while in New Zealand, India and Cambodia, they topped 50%. In Japan and China, taxes almost represent 50% of total subnational revenue.

Symmetrically, grants and subsidies represent less than 30% of total subnational revenue in New Zealand, India and Cambodia, while they represent more than 50% in Korea, Mongolia, Thailand, the Philippines Sri Lanka and Indonesia.

Tariffs and fees represent more than 9% of total subnational revenue (the world average) in only three countries, New Zealand, Australia and the Philippines. For the remaining nine countries in the Asia-Pacific region, the sample average is 3.5%, less than half of the world average.

The share of property income is also low in Asia Pacific, reaching on average 1.9% of total subnational revenue, compared to 2.3% for the world average. However, this Asia-Pacific average is driven by Australia, New Zealand and China. In Australia and Viet Nam, subnational governments benefit from royalties from mineral exploitation. In New Zealand, subnational governments’ revenues are diverse and come to a higher degree from service charges and fees as well as permits and licences. In China, the sale of land-use rights is a powerful subnational tool. When these three countries are removed from the sample, the sub-sample average drops to 0.5%.

Finally, for all Asia-Pacific countries, the share of other revenues, including social contributions, are below the world average of 4.7%.

Figure 4.4. Breakdown of subnational sources of revenue in Asia Pacific, 2016

Subnational spending responsibilities

The bulk of subnational expenditure in Asia-Pacific countries went to economic affairs and transport (20.1% of subnational expenditure on average), general public services (18.0%) and education (18.0%). Figure 4.5 shows that these three categories are the main subnational spending categories in every global region. Interestingly, the importance of economic affairs and transport in subnational spending is higher in Asia Pacific than anywhere else in the world, indicating that subnational governments play a more prominent role in economic development in many countries of this region.

Figure 4.5. Share of COFOG category in subnational public expenditure by geographic region, 2016

Figure 4.6. Share of COFOG category in subnational public expenditure by country, 2016

Public investment at national and subnational levels

One specificity of Asia Pacific compared to other regions of the world is the high level of public investment at national and subnational level.

At national level, in 2016, in Asia Pacific, 19.3% of the total public expenditure was allocated to direct investment on average, compared to 14.3% on average in the world, accounting for 4.8% of GDP versus 4.3% at global level (Figure 4.7). This high share of direct public investment is more than double the Latin American and European averages.

Figure 4.7. Public direct investment as a percentage of GDP and total public expenditure by geographic area, 2016

By countries, in 2016, direct public investment ranges from 2.8% in Sri Lanka to 11.6% in Malaysia. For almost three-fourth of the Asia-Pacific countries, the share of their public expenditure allocated towards direct investment is higher than the world average. While seven countries allocate between 20% and 40% of their total public expenditure to direct investment, five countries allocate between 14% and 18%. Australia, New Zealand and Japan are the only countries with a share below the world average, around 10.5%. In terms of percentage of GDP, while Mongolia and Malaysia allocate around 10% to direct investment; Sri Lanka, Australia, Indonesia allocate less than 3% (Figure 4.8).

Figure 4.8. Public direct investment as a percentage of GDP and total public expenditure by country, 2016

At subnational level, public investment is also high. Asia-Pacific subnational governments are key public investors. In 2016, 40.6% of public direct investment was carried out at the subnational level in Asia Pacific, compared to the world average of 36.6%. The subnational direct investment represents a higher share of GDP, at 1.8% in 2016 in Asia Pacific, compared to 1.5% in Latin America, 1.3% in Europe and Euro-Asia, and 0.9% in Africa. High disparities exist among Asia-Pacific countries regarding subnational direct investment (Figure 4.9). Subnational direct investment represented more than 50% of public direct investment in 7 out of 13 countries, and more than 60% in Australia, India, Japan, and Viet Nam. In terms of share of GDP, subnational direct investment exceeded 4% of GDP only in Viet Nam; in Australia, India, Japan and Korea, it ranged between 2% and 3%. Subnational direct investment was the lowest in Cambodia, Malaysia and Philippines, both as a share of public direct investment, below 10%, and as a share of GDP, below 1%; closely followed by Sri Lanka.

Figure 4.9. Subnational direct investment as a percentage of GDP and public direct investment by country, 2016

Real degree of autonomy more limited than what fiscal indicators indicate

When taking into account other dimensions of subnational autonomy, many countries in Asia Pacific are more centralised than what fiscal indicators indicate, as subnational responsibilities are highly regulated by the central government and resources are often shared with the central government. This is, for example, the case in countries like China, India, Indonesia, Korea, Malaysia, Mongolia, Thailand, and Viet Nam; China and Viet Nam represent a special case of centralised countries since they have a one-party political system. Fiscal indicators tend to over-estimate the real autonomy of subnational governments because spending allocations to subnational governments can be mandated and controlled by the central government. For instance, subnational governments can act as “paying agents” for the central government (OECD, 2019[3]; Musgrave, 1959[17]). As an example, public spending is highly decentralised in China, but the central government maintains high control over it. On the contrary, a relatively large degree of subnational autonomy exists in Australia, Indonesia, Japan, New Zealand and the Philippines.

In some countries such as India, Malaysia and Sri Lanka, a concurrent list of competencies exists between the central government and the highest tier of subnational governments. In India, for instance, the union list describes the 97 exclusive competencies of the central government; the concurrent list covers the 47 items shared between the central government and the states, and the state list describes the 66 states’ competencies. In Cambodia, Korea, Sri Lanka and Thailand, the central government has the power to intervene directly in subnational affairs. Finally, central governments’ control over subnational governments can be exerted through the appointment of subnational leaders, as in China, Malaysia, Mongolia and Sri Lanka.

Typical challenges in subnational capacities

Decentralising can bring various positive outcomes, but they are not given, and several pre-conditions are essential for subnational governments to have adequate capacities that enable them to fulfil their responsibilities effectively. Subnational government capacities are strongly associated with the degree of institutional quality in the country, the level of economic development, the level of human development indicators, the level of regional disparities within the country and the scale of the jurisdictions (see Chapter 1). The literature suggests that decentralisation might foster convergence when institutional quality is high but tend to exacerbate disparities in a low-quality environment, fuelling local capture (OECD, 2019[3]). The levels of economic and human development are also an important dimension because the higher the development of the country is, the higher the subnational governments’ abilities to raise revenue and to build strong capacities are. The level of regional disparities and the scale of subnational jurisdictions also matter for subnational capacity building. Up to a certain point, the larger the jurisdiction, the larger the tax base is, and the larger the economies of scale are, thus enabling larger capacities (Ter-Minassian, forthcoming[18]).

In its 2019 publication Making Decentralisation Work, the OECD highlights the main challenges related to decentralisation, notably for subnational capacity building. They encompass the unclear assignment of responsibilities, the weakness of fiscal decentralisation, poor horizontal and vertical co-ordination, and the lack of administrative capacities, especially in terms of human resources regarding adequate staff and expertise (OECD, 2019[3]; OECD/CoR, 2015[19]). Some challenges, like the unclear assignment of responsibilities and weak fiscal decentralisation, regarding unfunded and/or under-funded mandates, are more frequent in countries with recent decentralisation reforms, as in several Asia-Pacific countries.

Broader enabling environment: Institutional quality

The broader institutional environment matters significantly for decentralisation to produce positive outcomes and for subnational governments to have the ability to build their capacities (OECD, 2019[3]). Empirical studies have found that the positive impacts of decentralisation are less significant in developing countries compared to developed ones, suggesting that the context matters, including the institutional quality that is traditionally higher in more developed countries (Martinez-Vazquez and McNab, 2006[20]; Baskaran, Feld and Schnellenbach, 2016[21]).

Figure 4.10 supports this finding. The linear relation between expenditure decentralisation and the level of GDP per capita9 is strong for European countries (R²=0.45) while it is relatively weak for Euro-Asia countries (R²=0.17); this relation does not exist for Asia-Pacific, African and Latin American countries (R²<0.05). This finding is further supported by the existence of a positive relationship at the world level between expenditure decentralisation and the Human Development Index (UNDP, 2019[22]) while it is relatively weak among the Asia-Pacific group, as shown in Figure 4.11.

Figure 4.10. GDP per capita and expenditure decentralisation by region, 2016

Figure 4.11. Human Development Index and expenditure decentralisation by region, 2016

Figure 4.12 shows the quality of governance in the Asia-Pacific region in 2016, according to the 2017 version of the World Governance Indicators (Kaufmann, Kraay and Mastruzzi, 2011[23]). These indicators encompass six dimensions: government effectiveness, control of corruption, political stability, regulatory quality, rule of law, and voice and accountability. In 2016, in Asia Pacific, the average cumulative governance score reached 15.7 out of 30, similar to the world average (15.8).10 This is above the African and Latin American averages (12.1 and 14.5, respectively), but below the European average (20.3). Heterogeneity in Asia is high, with New Zealand, Australia, Japan and Korea scoring above 19 and a majority of countries scoring below 15, such as Bangladesh, Cambodia, Nepal, China, the Philippines, Viet Nam and Thailand. The lowest averages are for political stability, and voice and accountability, two indicators for which one-third of the countries score below two out of five. In 2016, the level of political stability was the lowest in Bangladesh, the Philippines, Thailand and India, and the level of voice and accountability was the lowest in China, Cambodia, Viet Nam and Thailand.

Figure 4.12. Cumulative Governance Score in Asia-Pacific countries, 2016

The implications for decentralisation policies are important: when the capabilities and framework conditions are not in place, implementing decentralisation policies can produce adverse outcomes. The economic aspect of development, which allows for financial flexibility and revenue delegation to subnational governments, is as important as the human aspect of development, which encompasses the technical and administrative capacities of the central government. The lack of human development, especially in terms of education, hinders the ability of the central government to design and implement decentralisation reforms, which are among the most complex, and limits the pool of skilled workers who can be hired as public servants at the subnational level. Blöchliger (2016[24]) considers that good governance is complementary to decentralisation and enables it to produce positive outcomes in terms of economic development, well-being and political stability. Acemoglu and Johnson (2005[25]) find that high-quality institutions are often related to better design and implementation of public policies, such as decentralisation reforms. Besides, high-quality institutions enable a more efficient provision of public goods and services, translating into economic development. Additionally, Rodríguez-Pose and Di Cataldo (2014[26]) find that the capability to design and implement effective policies as well as keep corruption at bay, both characteristics of good governance, are conducive to innovation and efficiency. If these framework conditions are not met, decentralisation reforms might lead to unbalanced and/or partial decentralisation and can be more harmful than a centralised architecture, since subnational governments would lack the concrete means to fulfil their responsibilities, leading to the under-provision or low-quality provision of public goods and services (Enikolopov and Zhuravskaya, 2007[27]).

Size and scale

The size of subnational jurisdictions matters significantly for the ability of subnational governments to deliver their public goods and services efficiently. A significant variation in demographic size of municipalities is observed around the world. In the Asia-Pacific region, data from the World Observatory on Subnational Finance and Investment show that there are significant differences across countries concerning the average size of municipalities. Size can range from a highly fragmented municipal landscape with fewer than 2 000 inhabitants on average in Mongolia to highly populated municipalities with over 200 000 inhabitants on average, as in the case of China, Korea and Malaysia (see above).

Determining the optimal subnational unit size is a context-specific task; it varies not only by region or country but by policy area, as well. The efficient size differs between waste disposal, schools or hospitals. In Finland, research on scale benefits of expanding the size of local governments found that large municipalities were less efficient at service delivery and the optimal size was between 20 000 and 40 000 inhabitants (OECD, 2017[28]; Moisio, Loikkanen and Oulasvirta, 2010[29]). Yet in Japan, unit costs of public services bottomed out at about 120 000 inhabitants and increased at both higher and lower municipal sizes (OECD, 2017[28]; OECD, 2019[3]). In New Zealand, for instance, the government decided to align the regional boundaries with the limits of the drainage systems (OECD/UCLG, 2019[1]).

In practice, subnational governments are often too small to deliver public services or invest at the relevant scale. Besides, the lack of co-ordination across jurisdictions prevent them from taking advantage of economies of scale and from overcoming the limitations their insufficient scale represents (see Box 4.6). In the European Union, the 2015 OECD/Committee of Regions (CoR) survey highlighted horizontal co-ordination as a major challenge for most subnational governments. More than three-quarters of the subnational governments surveyed reported that they had no joint investment strategy with neighbouring cities or regions. In addition, the same percentage of subnational governments reported that the lack of incentives – including financial – to co-operate across jurisdictions was a problem (OECD/CoR, 2015[19]; Allain-Dupré, 2018[30]).

Challenges linked to fiscal decentralisation

The fiscal dimension is very often the weak or even missing link of decentralisation. One of the most frequent challenges, particularly in countries at an early stage of decentralisation, is the misalignment between responsibilities allocated to subnational governments and the resources available to them. Unbalanced decentralisation, when subnational governments have little fiscal autonomy and rely heavily on grants and/or shared taxes, is also common (OECD, 2019[3]). Several countries like China, India and Viet Nam, have a high degree of subnational government expenditure and revenue. However, in practice, their subnational governments have little autonomy as expenditures and revenues are highly constrained from the national level.

A high proportion of the revenue in subnational governments in most Asia-Pacific countries come from taxes, unlike in the rest of the world, but like Eurasia. However, the high proportion of taxes can be a misleading indication of fiscal autonomy since taxes can be shared with the central government or higher levels of governments, leading to limited decision power for subnational governments. In addition, like the rest of the world, the share of grants and subsidies in total subnational revenue remains high, reaching 50% on average for Asia-Pacific countries.

Most Asia-Pacific countries are characterised by a high degree of shared taxation. Shared taxes, often encompassing personal income tax (PIT), value-added tax (VAT), corporate income tax (CIT) or excise tax, are national taxes that are redistributed to subnational governments either discretionarily or according to allocative criteria. In China, four taxes are shared uniformly between the central government and the provinces: the VAT, CIT, PIT and the securities trading tax. Each province then decides how to distribute it among its lower tiers. In addition, local governments have no discretionary power over their own-source revenue as the central government determines the legislation and rate for each tax. In Mongolia, subnational governments also have little tax autonomy as any changes in tax base or rate must be approved by central authorities. In Korea, subnational taxes, even after the 2011 tax reform, are highly determined by the central government. This is also the case in Viet Nam. It must also be noted that in many Asia-Pacific countries, tax autonomy on local taxes is restricted; even the local own-source taxes are significantly controlled by the central government through various regulations, such as the imposition of minimal or maximal rates and obligations to apply certain exemptions, resulting in significant fiscal imbalances (OECD/UCLG, 2019[1]).

There are great fiscal imbalances across countries between the level of subnational expenditure and the level of subnational tax revenue, higher than in OECD countries, because of the low level of tax (OECD/KIPF, 2019[2])]. For instance, the subnational shares of public expenditure and tax revenue (shared and own-source taxes) are almost equivalent in India, Japan and Mongolia, while in Australia, Indonesia and Sri Lanka, the subnational share of public expenditure is more than double the subnational share of public tax revenue (Box 4.5).

Figure 4.13. Subnational government expenditure and tax revenue as a percentage of public expenditure and tax revenue, 2016

A certain degree of fiscal autonomy is needed for the accountability mechanism to thrive. Indeed, relying on own-source revenue, such as taxes, rather than grants and subsidies from the central government, improves the quality and efficiency of spending, targeted more at local needs. It also improves budget management and allows for longer-term planning since own-source revenue can be a reliable and foreseeable source of revenue compared to grants and subsidies, especially when they are allocated discretionarily by the central government (OECD/KIPF, 2019[2]).

Challenges linked to administrative decentralisation

In many countries around the world, the lack of clarity in the assignment of responsibilities and the overlap in the way functions are performed are the most important obstacles to effective decentralisation (OECD, 2019[3]). Unclear and overlapping responsibilities lead to costlier service delivery and policy making. They also hinder subnational capacity building (OECD, 2019[3]). Nevertheless, a certain degree of shared responsibilities among levels of government, especially for distributive functions and those with economies of scale or spillovers, is necessary to enable co-operation between the various levels and improve public service delivery. Since shared responsibilities bring a higher degree of complexity in the administrative structure, they require very clear assignments, however, such as level A and level B share a function, but level A is in charge of maintenance, and level B is in charge of service delivery. This need for clarity is even more relevant when uneven responsibilities are attributed across governments at the same level, bringing another degree of administrative complexity.

Unclear and/or overlapping assignment of responsibilities is an issue that affects both centralised and decentralised countries. For example:

In China, most transfers of responsibilities are not in the law.
In Australia, there are significant overlaps in mandates between the federal government and the states, even after the creation of the 2008 Intergovernmental Agreement on Federal Financial Relations (IGAFFR), which aims to clarify responsibilities between the federal government and the states. However, the risk of overlap is relatively low between the state and municipal levels.
In Indonesia, provinces should, in theory, take care of services that require horizontal co‑ordination and/or cannot be taken at the local level. However, in practice, there are several overlaps and not clear-cut delimitation between the provinces’ and local governments’ assignments. Also, the 2014 Village Law that recognises villages as self-governing bodies does not assign them clear responsibilities.
In the Philippines, the 1991 decentralisation reform brings several challenges. First, the 1991 Local Government Codes provide some specifications of the new responsibilities for each level of government, but the responsibilities of provinces, cities, municipalities and barangays overlap. Also, the various public agencies provide public services and public work at the local level, creating another layer of overlaps. In addition, the codes do not delegate the funds required for Local Government Units (LGUs) to carry out their new responsibilities.

Another major challenge linked to administrative decentralisation is the ability of subnational governments to hire enough staff with adequate competencies to provide public services effectively. There are many variations within countries in terms of capacities, as well as in developed ones (Tselios et al., 2012[31]; OECD, 2019[3]). For example, according to a survey of the OECD and the EU Committee of the Regions, two-thirds of the subnational governments (65%) reported that the capacity to design adequate infrastructure strategies is lacking in their city/region and more than half of the subnational governments (56%) reported a lack of adequate own expertise in infrastructure (OECD, 2019[3]).

Challenges linked to political decentralisation

In theory, decentralisation and accountability mechanisms both reinforce each other. Indeed, if accountability mechanisms are needed for decentralisation to yield positive outcomes, decentralisation can also improve accountability. A first requirement for the accountability mechanism to work is the election of subnational representatives. Then, a certain degree of citizen participation, open, transparent and accessible information regarding one’s jurisdiction’s planning, finance and projects, and a certain degree of own-resource revenue’s dependency are essential for local/subnational representatives to build trust. The capacity to build trust among its residents is fundamental for a subnational government to obtain the support of its local citizens, to develop long-term projects and to implement them when being re-elected. A well-functioning accountability mechanism also stimulates citizens’ public participation and helps subnational governments collect information about residents’ preferences that, in turn, help improve public service. In sum, subnational elected representatives are more aware of their residents’ preferences and can better build the government’s capacities accordingly, that is reinforced by the fact that these representatives answer to the residents and not to the central government. In some countries of Asia-Pacific, local representatives may be appointed by the central/federal government, instead of being elected. Local elections have sometimes been suspended.

Enabling conditions to foster subnational capacity building

The OECD has identified ten guidelines to help policy makers make decentralisation systems work more effectively (Box 4.6) and to enable subnational governments to build their capacities to function well. These guidelines apply mainly at the national government level since they encompass some preconditions regarding legal and constitutional frameworks, which concerns the prerogatives of the national government (OECD, 2019[3]). National governments are also in charge of delegating appropriate resources, financial, administrative and human, to subnational governments, and of supporting co-operation. In fact, a well-functioning decentralisation system implies a renewed role for the central government directed more towards strategic planning, co‑ordination and support to subnational capacity building.

Box 4.6. Ten guidelines to make decentralisation systems work more effectively

1. Clarify the responsibilities assigned to different government levels.

2. Ensure that all responsibilities are sufficiently funded.

3. Strengthen subnational fiscal autonomy to enhance accountability.

4. Support subnational capacity building.

5. Build adequate co-ordination mechanisms across levels of government.

6. Support cross-jurisdictional co-operation.

7. Strengthen innovative and experimental governance and promote citizen engagement.

8. Allow and make the most of asymmetric decentralisation arrangements.

9. Consistently improve transparency, enhance data collection and strengthen performance monitoring.

10. Strengthen national, regional development policies and equalisation systems and reduce territorial disparities.

Source: OECD (2019[3]), Making Decentralisation Work: A Handbook for Policy-Makers, OECD Multi-level Governance Studies, OECD Publishing, Paris, https://doi.org/10.1787/g2g9faa7-en.

Clarify the responsibilities assigned to different government levels, and ensure that all responsibilities are sufficiently funded

Clear and principled assignment of responsibilities across levels of government is crucial for governments to deliver on their mandates, to be held accountable by citizens and to build instituting mechanisms for intergovernmental, as well as beyond government, partnership and co-ordination. Clarity in the division of powers can prevent overlapping or inefficient provision of public services since each governing body knows what its prerogatives are (Box 4.7). Clarity in the division of powers can also encourage citizen engagement and prevent the development of democratic deficits since citizens know to which governing bodies they can address their requests or claims. Responsibility assignments should not only be clear and codified, but there should be clarity in the division of spending powers to ensure that revenue means are aligned with expenditure needs and other national objectives (Allain-Dupré, 2018[30]; OECD, 2019[3]).

Some countries of the Asia-Pacific region have tried to address this issue and conducted reforms towards greater clarification of responsibilities and functions, including China, Indonesia, Japan and New Zealand (Box 4.7).

Box 4.7. Guidelines 1 and 2: Clarify the responsibilities assigned to different government levels, and ensure that all responsibilities are sufficiently funded: some reforms’ examples

The way responsibilities are shared should be explicit, mutually understood and clear for all actors. Equally important is clarity in the different functions that are assigned within policy areas – financing, regulating, implementing or monitoring. Since multi-level governance systems are constantly evolving, a periodic review of jurisdictional assignments should be made to ensure flexibility in the system.
The way different responsibilities across policy areas are decentralised should be balanced, i.e. when the various policy functions are decentralised to a similar extent.

In China, the intergovernmental fiscal reform launched in 2016 addressed the long-standing misalignment of revenue and spending across levels of government. The core part of the reform provided guidelines that included a review of expenditure responsibilities, clarifying responsibility assignments that were traditionally not mentioned in the law. The State Council plan outlined three broad categories of spending allocation (Wingender, 2018[32]).

In Indonesia, following Law 32/2004, local governments gained broad responsibilities, making Indonesia one of the largest decentralised countries in the world. In 2014, the government enacted a new Local Government Law in (no. 23/2014) to replace 2004 law, which aimed to restructure decentralisation to make the public sector more effective. The new law aims to provide clearer guidance related to the distribution of governmental functions between the central and subnational governments. It has redefined the distribution of responsibilities across all levels of government, defining exclusive responsibilities for the central government, concurrent responsibilities and general affairs. It also established a list of obligatory and discretionary functions (OECD/UCLG, 2019[1]).

In Japan, the 1988 Municipal Government Act provides a reference framework for the distribution of responsibilities across levels of government, making a distinction between mandatory responsibilities (including some which are shared with central government or delegated) and optional responsibilities. The 1999 decentralization law eliminated opaque central decision-making on local responsibilities and clarified competences more generally. Subsequent waves of reform have continued to develop the goals of greater municipal autonomy, clearer delineation of responsibilities, and proper financing (OECD, 2017[28]; Allain-Dupré, 2018[30]).

In New Zealand, the Local Government Act 2002 redefined subnational government responsibilities and increased their autonomy regarding the activities they undertake. The LGA 2002 separated policy making from policy implementation and provided subnational governments with a general power of competence. This 2002 Act was amended in 2010 and 2014, in line with the Better Local Government New Zealand reform with the aim, among others, to clarify responsibilities between regional councils and territorial authorities (OECD, 2017[28]). The division of responsibilities between, regional councils and regional and territorial authorities follows the principle of separation between responsibilities related to planning and those which are related to service provision. Regional councils have generally no direct responsibilities for service delivery (except for Auckland and Wellington) and are primarily responsible in sectors related to environmental protection and natural environment, transport, green areas and water management. Territorial authorities generally have responsibilities related to local development and service delivery, local infrastructure and community development and amenities. There is a high degree of co-operation between regional and territorial councils, which have complementary roles (OECD/UCLG, 2019[1]).

Strengthen subnational fiscal autonomy to enhance accountability

In addition to the clear assignment of responsibilities, subnational governments need adequate revenue allocation that matches their spending responsibilities to deliver public services effectively and to support the accountability mechanism. Different empirical studies consistently show that revenue decentralisation is more conducive to regional development than spending decentralisation. A balanced basket of revenues and a large share of own-source revenues enable subnational governments to have more reliable and forecastable revenue streams and to better design long-term development plans (Box 4.8).

In addition, the more subnational governments have fiscal autonomy over their revenue, the more enhanced the accountability mechanism is. Indeed, local residents must have incentives to monitor and evaluate the efficiency of their local administrations for the accountability mechanism to work properly. The literature identifies two requirements to trigger resident incentives: subnational governments depend on own-resource financing, meaning that local residents pay taxes that are directly used by their local representatives; and residents have open access to information about their own jurisdiction’s outcomes as well as those of other jurisdictions (Weingast, 2014[33]; Faguet, 2014[34]).

Box 4.8. Guideline 3: Strengthen subnational fiscal autonomy to enhance accountability

Subnational governments should have a certain degree of autonomy in the design and delivery of their public service responsibilities within the limits set by normative regulations, such as minimum service standards.
Subnational governments need own-source revenues beyond shared tax revenues – and they need to develop other sources of revenue to have a balanced basket of revenues.

Source: OECD (2019[3]), Making Decentralisation Work: A Handbook for Policy-Makers, OECD Multi-level Governance Studies, OECD Publishing, Paris, https://doi.org/10.1787/g2g9faa7-en.

Several fiscal instruments are well suited at various subnational levels and can be enhanced to strengthen subnational own-source revenues. The property tax is a local tax by nature and is widespread in Asia Pacific and in the world (see Box 4.9). However, in many Asia-Pacific countries, the central government keeps tight control over the definition of the tax base and rate and often oversees its collection and administration. Some exceptions exist, such as in New Zealand, where the subnational autonomy in property tax is large (OECD/KIPF, 2019[2]).

Box 4.9. The property tax in Asia Pacific

Box 4.13 displays the share of the recurrent property tax in subnational total revenue and tax revenue in 2016. In 2016, in New Zealand, the recurrent property tax represented more than three-quarters of subnational tax revenue (85.8%), and 44% of total subnational revenue. In Cambodia and the Philippines, the recurrent property tax represented respectively 46.6% and 39.6% of their subnational tax revenue, but while it represented 36% of Cambodia’s total subnational revenue, it represented only 9% of the Philippines’ total subnational revenue. In Australia and Japan, it represented between 25% and 30% of subnational tax revenue and less than 15% of total subnational revenue. The recurrent property tax only represents 17% to 13% of subnational tax revenue in China, Thailand, Korea and Mongolia; it represents less than 1% in India.

Figure 4.14. The tax revenue and recurrent property tax as a percentage of total subnational revenue in Asia Pacific, 2016

The other common tax instruments at subnational level are motor vehicle taxes (e.g. Australia, India, Indonesia, Japan and Korea), excise taxes, in particular on fuel or domestic goods and services (Indonesia, Malaysia, New Zealand, Russia, Thailand, Viet Nam), local business taxes, licences, tax on payrolls and professional tax (Australia, India, Japan, Korea, Mongolia, Philippines, Viet Nam), local consumption taxes (Japan, Korea), land use taxes (China, Viet Nam), taxes on natural resources (Indonesia, Viet Nam) and other minor taxes such as an education tax (Korea, Philippines), or taxes on touristic activities and entertainment. Other countries also have specific taxes targeted specifically at cities (city planning tax in Japan) metropolitan cities (Korea) or even a special tax for the capital Ulaanbaatar in Mongolia (OECD/KIPF, 2019[2]).

This large diversity of tax experiences in Asia-Pacific countries presents an opportunity to further strengthen the autonomy of subnational governments over these fiscal instruments since the taxation structures already exist and do not need to be created from scratch.

The high degree of informality in some countries, such as Bangladesh, India, Nepal and Sri Lanka, and/or the lack of a sound tax-collection system can be however a significant barrier to sustaining subnational finance with taxation. However, in Cambodia, major improvements to address informality in the economy came with a significant tax revenue increase, from 9.6% of GDP in 2009 to 15.3% in 2016. In the Philippines, locally collected revenue grew by 37% between 2013 and 2016 thanks to capacity-training programmes supported by the central government and international donor agencies.

In addition to make the most of taxation at subnational level, other sources of revenue, such as user charges and fees, licences and permits, can be valuable fiscal tools to finance subnational public services. . There is an opportunity to diversify and develop other sources of revenue, especially since alternative means of finance are currently low in Asia Pacific, including property income, tariffs and fees, and other types of revenue. Examples of other types of revenue in Asia-Pacific countries are mentioned in Box 4.5.

Support subnational capacity building

Subnational capacities in terms of quantity of staff, level of expertise and scale of actions are key to addressing complex issues such as strategic planning, infrastructure investment, oversight of public goods and services procurement or production and delivery, etc. (Box 4.10). It entails subnational governments first having the constitutional/legal/traditional empowerment to effectively hire, fire and set terms of employment of own employees; that they can co-ordinate or co-deliver policies and programmes with other governments (horizontally and vertically) and beyond government stakeholders; they can carry out fiscal and financial management; and they can audit and evaluate own services. In addition to these institutional abilities to perform this range of actions, subnational governments require resources, competencies, skills, and organisation to deliver high-quality services effectively, efficiently and sustainably. Governments should seek to reinforce the capacities of public officials and institutions with a systemic approach, rather than adopting a narrow focus on technical assistance only. For instance, staff training on local public financial management should be established and mandatory for all staff covering these functions. Governments should also institute mandatory training for budgeting department staff in budget methods, budget formulation, budget execution, revenue analysis and strategic planning (OECD, 2019[3]).

Box 4.10. Guideline 4: Support subnational capacity building

Central government should assess capacity challenges in the different regions on a regular basis. Policies to strengthen capacities should be adapted to the various needs of territories. Governments should seek to reinforce the capacities of public officials and institutions in a systemic approach, rather than adopting a narrow focus on technical assistance.
Staff training in the basics of local public financial management should be established. Open, competitive hiring and merit-based promotion should be ensured.
Special public agencies accessible to multiple jurisdictions should be encouraged in areas of needed expertise (e.g. regional development agencies, PPP units).

Source: OECD (2019[3]), Making Decentralisation Work: A Handbook for Policy-Makers, OECD Multi-level Governance Studies, OECD Publishing, Paris, https://doi.org/10.1787/g2g9faa7-en.

When subnational governments have limited financial and human-resource capacities or when no co-ordination among various levels of governments has been possible to deal with externalities or economies of scale, outsourcing can be a solution. As an example, in India, municipalities are relying more and more on the private sector to fund infrastructure improvements and to deliver basic services. However, for a service to be outsourced, it must be intrinsically profitable while fulfilling the principle of equality in public service access. Another alternative to outsourcing is the development of public-private partnerships (PPPs). However, a limitation of PPPs is their administrative and judiciary complexity, which requires sound administrative capacities at the subnational level or support from strong central government or a higher subnational level.

Building adequate capacities takes time and requires the long-term involvement and commitment of both subnational – and central – governments. Indeed, central governments are in charge of ensuring that subnational governments have enough capacities. In Indonesia, for example, the two major reforms of 2001 and 2005 induced significant delegation to subnational governments of political autonomy, important responsibilities and capacities. Two-thirds of the central government’s workforce has been transferred to subnational governments together with several assets and resources.

Equally, central governments should assess on a regular basis the capacity challenges of their subnational governments and adapt their policies to strengthen capacities according to their territories’ specific needs. They can also issue standardised guidance documents in key areas, such as planning, project appraisal, procurement, monitoring and evaluation, to disseminate to subnational governments. This practice is highly cost-effective; it helps subnational governments structure their work and save time and money since they do not have to allocate staff for these tasks. This practice can also lead to harmonised procedures across a national territory, which can, in turn, support horizontal co-ordination.

In addition, central governments can create or support the creation of special public agencies accessible to multiple jurisdictions in areas of needed expertise to help support subnational capacities (e.g. PPP units, regional development agencies) (OECD, 2019[3]). In Korea, for example, the Public and Private Infrastructure Investment Management Centre (PIMAC) established a Public-Private Partnership Unit that provides technical assistance to subnational governments. There is also a PPP centre in the Philippines, but the support provided to local government units is currently limited, and these units are not yet ready to engage in PPP.

The megatrend regarding digitalisation could also be an opportunity. A central government might produce online courses to train subnational public servants on various matters (town planning, budget management, PPPs, infrastructure investment, development planning, etc.) and could organise regular meetings with web conference tools. An online platform could also be designed to report subnational governments’ progress on their various projects. This type of platform could also be an opportunity to share cross-jurisdiction experiences with the support of the central government.

In countries with regional layers of governments, states governments, often in charge of overseeing several local governments, also have the administrative expertise to support local governments. Various forms of co-ordination mechanisms exist, such as having a regional public servant dispatched in local governments for specific projects; cyclical meetings to discuss the various functions and projects under the local governments’ responsibilities; mixed teams of regional and municipal public servants working on a specific service delivery; consulting services available for local governments, establishing local government training institutes, etc.

In 2007, Chile created the Academia de Capacitación Municipal y Regional to strengthen subnational capacities. It aims to be a technical reference for subnational staff and to strengthen human resources in municipal and regional governments to support a broad spectrum of knowledge of use in various territorial situations. It provides free training, in-person and online training for public servants. In addition, a Fund for the Training of Municipal Public Servants was created in 2014, financing technical and professional studies for municipal personnel (OECD, 2017[35]).
In Korea, the Ministry of the Interior and Safety host the Local Government Officials Development Institute (LOGODI) since its creation in 1965. The institute contributes to the development of local administration through strengthening the capacity of Korean and foreign local government officials thanks to various education and training programs. With a strong focus on international cooperation, the institute, for instance, provided training on capacity building for local administration the third week of November 2020 to more than 100 Ugandan Officials (Korean Ministry of the Interior and Safety, 2020[36]).
In the Indian state of Kerala, the Kerala Institute of Local Administration (KILA) is an autonomous institution constituted under the Ministry of Local Self Government, government of Kerala, with the mandate of facilitating and accelerating the socio-economic development of the State through strengthening the Local Self Government Institutions (LSGIs). KILA is a training, research and consultancy institution providing capacity-building intervention on local governance and decentralisation through training programmes of local elected representatives and officials, facilitation and strengthening of decentralised planning processes, documentation of local governance best practices for dissemination, etc. It is organised around 6 regional centres, each focusing on various topics such as tribal development and natural resource management, human resource development and good governance (KILA, 2020[37]).
In Philippines, the University of the Philippines and the Congress established the Centre for Local and Regional Governance (CLRG) in 1965. The CLRG is affiliated to the University of Philippines as its research, training and consulting centre for local governments and is a constituent unit of the national College of Public Administration and Governance. The CLRG is responsible for developing and contributing to responsive knowledge on local and regional governance, providing consulting services in local and regional governance systems, effectively developing competencies of local governance participants and nurturing a community of practice on local and regional governance (CLRG, 2020[38]).
In the United States, the National League of Cities (NLC), a voluntary organisation of municipalities, has established an NLC University to impart online and face-to-face training in municipal governance. It also produces toolkits and other training materials of use to municipal leaders and officials.

Build adequate co-ordination mechanisms across levels of government

Vertical and horizontal co-ordination across levels of government is crucial for any decentralised system to function well, and for subnational governments to be able to build capacity. Vertical co-ordination helps to both align policies at all government levels and enable higher levels of government to support lower levels in capacity building and in fulfilling their mandates (Box 4.11). Especially in the context of shared rules, a dense network of national/regional/local political and bureaucratic interactions is necessary to deliver public services effectively and efficiently. Indeed, delivery methods must be consistent across all levels of government and should not overlap. Specific instruments and platforms exist to foster vertical co-ordination, from consultation mechanisms to joint decision-making processes; they include dialogue platforms, fiscal councils, standing commissions and intergovernmental consultation boards, and contractual arrangements (OECD, 2019[3]).

Box 4.11. Guideline 5: Build adequate co-ordination mechanisms across levels of government

Since most responsibilities are shared, it is crucial to establish governance mechanisms to manage joint responsibilities. Creating a culture of co-operation and regular communication is essential for effective multi-level governance and successful long-term reform. Tools for vertical co-ordination include, for example, dialogue platforms, fiscal councils, standing commissions and intergovernmental consultation boards, and contractual arrangements.
It is important to avoid multiplying co-ordination mechanisms with no clear role in the decision-making process.

Source: OECD (2019[3]), Making Decentralisation Work: A Handbook for Policy-Makers, OECD Multi-level Governance Studies, OECD Publishing, Paris, https://doi.org/10.1787/g2g9faa7-en.

Examples from countries who have worked in this area include:

In Italy, inter-governmental coordination mechanisms are well developed. Italy has three levels of "conferences" between the central and subnational governments, serving as fora for intergovernmental coordination: a Conference of State-Regions, a Conference of State-Municipalities and other Local Authorities and a Unified Conference of State-Regions-Municipalities and Local Authorities, which includes all the members of the two other conferences (OECD, 2017[28]).
In Spain, vertical coordination between the central government and the autonomous communities is conducted, on a voluntary basis, through the Conference of Presidents (Conferencia de Presidentes), created in 2004. Vertical coordination also takes place through sectoral conferences such as the Council of Fiscal and Financial Policy (Consejo de politica fiscal y financiera, CPFF) for economic, fiscal and financial matters. Vertical coordination between the central government and local governments takes place through the National Commission for Local Administration (Comisiòn Nacional de Administracion Local), which was created in 1985. Autonomous communities have their own fora for coordinating with local governments under their jurisdiction.
In Australia, the new National Federation Reform Council (NFRC) replaces the Council of Australian Governments (COAG) since June 2020, following an agreement between the Prime Minister of Australia and the Premiers and Chief Ministers. The NFRC, which has the National Cabinet at its core, also comprises the Council on Federal Financial Relations (CFFR) and the Australian Local Government Association (ALGA). The NFRC meets once a year to focus on national priority issues. It is intended that the new model will streamline processes, enabling improved collaboration, communication and effectiveness. Under the new structure, the National Cabinet will oversee seven ministerial reform sub-committees in select areas, consolidating the work of 19 ministerial forums and nine regulatory councils. These areas include rural and regional; skills; energy; housing; transport and infrastructure; population and migration; and health (OECD, Implementation Handbook for Quality Infrastructure Investment, Forthcoming).
In India, the government is currently promoting a new federalism paradigm consisting of co-operative and competitive federalism. The idea is to shift from a top-down/planning approach to a bottom-up approach that promotes experimentation, benchmarking and the sharing of experience across states. Several examples of vertical co-ordination arrangements already exist: the Inter-State Council (ISC); the National Institution for Transforming India, also called NITI Aayog; and the Finance Commission of India. The ISC, in particular, is specified in the Constitution as a platform for strengthening Centre-State and Inter-State relations. The NITI Aayog, which replaces the Planning Commission instituted in 1950, was established in 2015 to foster co-operative federalism (OECD/UCLG, 2019[1]).
During the COVID-19 pandemic, the countries most successful in addressing the crisis have had strong coordination mechanisms across levels of government (OECD, 2020_[14]). For instance, the Korean government has strengthened the “whole-of-government approach” in the fight against COVID-19. The Prime Minister chairs the Central Crisis Management Committee, on which are represented all relevant central government ministries, as well as Korea’s seventeen provinces and major cities. As many Asian countries, South Korea draws on its experience with the SARS epidemic in 2003. In Australia, the central government has introduced a National Cabinet to address health and economic issues related to managing the COVID-19 crisis and recovery, gathering the Prime Minister and the First Ministers of each Australian State and Territory. The Australian Health Protection Principal Committee, a parallel group composed of all state and territory Chief Health Officers and chaired by the national Australian Chief Medical Officer, advises the National Cabinet (OECD, 2020[4]; Institut Montaigne, 2020[39]).

Reaching the right scale by merging and/or cooperating

A central government can decide to delegate a function to a certain level of government, but it cannot ensure that the jurisdiction boundaries match the “benefit area” and enable the subnational government to exercise the function in the most efficient way. In order to overcome this limitation, several options exist (Box 4.12). If few subnational functions produce externalities and/or would benefit from economies of scale, subnational governments can co-operate on these few specific functions, such as inter-municipal transports and energy power plants, or can create cross-jurisdiction local enterprises.

Box 4.12. Guideline 6: Support cross-jurisdictional co-operation

Horizontal co-ordination can be carried out using specific matching grants, and by promoting inter-municipal and interregional co-operation. Metropolitan governance should be promoted as well. The legal system should allow such tools.
Rural-urban partnerships should be promoted as a form of cross-jurisdiction collaboration to enhance inclusive growth by bringing multiple benefits, such as expanding the benefits of agglomeration economies, to overcome co-ordination failures and strengthen capacity.

Source: OECD (2019[3]), Making Decentralisation Work: A Handbook for Policy-Makers, OECD Multi-level Governance Studies, OECD Publishing, Paris, https://doi.org/10.1787/g2g9faa7-en.

National governments – as well as state governments – may promote inter-municipal cooperation by improving legal frameworks, spreading the values and benefits of cooperation amongst local elected, and providing financial (grants and other financial incentives) and technical incentives for cross-jurisdiction co‑operation.

Some countries have made headway in this area, including:

In New Zealand, the Local Government Act (LGA) 2002 Amendment Bill encourages inter-municipal co-operation and shared services between local governments. In 2014, in line with the “Better Local Government New Zealand” the LGA was amended in particular to further encourage inter-municipal co-operation and shared services.
In Australia, policies fostering inter-municipal cooperation are under states’ responsibility. Various types of inter-municipal co-operation exist depending on the state: Regional Local governments (Western Australia), Regional Subsidiaries (South Australia), County Councils (NSW), etc. Shared services arrangements are promoted at both state and local levels throughout Australia (OECD/UCLG, 2019[1]). In France, inter-municipal co-operative units can also have their own sources of tax revenue.
In India, federated states have established “urban development authorities” to address the lack of horizontal co-ordination among municipalities.
In Mongolia, cooperation between territorial units has been encouraged by development agencies active in the country. Inter-soum models have been developed in health management, and Habitat III’s national report mentions other various inter-soum experiments in the provision of other services (OECD/UCLG, 2019[1]).
In Japan, inter-municipal cooperation is increasingly promoted, in particular through voluntary partnership agreements that are established under the Local Autonomy Act.
In Germany, inter-municipal co-operation is strongly encouraged by the Länder, which decide independently on the rules of establishing such bodies. There are around 4 530 municipal associations (Gemeindeverband), which have different forms and status: offices, joint municipalities, association of communities and syndicates. Syndicates in particular are special-purpose associations created to deliver standard local services such as waste management, water and wastewater or transport. They are widespread throughout Germany and are one of the most common and oldest forms of inter-municipal co-operation in the country.

Local public companies are another alternative to improve the efficiency of service delivery due to cost-effectiveness requirements and to foster expertise. Japan, a decentralised country, has a network of 8 398 local public companies active in public service delivery, especially in sewerage (43.2% of all local public companies), water supply (22.9%) and hospitals (7.5%). Likewise, Indonesia relies on local public enterprises with around 650 local public enterprises owned and managed by regencies and cities, and 108 local public enterprises owned and managed by provinces, as of 2014. The local enterprises include drinking water companies and marketplaces.

Allow and make the most of asymmetric decentralisation arrangements

In most Asia-Pacific countries, certain subnational bodies have different self-governing statuses. Some regions are more autonomous than others (e.g. China, India, Malaysia and the Philippines). Some large cities have the status of an intermediate or regional government (in, for example, Cambodia, China, Japan, Korea, Mongolia, Thailand and Viet Nam) and some urban areas have a different status than rural ones (for example in Bangladesh, India, Malaysia, Nepal, New Zealand, Sri Lanka and Viet Nam). These asymmetric decentralised arrangements are more common in Asia Pacific than in other regions of the world (OECD/KIPF, 2019[2]). These asymmetric arrangements are both an opportunity to address geographic, ethnic and socio-economic heterogeneity and allow some degree of differentiation, and a challenge, as it creates additional complexity in terms of territorial organisation and can result in unclear definitions of responsibilities across levels of government and overlap in public service provision.

To make the most out of asymmetric arrangements, several conditions must be met (Box 4.13). They should be considered as experimental and evolving and should be monitored on a regular basis to highlight good practices and enable modifications. Finally, the number of asymmetric arrangements within a country must be kept within reason to limit co-ordination costs and complexity (Allain-Dupré, Chatry and Moisio, forthcoming[40]).

Box 4.13. Guideline 8: Allow and make the most of asymmetric decentralisation arrangements

Asymmetric decentralisation should be supported by effective vertical and horizontal co-ordination mechanisms and needs to go hand-in-hand with an effective equalisation system. An asymmetric decentralisation approach should be based on dialogue, transparency and agreements between all main stakeholders, and be part of a broader strategy of territorial development.
The way asymmetric responsibilities are allocated should be explicit, mutually understood and clear for all actors. To the greatest extent possible, participation in an asymmetric arrangement should remain voluntary.

Source: OECD (2019[3]), Making Decentralisation Work: A Handbook for Policy-Makers, OECD Multi-level Governance Studies, OECD Publishing, Paris, https://doi.org/10.1787/g2g9faa7-en.

Conclusion

This chapter provides an overview of the diversity of institutional and fiscal arrangements for subnational governments in the Asia-Pacific region. It provides an international perspective on subnational capacity building, with concrete guidelines from the OECD publication Making Decentralisation Work: A Handbook for Policy-Makers. In all types of systems, for decentralisation to be the most effective, national governments need to support subnational governments in their effort to build sound capacities. National governments should embrace a renewed role, directed more towards strategic planning, providing adequate enabling conditions and co-ordination, to ensure a well-functioning decentralisation system.

Annex 4.A. Indicator tables for selected Asia-Pacific countries

	Area	Population	Population growth	Urban population	Urban population growth	Density	Population of capital city	GDP per capita	Real GDP growth	GFCF	HDI	Unemployment rate
	km²	Thousand inhabitants	Average annual %	% of total population	Annual %	Inhabitants /km²	% of total population	USD PPP	Annual %	% GDP	World rank	% labour force
Australia	7 741 220	24 598	1.5%	85.9%	1.7%	3.2	1.8%	48 460.0	2.3%	24.0%	3	5.6%
Bangladesh	147 630	164 670	1.2%	35.9%	3.2%	1 115.4	11.9%	3 868.8	7.3%	30.5%	136	4.4%
Cambodia	181 040	16 005	1.6%	23.0%	3.3%	88.4	12.2%	4 009.0	6.8%	21.9%	146	0.2%
China	9 634 057	1 386 395	0.5%	58.0%	2.7%	145.0	1.4%	16 806.7	6.9%	41.9%	86	4.1%
India	3 287 259	1 339 180	1.2%	33.6%	2.4%	407.4	2.1%	7 059.3	6.7%	28.5%	130	3.5%
Indonesia	1 910 931	263 991	1.3%	54.7%	2.3%	138.1	4.0%	12 283.6	5.1%	32.2%	116	4.2%
Japan	377 962	126 728	-0.1%	91.5%	-0.1%	335.4	29.6%	43 279.0	1.7%	23.5%	19	2.8%
Korea	99 461	51 446	0.4%	81.5%	0.4%	517.2	19.4%	38 350.3	3.1%	31.1%	22	3.7%
Malaysia	330 345	31 624	1.4%	75.4%	2.2%	95.7	23.9%	29 448.9	5.9%	25.0%	57	3.4%
Mongolia	1 564 120	3 076	1.9%	68.4%	1.7%	2.0	49.4%	12 918.4	5.3%	24.7%	92	6.4%
Nepal	147 180	29 305	1.2%	19.3%	3.2%	199.1	4.5%	2 696.7	7.5%	34.0%	144	2.7%
New Zealand	267 710	4 820	1.1%	86.5%	2.2%	18.0	8.5%	41 109.0	3.0%	24.0%	16	4.7%
Philippines	300 000	104 918	1.6%	46.7%	2.0%	349.7	12.8%	8 342.8	6.7%	25.0%	113	2.4%
Sri Lanka	65 610	21 444	0.5%	18.4%	1.5%	326.8	2.8%	12 826.6	3.3%	26.3%	76	4.2%
Thailand	513 120	69 038	0.4%	49.2%	1.8%	134.5	13.7%	17 872.2	3.9%	23.2%	83	0.9%
Viet Nam	330 967	95 541	1.1%	35.2%	3.0%	288.7	4.5%	6 775.8	6.8%	23.0%	116	2.1%

Source: Authors’ elaboration based on the 2019 World Observatory on Subnational Government Finance and Investment database, found at www.sng-wofi.org.

	Total expenditure		Direct investment		Total revenue		Tax revenue		Total debt		Financial debt
	% GDP	% CG	% GDP	% CG	% GDP	% CG	% GDP	% CG	% GDP	% CG	% GDP	% CG
Australia	16.4%	63.3%	2.1%	72.5%	16.3%	49.6%	5.5%	20.6%	19.8%	29.1%	8.5%	22.6%
Bangladesh	..	..	..	..	..	..	..	..	0.1%	0.3%	..	..
Cambodia	1.0%	4.3%	0.2%	2.8%	1.2%	7.9%	0.9%	..	..	..	..	..
China	21.6%	91.2%	..	..	18.2%	72.0%	8.7%	49.6%	20.6%	46.6%	..	..
India	17.1%	61.8%	2.7%	63.4%	14.6%	70.8%	10.6%	60.1%	21.3%	29.9%	21.3%	35.0%
Indonesia	8.1%	47.7%	1.8%	59.0%	8.1%	55.9%	1.3%	10.9%	0.3%	1.1%	0.0%	0.1%
Japan	15.5%	39.7%	2.6%	68.7%	15.5%	43.5%	7.4%	39.9%	33.9%	15.3%	32.2%	14.9%
Korea	13.8%	42.2%	2.9%	58.4%	14.3%	41.4%	4.7%	24.1%	4.2%	7.6%	2.1%	4.3%
Malaysia	2.4%	7.8%	0.8%	7.2%	2.4%	13.4%	..	..	0.5%	0.6%	0.5%	0.6%
Mongolia	9.7%	24.8%	1.9%	19.6%	9.1%	35.9%	3.3%	22.7%	..	..	..	..
Nepal	..	..	..	..	..	..	..	..	..	..	..	..
New Zealand	4.4%	11.4%	1.3%	32.1%	4.3%	10.8%	2.2%	7.1%	5.8%	11.5%	4.8%	13.1%
Philippines	3.1%	14.9%	0.4%	8.5%	4.0%	20.8%	0.9%	6.1%	0.6%	1.4%	0.6%	..
Sri Lanka	2.3%	11.8%	0.3%	11.7%	2.3%	16.4%	0.6%	4.8%	..	..	..	..
Thailand	4.0%	19.1%	1.6%	52.0%	4.1%	19.3%	1.4%	9.3%	1.0%	2.4%	0.1%	0.3%
Viet Nam	15.6%	54.2%	4.3%	71.4%	11.2%	45.8%	..	..	1.5%	2.3%	1.5%	2.3%
Regional average	9.6%	35.3%	1.8%	40.6%	9.0%	36.0%	4.0%	23.2%	9.1%	12.4%	..	..

Notes: CG: Central Government. Each category is the share of subnational governments within the general government for this category. For instance, Total expenditure (% CG) means subnational expenditure as a percentage of total public expenditure; Tax revenue (% CG) means subnational tax revenue as a percentage of total public tax revenue. The year of reference is 2016 but can vary country by country according to data availability. More information can be found on the database.

Source: Authors’ elaboration based on the 2019 World Observatory on Subnational Government Finance and Investment database, found at www.sng-wofi.org.

	System	Municipal level	Intermediate level	Regional or state level	Total number of SNGs
Australia	Federal	562 local government areas		6 states + 2 federal territories	570
Bangladesh	Unitary	Rural: 4 553 union of villages Urban: 11 city corporations + 324 municipalities	Rural: 489 sub-districts	Rural: 64 districts	5 441
Cambodia	Unitary	1 410 communes + 236 Sangkats	159 districts + 26 municipalities	24 provinces + 1 capital city	1 856
China	Unitary	2 851 counties	334 prefectures	31 provinces	3 216
India	Federal	262 771 rural local bodies + 4 657 urban local bodies		29 states + 7 union territories	267 464
Indonesia	Unitary	83 344 villages	416 regencies + 98 cities	34 provinces	83 892
Japan	Unitary	1 718 municipalities + 23 special wards within Tokyo		47 prefectures	1 788
Korea	Unitary	226 municipalities		17 regional entities	243
Malaysia	Federal	Rural: 98 district councils Urban: 12 city councils + 38 municipal councils		13 states	167
Mongolia	Unitary	1 568 communities + 152 neighbourhoods	330 regions + 9 districts	21 provinces + 1 capital city	2 081
Nepal	Federal	276 urban municipalities + 460 rural municipalities + 6 metropolitan cities + 11 sub‑metropolitan cities		7 provinces	760
New Zealand	Unitary	67 territorial authorities		11 regional councils	78
Philippines	Unitary	42 045 villages	145 cities + 1 489 municipalities	81 provinces + 1 autonomous region	43 761
Sri Lanka	Unitary	24 municipal councils + 41 urban councils + 276 rural councils		9 provincial councils	350
Thailand	Unitary	Pattaya City + 30 city municipalities + 178 town municipalities + 2 232 sub-district municipalities		75 Provincial Administrative Organisations (PAOs) + Metropolitan City of Bangkok	2 517
Viet Nam	Unitary	8 978 communes + 1 581 wards + 603 commune-level towns	546 rural districts + 46 urban districts + 51 district-level towns + 67 provincial cities	58 provinces + 5 centrally run cities	11 938

Notes: SNGs: Subnational governments. The year of reference is 2017/18 but can vary country by country according to data availability. More information can be found on the database.

Source: Authors’ elaboration based on the 2019 World Observatory on Subnational Government Finance and Investment database, found at www.sng-wofi.org.

References

[25] Acemoglu, D. and S. Johnson (2005), “Unbundling institutions”, Journal of Political Economy, Vol. 113/5, pp. 949-995.

[30] Allain-Dupré, D. (2018), “Assigning responsibilities across levels of government: Trends, challenges and guiding principles for policy-makers”, OECD Working Papers on Fiscal Federalism, No. 24, OECD Publishing, Paris, https://doi.org/10.1787/f0944eae-en.

[40] Allain-Dupré, D., I. Chatry and A. Moisio (forthcoming), Asymmetric decentralisation: Trends, challenges and policy implications, OECD Publishing, Paris.

[21] Baskaran, T., L. Feld and J. Schnellenbach (2016), “Fiscal Federalism, Decentralization, and Economic Growth: A Meta‐Analysis”, Economic Inquiry, pp. 1445-1463.

[38] CLRG (2020), Centre for Local and Regional Governance, https://localgov.up.edu.ph/ (accessed on 7 December 2020).

[6] ECDC (2020), COVID-19 situation update worldwide, as of 2 December 2020, https://www.ecdc.europa.eu/en/geographical-distribution-2019-ncov-cases.

[27] Enikolopov, R. and E. Zhuravskaya (2007), “Decentralisation and Political Institutions”, Journal of Public Economics, Vol. 91, pp. 2261-90.

[34] Faguet, J. (2014), “Decentralization and Governance”, World Development, Vol. 53, pp. 2-13, http://dx.doi.org/10.1016/j.worlddev.2013.01.002.

[11] Government of Thailand (2020), Cabinet Approves the Holding of Elections of PAO Members, https://thailand.prd.go.th/ewt_news.php?nid=10194&filename=index (accessed on 4 december 2020).

[13] Hooghe, L. et al. (2016), Measuring Regional Authority: A Postfunctionalist Theory of Governance (Vol. 1), Oxford University Press.

[39] Institut Montaigne (2020), COVID-19: L’Asie oriental face à la pandémie, https://www.institutmontaigne.org/ressources/pdfs/publications/covid-19-lasie-orientale-face-la-pandemie-note1.pdf.

[23] Kaufmann, D., A. Kraay and M. Mastruzzi (2011), The worldwide governance indicators: Methodology and analytical issues, pp. 220-246, http://info.worldbank.org/governance/wgi/index.aspx#home.

[37] KILA (2020), Kerala Institute of Local Administration: About Us, https://www.kila.ac.in/about-us/ (accessed on 7 December 2020).

[36] Korean Ministry of the Interior and Safety (2020), Local Government Officials Development Institute, https://logodi.go.kr/en/ (accessed on 7 December 2020).

[20] Martinez-Vazquez, J. and R. McNab (2006), “Fiscal decentralization, macrostability, and growth”.

[29] Moisio, A., H. Loikkanen and L. Oulasvirta (2010), Public services at the local level - The Finnish way.

[17] Musgrave, R. (1959), The Theory of Public Finance, A Study in Public Economy, McGraw-Hill.

[9] OECD (2020), Economic Outlook for Southeast Asia, China and India 2020 – Update: Meeting the Challenges of COVID-19, OECD Publishing, Paris, https://dx.doi.org/10.1787/e8c90b68-en.

[8] OECD (2020), OECD Economic Outlook, Volume 2020 Issue 2: Preliminary version, OECD Publishing, Paris, https://dx.doi.org/10.1787/39a88ab1-en.

[4] OECD (2020), The Territorial Impact of COVID-19: managing the crisis across levels of government, OECD Publishing, https://read.oecd-ilibrary.org/view/?ref=128_128287-5agkkojaaa&title=The-territorial-impact-of-covid-19-managing-the-crisis-across-levels-of-government.

[3] OECD (2019), Making Decentralisation Work: A Handbook for Policy-Makers, OECD Multi-level Governance Studies, OECD Publishing, Paris, https://dx.doi.org/10.1787/g2g9faa7-en.

[35] OECD (2017), Making Decentralisation Work in Chile: Towards Stronger Municipalities, OECD Multi-level Governance Studies, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264279049-en.

[28] OECD (2017), Multi-level Governance Reforms: Overview of OECD Country Experiences, OECD Multi-level Governance Studies, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264272866-en.

[24] OECD (2016), Decentralisation and Quality of Public Finance: Intermediate Report, OECD Publishing.

[12] OECD (forthcoming), OECD Rural Policy Reviews: Korea.

[19] OECD/CoR (2015), Infrastructure Planning and Investment across Levels of Government: Current Challenges and Possible Solutions, OECD Publishing.

[2] OECD/KIPF (2019), Fiscal Decentralisation and Inclusive Growth in Asia edited by Kim, Junghun; Dougherty, Sean, OECD Fiscal Federalism Studies, OECD Publishing, Paris/Korea Institute of Public Finance, Seoul, https://dx.doi.org/10.1787/25cf7545-en.

[1] OECD/UCLG (2019), 2019 Report World Observatory of Subnational Government Finance and Investment - Country profiles, OECD Publishing, Paris, https://www.oecd.org/cfe/regional-policy/Observatory-on-Subnational-Government-Finance-and-Investment.htm.

[26] Rodríguez-Pose, A. and M. Di Cataldo (2014), “Quality of government and innovative performance in the regions of Europe”, Journal of Economic Geography, Vol. 15/4, pp. 673-706.

[10] Smoke, P. (2019), “Fiscal decentralisation and inclusive growth: The role of institutional context in selected Asian economies”, in Fiscal Decentralisation and Inclusive Growth in Asia, OECD Publishing, Paris, https://dx.doi.org/10.1787/67eca5ea-en.

[18] Ter-Minassian, T. (forthcoming), “Challenges of Subnational Capacity Development”, OECD.

[31] Tselios, V. et al. (2012), “Income inequality, decentralisation, and regional development in Western Europe”, Environment and Planning A, http://dx.doi.org/10.1068/a44334.

[22] UNDP (2019), Human Development Index, http://hdr.undp.org/en/content/human-development-index-hdi.

[5] United Nations (2019), Average annual population change (%) 2030-2035, https://population.un.org/wpp/.

[15] United Nations (2019), World Population Prospects 2019 Database, https://population.un.org/wpp/Download/Standard/Population/.

[14] United Nations (2018), World Urbanization Prospects: The 2018 Revision, https://population.un.org/wup/Publications/Files/WUP2018-Report.pdf.

[33] Weingast, B. (2014), “Second Generation Fiscal Federalism: Political Aspects of Decentralization and Economic Development”, World Development, http://dx.doi.org/10.1016/j.worlddev.2013.01.003.

[32] Wingender, P. (2018), Intergovernmental Fiscal Reform in China, IMF Working Paper, https://www.imf.org/~/media/Files/Publications/WP/2018/wp1888.ashx.

[7] World Bank (2020), Macro Poverty Outlook for East Asia and the Pacific: Data Dashboard, https://www.worldbank.org/en/publication/macro-poverty-outlook/mpo_eap#sec4 (accessed on 9 November 2020).

[16] World Observatory on Subnational Government Finance and Investment (2019), Country profiles, http://www.sng-wofi.org/country-profiles/.

Notes

← 1. In the World Observatory, data on demographic and economic indicators are available for all 122 countries.

← 2. Indonesia experienced a severe centralisation trend between 1950 (14.0) and 1979 (5.9) when it reached its lowest RAI score. The RAI score remained constant until 1998 and the fall of the President Suharto.

← 3. Data here are sourced from the 2019 update of the country profiles and the database of the World Observatory on Subnational Government Finance and Investment. The year of reference of the territorial organisation data are 2017/18 but can vary country by country according to data availability. There is data for all 122 countries of the database at www.sng-wofi.org. More information can also be found at https://stats.oecd.org/viewhtml.aspx?datasetcode=SNGF_WO.

← 4. Australia, India, Japan, Korea, Malaysia, Nepal, New Zealand, Sri Lanka and Thailand.

← 5. Bangladesh, Cambodia, China, Indonesia, Mongolia, Philippines and Viet Nam.

← 6. The averages are non-weighted. For the share of subnational spending as a percentage of GDP and as a percentage of public spending, no data are available for Bangladesh and Nepal.

← 7. The year of reference of fiscal indicators is 2016 but can vary country by country according to data availability.

← 8. The sample is reduced to 14 since no data on fiscal indicators are available for Bangladesh or Nepal.

← 9. Here, the coefficient of determination (R2) determines to which extent the information embodied in the level of GDP per capita explains the level of subnational expenditure. In simple linear regression, the coefficient of determination is the square of the Pearson-correlation coefficient.

← 10. To provide a consistent analysis, only the countries present in the World Observatory are considered to draw figures from the World Governance Indicators’ database, which covers 214 countries and territories.

Featured topics

Agriculture and fisheries

Climate change

Development

Digital

Economy

Education and skills

Employment

Environment

Finance and investment

Governance

Health

Industry, business and entrepreneurship

Regional, rural and urban development

Science, technology and innovation

Society

Taxation

Trade

Energy

Nuclear energy

Transport

Featured topics

Agriculture and fisheries

Climate change

Development

Digital

Economy

Education and skills

Employment

Environment

Finance and investment

Governance

Health

Industry, business and entrepreneurship

Regional, rural and urban development

Science, technology and innovation

Society

Taxation

Trade

Energy

Nuclear energy

Transport

Countries A - C

Countries D - I

Countries J - M

Countries N - R

Countries S - T

Countries U - Z

Regional and global engagement

Countries

Countries A - C

Countries D - I

Countries J - M

Countries N - R

Countries S - T

Countries U - Z

Regional and global engagement

Publications

Publications

Featured publications

Data

Data

Featured data

News & Events

News & Events

Featured Events

About OECD

About

Engage with us

Work with us

Featured topics

Agriculture and fisheries

Climate change

Development

Digital

Economy

Education and skills

Employment

Environment

Finance and investment