Chapter 2. Political economy perspectives on intergovernmental fiscal system design and implementation in Asia

Paul Smoke
New York University, Robert F. Wagner Graduate School of Public Service

This chapter, originally prepared for the 2nd KIPF-OECD-ADB Roundtable of the Network on Fiscal Relations in Asia (RoNFRA), builds on work done for the 1st RoNFRA (Smoke, 2019[1]) and on other work done by the author on the political economy of decentralisation. A special thanks to Pietro de Biase for help with the charts.

Despite advances in many countries in Asia and beyond, public sector decentralisation has proven to be a defiantly stubborn reform in many countries during the several decades it has been widely pursued. Available empirical literature indicates that performance, more often than not, falls short of expectations to varying degrees.1

Perhaps most surprising is the uneven progress with fiscal decentralisation. There is a well-articulated set of public finance (fiscal federalism) principles for designing fiscal decentralisation and supporting reforms, and it is often used as an anchor for broader decentralisation. Yet even where these principles have been (or appear to have been) followed, performance is too often unexceptional or problematic. This chapter argues that this frustrating situation persists both because the conventional principles do not adequately consider certain key factors that inevitably shape fiscal decentralisation and because the principles are not strategically used. Underlying both of these problems is a set of multifaceted contextual political economy and implementation considerations that rarely get the level or quality of attention they deserve.

This chapter focuses on the nature of these neglected challenges to fiscal decentralisation and intergovernmental fiscal reform and their implications for how more successful reforms might be pursued. The relevant challenges are considerable, including the scope and diversity of public and intergovernmental structures and functions; the incentives faced by and the behaviours of national politicians and bureaucrats who shape rules of the intergovernmental fiscal system and how they are implemented; subnational political economy dynamics among elected local politicians, local government staff and citizens; and the persistent imbalance between reform design and reform implementation. These elements interact in a broader context that informs the choice of options for effective fiscal decentralisation. Insufficient inattention to these factors and dynamics can produce nontrivial shortcomings in reform design and/or implementation.

The next section provides an overview of the scope and diversity of intergovernmental systems. The chapter then turns to neglected political economy dynamics – in the national political and bureaucratic arenas, at the local level, and in the behaviour of international development agencies that may be central players in fiscal decentralisation reforms in aid-dependent countries. This is followed by a section arguing that insufficient attention is paid to implementation and another that taking better account of context and implementation challenges could facilitate crafting and executing more viable and sustainable reform. The closing section offers summary comments on how to better incorporate broader political thinking into fiscal decentralisation analysis.

Although fiscal decentralisation has established norms, there is also general recognition that a broader array of legal and institutional factors require attention in pursuing reform. An adequate intergovernmental framework is indispensable if fiscal decentralisation is to be effective and sustainable, but the details of the framework may vary considerably across countries because the organisation and role of subnational governments are diverse.

All countries need an adequate intergovernmental framework to define the functional realms of different levels of government and other provisions that support fiscal decentralisation. The basic framework typically covers principles and practices for sharing public functions and powers among levels of government, along with structures, processes, and resources that support their operation. These frameworks include administrative, fiscal and political dimensions of subnational governance, and they also specify relationships among and within different levels of government (e.g. inter-jurisdictional co-operation between intermediate and local tiers of government and among distinct jurisdictions within larger metropolitan areas).

The importance of blending administrative, fiscal and political reforms is well recognised since their failure to work together can limit their impact. Weak fiscal decentralisation, for example, can undercut the capacity of, and incentives for, subnational elected and appointed officials to perform. Robust fiscal powers alone are not likely to be used effectively, however, if they are not disciplined by proper administrative and political systems and processes. A substantial number of conceptual papers and empirical studies recognise institutional design as a key element affecting the ability of decentralisation and intergovernmental reforms to be effective.2

Regarding fiscal design, much of the literature is largely framed around the tenets of first-generation fiscal federalism regarding expenditure and revenue assignment to subnational governments, e.g. assigning public services with chiefly local benefits for provision and financing at the subnational level. Higher-level governments should provide, share in or oversee the services with wider effects and levy broad-based taxes that are fiscally or administratively inappropriate for subnational governments. The details and underlying logic for decentralising certain functions and resources are well documented in the fiscal federalism literature and are not repeated here.3

An important consideration that is sometimes not sufficiently recognised is the need for balance between upward and downward accountability in decentralised systems. Although much of the focus is on decentralising functions and developing downward accountability, upward accountability to ensure compliance with procedural and service delivery standards is also essential. Such regulatory and oversight functions are needed for an effective public sector, but they can create obstacles to good performance if they unduly limit the subnational discretion that is expected to generate benefits or are inconsistently or arbitrarily applied.

It should be noted that the relevant literature acknowledges that normative principles are not always followed in practice, as discussed in more detail below. There may be legitimate historical and contextual reasons why the principles cannot or should not be followed. In addition, even if principles are formalised, legal provisions may not provide clear operational details. Finally, whatever policies and procedures are formally adopted, there are often considerable challenges to effective implementation. Flaws in the overarching national framework and more detailed specification of design features and implementation approaches are often considered to be among the most critical constraints on effective fiscal decentralisation.4

Some weaknesses are attributed to the narrowly defined scope of traditional fiscal federalism. Second-generation approaches expanded the coverage of relevant concerns, but largely with respect to specific concerns, rather than holistically.5 Other critical elements of the framework that are expected to affect performance, however, are even more extensive.6 These include provisions for development planning, public financial management and fiscal responsibility, civil service and human resource management, sector-specific policy frameworks and government partnership with private sector actors in performing public functions. Non-decentralisation-specific factors are also part of the framework. The nature of property rights, for example, affects local property taxation, and governance provisions (e.g. elections and non-electoral mechanisms) and civil society (e.g. the right to information and assembly) create space for citizen engagement that can help to shape subnational government behaviour and performance. Thus, although it is not sufficiently common in practice, the literature emphasises that the intergovernmental framework in principle should synthetically cover a broad, multi-dimensional constitutional, legal and administrative framework as well as the mechanisms required for its implementation and enforcement.

The diversity of institutional arrangements for decentralisation and intergovernmental relations creates challenges for identifying best reform practices. Many countries have more than one level of subnational government, and these may use a mix of the forms of decentralisation (devolution, de-concentration, delegation) in varied ways.7 One form may dominate, but forms may differ across levels, e.g. devolution at one and de-concentration at another. In some countries, such as India, intermediate tiers (states or provinces) have more power than lower tiers (municipalities, districts, etc.), but in other cases, such as Indonesia, the opposite is true. Dimensions may also vary across levels, e.g. provinces may receive more fiscal resources, but lower tiers may be more politically decentralised with directly elected local legislatures, as in Cambodia.

Although much of the literature on developing countries focuses on local level devolution, local governments are rarely the only (and may not be the main) service providers – there are relations among levels and/or joint responsibility. Levels may be more independent, as in the Philippines, or more hierarchical (e.g. higher tiers control lower tiers), as in Sri Lanka. There may be other government (at any level), parastatal and/or private entities with specific functional responsibilities, and these may or may not directly interact with elected subnational governments. Other governmental or nongovernmental actors may even overtly infringe on the legally defined functional territory of subnational governments.

Thus, subnational government performance must be understood in terms of the institutional framework in a particular country and the formal and informal relationships among differentially empowered levels and other governmental and non-governmental actors that are involved in various public functions. Without understanding such issues, reformers may try to pursue infeasible policies based on normative principles, and it may be difficult to explain the observed performance of subnational governments satisfactorily or to interpret the factors that shape it.

As noted above, there has been broad recognition of the importance of context in shaping decentralisation reform, but the details and nuances of the political economy landscape rarely get the attention they deserve. This landscape incorporates electoral and legislative politics as well as bureaucratic politics. It also has national and subnational dimensions.

The mainstream framework is premised on the assumption that decentralisation is undertaken for specific productive reasons: to increase efficiency in generating and using public resources, to improve public services, to enhance governance, and to alleviate poverty, among others. It is widely accepted, however, that many countries adopt decentralisation reform as much or more for political reasons.8 Some are reactions to economic or political emergencies, including conflict situations, as in Cambodia, Indonesia and Nepal, that generate pressure and opportunities for change. If reforms are hurriedly adopted to react to a crisis, there may be a limited analysis of or agreement on either the broad approach to power sharing or specific, related policies.

Such political motivations do not indicate that developmental goals are not also valued by the actors pushing reform, and such developmental advances can, of course, also promote political goals. Nevertheless, official developmental justifications may be less immediately significant than crisis alleviation, power consolidation or other political goals. The nature of these concerns influences the diversity noted above in terms of which levels of government are empowered (such as Indonesia’s emphasis on local levels due to concerns about empowered provinces seeking to secede), the type and strength of autonomy granted to each level, and the nature and pace of the process through which reforms are rolled out and managed. Cambodia and the Philippines, for example, moved rather slowly with decentralisation compared to Indonesia.

Even if decentralisation is genuinely motivated by a perceived need to nurture political credibility or to mitigate conflicts, once formally adopted, implementation may be hindered by national actors seeking to preserve their powers (more below), and the forces underlying decentralisation can shift. Rapid changes in volatile political environments can alter reform momentum, as has occurred, for example, in Pakistan and Thailand. A crisis may pass, or a new crisis may arise, producing incentives to shift course. Even without major contextual or policy changes, reforms can stall or be reversed through formal actions or tacit modification of activities controlled by central actors who are threatened by decentralisation, as has occurred to some extent in many Asian countries. There are even documented instances of recentralisation.9

A range of diverse central government agencies is typically involved in detailed efforts to design and implement fiscal decentralisation and intergovernmental fiscal reforms.10 These agencies, however, often have at least somewhat conflicting views of decentralisation and the appropriate role of subnational governments in public affairs. This may result in the pursuit of dissimilar approaches to how the intergovernmental system is organised and operated.

There is commonly a central agency in charge of regulating subnational governments, such as a ministry of local government, home affairs, interior, etc. Other agencies oversee specific aspects of public administration, such as planning, finance and the civil service, among others. Such agencies may be wary about or opposed to subnational government autonomy, often (officially) due to concerns about capacity that are framed as likely to result in misuse of public resources. In addition, sectoral agencies – for education, health, transport, water, etc. – prioritise service delivery over enabling subnational governance. Any of these national actors may have authority to weaken or impede subnational empowerment, perhaps with positive intentions related to their specific functional missions, but often at least in part to protect their own powers and funding.

This dissimilar set of institutional actors with conflicting perspectives and interests may adopt incompatible subnational government policies and procedures. For example, a local government ministry may pursue measures to strengthen subnational governments, while a finance ministry, a civil service commission, or a sectoral/service delivery agency may develop policies that diminish the powers of subnational governments, perhaps even over responsibilities legally allocated to them. Such challenges have been documented in Cambodia, Indonesia and the Philippines, among others.11 Some form of co-ordination is almost always officially endorsed to improve the compatibility of policies adopted by different agencies, but success in creating an effective instrument for this purpose has been elusive, including in many Asian countries.12

In some cases, central ministries may even directly provide public services that are legally subnational government functions, or they may assign such responsibility to special districts, parastatals or private sector actors. These measures may be justified to offer better service provision in the near term, but they may be inconsistent with provisions in the intergovernmental framework and may have adverse longer-term consequences. If these decisions violate legal requirements, their basis – legitimate criteria versus arbitrary or politicised decisions – and the role of subnational governments in the process of designing them should be documented.

Moving on to development assistance, when international agencies/donors support different national agencies pursuing decentralisation and intergovernmental reform, they may create challenges, particularly in aid-dependent countries.13 Despite official assertions about pursuing aid effectiveness agendas – country ownership, alignment with national policies, building country institutions, harmonisation with other external actors, etc. – these agencies may privilege their own priorities/approaches and mandate use of their own accountability mechanisms, even creating parallel entities. Such practices can hinder unified system development and introduce significant administrative burdens on client governments. Meeting aid effectiveness principles is particularly challenging in an era that seems to favour rapid verified positive results more than reformed systems and processes that may be essential to sustain good performance.

Another consideration is that donors and their departments may have priorities just as diverse as government agencies. A finance ministry charged with ensuring effective public resource use will likely prioritise central oversight over subnational government autonomy. Similarly, sectoral agencies will tend to emphasise service delivery and deal with subnational governments only insofar as this promotes their goals. Some donors (or donor departments) may favour enabling community or private sector initiatives over improving subnational governments, regardless of official country policies. If donors pursue agreements with particular government agencies that share similar priorities but are not necessarily following national law, they can reinforce or aggravate the type of central government policy incoherence discussed above.14

There is considerable evidence – of uneven strength and quality – that how subnational governments use legal authorities depends on the relative power of local political actors – business leaders, ethnic groups, political factions, labour unions, civil society, etc. – as well as the incentives their influence generates for politicians and administrators.15 Such local dynamics may create challenges for reform regardless of the normative quality of a national framework. Under some conditions, robust autonomy may allow elite capture or politicised access to services or enforcement of revenue compliance. Corruption may be more or less likely at the subnational level depending on local social and political relationships and the extent to which national upward accountability mechanisms can constrain problematic local political economy dynamics.

Mainstream fiscal federalism assumes a political mechanism that allows citizens to express their preferences on subnational government decisions.16 Impartial competitive elections are framed as foundational in devolved systems. Many countries in Asia hold competitive (in varied ways and degrees) subnational elections, but there are systemic or contextual challenges. Subnational legislatures may be partly appointed or controlled by managers chosen by the national government. In addition, voters may have to choose candidates on closed party lists, or one political party may be strong enough to impede genuine competition. How subnational electoral processes operate affects accountability whatever the provisions of the framework, but electoral systems can vary in their effects on subnational service delivery, revenue generation and other aspects of subnational government. As is often the case, context matters, but not always in predictable ways.17

Another important consideration is that subnational elections are held only every few years at best, and so are a fairly inexact accountability instrument. Given this, the value of non-electoral accountability mechanisms is well recognised. These may include participatory planning and budgeting, public meetings, local public service oversight bodies, public service user committees, feedback bureaus, and social auditing of subnational government activities, among others. These civic engagement mechanisms, various forms of which are common across Asia, promote improved public knowledge and solicit input on how subnational resources are generated and expended. When they work as intended, they can encourage more effective and equitable use of funding, enhance service delivery performance, and augment local social capital.18

At the same time, there are nontrivial caveats. Civic engagement processes can be implemented superficially, and their potential advantages can also be undermined by political dynamics. Participatory budgeting, for example, can be organised to reflect normative principles but may be only incompletely adopted. Participation may be superficial or not taken seriously by local officials, thus having little effect on local decisions or outcomes. Civic engagement mechanisms can also be dominated by local elites, whether political parties, business interests, or civil society groups. Even specific policies to augment inclusivity, such as mandatory roles for under-represented groups (e.g. a fixed membership share for minority groups), need not generate much pressure on subnational government decisions or lead to results sought by participants.

An additional qualification is that meaningful use of non-electoral accountability instruments requires citizens to have the information, ability and motivation to use them productively. Subnational governments may make budget data available or offer consultative forums, but constituents must know how to access them and feel comfortable using them. Particularly in newly democratising environments, citizens or specific groups may encounter obstacles in using civic engagement mechanisms. These may include, for example, lack of awareness, limited access to guidance and assistance, or even explicit intimidation that limits the use or open expression of opinions.

A well-conceived national intergovernmental framework with appropriate upward accountability and incentives for subnational governments to behave responsibly can reduce unduly politicised behaviour, but subnational realities can overwhelm good policy.19 How all of these factors interact (selectively summarised in Figure 2.1) to support or constrain service delivery, revenue generation, etc. will influence whether citizens feel fairly treated, and, therefore, whether they will be inclined to participate in subnational elections, to pay subnational taxes, to engage with participatory mechanisms and generally to be the active citizens required for the benefits of decentralisation to be realised.

Much of the decentralisation literature privileges system design consistent with mainstream principles noted above. Although systems must be adequately designed, even a normatively flawless system must be operationalised, and in a way that reflects political economy realities and other factors beyond fiscal concerns, as discussed above. Recently there has been some growing interest in the literature about how to implement and sequence decentralisation in a pragmatic and potentially more sustainable way, including capacity development.20

Although intergovernmental fiscal system designs are usually fairly comprehensive, they are not necessarily well linked to the larger intergovernmental framework discussed above, and the detailed design and implementation of individual elements of fiscal decentralisation are often done separately. To a certain extent, fragmentation is inevitable given the different responsibilities and priorities of various parties and the fact that not everything can be done at once. Such challenges are well documented in Asia.21 At the same time, more could often be done to frame the targeted efforts in the context of the larger system and according to a planned reform trajectory intended to eventually bring the interdependent elements of an intergovernmental system together.

Another commonly underappreciated consideration is the common asymmetry of subnational governments. Decentralisation design and implementation are often relatively standardised. There may be policies for distinct treatment of categories – for example, states/provinces versus local governments or urban versus rural governments. What is rarely taken into account, however, is that jurisdictions within an individual category often differ in capacity and performance. Thus, uniformly empowering all from the start may be counterproductive, and expecting them to implement reform at the same pace can generate poor performance.

There is obviously a role for both national and subnational governments in implementing decentralisation. Usually, the centre has a comprehensive legal framework, but the extent to which this framework is developed and how it is used can vary. On the one extreme, a framework adoption approach assumes the main task is done when the framework is issued because it is expected to incorporate incentives to encourage central and local actors to comply. Such an approach presumes that subnational governments will “sink or swim” in adopting legislated reforms. At the other extreme is a controlled gradualist approach, in which the centre may have a broad framework but makes all decisions about the order and pace of rolling out the elements of decentralisation reform. Without fair use of clear criteria, this process may lead to stalled reforms and could block even high-capacity local governments from meeting their potential. There have been calls to pursue a more strategic approach (more below) to implementation.

The scope a subnational government has for deciding on how to implement a reform will depend in part on what the national government allows and what type of support it provides, as well as local political and bureaucratic dynamics noted above. However, subnational governments face similar choices to the national government – they can try to do a lot at once if the legal framework empowers them to do so, or they can advance more gradually. Their specific approach will depend on what they feel they need to do, what they think local political conditions can support, what they think they have the capacity to handle. There is, however, also a case to be made for rolling out subnational implementation of decentralisation reforms in a more strategic manner.

Perhaps most fundamental in considering strategy, the organisational and operational alterations involved in implementing decentralisation often require substantial modifications in the mindsets and conduct of all involved parties. Central agencies – perhaps contrary to their dispositions and interests – need to learn to relinquish certain powers and transition from primarily supervisory and control functions to oversight and support. Subnational governments must learn to assume new roles and collaborate (with other subnational governments and other actors). Subnational government staff and elected officials (under devolution) must learn to work together, and subnational officials (elected and appointed) must develop skills to engage with constituents. Citizens also need to develop an understanding of their rights and duties and learn how to hold subnational governments accountable. Donors, particularly in aid-dependent countries, must cultivate a measure of co-operation with each other – at some point if not immediately (so as to allow sufficient experimentation) – in order to responsibly support country systems and policies.

Such behavioural evolutions are politically and institutionally demanding and require time and effort to evolve. If too many changes are rolled out quickly without efforts to influence attitudes, to generate effective incentives and to develop appropriate capacity, decentralisation reform will be unlikely to be effectively established and institutionalised. On the other hand, if reform is too modest and gradual, it may produce little visible change and the key players may lose interest.

One other foundational aspect of implementation merits specific mention. Capacity building, which is covered elsewhere in this volume, is obviously essential to implement decentralisation reform. There is much literature on capacity building, but it is often embedded in treatments of specific aspects of subnational government reform, and for the most part, does not highlight particularly generalisable findings. Some critics have expressed concern about the dominant “supply-driven” (by national governments and international agencies) approach to capacity development. Detractors complain that it privileges a uniform and comprehensive technical approach and often fails to appreciate the fundamental underlying incentives and dynamics outlined in this chapter.22

Much subnational capacity development remains dominated by conventional classroom instruction. Initiatives to foster more “demand-driven” (requested by subnational governments for immediately needed skills) and “on the job” coaching are recommended but unevenly adopted. Both standardised and tailored approaches can serve a role, with broad overview training providing a foundational understanding of the system and basic skills, while efforts to respond to subnational requests for skills required for proximate commitments can serve a complementary role in targeting, refining and reinforcing capacity on the ground.

A further concern is that capacity development is often concentrated on developing the technical skills of subnational government civil servants. More consideration should be given to governance capacity that supports subnational politicians and civil society – beyond the often-mechanical participatory planning and budgeting exercises – which may help to promote more productive civic engagement.

So how can reformers pragmatically deal with the many issues raised above in ways that might approach reform? 23 As previously indicated, there are useful principles and tools and some evidence regarding fiscal decentralisation. There is, however, no comprehensive analytical framework to systematically guide policy makers in how to support viable decentralisation reform, so informed judgment and a sense of pragmatism are key.

As noted above, relatively narrow technical analyses to define decentralisation reforms have become pervasive in a world where expertise is increasingly specialised, and there are few incentives for more holistic diagnostics. Not all analyses to support intergovernmental reforms can comprehensively take into account the broad scope of decentralisation and all relevant country details. At the same time, it is productive to consider how to prioritise reforms in the context of the larger landscape. Enhanced appreciation of linkages among reforms and political economy dynamics could potentially promote more reflective, cohesive and practical application of basic reform principles and generate more robust systems and results. Wider appreciation for the value of this type of analysis has surfaced more generally in recent years in the political economy analysis of development assistance and “doing development differently” literature.24

All implementation strategies need common elements to achieve their goals – a suitably defined entry point, an expected trajectory to roll out reforms over time, productive incentives for the uptake of reforms, capacity building/support structures and a managerially-oriented monitoring and feedback mechanism. When a central government agency managing decentralisation is considering implementation, developing a strategy will require attention to these elements.

Finding a starting point for action ideally involves both a broader vision in which to position reforms and the selection of attainable priorities. In cases where decentralisation is new or contentious, this may mean focusing on more basic reforms that don’t unduly threaten current power bases, such as those examined in the political economy discussion above, or excessively strain existing capacity. At the same time, it is useful to select an undertaking that is adequately consequential and visible, and that can initiate movement in an appropriate direction on a potentially feasible trajectory. In Cambodia, for example, efforts began on a small scale at a lower level of government but incorporated important technical and political reforms.

A strategic approach also includes the identification of interested partners who are likely to take the effort seriously. Thus, if a particular ministry or agency is prepared to take some concrete steps to transfer powers to subnational governments or enhance their ability to collect or manage resources, it could make sense to begin the process with such partners, rather than with agencies that are not committed to supporting reform, such as a ministry that wants to retain control over their pre-decentralisation functions and resources.

As noted above, national reforms often assume that subnational governments have similar absorption capacity. Expecting those with limited capacity to bear major responsibilities invites failure, while excessively controlling the more capable governments squanders resources and undercuts subnational accountability. Asymmetric entry points can be beneficial, and some reforms may, to some extent, even be negotiated with individual subnational governments rather than framed as universal central dictates. Such negotiation places some responsibility on a specific subnational government unit to comply with what they consent to do.

In choosing and sequencing reforms, individual components would ideally be appropriately connected to the extent feasible, even if initially in a rudimentary manner. As noted above, a fragmented approach can generate reforms that superficially meet traditional normative values but in fact, may end up being deficient in their ability to yield desired results. Improved subnational revenue collection, for example, requires not only technical and managerial reforms but also governance outreach to those expected to pay.

Positive and negative incentives (rewards and penalties) have the potential to motivate central and subnational governments to behave as intended under reforms. Where many actors with different but interdependent functions are involved, a co-ordinating mechanism would ideally oversee and enforce implementation, compelling all parties to act according to legal mandates. As noted above, the political economy of such co-ordination mechanisms is often challenging, as with the National Committee for Subnational Democratic Development in Cambodia, the Decentralisation Support Facility in Indonesia, and the National Decentralisation Committee in Thailand. Their commonly deficient performance, however, also reflects unrealistic expectations and flawed design that can potentially be corrected if there is sufficient consensus about how to start and proceed with reforms, and there is the motivation or pressure to do so.

Creatively defined inducements may also help to facilitate subnational government adoption of decentralisation reforms. These include, for example, monitored accountability mechanisms (e.g. central government contracts with subnational governments to meet specific reform targets, a successful initiative in Rwanda); financial incentives for reform adoption/performance gains (e.g. compliance or performance-based grants that reward subnational government attainment of specific targets, which have been used in Bangladesh, Cambodia, Indonesia, Nepal and the Philippines, among others); and tournament-based approaches (e.g. competitive contests that bring recognition by rewarding notable achievements of particular subnational governments, which have been used in several countries, including the Philippines).25

Capacity building and technical assistance for both central and subnational actors are essential for reform given the new roles they will have to play under decentralisation, but as indicated earlier, they are often framed by central governments and international agencies in an excessively uniform and mechanical way. The above discussion of capacity building noted specific concerns – a preference for traditional classroom training and technical skills, with less emphasis on skills perceived as urgent by subnational governments (demand-driven) and more limited focus on governance capacity or relations between elected and appointed officials.

There is consensus – in principle if not in practice – on the need to nurture both subnational technical capacity (of government staff who will be executing public functions) and subnational governance capacity (of citizens, elected officials and staff to work together). But related efforts are rarely designed to facilitate such interaction in a meaningful way. Beyond conventional courses, “on the job” training can enhance skills and retention, but it is more challenging than classroom training and requires dedicated effort to develop and execute.

The reform path (asymmetric if appropriate), should ideally be directly linked to dedicated capacity and performance-enhancement efforts. Technical reforms can be tied to efforts that build capacity for particular functions being implemented during a specific time period. By using defined criteria, reforms can be sequenced in a transparent way so that they build on each other. This is demanding relative to conventional approaches. It is not easy to design such a scheme; it could become overly bureaucratic and be captured by politics. Nevertheless, there is a need to be more innovative in implementing decentralisation, and if approached carefully, this type of strategic approach could reduce arbitrary or politicised decisions and limit stalled reform.

For such an approach to work effectively, there must be transparent monitoring of progress and performance as new reforms are implemented. There have been many decentralisation monitoring efforts, but they are often not assertively used to try to adjust general approaches or task/location-specific reforms. Even if they are, analysis often focuses on adopting a “symptom treating” type of reform that will not solve more fundamental obstacles to reform. Still, monitoring mechanisms have untapped potential as managerial tools, especially if linked to a sufficiently robust and integrated diagnosis of problems. Although the type of desired good information and its use for making policy changes is not well established in many countries, there have been positive steps taken to move in this direction. Indonesia, for example, has used performance information to make a number of changes in intergovernmental fiscal regulations.

Although it is difficult to neatly synthesise all of the relevant aspects of implementation, Figure 2.2 selectively summarises commonly neglected considerations. It also highlights elements of a more strategic approach to implementing intergovernmental fiscal reform.

The above discussion of strategy is generic, and it is more focused on thinking about how to roll out broader fiscal decentralisation reform. In many countries, some fiscal decentralisation is already in place, but challenges with aspects of the system need to be addressed. If local revenue generation under-performs, for example, details are needed to pursue a strategic approach. Problems identified by reformers are often symptoms of underlying issues that also need to be addressed if the identified flaw is to be resolved. If, for example, a local tax is unproductive and the tax base is small and/or stagnant and the rate is low, common mainstream solutions might include expanding the base, improving assessment or increasing the rate. But if the more severe and immediate problem is collection, enhancing the base or raising the rate may have a limited effect on yield or even worsen the collection situation. Indonesia, for example, is known for its shift from assessment-based to collection-based property tax reform.

If low revenue yield is determined to be a priority, it is essential not only to document the nature and extent of the deficiency but also to determine the core causes. It would also be essential to understand if any remedial measures have already been attempted and by whom. If they have, analysts need to determine why they have not worked. If steps have not been taken, the reasons for inattention to the problem need to be established. Weak subnational revenue generation may result from a wide range of factors. These could include, for example, incomplete development of the details of a general policy mandated by enabling legislation; poorly designed policies (e.g. excessive exemptions to a revenue base, low tax rates or disincentives created by fiscal transfers); poorly structured revenue collection mechanisms; data deficiencies; insufficient legal revenue enforcement authority; and lack of citizen and business tax compliance, among others.

Beyond these basic determinations, a range of capacity and political economy considerations could be shaping observed deficiencies. Flaws in the national framework could result from an unintentional omission or error in design or overt political attempts to limit subnational government powers. Weak follow-up and implementation efforts may occur because national government agencies actively obstruct implementation of legally devolved revenues or do not provide subnational governments with needed training and technical support. Such behaviour could reflect limited central bureaucratic capacity, budgetary constraints, apathy or overt hostility to enhancing subnational revenue generation, among others.

If the tax policy is well designed and central support is solid, analysis might consider why subnational governments have not used available revenue authority and support mechanisms. Again, a range of factors could be relevant here, including staffing deficiencies, inadequate funding, a local political preference to rely on intergovernmental transfers, an aversion among constituents to pay subnational taxes, etc. If citizens and businesses are not complying with tax requirements, their behaviour may be rooted in affordability concerns, or they may be dissatisfied on various fronts – with service delivery, in their perception of how fairly revenues are assessed and collected, or with the overall performance and credibility of subnational governments.

Such meticulous investigations can be tiresome, but they can help to identify the main factors and dynamics underlying observed performance problems and to begin determining if and how they can be eased. Some potential concerns might be excluded swiftly, while others would need further examination. Additional investigations could focus in more detail on factors that may influence subnational government behaviour and performance: national ministry conduct (policy inconsistency, weak co-operation among executing agencies); relationships at subnational levels (interaction between intermediate and local tiers of government and among adjacent local governments); subnational electoral accountability (degree of competition, citizen perceptions of fairness); and non-electoral civic engagement processes that affect how residents perceive and interact with subnational governments (accessibility to processes, degree of genuine influence).

Fully exhaustive analysis of individual fiscal decentralisation elements is not practical, but there will often be room to improve on the status quo. The analysis can be selective but should be appropriately broad, inquisitive and adaptive. Even generating a broad feel for answers to basic issues can help to indicate types of further investigation required and to determine prospective remedies. A conversant analyst will learn to determine the boundaries of the evaluation so as to concentrate on factors that are most valuable for identifying tangible reforms.

Depending on who initiates the analysis and where it leads, concrete measures will involve different prime actors. Only the national government, for example, can address weak subnational empowerment or ministerial interference in subnational affairs (although subnational governments may be able to take steps within prevailing constraints). Motivated subnational governments can improve their own capacity, attempt innovative approaches and interact more intensely with constituents in defining and advancing reforms. Citizens themselves can pressure subnational governments to change how they do business. Such steps may generate support or rouse resistance, some of which can be expected, from others that may require further action. In some cases, the constraints identified may be insurmountable, such that the most logical avenues of reform are blocked, and less effective, but more promising, options may have to be considered.

It is hard to draw satisfying conclusions from the extensive and complex literature on, and experience with, fiscal decentralisation, and space constraints limited coverage of specific examples. A few observations, however, can be made. First, fiscal decentralisation is often more diverse and intricate than academics and practitioners may appreciate. There has been a resilient tendency – despite acknowledgement of the significance of context – to approach decentralisation as a predominantly technical reform based primarily on normative principles. There has also been a propensity to handle different aspects of reform as if they were isolated phenomena rather than part of an intrinsically integrated system.

Second, there are potent constraints on achieving sustainable decentralisation reform consistent with orthodox principles. Institutional arrangements that emerged in specific contexts may not respond to reforms based on normative conventions. Deficient attention to political economy dynamics (national, intergovernmental, subnational) and weak appreciation of the gap between official and genuine reform goals and its effects for effective reform are common. In addition, favouring design relative to implementation produces challenges for decentralisation reforms.

Third, empirical evidence on decentralisation is limited, fragmented and inconsistent, a reflection of the difficulty of systematically assessing such a complex reform that is of interest to a range of different actors with diverse views and interests. Some research does substantiate a variety of fundamental conceptual expectations about decentralisation (e.g. better governance and service delivery) to different degrees, but there is also evidence to document limited or negative impacts.

In the final analysis, there is no strong empirical basis for confident, detailed guidance about how to approach fiscal decentralisation beyond limited general prescriptions that are unsurprising and not operationally specific. The evidence does confirm the extent to which “context matters” in pursuing reform that may generate positive results. The literature offers instances of good performance, but also many examples of mainstream programmes inadequately tailored to specific circumstances or insufficiently cognisant of specific constraints that did not perform as expected.

Despite the inadequacy of empirical evidence, there are a few basic lessons to draw from experience. First, progress can be made in decentralisation reform if there is appropriately nuanced and pragmatically oriented contextual scrutiny. Examining the political economy and other contextual factors can contribute to a better understanding of why decentralisation reform has been or should be approached in a particular way. Political economy dynamics influence how policy is originally defined, how the entities involved act in the course of implementation, the strength and evolving nature of intergovernmental relations, and how reforms perform in specific subnational jurisdictions facing their own political and bureaucratic dynamics.

Second, there can be value in integrating the elements of decentralisation reforms where feasible. The tendency to pursue fragmented reforms can produce inconsistencies in the structure and operation of intergovernmental systems. Some contradictions are a function of inherent tensions between technical/fiscal and governance reforms, but even specific aspects of technical reform can be inconsistent, such as intergovernmental transfers that create disincentives for subnational governments to meet their revenue generation and expenditure responsibilities. There is room for national governments to act more productively and strategically in defining decentralisation policies and constructing incentives to improve the performance of subnational governments.

Third, the somewhat limited attention to careful implementation of decentralisation is noteworthy. Where strategies exist, they are often perfunctory, disjointed across national agencies and not executed as planned. A suitably gradual, pragmatically sequenced, and contextually adapted process could allow subnational governments, national actors and citizens actors to gain the experience and competencies required to assume their evolving roles. Even limited progress in low-capacity environments can help to establish a foundation and facilitate momentum for deeper reforms with greater promise to be sustained. Of course, it is also crucial to guard against reforms stalling early on, so a strategic approach must be able to drive continued progress. This is not easy terrain to navigate, but proper preparation and monitoring could support a more pragmatic approach to rolling out (and adapting as needed) sustainable fiscal decentralisation reform in Asian countries.


[57] Agrawal, A. and J. Ribot (2012), Assessing the Effectiveness of Democratic Accountability Mechanisms in Local Governance, USAID, Washington, DC.

[23] Ahmad, E. and G. Brosio (eds.) (2015), Handbook of Multilevel Finance, Edward Elgar, Cheltenham.

[5] Ahmad, E. and V. Tanzi (2002), Managing Fiscal Decentralization, Routledge, Oxford.

[7] Ahmad, J. et al. (2005), Decentralization and Service Delivery, World Bank, Washington, DC.

[45] Altunbas, Y. and J. Thornton (2012), “Fiscal Decentralization and Governance”, Public Finance Review, Vol. 40/1, pp. 66-85.

[81] Andrews, M., L. Pritchett and M. Woolcock (2017), Building State Capability: Evidence, Analysis, Action, Oxford University Press, Oxford.

[77] Antwi, K. and F. Analoui (2008), “Challenges in Building the Capacity of Human Resource Development in Decentralized Local Governments”, Management Research News, Vol. 31/7, pp. 504-517.

[25] Bahl, R. and R. Bird (2018), Fiscal Decentralization and Local Finance in Developing Countries: Development from Below, Edward Elgar, Cheltenham.

[40] Bahl, R., J. Linn and D. Wetzel (eds.) (2013), Cities in the National Institutional and Fiscal Structure, Lincoln Institute of Land Policy, Cambridge, MA.

[20] Bahl, R., J. Linn and D. Wetzel (eds.) (2013), Financing Metropolitan Governments in Developing Countries, Lincoln Institute of Land Policy, Cambridge, Mass.

[89] Bahl, R., J. Linn and D. Wetzel (eds.) (2013), Grant Financing of Metropolitan Areas, Lincoln Institute of Land Policy, Cambridge, MA.

[72] Bahl, R. and J. Martinez-Vazquez (2013), “Sequencing Fiscal Decentralization”, Annals of Economics and Finance, Vol. 14/2, pp. 623-670.

[66] Baiocchi, G. and E. Ganuzza (2014), “Participatory Budgeting as if Emancipation Mattered”, Politics and Society, Vol. 42/1, pp. 29-50.

[8] Bardhan, P. and D. Mookherjee (eds.) (2006), Decentralization and Local Governance in Developing Countries: A Comparative Perspective, MIT Press, Cambridge.

[73] Bird, R. and J. Martinez-Vazquez (eds.) (2014), Why Theory and Practice Differ: The Gap Between Principles and Reality in Subnational Revenue Systems, Edward Elgar, Cheltenham.

[2] Bird, R. and F. Vaillancourt (eds.) (1998), Fiscal Decentralization in Developing Countries, Cambridge University Press, Cambridge.

[27] Boadway, R. and L. Eyraud (2018), “Designing Sound Fiscal Relations Across Government Levels in Decentralized Countries”, IMF Working Paper, No. 18/271.

[30] Boex, J. and S. Yilmaz (2010), An Analytical Framework for Assessing Democratic Local Governance and the Local Public Sector, The Urban Institute, Washington, DC.

[86] Booth, D. and S. Unsworth (2014), Politically Smart, Locally Led Development, Overseas Development Institute, London.

[79] Brinkerhoff, D. and P. Morgan (2010), “Capacity and Capacity Development: Coping with Complexity”, Public Administration and Development, Vol. 30/1, pp. 2-10.

[36] Brosio, G. (2014), Decentralization and Public Service Delivery in Asia, Asian Development Bank, Manila.

[92] Cabbanes, Y. and B. Lipietz (2018), “Revisiting the Democratic Promise of Participatory Budgeting in Light of Competing Political, Good Governance and Technocratic Logic”, Environment and Urbanization, Vol. 30/1, pp. 67-84.

[43] Chatry, I. and R. Vincent (2019), “A global view of sub-national governments in Asia: Structure and finance”, in Fiscal Decentralisation and Inclusive Growth in Asia, OECD Publishing, Paris,

[10] Cheema, G. and D. Rondinelli (eds.) (2007), Decentralized Governance: Emerging Concepts and Practices, The Brookings Institution, Washington, DC.

[62] Connerley, E., K. Eaton and P. Smoke (eds.) (2010), Decentralization and Community Empowerment, Lynne Rienner, Boulder.

[58] Connerley, E., K. Eaton and P. Smoke (eds.) (2010), Elections and the Development of Local Democracy, Lynne Rienner, Boulder.

[17] Connerley, E., K. Eaton and P. Smoke (eds.) (2010), Making Decentralization Work: Democracy, Development and Security, Lynne Rienner Publishers, Boulder.

[15] Crawford, G. and C. Hartmann (2008), Decentralization in Africa: A Pathway Out of Poverty and Conflict?, Amsterdam University Press, Amsterdam.

[35] Dafflon, B. and T. Madies (2013), The Political Economy of Decentralization in Sub-Saharan Africa: A New Implementation Model in Burkina Faso, Ghana, Kenya, and Senegal, Agence Française de Développement, Paris and World Bank, Washington, DC.

[53] Development Partner Working Group on Decentralization and Local Governance (2011), Busan and Beyond: Localizing Paris Principles for More Effective Support to Decentralization and Local Governance Reforms, GIZ, Bonn.

[54] Dickovick, J. (2014), “Foreign Aid and Decentralization: Limitations on Impact in Autonomy and Responsiveness”, Public Administration and Development, Vol. 34/2, pp. 194-206.

[97] Dickovick, J. (2011), Decentralization and Recentralization in the Developing World: Comparative Studies from Africa and Latin America, Penn State Press, University Park.

[21] Dickovick, J. and J. Wunsch (eds.) (2014), Decentralization in Africa: A Comparative Perspective, Lynne Rienner, Boulder.

[44] Eaton, K., K. Kaiser and P. Smoke (2011), The Political Economy of Decentralization: Implications for Aid Effectiveness, World Bank, Washington, DC.

[71] Ebel, R. and D. Weist (2006), Sequencing Subnational Revenue Decentralization, World Bank, Washington, DC.

[98] Faguet, J. (2014), “Decentralization and Governance”, World Development, Vol. 53/1, pp. 2-13.

[24] Faguet, J. and C. Pöschl (2015), Is Decentralization Good for Development? Perspectives from Academics and Policy Makers, Oxford University Press, Oxford.

[70] Falleti, T. (2005), “A Sequential Theory of Decentralization: Latin American Cases in Comparative Perspective”, American Political Science Review, Vol. 99/3, pp. 327-346.

[91] Fan, V. et al. (2018), “Fiscal Transfers Based on Outputs or Outcomes: Lessons from the Twelfth and Thirteenth Finance Commission in India”, International Journal of Health Planning and Management, Vol. 33, pp. e210–e227.

[48] Frank, J. and J. Martinez-Vazquez (eds.) (2016), The Political Economy of Local Infrastructure Planning in Developing Countries, Routledge, Oxford.

[52] Fritzen, S. (2007), “Linking Context and Strategy in Donor Support for Decentralization: A Diagnostic Framework”, Public Administration and Development, Vol. 27/1, pp. 13-25.

[63] Goldfrank, B. (2012), “The World Bank and the Globalization of Participatory Budgeting”, Journal of Public Deliberation, Article 7.

[75] Green, A. (2005), Managing Human Resources in a Decentralized Context.” In East Asia Decentralizes: Making Local Government Work, World Bank, Washington, DC.

[67] Heller, P. and V. Rao (eds.) (2015), But Who Will Speak for the People? The Travel and Translation of Participatory Budgeting, World Bank, Washington, DC.

[78] Hope, K. (2009), “Capacity Development for Good Governance in Developing Societies: Lessons from the Field”, Development in Practice, Vol. 19/1, pp. 79-86.

[16] Ichimura, S. and R. Bahl (2009), Decentralization Policies in Asian Development, World Scientific, Hackensack.

[88] Imansyah, H. (ed.) (2012), Incentives for Local Service Delivery: International Experience and Options for Indonesia, University of Indonesia Press, Jakarta.

[82] Khemani, S. (2019), “What is State Capacity?”, Policy Research Working Paper, No. 8734, World Bank, Washington, DC.

[28] Kim, J. and S. Dougherty (eds.) (2019), Fiscal Decentralisation and Inclusive Growth in Asia, OECD Fiscal Federalism Studies, OECD Publishing, Paris/Korea Institute of Public Finance, Seoul,

[83] Levy, B., V. Fritz and R. Ort (2014), Problem-Driven Political Economy Analysis: The World Bank’s Experience, World Bank, Washington, DC.

[3] Litvack, J., J. Ahmad and R. Bird (1998), Rethinking Decentralization in Developing Countries, World Bank, Washington, DC.

[65] Lund, J. and M. Saito-Jensen (2013), “Revisiting the Issue of Elite Capture of Participatory Initiatives”, World Development, Vol. 46, pp. 104-112.

[50] Malesky, E., C. Nguyen and A. Tran (2014), “The Impact of Recentralization on Public Services: A Difference-in-Differences Analysis of the Abolition of Elected Councils in Vietnam”, American Political Science Review, Vol. 108/1, pp. 144-168.

[18] Martinez-Vazquez, J. (2011), Fiscal Decentralization in Asia: Challenges and Opportunities, Asian Development Bank, Manila.

[19] Martinez-Vazquez, J. and F. Vaillancourt (eds.) (2011), Decentralization in Difficult Environments, Edward Elgar, Cheltenham, UK and Northampton, MA.

[80] McGill, R. (2010), “Building Capacity for Local Government to Perform”, Commonwealth Journal of Local Governance, Vol. 6, pp. 90-106.

[90] Mukherjee, A. (2014), Fiscal Federalism and Intergovernmental Transfers for Financing Health in India, Center for Global Development, Washington, DC.

[38] Oates, W. (2005), “Toward a Second-Generation Theory of Fiscal Federalism”, International Tax and Public Finance, Vol. 12/4, pp. 349–373.

[34] Oates, W. (1972), Fiscal Federalism, Harcourt, Brace, Jovanovich, New York.

[42] OECD and UCLG (2019), Subnational Governments Around the World: Structure and Finance, OECD Publishing, Paris and United Cities and Local Governments, Barcelona.

[41] OECD and UCLG (2016), Subnational Governments Around the World: Structure and Finance, OECD Publishing, Paris and United Cities and Local Governments, Barcelona.

[47] Öjendal, J. and A. Dellnäs (eds.) (2013), Decentralizing for Development: The Developmental Potential of Local Authorities and the Limits of Politics-Driven Decentralization Reforms, United Nations University Press, Tokyo.

[64] Öjendal, J. and A. Dellnäs (eds.) (2013), Participatory Budgeting and Local Governance, United Nations University Press, Tokyo.

[49] Öjendal, J. and A. Dellnäs (eds.) (2013), Recentralization in Developing Countries: Forms, Motivations and Consequences, United Nations University Press, Tokyo.

[68] Öjendal, J. and A. Dellnäs (eds.) (2013), Sanctions, Benefits and Rights: Three Faces of Accountability, United Nations University Press, Tokyo.

[46] Öjendal, J. and A. Dellnäs (2013), The Imperative of Good Local Governance: Challenges for the Next Decade of Decentralization, United Nations University Press, Tokyo.

[31] Öjendal, J. and A. Dellnäs (eds.) (2013), Understanding Decentralization: Key Issues for Successful Design, United Nations University Press, Tokyo.

[93] Ponce-Rodríguez, R. et al. (2018), “Rethinking the Political Economy of Decentralization: How Elections and Parties Shape the Provision of Local Public Goods”, Publius: The Journal of Federalism, Vol. 48/4, pp. 523–558.

[22] Rao, S., Z. Scott and M. Alam (2014), Decentralisation and Local Government: Topic Guide, GSDRC, University of Birmingham, Birmingham.

[84] Rocha Menocal, A. (2014), Getting Real about Politics: From Thinking Politically to Working Differently, Overseas Development Institute, London.

[29] Rodden, J. and E. Wibbels (eds.) (2019), Decentralized Governance and Accountability: Academic Research and the Future of Donor Programming, Cambridge University Press, Cambridge.

[60] Rodden, J. and E. Wibbels (eds.) (2019), Leadership Selection Rules and Decentralized Governance, Cambridge University Press, Cambridge.

[59] Rodden, J. and E. Wibbels (eds.) (2019), Traditional Leaders, Service Delivery, and Electoral Accountability, Cambridge University Press, Cambridge.

[51] Romeo, L. (2003), “The Role of External Assistance in Supporting Decentralization Reforms”, Public Administration and Development, Vol. 23/1, pp. 89-96.

[55] Scott, Z. (2009), Decentralization, Local Development and Social Cohesion: An Analytical Review, University of Birmingham, GSDRC, Birmingham.

[85] Shah, A. (2010), Sponsoring a Race to the Top: The Case for Results-Based Intergovernmental Finance for Merit Goods, World Bank, Washington, DC.

[61] Shah, A. (ed.) (2007), Participatory Budgeting, World Bank, Washington, DC.

[69] Shah, A. and T. Thompson (2004), “Implementing Decentralized Local Governance: A Treacherous Road with Potholes, Detours, and Road Closures”, Policy Research Working Paper, No. 3353, World Bank, Washington, DC.

[1] 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,

[74] Smoke, P. (2015), “Managing Public Sector Decentralization in Developing Countries: Moving Beyond Conventional Recipes”, Public Administration and Development, Vol. 35/4.

[37] Smoke, P. (2015), “Rethinking Decentralization: Assessing Challenges to a Popular Public Sector Reform”, Public Administration and Development, Vol. 35/1, pp. 97-112.

[4] Smoke, P. (2001), Fiscal Decentralization in Developing Countries: A Review of Current Concepts and Practice, United Nations Research Institute for Social Development, Geneva.

[9] Smoke, P., E. Gomez and G. Peterson (eds.) (2006), Decentralization in Asia and Latin America: Towards a Comparative Interdisciplinary Perspective, Edward Elgar, Cheltenham.

[33] Sow, M. and I. Razafimahefa (2017), “Fiscal Decentralization and Fiscal Policy Performance”, IMF Working Paper, No. 17/64, International Monetary Fund, Washington, DC.

[87] Steffensen, J. (2010), Performance Based Grant Systems: Concept and International Experience, United Nations Capital Development Fund, New York.

[14] UCLG (2016), Co-Creating the Urban Future: Fourth Global Report on Decentralization and Local Governance, United Cities and Local Governments, Barcelona.

[13] UCLG (2013), Access to Basic Services. Third Global Report on Decentralization and Local Governance, United Cities and Local Governments, Barcelona.

[12] UCLG (2010), Financing Local Government: The Challenges of the 21st Century. Second Global Report on Decentralization and Local Governance, United Cities and Local Governments, Barcelona.

[11] UCLG (2007), Decentralization and Local Democracy in the World. First Global Report on Decentralization and Local Governance, United Cities and Local Governments, Barcelona.

[76] UNCDF (2005), Delivering the Goods: Building Local Government Capacity to Achieve the Millennium Development Goals, UN Capital Development Fund, New York.

[26] UNDESA and UNCDF (2017), Financing Sustainable Urban Development in the Least Developed Countries, UN Department for Economic and Social Affairs Financing for Development Office and UN Capital Development Fund, New York.

[32] UN-Habitat (2015), The Challenge of Local Government Financing in Developing Countries, UN-Habitat, Nairobi.

[94] USAID (2013), Evidence Review Paper: Responsiveness and Accountability in Local Governance and Service Delivery, United States Agency for International Development, Washington, DC.

[39] Weingast, B. (2014), “Second Generation Fiscal Federalism: Political Aspects of Decentralization and Economic Development”, World Development, Vol. 53/1, pp. 14-25.

[96] World Bank (2015), East Asia’s Changing Urban Landscape: Measuring a Decade of Spatial Growth, The World Bank, Washington, DC,

[6] Wunsch, J. and D. Olowu (eds.) (2003), Local Governance in Africa: The Challenge of Decentralization, Lynne Rienner Publishers, Boulder.

[56] Yilmaz, S., Y. Beris and R. Serrano-Berthet (2010), “Linking Local Government Discretion and Accountability in Decentralization”, Development Policy Review, Vol. 28/3.

[95] Yoshino, N. and P. Morgan (eds.) (2017), Central and Local Government Relations in Asia: Achieving Fiscal Sustainability, Edward Elgar, Cheltenham.


← 1. Reviews of the extensive literature on decentralisation and intergovernmental relations from various perspectives include, for example: Bird and Vaillancourt (1998[2]); Litvack, Ahmad and Bird (1998[3]); Smoke (2001[4]); Ahmad and Tanzi (2002[5]); Wunsch and Olowu (2003[6]); Ahmad et al. (2005[7]); Bardhan and Mookherjee (2006[8]); Smoke, Gomez and Peterson (2006[9]); Cheema and Rondinelli (2007[10]); United Cities and Local Governments (2007[11]), (2010[12]), (2013[13]), (2016[14]); Crawford and Hartmann (2008[15]), Ichimura and Bahl (2009[16]); Connerley, Eaton and Smoke (2010[17]); Martinez-Vazquez (2011[18]); Martinez-Vazquez and Vaillancourt (2011[19]); Bahl, Linn and Wetzel (2013[20]); Faguet (2014[98]); Dickovick and Wunsch (2014[21]); Rao, Scott and Alam (2014[22]); Ahmad and Brosio (2015[23]); Faguet and Poschl (2015[24]); Bahl and Bird (2018[25]); UNDESA and UNCDF (2017[26]); Yoshino and Morgan (2017[95]); Boadway and Eyraud (2018[27]); Kim and Dougherty (2019[28]); and Rodden and Wibbels (2019[29]). More specific references to the substance of some of these diverse studies will be provided later.

← 2. Examples of synthetic frameworks (from varying perspectives) include: Bardhan and Mookherjee (2006[8]); Cheema and Rondinelli (2007[10]); Boex and Yilmaz (2010[30]); Connerley, Eaton and Smoke (2010[17]); Martinez-Vazquez and Vaillancourt (2011[19]); Bahl, Linn and Wetzel (2013[20]); Manor (2013[31]); Faguet (2014[98]); Ahmad and Brosio (2015[23]); UN-Habitat (2015[32]); Bahl and Bird (2018[25]); Sow and Razafemahefa (2017[33]); Boadway and Eyraud (2018[27]); and Rodden and Wibbels (2019[29]).

← 3. Fiscal federalism was originally framed by Oates (1972[34]), a cornerstone of the fiscal decentralisation literature.

← 4. See, for example, Martinez-Vazquez and Vaillancourt (2011[19]); Bahl, Linn and Wetzel (2013[20]); Dafflon and Madies (2013[35]); Brosio (2014[36]); Faguet and Pöschl (2015[24]); Smoke (2015[37]); Bahl and Bird (2018[25]); and Rodden and Wibbels (2019[29]).

← 5. See, for example, Oates (2005[38]) and Weingast (2014[39]) on second-generation fiscal federalism.

← 6. See, for example, Cheema and Rondinelli (2007[10]); Connerly, Eaton and Smoke (2010[17]); Manor (2013[31]); Öjendahl and Dellnäs (2013[46]); Smoke (2013[40]) and (2015[37]).

← 7. There are a few comparative overviews of diversity, such as OECD and UCLG (2016[41]) and (2019[42]), or attempts to compare selected countries on certain aspects, e.g. Chatry and Vincent (2019[43]); Smoke (2013[40]) and (2019[1]).

← 8. Examples of broader work on the political economy of decentralisation include Bardhan and Mookherjee (2006[8]); Smoke, Gomez and Peterson (2006[9]); Connerley, Eaton and Smoke (2010[17]); Eaton, Kaiser and Smoke (2011[44]); Altunbas and Thornton (2012[45]); Dafflon and Madies (2013[35]); Öjendahl and Dellnäs (2013[46]); Romeo (2013[47]); Faguet (2014[98]); Romeo and Smoke (2016[48]); and Ponce-Rodríguez et al. (2018[93]).

← 9. Some instances of recentralisation are covered in Dickovick (2011[97]); Smoke (2013[49]); Malesky, Nguyen and Tran (2014[50]).

← 10. Various aspects of the bureaucratic dynamics surrounding decentralisation are elaborated in Litvack, Ahmad, and Bird (1998[3]); Connerley, Eaton and Smoke (2010[17]); Eaton, Kaiser and Smoke (2011[44]); and Smoke (2015[37]).

← 11. See Smoke (2019[1]) for a summary of relevant literature.

← 12. Eaton, Kaiser and Smoke (2011[44]) discuss this problem.

← 13. Reviews of donor behaviour related to decentralisation are summarised in Romeo (2003[51]), Fritzen (2007[52]); Development Partner Working Group on Decentralization and Local Governance (2011[53]); Eaton, Kaiser and Smoke (2011[44]); and Dickovick (2014[54]).

← 14. See Eaton, Kaiser and Smoke (2011[44]); and Development Partner Working Group on Decentralization and Local Governance (2011[53]) for examples.

← 15. See, for example, Scott (2009[55]); Connerley, Eaton and Smoke (2010[17]); Yilmaz, Beris and Serrano-Berthet (2010[56]); Eaton, Kaiser and Smoke (2011[44]); Agrawal and Ribot (2012[57]); Öjendal and Dellnäs (2013[46]); and Rodden and Wibbels (2019[29]).

← 16. Politics are more explicit in second-generation fiscal federalism; see, for example, Oates (2005[38]) and Weingast (2014[39]).

← 17. See, for example, Bland (2010[58]); Ponce-Rodríguez et al. (2018[93]); Baldwin and Raffler (2019[59]); and Grossman (2019[60]).

← 18. Reviews of selected experiences with participatory mechanisms are found for example in: Shah (2007[61]); Brinkerhoff and Azfar (2010[62]); Goldfrank (2012[63]); Blair (2013[64]); Lund and Saito-Jensen (2013[65]), Baiocchi and Gannuza (2014[66]); Baiocchi (2015[67]); and Cabbanes and Lipietz (2018[92]).

← 19. Reviews of decentralised accountability and governance are found in Bardhan and Mookerjee (2006[8]); Cheema and Rondinelli (2007[10]); Boex and Yilmaz (2010[30]); Connerley, Eaton and Smoke (2010[17]); Yilmaz, Beris and Serrano-Berthet (2010[56]); Agrawal and Ribot (2012[57]); Grindle (2013[68]); Öjendal and Dellnäs (2013[46]); USAID (2013[94]); Faguet (2014[98]); Rodden and Wibbels (2019[29]).

← 20. How to approach the implementation of decentralisation is considered by Shah and Thompson (2004[69]); Falleti (2005[70]); Ebel and Weist (2006[71]); Martinez-Vazquez and Vaillancourt (2011[19]); Bahl and Martinez-Vazquez (2013[72]); Dafflon and Madies (2013[35]); Smoke (2014[73]) and (2015[74]); Sow and Razafimahefa (2017[33]); and Boadway and Eyraud (2018[27]).

← 21. See, for example, World Bank (2015[96]) and Smoke (2019[1]).

← 22. See, for example, some relevant discussions in Green (2005[75]); UNCDF (2005[76]); Antwi and Analoui (2008[77]); Hope (2009[78]); Brinkerhoff and Morgan (2010[79]); and McGill (2010[80]). Broader treatments of capacity and how to think about it are provided in Andrews, Pritchett and Woolcock (2017[81]); and Khemani (2019[82]).

← 23. This section is partly based on Smoke (2015[74]).

← 24. See, for example: Andrews, Pritchett and Woolcock (2017[81]); Booth and Unsworth (2014[86]); Levy, Fritz and Ort (2014[83]); and Rocha Menocal (2014[84]). See also

← 25. See, for example, Shah (2010[85]); Steffensen (2010[87]); Lewis and Smoke (2012[88]); Shah (2013[89]); Mukherjee (2014[90]); and Fan et al. (2018[91]).

Metadata, Legal and Rights

This document, as well as any data and map included herein, are without prejudice to the status of or sovereignty over any territory, to the delimitation of international frontiers and boundaries and to the name of any territory, city or area. Extracts from publications may be subject to additional disclaimers, which are set out in the complete version of the publication, available at the link provided.

© OECD 2020

The use of this work, whether digital or print, is governed by the Terms and Conditions to be found at