The fiscal decentralization process varies by country. Unitary and federal governments differ in the way that central governments interact with local governments. In unitary governments, some degree of decentralization is possible, depending on the framework for interaction among central and subordinate levels of government (e.g., central, provincial, and district levels). In comparison, federal governments, by definition, are composed of subnational governments with greater autonomy granted in legislation. Decentralized decision-making allows for local participation in development in line with national objectives.
Public services are allocated to local or regional governments based on the combination of different criteria. These may include population size, fiscal capacity, development needs, and idiosyncratic socio-economic features. Specific functions are usually shared between higher levels of government, which exercise a regulatory or policy role, and lower levels of government responsible for service delivery. For example, the technical specifications for bridge construction might come from a higher level of government, with the local government being in charge of construction and maintenance at the local level.
The allocation of resources to local governments must be consistent with their expenditure responsibility, as this will result in better accountability for service delivery. Additional institutional arrangements are needed for intergovernmental coordination, planning, budgeting, financial reporting, and implementation. For such arrangements to work, transparency and accountability to local constituencies should be encouraged by adequate reporting and oversight of local fiscal performance. By contrast, detailed central control may render decentralization ineffective. Clear rules for the transfer and use of fiscal resources combined with appropriate coordination between different levels of government could go a long way towards ensuring efficient use of resources thanks to more effective service delivery.
Pros and cons of fiscal decentralization
Two questions are useful for understanding the pros and cons of fiscal decentralization: (1) Why are some countries more fiscally decentralized than others? and (2) What are the potential effects of decentraliza-tion on the quality of the delivery of public goods? If decentralization were the result of exogenous factors such as median voter income, the degree of demo-graphic heterogeneity, or the type of political institutions involved, there would be little room for the national government to promote decentralization from above. If instead decentralization is seen as the best way to ensure optimal public service delivery, it should be treated as an integral part of fiscal policy.
Arguments in favor of fiscal decentralization
One of the powerful normative arguments in favor of decentralization is the proper representation of local voters. Limited knowledge of individual preferences by central government officers has often translated into sub-optimal decisions on the best allocation of public resources. This line of thought is an extension of Hayek’s lucid defense of the market economy. In this view, local bureaucrats and political leaders are likely to know more about people’s preferences than are central government officers. A variation of this argument is that, while centralization is associated with the homogenous provision of public services for all citizens across the nation, this may collide with heterogeneous local preferences across jurisdictions. Thus, devolution of fiscal and political powers to lower tiers of government would be a way to promote an efficient allocation of resources through which local demands are fully understood and properly attended to at the local (subnational) level of government.
A second strand of normative arguments sees local political jurisdictions as operating in the same way as an efficient and competitive private market. Local governments come to exist in the same way as private firms enter a particular industry, in response to demands from a particular community. Likewise, they can eventually be removed from the “local public goods industry” if the administrative cost of operating a particular jurisdiction becomes too high with respect to residents’ tax-revenue-generation capacity. Local jurisdictions may become “spatialized” by providing specific types of local services in the same way as different firms produce differ-ent varieties in line with differences in consumer desires and in the intended use of market products. Autonomous jurisdictions would compete in the provision of local public services tailored for differ-ent communities based on their own preferences. Conceptually, voters would “reveal their preference” for local public goods through the political process, by supporting (penalizing) good (poor) performance of local governments, holding them accountable for the quality of local service delivery. This “accountability advantage” would be diluted at the national level, in which political responsibilities are widespread, and voters’ capacity to oversee government performance is weaker.
An alternative view depicts the government as a “rent seeking” economic agent trying to take advantage of its power to extract Ricardian rents from taxpayers. Following this line of reasoning, decentralization may be seen as an institutional device to protect individuals from governments,6 in the same way as central banks safeguard monetary policy from undue influence from government. National governments are regarded as being no different from private monopolies, to the extent that they are the sole entities collecting taxes and/or delivering public goods. As a result, they would tend to charge higher prices (taxes) and deliver lower-quality public goods relative to a competitive situation. As voters become more aware of the dangers of the “Leviathan,” this theory predicts that the median voter will automatically bias his political preferences towards more decentralized institutional arrangements. This partly explains why high-income countries tend to be more decentralized. In line with this hypothesis, an extensive empirical literature has attempted to provide evidence that fiscal decentralization leads to a lower tax burden, with inconclusive results.
Arguments against fiscal decentralization
Alternative views challenge three major assumptions of theories that treat the dynamics of the “local public goods industry” as equivalent to those of private competitive markets. These alter-native views challenge the assumptions that (1) local governments are genuine representatives of their constituencies and try to compete to please residents through the best possible combination of prices (taxes) and local public goods quality; 2) the “local public goods industry” is broadly competitive, with no externalities among jurisdictions, and with all local costs and benefits fully internalized by local residents, and 3) local voters not only vote at the ballot box, but also do so with “their feet” by choosing the jurisdiction where they want to live, taking into account the type of local government they want to have.
At the core of these challenges is the fact that the revelation of citizens’ preferences may be rather imperfect, seeing as how many national and local governments do not face the necessary incentives to compete and/or maximize the welfare of their constituencies. This is because a well-functioning democracy is not always in place at both the national and the local levels, especially in develop-ing countries. Moreover, decentralization in imperfect democracies could result in excessive closeness between local authorities and big local private interests, which may lead to corruption and elite capture. Finally, there is some empirical evidence pointing to stronger professional qualifications and better availability of information for central government officers relative to local authorities.
Additionally, the assumption of high mobility across jurisdictions is not confirmed by empirical evidence. First, family and labor ties may prevent mobility from occurring in a Tiebout fashion. In other words, for various reasons, citizens may be unable to change locations in order to maximize their individual utility. Political scientist Daniel Treisman argues that poorly performing local governments are unlikely to be disciplined by residents’ mobility,
as the quality of local services is mainly capitalized into property prices, making “voting with their feet” a costly affair for homeowners and tenants. Moreover, subnational governments may not have the necessary incentives to target the local poor, for which mobility is more costly. In addition, excessive mobility across jurisdictions could also be a problem, as it may translate into additional fiscal pressures on local governments showing above-average performance because of an increasing number of potential beneficiaries coming from other jurisdictions.
The potential for fiscal imbalances resulting from fiscal decentralization deserves separate mention. A moral hazard problem may arise if local governments expect a “guarantee” from the national government at the time they decide on tax bases and tax rates, leading to over-borrowing. A moral hazard problem arises because of the possibility of a bailout of local governments by the central government. This is more likely in countries with weak and poorly designed institutions, often found in developing countries.
The case for fiscal decentralization
The case for decentralization is necessarily empirical. Recognizing that government functions encompass a wide range of specific tasks, Letelier and Saez postulate that two effects should be distinguished when it comes to the optimal degree of decentralization. On the one hand, if power is delegated to jurisdictions that are too small, a loss in economies of scale could result (“scale effect”), raising the cost of public goods and rendering decentralization a cost-inefficient solution. On the other hand, the “Von Hayek effect” over time would lead to better knowledge by local governments about people’s needs as decentralization takes root and allows collective demands to arise from below. Thus, the optimum degree of decentralization would depend on the balance between both effects.