XII. Transfers to Local Governments

Ke-young Chu, and Richard Hemming
Published Date:
September 1991
  • ShareShare
Show Summary Details

What are the advantages and disadvantages of decentralizing government expenditure?

What criteria are commonly used in assessing the efficiency of decentralization?

Should revenue collection also be decentralized?

What is the intended purpose of revenue sharing (block) grants, matching grants, and special purpose grants?

In many countries, grants from the central government to local governments are the primary means through which the central government provides local services. A typical allocation of responsibilities is for the central government to focus on primarily national concerns, such as defense and social insurance; regional governments to handle transportation and higher education; and the local governments to have responsibility for primary and secondary education, police and fire protection, local parks, and sanitation services. Although this degree of decentralization is more often observed in industrial countries, some developing countries have significant levels of decentralization. In India, for example, local governments account for over two fifths of consolidated government expenditures. Intergovernmental transfers account for nearly one fifth of central government spending in India and constitute 37 percent of total local government resources. There is also a significant degree of decentralization in other developing countries—see Table 1.

Table 1.Intergovernmental Grants in Developing Countries, 1986
Local and Regional Govt. ExpenditureCentral Govt. ExpenditureOf which: Intergovernmental GrantsCentral Govt. Transfers to Local Govts.Grant Receipts by Local and Regional Govts.
(In percent of GDP)(In percent of expenditure)(In percent of receipts)
Source: IMF Government Finance Statistics Yearbook.
Source: IMF Government Finance Statistics Yearbook.

In any country with decentralized functions, grants are the principal means through which the central government can influence local government decisions. However, the existence of intergovernmental grants raises questions regarding the costs and benefits of decentralization of government responsibilities. Why should the central government resort to indirect controls of local government decisions (i.e., grants), instead of assuming direct responsibility for local services or simply issuing guidelines to the local governments? Furthermore, if the central government needs to control local decisions in this manner, why decentralize?

There are some well-established advantages of decentralization of expenditure, both in terms of the decisions regarding the mix of local services to provide and the actual production of the services. However, in most cases, revenue is more efficiently collected in a centralized fashion. Therefore, a tension exists between efficient revenue collection and efficient service delivery, and grants serve the purpose of bridging the gap. Grants allow a country to rely on normally more efficient centralized tax collection agencies while still availing itself of the advantages of decentralized expenditure decisions. Additionally, grants afford the centralized authority the opportunity to influence local expenditure decisions to ensure that, when appropriate, such decisions take account of national interests.

Decentralization of Government Expenditure

Although there is little controversy as to the need for government within a market-oriented economy, there is strong disagreement over the proper scope of government. Considerable difficulty is involved in choosing both the right level of government expenditure (and taxation) and the mix of government services that maximizes social welfare, as determined by the needs and preferences of society. One means of improving the decision-making process is through the decentralization of certain government functions. Government functions that lend themselves to decentralization are referred to as local public goods (LPGs). The efficiency advantages of decentralization of LPGs are varied and quite significant. The first advantage of decentralization is the possibility of having regional variety in the mix and level of LPGs, which can greatly enhance social welfare. For instance, in a high-crime area where most of the housing is constructed of stone, the local population will desire high police expenditures and minimal spending on fire prevention. Alternatively, in one locality the road network might be in particular need of repair while in another region an upgrade of the water and sanitary services might be preferable. Once it is accepted that regional variety is desirable, the second advantage of decentralization is in the area of preference revelation. Regionally based governments are in a better position to determine the preferences of the local population. Research has indicated that preference revelation problems encountered in public finance are greatly diminished as the size and heterogeneity of the population decreases.

The third advantage of decentralization is that competition, proximity, and transparency provide a strong motivation for local governments to be more responsive to the desires of the public. In a democracy, there is competition among political parties at the national level which engenders responsiveness to the public. However, at the regional level, competition for residents and businesses provides an additional incentive for the local authorities to be responsive to the public. Furthermore, the smaller size of the operation makes it much easier for the public to monitor local governments and to influence local policies. The fourth advantage stems from scale advantages—smaller bureaucracies are easier to administer efficiently. In general, there is an absence of economies of scale in centralizing the supply of LPGs.

The advantages to decentralization are offset somewhat by three factors. The potential gains from decentralization are likely to be lower in a geographically small country—there is normally less regional diversification of needs. Second, decentralization can increase the scope for corruption within government since the number of officials in a position to benefit from corruption increases. Furthermore, large, centralized bureaucracies are easier to monitor than a series of decentralized offices. Finally, decentralization tends to increase the demand for skilled administrators by the government sector. Since each autonomous regional and local authority requires its own administrative staff, in a country where capable public management is scarce, a greater degree of centralization might enable the conservation of scarce management resources.

National Public Goods and Inter-Regional Externalities

While it is often preferable to decentralize the functions of government that are local in nature, there are strong reasons to advise against decentralizing all government activities. There are no advantages to local supply of national public goods, that is, government services which provide nonrival benefits to all regions. In fact, technically speaking, the same kind of free-rider problems that exist with the private provision of public goods will emerge. Local control of national public goods will almost certainly lead to inadequate provision, given the needs and desires of society. For instance, a local government is not likely to be willing to fund expenditures on a highway used substantially by nonresidents to travel through its locality. A similar, nearly identical, issue arises when essentially local public goods have strong interregional externalities which require national coordination. There are a number of possible solutions, depending on the severity of the externality. When the externality is quite substantial compared to the local component of the program, centralized provision is often a sensible solution. In other cases, the externality may be insignificant enough to warrant no action on the part of the central government. However, in many cases, the advantages of decentralization need not be forgone, but the externality is significant enough to warrant centralized action. In these circumstances, the use of grants (see below)—and occasional sanctions when negative externalities exist—is the prescribed solution.

Centralized Tax Collection and Deficit Financing

As is the case with expenditure activities, there are revenue devices that are national in nature and others that are local. However, in general, it is more efficient to centralize collection of income taxes, trade taxes, and taxes on goods and services. The reasons for the greater efficiency with centralized revenue collection relate to both administration and economic efficiency.

Centralized tax collection is often less costly to administer for informational reasons. A centralized system requires only one file on each economic entity, while a decentralized system could lead to multiple agencies collecting essentially the same information. It is also more difficult to assess taxable income with decentralized tax collection. Many residents and companies normally acquire income in multiple regions outside the jurisdiction of the tax authority. A locality is always at a disadvantage in collecting data on events that occur outside its jurisdiction.

The economic efficiency reasons for centralized tax collection revolve around the issue of locational distortions. A high tax in one region can lead to relocation of economic activity to another region. Tax-induced relocation can either lead to a lower output—for instance, when a firm locates farther away from a port than is economical because of high taxes in the port district, or cause a direct decrease in economic well-being—such as when a family lives in a less desirable location to avoid high taxes elsewhere. High local taxes have been found to have a severe impact on the economic health of a region. Distortions of this sort can be avoided with a national tax.

At least one local tax does not have these problems—land-value property taxation. A property tax that is based purely on the value of land is an efficient local tax for two reasons. Such a tax has to be administered at the local level since property values vary from district to district and region to region. Furthermore, because land is immobile, the tax causes no relocation or economic inefficiency. In a competitive market setting, the tax is capitalized into the value of the property (the sale price of the property falls), and thus the tax does not affect relative prices. In contrast, a traditional property tax on both the value of land and the value of buildings does distort economic decisions by discouraging building. Local user fėes are another nondistorting potential source of local revenue. When local governments provide goods that are partially private in nature, there is often a valid reason to impose a fee on the users—see the note on Pricing and Cost Recovery for further discussion.

With respect to deficit financing, this too is generally the prerogative of the centralized authority. In many countries, localities are prohibited from incurring a budget deficit. In other countries, bond financing is restricted to the funding of capital projects. For several reasons, the scope for localities and regional governments to use deficit financing is more limited than for the central government. First, a locality’s ability to repay a loan is constrained by the limitations on tax instruments. Second, unlike most central governments, localities cannot print money. Finally, localities operate in a more open economic environment. Thus, a large overhanging debt can be a serious deterrent to the economic vitality of a region because businesses and households will avoid communities where the prospect of raising taxes in the future is high. Therefore, local borrowing should be closely linked to the ability to repay and to the timing of benefits (such as for capital projects). For these reasons, market forces normally limit the ability of local and regional governments to use deficit financing. If market forces do not impose constraints on the ability of localities to borrow, the central government often steps in to limit budget deficits.

Intergovernmental Grants

Block grants

A block grant or a general purpose grant is the principal means of increasing the level of funds available to local governments. Block grants are usually disbursed through a formula based on the goals of the central government. For instance, the grants are normally inversely proportional to per capita community income and positively correlated with community population. Therefore, they are a means of providing both general revenues to localities to fund LPGs and of redistributing funds in favor of more needy localities, which enables the locality to increase its level of services or to lower local taxes.

Matching grants

A matching grant is normally related to a specific functional category of expenditure, such as primary education or health clinics. The size of the grant is variable and depends upon the level of local expenditure on the given program. The purpose of these grants is to encourage localities to spend more on a targeted service both by providing funds and by providing the locality with an incentive to use more of its own resources for this purpose.

Special purpose grants

Special purpose grants are restricted-use funds offered to local governments. The locality must use the funds for the specifically designated purpose prescribed by the central government in order to qualify for the grant. Economists do not generally favor special purpose grants because they are either ineffective or inefficient. When a special purpose grant pays for a program that the locality intended to provide anyway, they are simply equivalent to block grants. If the specific purpose grant is for a program the locality intended to fund, but the amount exceeds the locality’s intended budgetary allocation, the locality will often use all the available funds for this purpose. In this case, the special purpose grant is equivalent in part to a block grant and has the effect of forcing an increase in local expenditures on the program in question. However, unlike a matching grant, the locality will normally spend less of its own funds on the program. Finally, when a specific purpose grant is provided for a program that the locality did not intend to fund and does not consider to be very desirable, the locality will sometimes simply not apply for the funds (this is common in the United States). In conclusion, economic theory suggests that both block grants and matching grants, are preferable to special purpose grants.

The impact of grants

The impact of grants on local expenditure patterns has been found to be extremely large in a wide variety of settings and over a long period of time. Block grants have been found to be used by localities to increase the level of local government services; they do not seem to induce localities to lower local taxes. Likewise, matching grants have proven to be extremely effective in increasing local expenditures. While most of the empirical studies of the impact of grants used data from Europe, the United States, and Canada, the same effect has been observed elsewhere. An example, a matching grant program to help fund local infrastructure in Ghana yielded a local response far in excess of expectations. Numerous theoretical explanations have been offered for the unexpected potency of grants, however, there is as yet no consensus as to the explanation of this so-called flypaper effect.

Country Illustration

Fiscal federalism in Brazil

Brazil has 24 state governments and nearly 4,000 municipalities, in addition to the central government. The distribution of revenue and expenditure among the various levels of government is shown in Table 2. State and local governments collect a little under one third of total tax revenue, most of which is accounted for by a state-level value-added tax and local property taxes. Revenue collections are, however, insufficient to cover local expenditure, which necessitates intergovernmental grants. Some of these grants reflect revenue-sharing arrangements: for example, both state and local governments receive a mandated (and steadily rising) share of central government income tax and excise duty revenue while local governments receive a fixed share of VAT revenue from the states. Block grants are also fairly common; some use is made of special purpose grants; but matching grants are unusual.

On the expenditure side, the central government is primarily responsible for debt service, subsidies, and net lending. Both state and local governments are an integral part of public administration, and account for the bulk of expenditure on wages and salaries. They also undertake major capital projects, partly financed through loans from the central government. As regards the functional breakdown of expenditure, the central government is responsible for military spending, national transportation and communications, support for industry and commerce, and social security. State governments provide police services, local transportation, energy and education and health services. Localities are primarily responsible for housing. This pattern of expenditure is similar to that found in other countries with decentralized government. The indications are that the decentralization of both revenue collection and expenditure is likely to be increased, after a period of centralization since the mid-1970s.

Table 2.Brazil: Government Revenue and Expenditure, 1986(In percent of GDP)
Level of Government
Total revenue30.422.49.13.4
Tax revenue23.916.86.60.5
Income taxes12.011.80.3
VAT and excises10.
Nontax revenue6.
Intergovernmental grants4.
Expenditure and net lending44.835.510.13.6
Current expenditure33.
Of which: Wages and salaries6.
Interest payments12.612.00.60.1
Intergovernmental grants4.83.41.4
Capital expenditure4.
Net lending8.
Source: IMF Government Finance Statistics Yearbook.
Source: IMF Government Finance Statistics Yearbook.

    Mieszkowski Peter M. and William H. Oakland Fiscal Federalism and Grants-in-Aid (Washington: The Urban Institute1979).

    Mueller Dennis C. Public Choice II (New York: Cambridge University Press1989).

    Tiebout Charles M. “The Pure Theory of Local Expenditures,” Journal of Political Economy 641956.

    Winer Stanley L. “Some Evidence on the Effects of the Separation of Spending and Tax Decisions,” Journal of Political Economy 911983.

    • Search Google Scholar
    • Export Citation

    Other Resources Citing This Publication