Week 5 - Assignment: Analyze a Case Study on Supervising Employees and Week 8 - Assignment: Create a Brochure on Organizational Planning

Week8r3.pdf

Home >Business & Finance homework help >Management homework help >Week 5 - Assignment: Analyze a Case Study on Supervising Employees and Week 8 - Assignment: Create a Brochure on Organizational Planning

Intergovernmental Fiscal Relations: Coordination Failures and Fiscal Outcomes

LUIZ R. DE MELLO, JR

This article estimates the fiscal impact of coordination failures in intergovernmental fiscal relations. The coordination failures considered here are due to agency problems arising from the delegation of fiscal powers to sub-national governments, and "com- mon pool" problems associated with funding decentralised government spending through intergovernmental transfers. Particular attention is focused on the trade-off between coordination and fiscal decentralisation. Evidence provided for a sample of thirty countries suggests that coordination failures are likely to result in a deficit bias in decentralized policy making, particularly in the case of developing countries, for which the benefits of decentralization may be over-stressed. Developed countries were found to be less adversely affected by coordination failures and have therefore managed to pursue fiscal consolidation in a decentralized setup.

INTRODUCTION

The problem of coordinating intergovernmental fiscal relations in decentralised gov- ernments has puzzled theoreticians and practitioners in recent years.' This is because, in general terms, policy outcomes may suffer from coordination failures in intergov- ernmental fiscal relations. These failures induce sub-national governments to spend inefficiently and beyond their means vi'hen fiscal policy is designed and implemented in a decentralised fashion. Policy failures tend to manifest themselves in terms of a deficit bias in decentralised policy making and higher costs of borrowing, given the risk premium associated with a higher probability of default.^ The most important sources of coordination failures are moral hazards arising from the delegation of fiscal

Luiz R. de Mello, Jr., University of Kent, UK. The author is now at the Fiscal Affairs Department, International Monetary Fund, 700 19th St. NW Washington, D.C. 20431. The author is indebted to an anonymous referee and the participants of ECLAC's 1998 Regional Seminar on Fiscal Policy for helpful comments. The usual disclaimer nevertheless applies.

de Mello / Intergovernmental Fiscal Relations: Coordination Failures and Fiscal Outcomes

powers to sub-national governments, and "common pool'" problems associated with funding decentralised government spending through transfers from common pools of sharable tax revenues.

In a nutshell, moral hazards in decentralised fiscal policy making, or agency prob- lems in general, are due to the asymmetry of information on the costs and benefits of government spending between the center and the sub-national governments to which fiscal powers are delegated. When important tax bases are devolved to sub-national governments, the central government may be unable to monitor how efficiently sub-national governments utilize their tax bases. The essence of "common pool" prob- lems is that the tax bases that are efficient and simple to administer by local govern- ments are few and poor.^ Due to the possibility of tax exportation, externalities in the provision of public goods and services, factor mobility and economies of scale, broad tax bases are best managed by the central government. It is therefore necessary for the center to share part of its tax revenues with sub-national governments to bridge the gap between the revenues mobilised in sub-national jurisdictions and decentralized spend- ing, hence correcting vertical fiscal imbalances. However, in this case, sub-national governments face the incentive to under-utilize their local tax bases to minimise the costs of decentralised provision borne by local taxpayers, and over-utilize national sharable tax bases.'*

Critics to fiscal decentralisation also argue that giving new budgetary rights and responsibilities to sub-national governments may be inconsistent with the task of si- multaneously promoting institutional clarity and transparency in budgeting and overall fiscal policy making, so that spending matches revenues at the sub-national level.^ Fiscal decentralisation may therefore aggravate fiscal imbalances and consequently endanger overall macroeconomic stability, unless sub-national governments are com- mitted to fiscal discipline, and the decentralisation package includes incentives for prudence in debt and expenditure management.^ The imposition of stringent con- straints on sub-national indebtedness and effective monitoring of sub-national fiscal positions are additional important prerequisites for successful fiscal decentralisation, in addition to the availability of expertise at the sub-national level and the ability of sub-national governments to manage efficiently an increased volume of resources.

Despite these shortcomings, advocates of fiscal decentralisation stress the potential benefits of devolving fiscal responsibilities to sub-national levels of government on the grounds of increased efficiency in service delivery and reduced information and trans- action costs associated with the provision of public goods and services.' Streamlining public sector activities and encouraging the development of local democratic traditions are also regarded as important objectives of fiscal decentralization programs. Also, the literature on fiscal federalism sustains that the main advantage of decentralized, rather than centralized, provision is that, under subsidiarity, local preferences and needs are likely to be best met by local, rather than national, governments.* Accountability and transparency in expenditure management can also be enhanced by bringing spending assignments closer to revenue sources, and hence to the median voter.

Public Budgeting & Finance / Spring 1999

In short, with regard to the three traditional Musgravian functions of government, the critics of fiscal decentralisation stress the difficulties related to macroeconomic stability and redistribution, while advocates of decentralisation stress its benefits in terms of allocative efficiency.^ Against this background, the objective of this article is to shed more light on the relationship between fiscal decentralisation and policy out- comes by providing empirical evidence for a sample of thirty countries for which data on government spending and fiscal balances are disaggregated between central and sub-national governments in the IMF's Government Financial Statistics. Motivation for the study of coordination in intergovernmental relations, as a prerequisite for sound macroeconomic governance and fiscal consolidation, is found in Roubini and Sachs (1989), Roubini (1991), and Fukasaku and de Mello (1998). Further motivation for this paper can be found in the recent developments in the consolidation of the Euro- pean Union. In Europe, deeper integration calls for higher degrees of decentralisation in fiscal policy making, given the need to match public sector provision with local preferences and needs, and commitment to fiscal conservatism as a result of monetary union.'" This paper stresses the challenges involved in the federalisation of fiscal policy making in Europe at the same time as avoiding excessive government deficits and hence preserving monetary discipline.

The remainder of the article is organized as follows. Section 2 provides further details of the sources of coordination failures examined in this article. Section 3 analyses a number of fiscal indicators in a sample of developed and developing coun- tries, with the aim to compare and contrast the extent of fiscal decentralisation in different economies, and to highlight a few stylised facts. Emphasis is placed on the relationship between fiscal decentralisation, coordination failures, and fiscal outcomes. Section 4 presents empirical evidence, and Section 5 concludes.

THEORETICAL CONSIDERATIONS: THE GENERAL ARGUMENT

The focus of this article is on fiscal decentralisation, given that coordination failures in intergovernmental fiscal relations may lead to undesirable fiscal outcomes. When fis- cal decentralization drives a wedge between expenditures and sources of finance, sub- national governments face the incentive to overspend as a result of a "common pool" resource problem. As suggested earlier, this problem arises because the social returns of government spending exceed both private returns and social costs, and hence sub-national governments favour spending in their jurisdictions financed by a common pool of tax revenues." In this case, the burden of providing public goods and services can be shared across government jurisdictions, whereas the benefits of public sector spending can be internalised and hence generate a political payoff to sub-national governments. Also, overspending can be attributed to "common pool" funding because of competition among sub-national governments to secure a larger portion of sheirable funds in the form of grants and transfers from the central government.

An additional type of "common pool" problem which has an immediate adverse

de Mello / Intergovernmental Fiscal Relations: Coordination Failures and Fiscal Outcomes

impact on macroeconomic stability is the following. In the case of rigid revenue- sharing arrangements, in which transfers are automatic, every time a central govern- ment raises taxes to reduce its own fiscal deficit, sub-national governments receive a corresponding revenue benefit which they are free to spend. This tends to inhibit the potential for reducing the consolidated fiscal deficit by increasing the tax burden, when revenues are shared vertically across government levels. This is also the case of na- tional expenditure programs that are sponsored and fully funded by the central govern- ment, without sub-national co-funding. Incentives to delay adjustment is another con- sequence of the "common pool" problem, since individual jurisdictions have limited incentives to act alone and strong incentives to free ride, if the burden of fiscal re- trenchment can be shared across jurisdiction, i.e., borders.'^

In terms of agency problems, it is well known that delegation allows the center to provide goods and services according to local market information and, by separating responsibilities, more powerful incentives to encourage efficient provision can be put in place. Because local jurisdictions can identify community preferences more easily, decentralised policy making tends to reduce information costs.'^ Delegation brings the government closer to the people, increases accountability in service delivery, and strengthens society's scrutiny of public sector actions. An agency problem between society and the government can therefore be solved, or minimised, by decentralizing fiscal responsibilities and devoling spending assignments to local governments. Never- theless, by reducing the "informational distance" between the government and society, fiscal decentralisation tends to increase the distance between the center and the decentralised agencies or local governments to which fiscal responsibilities are de- volved. This loss of control of the center over decentralized agencies or sub-national governments entails an efficiency loss in delegation, which increases with the informa- tional distance between both government levels, and renders the acquisition and pro- cessing of information inside the government more costly.•'* In particular, the devolu- tion of tax bases to sub-national jurisdictions is likely to create a moral hazard in decentralised fiscal policy making, given that the central government may be unable to monitor how efficiently sub-national governments utilise their tax bases.

Both problems—"common pool" and agency—are therefore intertwined, given that decentralisation implies delegation of spending functions to sub-national jurisdictions and creates vertical imbalances in financing. Policy recommendations to solve one of these problems may well exacerbate the other. Against this background, in providing public goods and services, there seems to be a trade-off between coordination, which requires some degree of centralisation in policy making, and the need for information on local needs and preferences, which is better extended at the local level.'^ A strong case can be made in favor of centralized provision and policy making as far as dis- tributive and macroeconomic stabilisation policies are concemed. In this case, effi- ciency gains, which are the main advantages of decentralisation, may be dwarfed by the need to ensure good macroeconomic governance and fiscal discipline in a decentralised government structure. Also, both types of policy tend to be nationwide in

Public Budgeting & Finance / Spring 1999

scope, rather than regional, particularly in developing countries, thus rendering the potential gains of decentralisation in terms of allocative efficiency less promising vis- k-vis the risks involved in macroeconomic governance.'^

EMPIRICAL EVIDENCE

The Testable Hypotheses

Turning to empirical evidence, preliminary information on fiscal indicators is provided in Tables 1 and 2 for the sample of thirty countries under examination in this article. For the sake of international comparisons, a few stylised facts can be highlighted. First, governments tend to be smaller in Latin America, and particularly Asia, than in OECD countries. It is widely accepted that the demand for public goods and services increases with income, such that government spending tends to be larger in richer countries than in their poorer counterparts, ceteris paribus. Central government spend- ing ratios range from 20 percent of GDP in Asia, to 40 percent in the European countries of the OECD sample. The sub-national share of total government spending is below 5 percent in Asia, and ranges from 10 to 40 percent in Latin America, and from 12 to 60 percent in the OECD sample. In relative terms, sub-national governments tend to be large in countries where the central government is small in both OECD and Latin American countries. This is nevertheless not true in Asia (with the exception of India), where countries with small central governments also tend to have small sub-national governments. In the OECD sample and in Latin America, unlike Asia, a reduction in central government spending is achieved chiefiy by delegating public sector functions and spending responsibilities to sub-national governments, thus increasing their share in total government spending.

Second, with regard to the sources of finance of sub-national spending, in Asia, sub-national governments rely heavily on transfers from the center. In the OECD area, there is a clear-cut distinction between the federations, where emphasis is placed on local tax revenue mobilization, and the European countries (as well as Australia), where intergovernmental transfers prevail as the main source of finance of sub-na- tional spending. The picture is less clear cut in Latin America, where fiscal outcomes tend to be poor, nationally and sub-nationally. Sub-national fiscal imbalances tend to be limited in the OECD area, despite relatively high sub-national spending shares. Asia's fiscal centralism is associated with limited fiscal imbalances, nationally and sub-nationally.

Against this background, important empirical questions to be asked are, first, whether decentralized fiscal policy making leads to a deterioration of sub-national finances and, second, whether such deterioration worsens the fiscal position of the central government. As suggested by Figure 1, fiscal outcomes tend to be worse in bigger governments, nationally and sub-nationally. On the other hand, in Figure 2 (Panel A), an increase in the sub-national share of total government spending does not

de Mello / Intergovernmental Fiscal Relations: Coordination Failures and Fiscal Outcomes

TABLE 1 Public Finance Indicators

Country Consolidated Central Government Sub-national Governments

Government Size Overall Balance Government Size Overall Balance

Avrg. St.Dev. Avrg. S t D e v . Avrg. St.Dev. Avrg. St. Dev.

Argentina 1 (1974-85) Argentina 2 (1986-92) Bohvia Brazil 1 (1983-89) Brazil 2 (1990-93) Chile .1(1974-80) Chile 2 (1981-88) Colombia Mexico Peru India Indonesia Malaysia Philippines Thailand 1 (1972-80) Thailand 2 (1981-94) SouthAfrica 1(1977-83) South Africa 2 (1984-93) Australia Austria Belgium Canada Denmark Finland France Germany Iceland Italy Netherlands Norway 1 (1972-78) Norway 2 (1979-94) Spain 1 (1980-89) Spain 2 (1990-93) Sweden Switzerland United Kingdom United States

16.1 12.2 16.0 36.3 32.5 32.8 25.2 13.2 19.4 16.9 15.0 20.1 28.1 13.8 15.5 17.2 24.0 32.3 24.8 36.9 48.5 22.5 38.4 31.1 41.1 29.7 29.5 42.8 52.4 36.6 39.8 26.2 28.8 38.8 18.4 31.1

2.7 2.5 5.7

15.1 5.6 4.4 5.3 1.5 5.8 2.1 2.3 2.5 4.0 4.2 1.5 2.3 2.7 2.6 3.1 3.9 6.0 2.1 4.5 5.9 4.6 3.0 3.0 7.3 4.4 2.3 2.0 5.9

16.2 7.3 2.8 3.4 1.8

-5.3 -1.0 -4.4 -3.8 -5.0 -1.6

0.5 -1.2 -5.0 -3.2 -6.3 -1.7 -5.3 -1.9 -2.9 -0.7 -A.6 -5.6 -1.6 -3.8 -7.0 -3.2 -0.7 -2.6 -2.3 -1.4 -3.2

-11.2 -3.7 -4.1 -0.3 -3.3 -3.6 -4.6

0.0 -3.2 -3.2

2.5 1.3 8.1 2.4 3.7 5.8 2.2 2.1 2.3 2.2 1.7 1.6 5.0 1.4 1.7 3.8 1.3 2.7 1.8 1.7 3.2 2.1 3.2 4.7 1.9 1.2 1.3 1.9 2.3 2.6 3.0 2.3 2.5 4.4 0.5 2.6 1.4

8.1 9.0 3.4 5.7

19.0 1.0 2.6 5.2 3.3 3.5

12.4 2.8 6.2 1.6 3.6 1.5 9.9 9.9

17.1 16.4 7.1

30.1 31.7 18.2 8.3

22.1 8.0

11.1 17.9 22.7 18.8 7.9

12.6 24.6 16.2 13.4 17.5

1.5 0.6 1.5 3.5 0.5 0.2 0.6 0.9 0.7 0.9 1.8 0.4 0.9 0.2 0.6 0.4 0.6 3.7 2.3 1.6 0.9 2.7 2.7 2.4 1.1 1.4 2.2 3.4 1.3 1.4 1.1 2.8 6.9 1.8 8.7 1.2 1.0

-2.9 -2.8 -0.1 -0.7 -2.5

0.0 -0.2

0.1 -0.3

0.0 -2.7

0.0 -0.8

0.1 0.1 0.1

-0.9 -0.4 -1.0 -0.5 -0.6 -2.2 -0.1 -0.1 -0.6 -1.4 -0.1 -1.0 -0.9 -1.4 -0.9 -0.4 -1.1 -0.8 -0.6 -0.7

0.9

1.5 2.1 0.3 0.5 0.5 0.1 0.4 0.3 1.3 0.1 0.6 0.1 0.9 0.1 0.0 0.1 0.2 0.3 0.8 0.5 0.7 1.2 0.6 0.4 0.3 0.6 0.4 0.6 1.0 0.4 0.8 0.3 0.8 0.9 0.7 1.1 0.4

Note: Countries for which a sizeable discrepancy (over 5 percentage points) in government size was observable are presented in two sub-samples.

Public Budgeting & Finance / Spring 1999

TABLE 2 Inter-Governmental Fiscal Indicators

Argentina 1 Argentina 2 Bolivia Brazil 1 Brazil 2 Chile 1 Chile 2 Colombia Mexico Peru India Indonesia Malaysia Philippines Thailand 1 Thailand 2 South Africa 1 South Africa 2 Australia Austria Belgium Canada Denmark Finland France Germany Iceland Italy Netherlands Norway 1 Norway 2 Spain 1 Spain 2 Sweden Switzerland United Kingdom United States

Sub-national Fiscal Autonomy

Tax Autonomy

Avrg.

82.2 83.3 52.6 54.8 48.3 40.7 32.1 39.7 78.2

8.7 45.8 15.4

35.1 23.3 53.5 16.3 16.5 32.8 52.0 33.7 56.1 42.9 47.7 42.7 54.4 82.3 13.2 6.7

46.8 48.1 48.4 31.0 58.6 54.6 25.5 48.2

St. Dev.

6.8 12.1 13.8 4.4 3.8 8.1 2.7 5.0 5.6 2.1 1.3 3.2

4.0 2.8

13.3 1.2 3.6 2.3 4.0 2.8 4.2 3.5 2.8 4.5 2.5

12.3 7.3 2.4 0.8 4.2

15.1 1.4 1.9 1.2

15.7 1.7

Non-Tax Autonomy

Avrg.

17.6 15.9 35.1 13.0 15.9 15.9 36.7

8.6 15.9 17.7 11.6 5.9

25.2 5.1 9.5

30.9 25.8 17.5 20.7

7.7 14.1 8.1

17.1 18.5 21.5 13.4 8.5

13.4

11.9 15.5 8.5

17.3 21.4 17.3 23.0

St. Dev.

6.4 11.4 12.2 4.0 3.6 3.6 9.0 1.9 5.2 3.3 1.3 1.0

3.4 1.1 1.7 2.1 5.9 3.1 1.6 1.1 0.8 0.6 2.0 2.0 1.2 7.7 1.3 5.2

3.9 4.3 0.7

15.8 2.2 3.1 1.9

Sub-national Fiscal Dependency

Avrg.

6.8 31.7 34.4 43.1 30.1 47.8

6.0 72.5 42.5 78.7 21.7 35.1 71.5 35.6 49.4 55.7 47.4 26.3 58.6 28.6 47.3 31.0 39.3 23.1

6.7 77.3 78.5 19.6 39.1 35.1 51.2 24.6 23.3 53.1 31.1

St. Dev.

4.4 2.1 2.1 9.9 8.2 4.5 4.4 3.7 2.0 3.7 3.9 4.0 3.2

14.9

1.1 8.8 5.0 3.5 3.3 2.9 3.7 3.3 5.6 2.2 0.7 6.3 5.4 4.7 1.1

18.3 18.3

1.9 1.6 4.4 2.6

Sub-national Spending Share

Avrg. !

33.4 42.7 17.0 31.7 37.3

3.2 8.0

28.1 16.1 17.7 45.4 11.7 18.0 10.1 19.1 7.9

28.5 21.8 40.7 30.8 12.0 61.0 45.3 39.3 17.9 42.7 21.2 20.6 57.2 38.7 32.1 19.5 38.0 39.4 54.4 26.0 44.1

5t. Dev.

3.2 4.0 4.6 2.2 3.8 0.7 1.0 1.8 5.8 4.2 1.3 1.5 1.8 2.3 3.2 1.8 1.9 4.5 1.0 0.9 1.1

12.2

1.8 0.8 5.4 1.9 5.3 2.2 0.4 0.4 0.6 4.7

17.0 3.7 2.2 2.6 1.9

Note: In the case of Argentina, tax and non-tax revenue figures include intergovernmental transfers.

de Mello / Intergovernmental Fiscal Relations: Coordination Failures and Fiscal Outcomes

seem to affect the central government's fiscal position, which suggests that fiscal decentralization may be a solution to the rather disappointing picture in Figure 1. However, this conclusion does not hold if attention is restricted to the subsample of developing countries, in which an increase in sub-national government spending (as a share of total government spending) tends to worsen the fiscal position of the central government (Panel B ) . " This negative correlation reveals an interesting relationship between policy outcomes and fiscal decentralisation in developing countries, which will be examined in greater detail below.

Intergovernmental Coordination Failures and Fiscal Outcomes

Because coordination failures cannot be measured directly, their impact on fiscal outcomes can be estimated indirectly when proxies for the potential sources of coordi- nation failures are available. The estimating strategy here consists of regressing the central government's budget balance, measured as the ratio of the fiscal deficit to GDP, on a set of regressors of two types: (i) fiscal decentralisation indicators, which are expected to proxy for the potential sources of intergovernmental coordination failures; and (ii) a number of control variables which have become standard in the public finance literature.'*

The fiscal decentralisation indicators used here are as follows:

• Sub-national government spending (C) measures the extent of fiscal decentralisation. A high sub-national share of total public sector spending is expected to increase the scope for coordi- nation failures in intergovernmental fiscal relations, given the larger amount of resources controlled by sub-national jurisdictions.

• Sub-national tax autonomy (P). This variable measures the ability of sub-national govern- ments to finance spending in their jurisdictions through local revenue mobilisation, and is therefore likely to be a good proxy for moral hazards in decentralised policy making.

• Sub-national dependency on intergovernmental transfers (TR'), or vertical imbalances. This variable is a proxy for coordination failures due to "common pool" problems by, it defines the gap between sub-national expenditure functions and revenue capacity, or between the benefits and costs of government spending.

The control variables used here are as follows:'^ 1. Money creation (Am). The incorporation of money creation in the estimating

equation serves a number of purposes. First, it may be argued that decentralisation may foster central bank independence

and ensure arm's length transactions between the government and the banking sector, by increasing transparency in the assignment of public sector functions, including that of the central bank.̂ ^ In this case, fiscal decentralisation may be conducive to mon- etary discipline. Second, if decentralised policy making leads to a deficit bias, fiscal imbalances may be financed by money creation. The deficit bias may be due to the need to finance unftinded transfers to sub-national governments; lack of control over sub-national spending; and soft budget constraints at the sub-national level, with im-

PubUc Budgeting & Finance / Spring 1999

FIGURE 1

I •(5

m "c E

i/e rn

o O

c 0) O

1 1

0.5

0 •

(

-0.5

-1

-1.5

-2 -

-2.5 -

-3 -

Sub-national Government Size and Full Sample

•

• • • • * • • •

» • • 5 10 15 20 " ^ — * * ^ ^ /

•

Sub-national Government Size

Fiscal

• •

Position-

30 35

•

de Mello / Intergovernmental Fiscal Relations: Coordination Failures and Fiscal Outcomes 11

FIGURE 2

O c ro ro cn c a>

I (3 "TO

0 (

-2

-4

- 0

-8

- 1 2 -

Sub-national Government Size and Central Government Fiscal Position-Full Sample

• } 1 1 A 1 1

1 % •IO 15 20 - 25 • • • • •

• • ^ ^ •

• • • : • • ' •

•

Sub-national Government Size

1 -H

30 * 35

1 T

m -2

F -3 c o -4 S O -5 ro ^ -6

-8

Sub-national Government Size and Central Government Fiscal Position- Developing Country Sample

Sub-national Government Size

12 Public Budgeting & Finance / Spring 1999

plicit bailouts to troubled government-owned financial institutions and sub-national governments. Also, because monetary policy is carried out by the central government, inflationary finance can be regarded as a tax levied by the center. As a result, an increase in the central government's deficit due to failures in intergovernmental fiscal relations can be collected by a rise in explicit taxes (financed by the center and not shared with sub-national governments), and a rise in the inflation tax. Third, the rate of money creation may be a good proxy for the interest rate, and hence the actual cost of financing fiscal imbalances and servicing the national debt. Finally, money creation may be an indicator of financial deepening: if domestic capital markets are shallow, the government may have few non-inflationary options to avoid lax monetary policy and finance fiscal imbalances.

2. GDP growth (Ay ). The rate of growth of the economy provides an indication of the cyclicality of flscal policy; fiscal imbalances tend to be smaller in periods of expansion as a result of increased revenues. On the other hand, GDP growth may also provide an indicator of volatility, and hence the ability of the government to cushion adverse domestic or external shock by increasing public spending counter-cyclically.

3. Terms of trade (77). Openness to foreign trade has become a standard control variable in government size equations.^' However, when the dependent variable is fiscal outcomes, rather than government spending as a share of GDP, international relative prices may be a better control variable.22 This is because tax and non-tax revenues may accrue to the government as a result of an improvement in the terms of trade. Quasi-fiscal revenues also accrue to the government as a result of an improve- ment in the terms of trade due to favourable exchange rate movements, particularly when government revenues are denominated in foreign currency and expenditures/ liabilities are denominated in domestic currency. This is the case of natural resource- exporting countries, for instance.

4. Age dependency ratio (Dep). The age dependency ratio proxies for the long-run social security liabilities of the public sector: the larger the dependent population relative to the total working-age population, the greater the demand for public sector welfare provision (primary education, health care, pensions, etc.), and the long-run welfare-related liabilities of the public sector.

The basic structure of the regression model is as follows:

(1) di, = ao + a\Gif -1-

In equation (1), i indexes the countries in the panel, t denotes time, d^ is the central government's balance (budget deficit as a share of GDP), and e,, is an error term. Theory suggests that «] > 0 (due to intergovernmental coordination failures in gen- eral), a2 > 0 (due to moral hazards in decentralised policy making), 03 > 0 (due to "common pool" problems), 04 >< 0 (negative if deficits are monetised; and positive, if the variable captures debt-servicing costs), as < 0 (if fiscal policy is counter-cycli- cal), a6 < 0 (due to quasi-fiscal revenues), a, > 0 (due to the accumulation of welfare- related long-term liabilities).

de Mello / Intergovernmental Fiscal Relations: Coordination Failures and Fiscal Outcomes 13

EVIDENCE

All equations are estimated for a panel of thirty countries for which basic fiscal indicators are presented in Tables 1 and 2. The variables in the panel are constructed for five-year averages in the period 1970-95 to smooth out transitory fluctuations in the data. Preliminary descriptive statistics are provided in Tables Al and A2 in the Appendix. The use of panel analysis serves the purpose of combining the time series and the cross-sectional dimensions of the decentralization process, rather than simply exploring one of these two aspects separately. Because fiscal decentralization has evolved in different countries at different paces, particularly in developing countries, the time series dimension of the data deserves special attention. Crosssectional varia- tions in the data are more relevant in the case of more mature, established federations, particularly in the developed world, in which intergovemmental fiscal relations and budgetary institutions have changed less markedly over time.

The results of the estimations are reported in Tables 3 to 5. For the full sample of thirty countries, the parameter estimates reported in Table 3 are in general correctly signed, of reasonable magnitudes, and robust to different model specifications. The preliminary findings reported in Models 1 and 2 provide prima facie evidence that fiscal outcomes are likely to be affected by coordination failures due to "common pool" and moral hazard problems, as hypothesised in previous sections.23 In Model 3, an interaction term was included in the regression to measure the combined impact of sub-national dependency and spending shares. In this case, the finding suggests that sub-national dependency has a more detrimental impact on fiscal outcomes when sub-national govemments are large relative to the central govemment in terms of spending. The interaction term is also robust to the inclusion of the sub-national spending share as a regressor in Model 4. As suggested in Figure 2 (Panel B), an increase in sub-national spending shares tends to worsen fiscal outcomes. As for the control variables, GDP growth and money creation are correctly signed and of reason- able magnitudes across model specifications.^'* Although correctly signed, the age dependency ratio and the terms of trade failed to be statistically significant, when group dummies are included in the regression.^^

Given the findings above, it can be argued that developed, mature decentralised govemments may exhibit different policy dynamics to those of govemments undergo- ing significant institutional changes towards fiscal decentralisation. In this case, be- cause the relationship between decentralisation indicators and fiscal outcomes may differ across broad groups of countries, the models in Table 3 were reestimated sepa- rately for sub-samples of seventeen OECD countries and thirteen non-OECD coun- tries.26 It is also fair to expect greater cross-country data variation in the non-OECD sample. The resuhs are reported in Tables 4 and 5 and reveal interesting differences between these two groups of countries.

First, sub-national tax autonomy worsens fiscal outcomes in the OECD sample, but no statistically significant causal relationship was found in the non-OECD sample.

Public Budgeting & Finance / Spring 1999

TABLE 3 Decentralised Fiscal Outcomes, Full Sample (Dependent Variable:

Central Government Balance)

Intercept

Log Sub-national Tax Autonomy

Lagged Log Sub. Tax Autonomy

Log Sub-national Dependency

Lagged Log Suh-nat. Dependency

Model 1

0.12** (3.713)

Interaction Term (Sub-nat. Dep. x Sub- nat. Spending Share) Log Sub-national Spending Share

GDP Growth

Lagged Money Creation

Log Dependency Ratio

Log Terms of Trade

FTest

Nobs. Estimation Method Group Dummies LM Test

Hausman Test

-0.97 (-1.588) -0.07+

(-1.817) -.10

(-0. 106) -0.02

(-0.298) 0.30 2.61

[34.115] 150 OLS Yes

5.94 [0.014] 13.22 [0.021]

Model 2

0.77* (2.230)

-0.11+ (-1.811) -0.06+

(-1.697) 0.17

(0.161) 0.01

(0.203) 0.21 2.20

[34.115] 150 OLS Yes

6.72 [0.009] 13.39 [0.020]

Model 3

0.10* (2.089) -0.32**

(-3.437) -0.72

(-1.521) 0.40**

(4.409)

-0.12* (-2.005) -0.07+

(-1.940) 0.26

(0.252) -0.75

(-0.126) 0.33 2.99

[37.112] 150 OLS Yes

0.64 [0.425] 23.57 [0.003]

Model 4

0.04 (0.462) 0.10*

(2.021) -0.31**

(-3.313) -0.07 (-1.503)

0.34** (3.096) 0.02

(0.225) -0.11+

(-1.793) -0.07+

(-1.968) 0.07

(0.065)

0.32 2.93

[38.111] 150 OLS Yes

0.77

Model 5

44.57

(1.622) -O.08

(-1.229) 0.10*

(2.395) -O.40**

(-4.389) -0.09*

(-2.227) 0.46**

(4.579) 0.12*

(2.174) -0.003

(-0.525) -O.005

(-0.182) 1.11*

(2.262)

0.23 5.93

[9.140] 150 OLS No

[0.378] 23.54 [0.005]

Note: 5-year average panels, 1970-95. Heteroscedasticity-consistent r-statistics are reported in parentheses. (**), (*), and (+) indicate significance at the 1, 5, and 10 percent levels. LM and Hausmann tests confirm the appropriateness of the fixed-effect (OLS) estimators.

de Mello / Intergovernmental Fiscal Relations: Coordination Failures and Fiscal Outcomes 15

TABLE 4 Decentralized Fiscal Outcomes, OECD Sample

(Dependent Variable: Central Government Balance)

Model 1 Model 2 Model 3 Model 4

Intercept

Log Sub-national Tax Autonomy

Lagged Log Sub-nat. Tax Autonomy

Log Sub-national Dependency

Lagged Log Sub-nat. Dependency

Interaction Term (Sub-nat. Dep. x Sub- nat. Spending Share) Log Sub-national Spending Share

Lagged GDP Growth

Log Dependency Ratio

FTest

Nobs. Estimation Method Group Dummies LM Test

Hausman Test

1.18-1- (1.944) 0.21**

(4.730) 0.08+

(1.876)

1.95* (2.042) 0.26

85 GLS No

0.33 (0.470)

0.08-1- (1.784)

0.79 (0.714) 0.03

85 GLS No

1.41** (2.846)

0.20** (4.064)

- 0 . 4 7 * * (-5.128) -0.12**

(-2.718) 0.56**

(6.534)

2.43** (3.159) 0.40

12.38 [5.79]

85 OLS No

1.17 [0.278] 11.93 [0.036]

1.69** (3.398)

0.11* (2.521) -0.42**

(-4.206)

0.42** (3.871) 0.15*

(2.272) -0.03

(-0.537) 2.82**

(3.598) 0.39 9.89

[6.78] 85 OLS No

2.47 [0.116] 10.83 [0.094]

16 Public Budgeting & Finance / Spring 1999

TABLE 5 Decentralized Fiscal Outcomes, Non-OECD Sample (Dependent Variable: Central Government Balance)

Intercept

Log Sub-national Tax Autonomy

Lagged Log Sub-nat. Tax Autonomy Log Sub-national Dependency

Interaction Term (Sub-nat. Dep. x Sub-nat. Spending Share) Interaction Term (Sub-nat. Tax Aut. X Sub-nat. Spending Share) Log Sub-national Spending Share

Money Creation

GDP Growth

Lagged GDP Growth

Lagged Money Creation

Lagged Log Terms of Trade

Log Dependency Ratio

FTest

Nobs. Estimation Method Group Dummies LM Test

Hausman Test

Model 1

0.50 (1.045)

0.18 (0.502)

0.08* (2.192) -0.19**

(-3.006)

0.15* (2.108)

0.21

65 GLS No

Model 2

0.67 (1.506)

0.05+ (1.767)

0.07* (2.057) -0.19**

(-3.204)

0.13* (2.019) -0.13*

(-2.025) -0.53

(-0.495) 0.25

65 GLS No

Model 3

0.40 (0.547)

-0.05 (-0.602)

0.05 (0.608)

0.01 (0.220) 0.06

(1.546) -0.19**

(-3.038) 0.13*

(1.920)

-0.13* (-1.925) -0.51

(-0.445) 0.24

65 GLS No

Model 4 1

0.61 (1.021) 0.01

(0.176)

-0.01 (-.175) 0.05

(0.852) 0.07*

(2.159) -0.17**

(-2.812) 0.14*

(2.060)

-0.14* (-2.056) -0.46

(-O.430) 0.13 2.29

[8.56] 65 OLS No

1.58

Models

-0.31* (-2.645)

0.31** (2.661) -0.12

(-0.937) -0.02 -0.541) -0.26**

(-3.638) 0.10

(1.500)

-0.87 (-0.700)

0.26 2.13

[20.44] 65 OLS Yes

[0.208] 16.60 [0.034]

Note: 5-year average panels, 1970-95. Heteroscedasticity-consistent t-statistics are reported in parentheses. (**), (*), and (+) indicate significance at the 1, 5, and 10 percent levels. LM and Hausmann tests confirm the appropriateness of the fixed-effect (OLS) estimators.

de Mello / Intergovernmental Fiscal Relations: Coordination Failures and Fiscal Outcomes 17

except when an interaction term was included to capture the combined effect of tax autonomy and sub-national govemment size. The latter finding is suggestive of effi- ciency losses involved in the devolution of tax bases to sub-national govemments. Also, the transfer of revenue sources to sub-national govemments, following the devo- lution of expenditure functions, may have deprived the centre of important revenue sources in these countries, thus generating imbalances at the center.^' As suggested above, because local tax bases are narrow, sub-national spending cannot increase without the devolution of broader tax bases, unless vertical imbalances are allowed worsen, thus exacerbating "common pool" problems.

Second, in line with the finding above, "common pool" problems tend to worsen fiscal outcomes in the non-OECD sample. Nevertheless, in the OECD sample, the dependency of sub-national govemments on intergovemmental transfers tends to im- prove fiscal outcomes, as long as sub-national spending is not too large relative to that of the central govemment. In this case, in the OECD sample, vertical imbalances, rather than measuring the extent of "common pool" problems, may provide evidence of the ability of central govemments to put a cap on sub-national spending by increas- ing their dependency on intergovemmental transfers, and hence improve fiscal out- comes.^* Stricter control of the center over sub-national finances and stringency of sub-national balanced budget requirements may also limit the scope for "common pool" problems in the OECD countries.^^

Third, as for the control variables, the more important distinction between the two sub-samples is the fact that the age dependency ratio tends to worsen fiscal outcomes in the OECD sample, but not in the non-OECD sample. In the latter sample, GDP growth, money creation and the terms of trade were found to affect fiscal outcomes more strongly. GDP growth and the terms of trade were found to improve fiscal outcomes, as expected, whereas money creation was found to worsen fiscal imbal- ances, thus reflecting the cost of debt servicing.

We can now proceed to address a final issue of interest in this article. It can be argued that, to assess the impact of coordination failures on fiscal outcomes, the relevant equations should be estimated for each govemment level simultaneously, thereby taking account of the information in the covariance of the error terms in each equation. The results are reported in Table 6 and confirm the findings above. Given the possibility of coordination failures due to moral hazards in decentralised policy mak- ing, sub-national tax autonomy was found to worsen fiscal outcomes at the sub-national govemment level in the full sample and in the non-OECD sample, and at both govemment levels in the OECD sample. Evidence of coordination failures due to vertical imbalances in intergovemmental fiscal relations was found in the non-OECD sample, in which sub-national dependency on intergovemmental transfers was found to worsen fiscal outcomes at the central govemment level. The converse was found in the full sample and in the OECD sample, as above. As for the control variable, in the non-OECD sample, money creation and the terms of trade were also found to worsen

18 Public Budgeting & Finance / Spring 1999

fiscal outcomes. In both the OECD and the full samples, fiscal outcomes were found to worsen due to the social security liabilities associated with high dependency ratios.

CONCLUDING REMARKS

The objective of this article was to identify the sources of coordination failures due to moral hazards and "common pool" problems in decentralised fiscal policy making. A number of different countries, both developed and developing, have implemented public sector reform programs with varying degrees of success in recent years. An important aspect of these programs has been fiscal decentralization, which consists of reassigning expenditure functions and revenue sources to lower tiers of govemment and decentralising fiscal policy making and implementation across govemment levels. Efficiency gains, reduction in operating costs, and improved public sector performance are the expected gains of fiscal decentralisation. Loss of control over sub-national finances and coordination failures in intergovemmental fiscal relations, and ensuing fiscal and monetary disarray, are among its pitfalls.

By focusing on a sample of thirty countries in the period 1970-95 and by estimating the impact of coordination failures on fiscal outcomes, this article contributes to the growing literature on the political economy of policy making and the positive, rather than normative, theory of govemment behavior. The empirical evidence reported here indicates that the merits of fiscal decentralization have to be weighed against the risks involved in increasing sub-national spending power at the expense of higher levels of government. This is likely to increase the scope for coordination failures in decentralised policy making and put pressure on sub-national finances. As a corollary, our results also suggest that correcting these failures by imposing fiscal discipline at sub-national levels of govemment, by market forces or institutional design, is a cmcial prerequisite for successful decentralisation. Intergovemmental cooperation may be en- couraged via such expedients as the enforcement of fiscal contracts in the legislature, centralisation of budget-making, incentives for fiscal restraint, and punishment for excessive largesse.^"

In general, fiscal outcomes were found to be affected by sub-national tax autonomy and dependency on intergovemmental transfers, proxying for, respectively, the extent of moral hazards and "common pool" problems in decentralised policy making. In the OECD area, fiscal institutions may be better designed such that vertical imbalances in intergovemmental relations are not likely to worsen fiscal outcomes. In fact, as sug- gested above, OECD countdes have had higher sub-national spending shares for a much longer span of time than most developing countries examined here, without significant fiscal imbalances at the centre. In these countdes, more stringent control of sub-national finances is believed to have prevented the deterioration of national and sub-national fiscal positions and kept intergovemmental coordination failures in check. OECD countries, unlike the developing countdes in the sample, are therefore believed

de Mello / Intergovernmental Fiscal Relations: Coordination Failures and Fiscal Outcomes 19

TABLE 6 Decentralised Fiscal Outcomes (SURE Estimations)

Log Sub-national Tax Autonomy Lagged Log Suh-nat. Tax Autonomy Log Sub-national Dependency Lagged Log Sub-nat. Dependency Interaction Term (Suh. Dep. x Sub. Spending Share) Log Suh-national Spending Share Money Creation

GDP Growth

Lagged GDP Growth

Log Terms of Trade

Lagged Log Terms of Trade

Log Dependency Ratio

DW Nobs.

Full Sample

Central Gov.

Balance

-0.83 (-1.483)

0.10* (2.385)

-0.41** (-4.651) -0.09*

(-2.382) 0.47**

(4.873) 0.11*

(1.962) 0.06+

(1.947)

-0.02 (-0.486)

0.40+ (1.842) 0.24 1.71

150

Sub-nat. Gov.

Balance

0.20** (3.002)

0.12** (2.594)

0.67** (8.326)

-0.11 (-1.362)

0.60+ (1.834) 0.76 1.58

150

OECD Sample

Central Gov.

Balance

0.16** (2.770) -0.40**

(^.073) -0.13**

(-2.436) 0.48**

(4.059) 0.01

(0.186)

-0.05 (-0.939)

0.39+ (1.648) 0.34 1.13

Sub-nat. Gov.

Balance

0.96** (4.966) -0.02

(-0.379)

0.02 (0.097)

-0.04 (-0.473)

0.79* (2.206) 0.71 0.71

Non-OECD Sample

Central Gov.

Balance

0.05+ (1.797)

0.08* (2.565) - 0 . 1 8 * *

(-3.269) 13.57* (2.210) 0.13

(1.415) -0.13*

(-2.214) -0.69

(-0.719) 0.17 2.23

Sub-nat. Gov.

Balance

0.15* (2.383)

0.82** (11.34)

0.22** (3.156)

-0.41** (-3.428)

-2.36+ (-1.770)

0.86 1.95

to be better equipped to reap the benefits of fiscal decentralisation, while maintaining fiscal discipline, at the same time.

In most developing countdes, on the other hand, local revenue mobilisation may not have been encouraged by their systems of intergovemmental transfers, and vertical imbalances have given dse to "common pool" problems, which tend to worsen fiscal

20 Public Budgeting & Finance / Spring 1999

outcomes at the central govemment level. In fact, a number of developing countdes have recently gone a long way in fiscal decentralisation, by raising sub-national spend- ing ratios substantially, devolving revenue sources and expenditure functions to sub-national jurisdictions, and granting significant autonomy in policy making at the sub-national level. However, the devolution of tax bases to sub-national govemments may have reduced the efficiency of tax instruments, and created moral hazards in decentralized policy making. The transfer of spending assignments to sub-national govemments may not have been matched by an proportional reduction in the spending share of the center in these countries. Finally, when fiscal decentralization is carded out in a haste, as in the case of many developing countries, sub-national fiscal imbalances may also be attributed to insufficient expertise building in local and state govemments to handle larger resources and to deal effectively with expenditure management.

NOTES

1. J. Poterba, "Budget Institutions and Fiscal Policy in the U.S. States," American Economic Review, 86 (1996), pp. 395-400.

2. J. Poterba and K. S. Reuben, "State Fiscal Institutions and the U.S. Municipal Bond Market," NBER Working Paper No. 6237, 1997.

3. R. Bird, Tax Policy and Economic Development, Baltimore, MD, John Hopkins University Press, 1992.

4. R. P. Inman and D. L. Rubinfeld, "Designing Tax Policy in Federalist Economies: An Overview," Journal of Public Economics, 60 (1996), pp. 307-334.

5. R. Boadway, S. Roberts and A. Shah, 'The Reform of Fiscal Systems in Developing and Emerging Market Economies: A Federalist Perspective," Policy Research Working Paper No. 1259, World Bank, 1994.

6. R. Prud'homme, "On the Dangers of Decentralization," World Bank Research Observer, August 1995, pp. 201-210. J. Huther and A. Shah, "A Simple Measure of Good Governance and its Applica- tion to the Debate on the Appropriate Level of Fiscal Decentralization," World Bank, mimeo, 1996. T. Ter-Minassian, "Decentralization and Macroeconomic Management," in L. R. de Mello and K. Fukasaku, eds.. Fiscal Decentralization, Inter-Governmental Fiscal Relations and Macroeconomic Governance, Paris: OECD Development Center, 1999.

7. World Bank, The World Development Report, Washington, D.C: The World Bank, 1997. V. Tanzi, "The Changing Role of the State in the Economy: A Historical Perspective," in L. R. de Mello and K. Fukasaku, eds.. Fiscal Decentralization, Inter-Governmental Fiscal Relations and Macroeco- nomic Governance, Paris: OECD Development Center, 1999.

8. M. Olson, Jr., "The Principle of Fiscal equivalence: The Division of Responsibilities among Differ- ent Levels of Govemment," American Economic Review, 59 (1969), pp. 479-487. W. Oates, Fiscal Federalism, New York: Harcourt Brace Janovich, 1972. W. Oates, The Political Economy of Fiscal Federalism, Lexington, D.C: Heath, 1977. D. Wildasin, "Nash Equilibria in Models of Fiscal Com- petition," Journal of Public Economics, 35 (1988). D. Wildasin, "Introduction: Fiscal Aspects of Evolving Federations," International Tax and Public Finance, 3 (1996), pp. 121-135.

9. R. P. Inman and D. L. Rubinfeld, "Rethinking Federalism," Journal of Economic Perspectives, 11, (1997), pp. 43-64.

10. Boltonetal. 11. For further details on the social aspects of "common pool" problems, see B. Weingast, K. Shepsle

de Mello / Intergovernmental Fiscal Relations: Coordination Failures and Fiscal Outcomes 21

and C. Johnsen, "The Political Economy of Benefits and Costs: A Neo-Classical approach to Dis- tributive Politics," Journal of Political Economy, 89 (1981): pp. 642-64.

12. For further details, see A. Alesina and A. Drazen, Why Are Stabilizations Delayed?", American Economic Review, 81 (1991).

13. R. Radner, "The Organization of Decentralized Information Processing," Econometrica, 61 (1993), pp. 1109-46. P. Bolton and M. Dewatriport, "The Firm as a Communication Network," Quarterly Journal of Economics, CIX (1994): 809-40. D. Martimort "The Multiprincipal Nature of Govem- ment," European Economic Review, 40 (1996): pp. 673-85.

14. J. Tirole, "The Intemal Organization of Govemment," Oxford Economic Papers, 46 (1994): 1-29. G. Gilbert and P. Picard, "Incentives and the Optimal Size of Local Territories," European Economic Review, 40 (1996).

15. B. Caillaud, B. Jullien and P. Picard, "Hierarchical Organization and Incentives," European Eco- nomic Review, 40 (1996): pp. 687-95.

16. V. Tanzi, "Fiscal Federalism and Decentralization: a Review of Some Efficiency and Macroeco- nomic Aspects," Annual Bank Conference on Development Economics, Washington, D.C: The World Bank, 1995.

17. The negative relationship prevails even if the outliers are removed from the sample. 18. N. Roubini and J. Sachs, "Political and Economic Detemiinants of Budget Deficits in the Industrial

Economies," European Economic Review, 33 (1989): 903-38. A. Alesina, G. D. Cohen and N. Roubini, "Electoral Business Cycle in Industrial Democracies," European Journal of Political Economy, 9 (1993): 1-23. B. Eichengreen and T. Bayoumi, "The Political Economy of Fiscal Restrictions: Implications for Europe and the United States," European Economic Review, 38 (1994): pp. 783-91. A summary of data sources and variable descriptions is provided in Table A3 in the Appendix.

19. Additional variables that are expected to affect fiscal outcomes are related to electoral systems and budgetary institutions. These political economy variables do not normally survive if other, more powerful regressors, such as the fiscal decentralization indicators considered here, are included in the estimating equation. For further details, see N. Roubini and J. Sachs, "Political and Economic Determinants of Budget Deficits in the Industrial Economies"; Nouriel Roubini, "Economic and Political Determinants of Budget Deficits in Developing Countries," Journal of International Money and Finance, 10 (1991): pp. S49-S72; V. Grilli, D. Masciandaro and G. Tabellini, "Institutions and Policies," Economic Policy, 6 (1991): pp. 341-91; A. Alesina, G. D. Cohen and N. Roubini, "Elec- toral Business Cycle in Industrial Democracies," S. A. Borelli and T. J. Royed, "Govemment 'Strength' and Budget Deficits in Advanced Economies," European Journal of Political Research, 28 (1995): pp. 225-60; J. Von Hagen and I. Harden, "Budget Processes and Commitment to Fiscal Discipline," European Economic Review, 39 (1995): pp. 771-79. For the specific case of Latin America, see A. Alesina, R. Hausniann, R. Hommes and E. Stein, "Budget Institutions and Fiscal Performance in Latin America," NBER Working Paper No. 5586,1996.

20. J. Huther and A. Shah, "A Simple Measure of Good Govemance and its Application to the Debate on the Appropriate Level of Fiscal Decentralization."

21. D. Rodrik, "Why Do More Open Economies Have Bigger Govemments?," NBER Discussion Paper No. 5537, April 1996.

22. Inter-American Development Bank, "Overcoming Volatility in Latin America," in Report on Eco- nomic and Social Progress in Latin America, Washington, D.C: John Hopkins University Press, 1995.

23. The non-tax autonomy indicator was also experimented with, but failed to be statistically significant at classical confidence intervals. Non-tax revenues tend to be volatile, given that they comprise natural resource rents, and are therefore likely to worsen fiscal outcomes.

24. The initial debt-to-GDP ratio was also included in the regression but was found not to be statistically significant at classical confidence intervals. The standard deviation of GDP growth was also experi-

22 Public Budgeting & Finance / Spring 1999

mented with to measure economic volatility but the results (not reported) were found to be very similar to the ones presented with GDP growth, and therefore omitted.

25. Country dummies in fixed-effects estimations serve the purpose of controlling for unobservable country-specific factors that are likely to affect fiscal outcomes, in addition to the fiscal indicators used here, such as political risk, policy and institutions.

26. Given that Mexico has not been an OECD Member Country during most of the sample period under examination, and in the face of the country's socioeconomic indicators, it is included in the non-OECD sample.

27. V. Tanzi, "Fiscal Federalism and Decentralization: A Review of Some Efficiency and Macroeco- nomic Aspects." This is precisely the case of VAT in Brazil, which was devolved to middle-tier governments in the late 1980s. See J. R. R. Afonso, "Descentralizacao Fiscal, Efeitos Macroecono- micos e Funcao de Estabilizacao: o Caso (Peculiar) do Brazil," paper presented at the VII Regional seminar of Fiscal Policy, ECLAC, Santiago, Chile, 1996; C E . McLure, Jr., "Topics in the Theory of Revenue Assignment," in M. I. Blejer and T. Ter-Minassian, Macroeconomic Dimensions of Public Finance: Essays in Honor of Vito Tanzi, London: Routhledge, 1997; L.R. de Mello, Jr., "Fiscal Federalism and Macroeconomic Stability in Brazil: Background and Perspectives," in L. R. de Mello and K. Fukasaku, eds.. Fiscal Decentralization, Inter-Governmental Fiscal Relations and Macroeco- nomic Governance, Paris: OECD Development Center, 1999.

28. Australia has an interesting fiscal arrangement: although it is a federation, sub-national governments have limited taxing powers, as shown in section "The Testable Hypothesis." Taxes are collected by the center and subsequently transferred to sub-national jurisdictions for equalization purposes. How- ever, high dependency ratios have not been translated into sizeable sub-national fiscal imbalances.

29. See B. Eichengreen and T. Bayoumi (1994), for further details based on evidence for the U.S. states. 30. For further details, see J. von Hagen and I. Harden, "National Budget Processes and Fiscal Perfor-

mance," IMF working Paper, 1996. For a review of the political economy aspects of budget enforce- ment, delegation and fiscal contracts, see M. Hallerberg and J. von Hagen, "Electoral Institutions and the Budget Process," in K. Fukasaku and R. Hausmann, eds.. Democracy, Decentralization, and Deficits in Latin America, OECD, 1998.

de Mello / Intergoveramental Fiscal Relations: Coordination Failures and Fiscal Outcomes 23

Variable

Sub-nat. Tax Autonomy (%) Sub-nat. Non-Tax Autonomy (%) Sub-nat. Dependency (%) Central Gov. Size (%) Central Gov. Balance (%) Sub-nat. Spending Share (%) Sub-nat. Gov. Balance (%) Age Dependency Ratio (%) GDP growth (%) Money Creation (%) Terms of Trade

APPENDIX

TABLE A l Descriptive Statistics

Mean St. Deviation

42.76 16.65 17.23 10.04 37.43 20.44 27.93 11.41 -3.32 3.22 30.42 15.28 -0.72 0.98

0.62 0.18 3.42 2.35

66.85 260.64 99.67 21.03

TABLE A2 Simple Correlations

Minimum

4.13 4.55 2.66

10.10 -15.81

2.25 ^ . 0 8

0.00 - 1 . 5 7 - 7 . 5 0 43.40

Maximum

93.55 52.21 84.48 56.70

3.20 79.64

1.19 1.00

11.76 2703.00

194.30

Tax Aut.

Non Tax Aut.

Sub. Nat. Dep.

Cent. Gov. Size

Cent. Gov. Bal.

Sub. Gov. Size

Sub. Gov. Bal.

Age Dep. Ratio

GDP Gr.

Money Gr.

Tax Autonomy Non-tax Autonomy Sub-national Dependency Central Gov. Size Central Gov. Balance Sub-nat. Gov. Spending Share Age Dep. Ratio GDP Growth Money Creation Terms of Trade

1.00 0.12

-0.69 -0.33 0.08

-0.01 0.06 0.05 0.05

-0.27

1.00 -0.42 -0.20 0.11

-0.07 0.07

-0.17 0.18

-0.20

1.00 0.20

-0.04 -0.02 0.21 0.21

-O.IO 0.26

1.00 -0.29 -0.16 0.03

-0.51 -0.21 0.26

1.00 -0.02 -0.62 0.31 0.11 0.11

1.00 -0.50 -0.30 -0.15 0.20

1.00 0.45 1.00 0.17 -0.06 1.00

-0.30 -0.12 -0.30 1.00

24 Public Budgeting & Finance / Spring 1999

TABLE A3 Data Summary

Variable Description Source

Sub-nat. Tax Autonomy

Sub-nat. Non-Tax Autonomy

Sub-nat. Dependency

Government Size

Government Balance

Sub-nat. Spending Share

Age Dependency Ratio

GDP growth Money Creation Terms of Trade

Ratio of tax revenue to total revenue of sub- GFS, IMF national govemments (tax, non-tax, inter-governmental transfers, and capital revenue net of grants). Ratio of non-tax revenue (rents, fees, etc.) GFS, IMF to total revenue of sub-national govemments. Ratio of inter-govemmental transfers to total GFS, IMF revenue of sub-national governments. Ratio of total govemment spending to GDP, GFS, IMF per govemment level. Govemment balance as a share of GDP, per GFS, IMF govemment level. Ratio of sub-national govemment spending GFS, IMF to total govemment spending. Ratio of population aged below 15 and above 65 years in total working-age population. World Bank Annual rate of growth of real GDP. GFS, IMF Annual rate of growth of M2. World Bank Ratio of export prices to import prices (base = 1987). World Bank

Note: Sample 1970-95.

de Mello / Intergovernmental Fiscal Relations: Coordination Failures and Fiscal Outcomes 25