Chapter

10 Too Large and Too Small Governments

Editor(s):
Ke-young Chu, Sanjeev Gupta, and Vito Tanzi
Published Date:
May 1999
Share
  • ShareShare
Show Summary Details
Author(s)
Alberto Alesina

Introduction

In today’s world, government intervention in the economy is broad and multifaceted. One of government’s important roles is to provide social safety nets and avoid excessive market-induced levels of social inequality.

At one end of the spectrum, in most industrialized countries, governments try to do too much. Their attempts to provide excessive social insurance are becoming counterproductive; in fact, the tax burden required to support these extensive transfer programs reduces growth and increases unemployment. At the other end, many governments of low-income countries do not do enough, either to provide infrastructure or to reduce inequality, and what they do is often inefficient and corrupt. In these countries, governments generally cannot produce even the most basic public goods and infrastructure needed for development.

At both ends of the spectrum, governments fail. This polarization is the result of a tendency to slip into two political equilibria. In advanced industrialized economies, an increasing share of the population—particularly the unemployed, pensioners, and, of course, public employees—relies on government transfers or salaries as a source of income. The larger this group, the stronger the political demand for public programs; however, the larger the programs, the higher the tax burden needed to support them. In turn, an excessive tax burden slows the growth of the economy, leading to an increase of the share of the population relying on government transfers. As a result, we have a vicious circle, which helps explain the so-called eurosclerosis.

Many developing countries, particularly in Latin America and Africa, are experiencing an opposite vicious circle. Inefficient tax systems and the lack of capacity or will to collect taxes have led to rampant tax evasion and the growth of a black economy, which operates outside the legal system without infrastructures and public enforcement of contracts. In turn, the lack of tax revenues has led to poor public infrastructures, reducing the incentives for participants in the black economy to come out and take advantage of the public infrastructures, embrace the legal system and “law and order,” and pay taxes. Thus, poor public policies have led to tax evasion, and the reduction of tax revenue has made public policies even worse. In addition, the relatively small tax revenue (and foreign aid) have been misdirected or wasted and have not reached the truly needy or supported the development of efficient infrastructures. The very poor quality of public institutions and policies in developing countries coupled with the excessive size but relatively higher efficiency of governments in members of the Organization of Economic Cooperation and Development (OECD) point to an interesting empirical implication: the relationship between the size of government and growth may be complex and not linear. This is precisely the finding of La Porta and others (1998).

In summary, the picture that emerges of the role of government as a provider of social insurance is rather bleak. In many cases, governments do too much—defeating their own purpose—or are incapable of providing a minimum of infrastructure or a well-directed system of social insurance, leading to the appalling phenomena of poverty and income inequality, especially in Latin America and sub-Saharan Africa.1

This paper provides some evidence on the size, composition, and efficiency of government programs in some industrialized and developing countries, then discusses the two equilibria highlighted in this introduction. It concludes by drawing the policy implications, particularly for the IMF.

Size and Composition of the Public Sector

In the last few decades, the size of government has increased.2 The most spectacular phenomenon has been the industrialized economies’ rapid expansion of transfer programs—the so-called welfare state. In fact, since the early 1960s, the composition of government outlays in industrialized economies has shifted from consumption of goods and services toward transfers to households and government wages.3

Figure 10.1 shows the evolution of total government outlays, primary expenditure, and total revenues as a share of GDP for an average of 20 OECD countries from the early 1960s to 1994.4 The size of government has increased substantially. Even more striking are the changes in the composition of government outlays. Figure 10.2 plots several components of government spending as a share of GDP: transfers, government wages, nonwage government consumption, and public investment. Figures 10.3 and 10.4 plot, respectively, the ratio of transfers to total primary spending and the ratio of transfers and government wages to total primary spending. The picture that emerges from these plots speaks for itself: governments in OECD countries are becoming larger, and increasingly turning themselves into redistributive machines. Although public investment shows a declining trend, transfers show a sharply upward movement. As a result, in the 1990s, about three-fourths of total spending has been for transfers and public wages. This implies that an increasingly larger fraction of the population derives its main source of income (wages, pensions, unemployment compensations, welfare subsidies) from the public sector, and is dependent on it.

Figure 10.1.OECD Countries: Expenditures and Revenues

(As a percent of GDP)

Source: Organization for Economic Cooperation and Development.

1Primary expenditure is total expenditure less interest payments.

Figure 10.2.OECD Countries: Composition of Expenditures

(As a percent of GDP)

Source: Organization for Economic Cooperation and Development.

Figure 10.3.OECD Countries: Transfers

(As a percent of primary expenditure)1

Source: Organization for Economic Cooperation and Development.

1Primary expenditure is total expenditure less interest payments.

Figure 10.4.OECD Countries: Transfers and Wage Spending

(As a percent of primary expenditure)1

Source: Organization for Economic Cooperation and Development.

1Primary expenditure is total expenditure less interest payments.

The counterpart of this increasingly redistributive public sector is a growing tax burden, which reduces growth and thus increases unemployment, as some recent papers have documented. For example, Fuente (1997) presents convincing empirical evidence of the negative effect of larger government size on growth in OECD countries.5Alesina and Perotti (1997) show the negative effect on relative unit labor costs (international competitiveness) of high tax rates, particularly in countries with unionized labor markets. These authors show that even though the effects of labor tax increases on labor supply may be small in competitive labor markets, they could be much higher in unionized labor markets, with certain institutional features typical of Western Europe. They compute that a 1 percent increase in labor income taxes may lead to as much as a 2 percent increase in relative labor costs. Unemployment is increased even more by labor regulations and public policies that tend to be overprotective of employed union members—the so-called insiders. The increasing number of “outsiders” are left in the hands of the government and the average taxpayer.6

Reliable and comparable breakdowns of government spending in developing countries are not readily available. The welfare state, however, does not appear to be expanding in the developing world. In these countries, the size and composition of government expenditure vary, and it is difficult to make international comparisons. In many developing countries, the size of government is quite small—often very small; however, small does not necessarily mean efficient. In fact, the quality of spending in many developing countries is quite low. As a result, even the supposedly “pro-poor” types of spending do not do much to reduce poverty and income inequality. Tanzi noted as far back as 1974 that in Latin America, “first it appears that even the supposedly pro-poor social-type expenditure has little effect on income distribution. Second, the group that seems to be getting the greatest advantage from public spending is the urban middle class” (Tanzi, 1974, p. 81).

In two recent papers (Alesina, 1997; 1998), I reviewed the large body of literature that has followed Tanzi’s insight of 1974. I reach similar—and disappointing—conclusions: in Latin American and sub-Saharan African countries, and even in several Asian countries (e.g., the Philippines and Indonesia), public consumption and transfers are often mistargeted, do not reduce income inequality, and largely support special interests, as defined socially, geographically, or occupationally. This disappointing conclusion is reached by reviewing several case studies of public spending on health, education, and social security. For example, Pradhan (1996) shows that in a sample of several developing countries, neither the spending on health nor that on education reaches the bottom of the income ladder.7 In most countries, most social spending benefits the wealthiest 40 percent of the population. Graham (1994) shows how public support of education is highly regressive and ineffective in Senegal. Angell and Graham (1995) reach a similar conclusion for Venezuela: “The problems in Venezuela’s social sectors stem from decades of resource misallocation rather than from fiscal constraints” (p. 212). Work by Aspe and Sigmund (1984) highlights the pro-city bias in welfare spending in Mexico, which has generated an enormous geographical difference in the coverage and quality of public services and infrastructures.

A more recent volume, edited by van de Walle and Nead (1995), presents case studies on the effects of welfare spending, with particular attention to Indonesia, Peru, Pakistan, the Philippines, Sri Lanka, India, and Malaysia. The conclusion of these case studies is that welfare spending in these countries is mistargeted and not particularly progressive—if not plainly regressive—in addition to being inefficiently administered. For example, Pitt, Rosenzweig, and Gibbons state that spending on various public education programs in Indonesia “cannot account for a large part of the actual growth in human capital outcomes … in the 1980s” (p. 145). As for public health spending in Indonesia, Deolalikar says that “the evidence indicates that increased government expenditure is actually associated with lower use of health services by the children of the poor” (p. 283). And van de Walle states that health programs in Indonesia “are not particularly well targeted. Indeed the uniform (untargeted) provision of lump-sum transfers to the whole population would have been much more progressive” (p. 249). The evidence on all countries covered in this volume leads to similar discouraging conclusions.

Rampant corruption and bureaucratic inefficiency make mistargeting even worse. Every available measure of corruption is inversely related to per capita income, whereas every measure of bureaucratic efficiency, rule of law, and contract enforceability is directly related to per capita income (Mauro, 1995; and Barro, 1997). Table 10.1 shows some simple cross-country correlations for the period 1960–92 between income, growth, and government efficiency. The level of income is strongly linked to the quality of the bureaucracy, rule of law, and democracy. Richer countries have lower levels of corruption, and lower levels of ethnic fractionalization and political instability (revolutions). Political stability and institutional quality are also linked to growth.

Table 10.1.Correlations Between Income, Growth, and Government Efficiency, 1960–92(In a sample of 78 countries)
Measures of Government EfficiencyGrowthPer Capita Income (1960)
Ethnic fractionalization–0.23–0.54
Rule of law0.450.75
Corruption0.360.77
Bureaucratic quality0.440.77
Democracy0.220.75
Revolutions–0.27–0.36
Note: Number of observations: 78 countries.Sources: Growth: percent growth rate in real per capita GDP (Summers and Heston, 1991). Per capita income 1960: log of real per capita GDP in 1960 (Summers and Heston, 1991). Ethnic fractionalization: index of ethnolinguistic fractionalization (Mauro, 1995). Rule of law: index of enforcement of rule of law from 1 to 6; higher number means more rule of law (International Country Risk Guide (ICRG)). Corruption: higher values mean less corruption (ICRG). Bureaucratic quality: index of efficiency of bureaucracy; higher values mean more efficiency (ICRG). Democracy: index measuring the extent of civil and political rights; higher values mean more rights (Freedom House). Revolutions: number of revolutions and coups (Banks, 1994). For more details on the data, see Alesina (1998).
Note: Number of observations: 78 countries.Sources: Growth: percent growth rate in real per capita GDP (Summers and Heston, 1991). Per capita income 1960: log of real per capita GDP in 1960 (Summers and Heston, 1991). Ethnic fractionalization: index of ethnolinguistic fractionalization (Mauro, 1995). Rule of law: index of enforcement of rule of law from 1 to 6; higher number means more rule of law (International Country Risk Guide (ICRG)). Corruption: higher values mean less corruption (ICRG). Bureaucratic quality: index of efficiency of bureaucracy; higher values mean more efficiency (ICRG). Democracy: index measuring the extent of civil and political rights; higher values mean more rights (Freedom House). Revolutions: number of revolutions and coups (Banks, 1994). For more details on the data, see Alesina (1998).

Tanzi (1998) convincingly argues that corruption imposes significant economic costs, contrary to some economists’ argument that some amount of corruption facilitates transactions between the private and the public sectors of the economy. The author of this paper, however, acknowledges the difficulty of eradicating corruption, an argument consistent with this paper’s view that a high degree of corruption and tax evasion is a form of inefficient equilibrium.

In an important recent paper, Gupta, Davoodi, and Alonso-Terme (1998) attempt to measure statistically the impact of corruption in a large sample of countries. They show that higher degrees of corruption increase income inequality. A striking finding is that “a worsening in the corruption index of a country by one standard deviation … is associated with the same increase in the Gini coefficient as a reduction in the average secondary schooling of 2.3 years” (p. 4). This result is quite strong, since secondary schooling is one of the variables most strongly associated with inequality (Perotti, 1996). A vast body of empirical literature has established that high levels of income inequality are harmful to growth;8 therefore, higher corruption, by increasing inequality, would also have a negative effect on growth.

In summary, poor countries have misdirected spending programs, relatively high levels of corruption, and inefficient bureaucracies. Viewed together, these observations present an interesting empirical implication for the economic effect of government size. On average, the larger governments of OECD countries are more efficient than the smaller governments of developing countries; however, the former impose a higher tax burden. The empirical relationship, therefore, between government size and economic success (say, economic growth) is complicated by this direct relationship between efficiency and size of the government sector.9

Two Vicious Circles

In the industrialized economies, retirees live longer, are more numerous, and receive an increasingly larger fraction of GDP as their pensions. In several of these countries, a good example being Italy, the weight of pensions is the greatest obstacle to achieving a permanent budget consolidation; and other forms of social insurance are also squeezed by pensions. Pay-as-you-go systems have created a large constituency of the elderly, who remain opposed to pension reform. Retirees are normally relatively wealthy and have more time to engage in political action; as a result, their political influence may exceed their voting share in the population. This effect is reinforced in situations where retirees are active members of the union movement and therefore influence union policies on welfare reforms.10

Similar arguments apply to public employees and bureaucracies. In many countries, public employment is used as an indirect means of distributing resources among politically influential lobbies.11 In many industrialized countries, public employment has virtually permanent tenure because it is so difficult to dismiss public employees.

Casual observations suggest that public employee union members, like retirees, have political influence that exceeds their voting share in the population. Recent findings by Alesina, Perotti, and Tavares (1998) are consistent with this hypothesis. These authors examine the electoral effects of fiscal policy and find that governments of OECD countries that engage in fiscal adjustments are typically not punished by the voters. In addition, fiscal contractions based on cuts in transfers and public wages appear to be electorally rewarded!12 These results are very different from the standard perception that reductions in deficits are politically costly. One interpretation is that the constituencies that benefit the most from an overextended public sector and welfare system have enough influence to block any welfare reform and fiscal adjustments; however, the governments that “bite the bullet” and take the risk are not punished at the ballot box, at least not systematically.

The existence of a “blocking coalition” of public sector unions (often supported by other unions) and retirees may be particularly harmful; the recent literature has consistently reached the following conclusion:13 fiscal adjustments that lead to a more permanent consolidation of public budgets are those that rely primarily or exclusively on spending cuts, and especially cuts in transfer programs and government wages. Therefore, a coalition of public sector unions and retirees would block exactly the type of spending cuts that are necessary.

In summary, a large and politically influential coalition derives its income from transfers and public wages. In turn, the internal dynamics of entitlement programs lead to an increase in the tax burden. The latter leads to a reduction in the rate of growth, making the welfare problem (unemployment and early retirement) even worse, which, in turn, leads to the vicious cycle. Even “culture” and attitudes play a role. In many OECD economies, a culture of welfare leads to an expectation that the taxpayers have to support a large fraction of non-workers. To the extent that this becomes part of the “norm,” it has perverse incentive effects: when a large fraction of the population enjoys lifelong public employment, pensions unrelated to their contributions, and generous unemployment compensations, it is difficult to eradicate the attitude of dependency.

An opposite vicious circle occurs in several developing countries. In many cases, a relevant fraction of the economy operates in the black, or at least in the gray, area; many businesses and individuals prefer to escape taxes and regulations even at the cost of not being able to use the public infrastructures available to the formal sector. The more inefficient the public sector, the lower the benefits for the formal sector and the more attractive the informal options.

Loayza (1996), for example, studies a two-sector model of economies with formal and informal sectors. He shows that, depending on parameter values, the economy can converge to an equilibrium with a large informal sector, low tax revenues, low enforcement of tax collection, and poor public services, or to an equilibrium with the opposite features. The informal sector is less productive than the formal, and specializes in low value-added activities. The two equilibria with a high or low share of the informal economy can thus be ranked in terms of efficiency.

The larger the informal sector, the more difficult it is for the government to collect taxes. With low tax revenues, the government is forced to keep spending low. Because the spending programs that are cut are not always the most inefficient, but those without vocal support, low tax revenues also reduce the efficiency of public goods and infrastructures. This results in a vicious circle of low revenues leading to an inefficient government, creating an even larger incentive for the economy to become informal, and thus a further lowering of tax revenue. Johnson, Kaufman, and Shleifer (1997) present a useful model of this interaction and apply it to analyze the emergence of a black economy in Eastern Europe. Among the infrastructures that the informal economy cannot take advantage of, these authors emphasize, is the provision of “law and order.” To the extent that law and order are inefficiently provided for the formal economy, the incentive to operate informally is higher.

Tax evasion is not a prerogative of developing countries; it is present in OECD countries as well.14 However, even the dynamics of tax evasion and tax collection exhibit multiple equilibrium features. If tax evasion is low, the probability of being caught for a given enforcement effort is higher, thus reducing the incentive to evade; the higher the tax evasion, the lower the risk of detection—thus increasing the incentive to evade. Therefore, tax evasion is another force pushing toward two polarized equilibria. Cultural attitudes toward evasion are important. If the public is used to avoiding payment of taxes, tax evasion loses any social stigma and is viewed as a smart business practice. It then becomes politically costly—even for a well-meaning government—to prosecute tax evaders. When the evaders become a majority, or close to it, there is no political support for tax reforms.

In summary, some governments “fail” because they are too large, others because they are too small and inefficient. Self-sustained politicoeconomic forces push governments toward these two “extremes,” both of which can be inefficient equilibria.

Policy Implications for the International Monetary Fund

The IMF often advises countries to reduce their budget deficits. Fiscal stabilizations are frequently critical components of country programs. Even in countries receiving no direct lending from the IMF, its judgment on fiscal adjustments is often viewed as a strong indicator of the countries’ progress toward fiscal stability.

Traditionally, the IMF has placed almost exclusive emphasis on the achievement of a balanced budget; it has been less interested in how this target is reached, whether by increasing revenues or cutting spending, and if the latter, on which spending programs. However, the discussion above shows that how a fiscal adjustment is achieved may be as important as the size of the adjustment itself (measured by the amount of reduction of budget deficits). The IMF should focus just as much (or even more) on the size of taxes and spending over GDP as on the size of the deficit over GDP. Abed and others (1998) review the features of IMF-sponsored fiscal adjustments in several developing countries in the past 10 years. This report does illustrate an IMF effort to address compositional issues but acknowledges that “some countries with low initial revenue effort failed to improve revenue performance” (p. 4); and, on the expenditure side, it states that “shortcomings in expenditure management have persisted” (p. 4).

In OECD countries, a large, recently assembled body of evidence shows that the fiscal adjustments that are long-lasting and not contractionary—even in the short run—are those based exclusively on spending cuts, particularly on transfers and government wages.15 For example, Alesina, Perotti, and Tavares (1998) study the effects of fiscal adjustments made in 19 OECD countries from the early 1960s to 1995. The authors define a fiscal adjustment as a year in which the primary cyclically adjusted budget deficit falls by at least 1.5 percent of GDP;16 they define a “successful” adjustment as one followed in the next three years by a fall in the debt-to-GDP ratio of at least 5 percent of GDP. Successful and unsuccessful adjustments look very different. In the successful cases, more than two-thirds of the deficit reductions came from spending cuts; in the unsuccessful, only one-fourth. In the successful cases, more than half the large spending cuts were in transfers and government wages; in the unsuccessful, the cuts were virtually all in public investment. The authors also show that successful adjustments were not associated with recessions, whereas the unsuccessful generated short-run downturns and did not lead to a permanent consolidation of the budget, making another adjustment unavoidable.

A good example is Ireland. In the early 1980s, this country made a tax-based attempt to reduce budget deficits, which was largely unsuccessful. A second attempt followed in the late 1980s. This time, all the adjustment was on the spending side: the ratio of primary spending to GDP fell from 43 percent to 35 percent. The tax burden was actually reduced by 1 percentage point of GDP. Spending cuts were mostly in transfers and government activities; for example, in about three years (1987–89), public employment was reduced by about 10 percent, from 300,000 to 270,000 employees. Since the beginning of this second adjustment, the Irish economy has experienced the highest growth rate in the OECD and has been nicknamed the “tiger of Europe.” In about ten years, the debt-to-GDP ratio has been almost halved, from 110 percent of GDP in 1987 to about 60 percent of GDP at present.

Similar conclusions on the composition of successful fiscal adjustments may not apply to developing countries with “small governments” that experience difficulty in raising revenues. In these cases, fiscal adjustments could provide the opportunity for governments to improve their ability to collect taxes and reduce tax evasion by means of tax and regulatory reforms, thus bringing more of the informal sector into the formal, tax-paying sector. Preliminary results by Gavin and Perotti17 find that, in Latin America, fiscal adjustments that are more likely to be successful and long-lasting are tax-based—this conclusion is exactly the opposite of that reached by the empirical literature on OECD countries. These preliminary results are, however, consistent with the argument that although most OECD governments spend too much, many Latin American governments do not collect enough taxes because of tax avoidance.

However, the conclusion that many developing countries have to raise more revenue does not imply that they simply have to increase spending on existing programs. Their task is more daunting: to reform and redirect spending programs. For example, in many countries public employment (a major component of spending) is an indirect (albeit inefficient) way to transfer income to certain groups rather than a way to provide public services.18 In fact, La Porta and others (1998) find that the share of public wages in total spending is inversely related to several measures of quality of public policies and outcomes.

The available evidence on foreign aid suggests that simply increasing government revenues in developing countries is not enough. Results by Boone (1995) and Burnside and Dollar (1997) show that most foreign aid has been wasted, has increased government consumption rather than investment, and has not promoted the adoption of “better” macroeconomic policies.19

Improvements in a government’s ability to reduce poverty, reduce income inequality, and increase growth should take different forms in different countries. In OECD countries, there is no doubt that these can be better achieved if the coverage and benefits of the welfare state are greatly reduced, particularly in two areas: pensions and unemployment benefits.20 The move from a pay-as-you-go to a fully funded system is overdue in many OECD countries. This move would not only improve economic efficiency but be more equitable, in an intergenerational sense. The nontrivial problems of transition will have to be solved with regard for each country’s specificities, but the general principle is the same. Labor market policies must also be reformed. Excessive employment protection together with high unemployment compensation create labor market rigidities. If governments intervened less in labor markets and reduced the tax burden, growth performance—and even income distribution—would improve.

Many developing countries have even more difficult problems to solve. For some, improving government performance to achieve social goals means switching from a tax-evading economy with no infrastructure and with mistargeted safety nets to a tax-paying formal economy with relatively small but efficient social safety nets. However, many of these reforms are not likely to be successful without a reduction in corruption and bureaucratic inefficiency in these countries. This is precisely why the IMF and World Bank have placed so much emphasis on “governance” in the last few years.

As noted by Tanzi (1998), improving “governance” is not an easy task, and the monitoring of the process by an outside organization is extremely difficult. However, as discussed, these improvements are critical to both reducing inequality and improving efficiency and growth. What can the IMF and World Bank do in this regard? One possibility, discussed in more detail in Alesina (1997), is to impose more stringent institutional requirements on countries that require assistance from the IMF and World Bank. These organizations should consider withdrawing technical and financial assistance from governments that do not satisfy minimum standards of efficiency and transparency and are excessively corrupt. This approach might generate domestic incentives for reform and improvement in governance. The option of continuing assistance may simply be to postpone the necessary “change of regime.”

In summary, the necessary fiscal reforms will vary by country. The IMF has recently been accused, sometimes justifiably, of providing the same policy prescription to every country. Although some of the IMF’s critics have been too harsh, their point has some validity.21 This paper suggests that the IMF support different fiscal programs in different countries, both for fiscal stabilization and for longer-run structural fiscal reforms.

Furthermore, the IMF’s insistence on deficit-reduction policies, although perfectly appropriate in most cases, should not overshadow the level of spending and taxes. For some countries, the cure for the deficit disease may be worse than the disease itself. Consider, for example, the convergence criteria for the European Monetary Union: all fiscal criteria were based on deficit targets. These criteria have encouraged—or forced—overdue reductions in deficits; however, an overwhelming proportion of these recent fiscal adjustments have occurred on the revenue side, further increasing an already high fiscal burden.22 In Italy, where almost all of the fiscal adjustment has been on the revenue side, the tax-to-GDP ratio increased from 42 percent in 1992 to about 49 percent in 1997. On the spending side, the transfer-to-GDP ratio continued to increase throughout the adjustment, and the public-wage-to-GDP ratio did not fall. As in many other OECD countries, the reduction of the size of Italy’s government is as, if not more, important than reductions in the debt-to-GDP ratio. In the next decade, the priority should be to make substantial, balanced budget cuts in taxes and spending.

Another policy implication of these arguments about multiple and inefficient equilibria on government size is the “speed” of fiscal reforms. For both OECD economies and developing countries with large informal economies, a “cold-turkey” approach is probably the best. Both types of governments need to converge to a different equilibrium. For OECD countries, the change of regime should signal a significant reduction of the role of government as provider of guaranteed income to those not in the labor force or the unemployed. The welfare state, as envisioned in the 1960s and 1970s in many European countries, simply cannot survive. For developing countries, a change of regime is needed to provide more incentives for choosing the formal over the informal economy. The most important—and difficult—step is to improve tax collection.

Improving tax collection will require a coordinated effort: governments must increase their efficiency in providing services while reducing tax evasion, thereby expanding the tax base without increasing tax rates. This will not be easy; it will require a change of regime to lend credibility to the reforms.23 Once the change is made, the convergence to a new equilibrium should happen relatively fast: when the vicious circle of poor public services, tax evasion, and the black economy is reversed, government efficiency should quickly improve.24

The above-mentioned fiscal reforms may create the need to improve social safety nets; however, one should not overestimate the cost of the reforms. Well-crafted policy reforms may begin to deliver social returns relatively quickly, as seen in Latin America and Eastern Europe. This suggests that one should not become “obsessed” with social safety nets: market opportunities work pretty well and pretty fast. However, some amount of social safety nets will be necessary to ensure a successful reform package. The economics of social safety net design is relatively well understood (see, e.g., the recent volume edited by Chu and Gupta, 1998). Political economy issues are more difficult to resolve. For example, Graham (1994) documents that even well-intended social programs have ended up supporting relatively privileged groups rather than the really poor. Graham shows how government bureaucracies and the urban lower-middle class have been excessively protected relative to the rural poor in many Latin American countries. Chu and Gupta (1998) define this phenomenon as the “middle class capture” of social safety nets (p. 9).

In principle, appropriate targeting could solve the problem; in practice, however, targeting has limitations. First, because of the middle class capture, targeting to the really poor may be politically unfeasible; that is, programs that target disadvantaged minorities may not be politically feasible if the country has great ethnic and linguistic diversity. Second, identifying and reaching the very poor, particularly those in urban areas and in a country with a large informal economy, maybe quite difficult. Furthermore, complicated and highly targeted programs may be technically difficult to implement and monitor, creating incentives for distortions and corruption. Tax-spending schemes must therefore be simple if they are to be effective. The choice between simple, broad-based programs and targeted programs may vary among countries. This choice must not be based solely on economic factors but should take into account the political equilibrium needed to support and implement the programs. Some programs may therefore have to be more broadly applied—and less targeted—than is optimal, in order to ensure political support.

References

    AbedGeorge T. and others1998Fiscal Reforms in Low-Income Countries: Experience Under IMF-Supported Programs IMF Occasional Paper No. 160 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

    AlesinaAlberto1997“The Political Economy of High and Low Growth,”paper prepared for the Annual World Bank Conference on Development EconomicsApril 30 and May 1 (Washington: World Bank).

    • Search Google Scholar
    • Export Citation

    AlesinaAlberto1998“The Political Economy of Macroeconomic Stabilizations and Income Inequality: Myths and Reality,” in Income Distribution and High-Quality Growthed. by VitoTanzi and Ke-youngChu (Cambridge, Massachusetts: MIT Press).

    • Search Google Scholar
    • Export Citation

    AlesinaAlberto and SilviaArdagna1998“Tales of Fiscal Adjustments,”Economic Policy No. 27 (October) pp. 487545.

    AlesinaAlbertoRezaBaqir and WilliamEasterly1997“Public Goods and Ethnic Divisions,”NBER Working Paper No. 6009 (Cambridge, Massachusetts: National Bureau of Economic Research).

    • Search Google Scholar
    • Export Citation

    AlesinaAlberto and DavidDollar1998“Who Gives Foreign Aid to Whom and Why?”NBER Working Paper No. 6612 (Cambridge, Massachusetts: National Bureau of Economic Research).

    • Search Google Scholar
    • Export Citation

    AlesinaAlberto and RobertoPerotti1997“The Welfare State and Competitiveness,”American Economic ReviewVol. 87No. 5 pp. 92139.

    • Search Google Scholar
    • Export Citation

    AlesinaAlberto and JoséTavares1998“The Political Economy of Fiscal Adjustments,”Brookings Papers on Economic Activity: 1 Brookings Institution pp. 197266.

    • Search Google Scholar
    • Export Citation

    AngellAlan and CarolGraham1995“Can Social Reform Make Adjustment Sustainable and Equitable? Lessons from Chile and Venezuela,”Journal of Latin American StudiesVol. 27Part I (February) pp. 189219.

    • Search Google Scholar
    • Export Citation

    Aspe ArmellaPedro and Paul E.Sigmundeds.1984The Political Economy of Income Distribution in Mexico (New York: Holmes and Meier).

    • Search Google Scholar
    • Export Citation

    BanksArthur S.1994“Cross-National Time Series Data Archive,”Center for Social Analysis (Binghamton, New York: State University of New York).

    • Search Google Scholar
    • Export Citation

    BarroRobert J.1989“Economic Growth in a Cross Section of Countries,”NBER Working Paper No. 3120 (Cambridge, Massachusetts: National Bureau of Economic Research).

    • Search Google Scholar
    • Export Citation

    BarroRobert J.1991“World Interest Rates and Investment,”NBER Working Paper No. 3849 (Cambridge, Massachusetts: National Bureau of Economic Research).

    • Search Google Scholar
    • Export Citation

    BarroRobert J.1997Determinants of Economic Growth: A Cross-Country Empirical Study (Cambridge, Massachusetts: MIT Press).

    BoonePeter1995“Politics and the Effectiveness of Foreign Aid,”NBER Working Paper No. 5308 (Cambridge, Massachusetts: National Bureau of Economic Research).

    • Search Google Scholar
    • Export Citation

    BurnsideCraig and DavidDollar1997“Aid, Policies, and Growth,”World Bank Policy Research Working Paper No. 1777 (Washington: World Bank).

    • Search Google Scholar
    • Export Citation

    ChuKe-young and SanjeevGuptaeds.1998Social Safety Nets: Issues and Recent Experiences (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

    Freedom Housevarious issues“Freedom in the World: Political Rights and Civil Liberties” (New York: Freedom House).

    FuenteAngel de la1997“Fiscal Policy and Growth in the OECD,”CEPR Discussion Paper No. 1755 (London: Centre for Economic Policy Research).

    • Search Google Scholar
    • Export Citation

    GavinMichael and RobertoPerotti1997“Fiscal Policy in Latin America,”NBER Macroeconomics Annualed. by B.Bernanke and J.Rotemberg (Cambridge, Massachusetts: MIT Press).

    • Search Google Scholar
    • Export Citation

    GrahamCarol1994Safety Nets Politics and the Poor: Transitions to Market Economies (Washington: Brookings Institution).

    GuptaSanjeevHamidDavoodi and RosaAlonso-Terme1998“Does Corruption Affect Income Inequality and Poverty?”IMF Working Paper 98/76 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

    International Country Risk Guide (ICRG)various issuesPolitical Risk Services (Syracuse, New York).

    International Monetary Fund1996World Economic Outlook (Washington: IMF) May.

    JohnsonSimonDanielKaufman and AndreiShleifer1997“The Unofficial Economy in Transition,”Brookings Paper on Economic Activity: 2 Brookings Institution pp. 159239.

    • Search Google Scholar
    • Export Citation

    KraemerMoritz1997“Electoral Budget Cycles in Latin America and the Caribbean: Incidence, Causes and Political Futility,”Inter-American Development Bank Working Paper Series 354 (Washington: Inter-American Development Bank).

    • Search Google Scholar
    • Export Citation

    La PortaF. and others1998“Fundamental Determinants of Government Performance” (unpublished).

    LindbeckAssar and DennisSnower1984“Involuntary Unemployment as an Insider-Outsider Dilemma,”University of Stockholm Institute for International Economic Studies Seminar Papers No. 282 (July) pp. 145.

    • Search Google Scholar
    • Export Citation

    LoayzaNorman1996“The Economics of the Informal Sector: A Simple Model and Some Empirical Evidence from Latin America,”Policy Research Working Paper 1727 (Washington: World Bank).

    • Search Google Scholar
    • Export Citation

    MauroPaolo1995“Corruption and Growth,”Quarterly Journal of EconomicsVol. 110No. 2 (August) pp. 681712.

    McDermottC. John and Robert F.Wescott1996“An Empirical Analysis of Fiscal Adjustments,”Staff PapersInternational Monetary FundVol. 43 (December) pp. 72553.

    • Search Google Scholar
    • Export Citation

    Organization for Economic Cooperation and DevelopmentEconomic Outlook (Paris: OECD) various issues.

    PeltzmanSam1992“Voters as Fiscal Conservatives,”Quarterly Journal of EconomicsVol. 107 (May) pp. 32761.

    PerottiRoberto1996“Growth, Income Distribution, and Democracy: What the Data Say,”Journal of Economic GrowthVol. 1No. 2 (June) pp. 14986.

    • Search Google Scholar
    • Export Citation

    PradhanSanjay Kumar1996“Evaluating Public Spending: A Framework for Public Expenditure Reviews,”World Bank Discussion Paper No. 323 (Washington: World Bank).

    • Search Google Scholar
    • Export Citation

    RostagnoMassimo and FrancescaUtili1998“The Italian Social Protection System—The Poverty of Welfare,”IMF Working Paper 98/74 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

    SummersRobert and AlanHeston1991“Penn World Table (Mark 5): An Expanded Set of International Comparisons, 1950–1988,”Quarterly Journal of EconomicsVol. 106 (May) pp. 32768.

    • Search Google Scholar
    • Export Citation

    TanziVito1974“Redistributing Income Through the Budget in Latin America,”Banca Nazionale del Lavoro Quarterly ReviewVol. 27No. 108 pp. 6587.

    • Search Google Scholar
    • Export Citation

    TanziVito1998“Corruption Around the World: Causes, Consequences, Scope, and Cures,”IMF Working Paper 98/63 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

    TanziVitoed.1982The Underground Economy in the United States and Abroad (Lexington, Massachusetts: Lexington Books).

    TanziVito and LudgerSchuknecht1997“Reconsidering the Fiscal Role of Government: The International Perspective,”American Economic ReviewVol. 87No. 2 (May) pp. 16468.

    • Search Google Scholar
    • Export Citation

    TanziVito and ParthasarathiShome1995“A Primer on Tax Evasion,”Staff PapersInternational Monetary FundVol. 40 (December) pp. 80728.

    • Search Google Scholar
    • Export Citation

    van de WalleDominique and KimberlyNeadeds.1995Public Spending and the Poor: Theory and Evidence (Baltimore, Maryland: Johns Hopkins University Press for the World Bank).

    • Search Google Scholar
    • Export Citation
Note: The author wishes to thank Silvia Ardagna, Olivier Blanchard, Ke-young Chu, and Andrei Shleifer for their useful comments.
1This is, of course, a stylized conclusion. The level of income is not the only determinant of institutional quality and government performance. Important differences on this point remain between the industrialized and developing countries. For a recent insightful discussion of this issue, see La Porta and others (1998).
2See, in particular, Tanzi and Schuknecht (1997) on this point.
3Another component of government outlays—interest payments—has increased in the last two decades owing to the large accumulation of public debt in several OECD countries.
4The countries are Australia, Austria, Belgium, Canada, Denmark, Finland, France, Germany, Greece, Ireland, Italy, Japan, the Netherlands, Norway, Portugal, Spain, Sweden, Switzerland, the United Kingdom, and the United States. All the data are from the OECD.
5For evidence along similar lines on a larger sample of countries, see Barro (1991); Alesina (1997); and the references cited therein.
6For a classic discussion of the insiders/outsiders problem, see Lindbeck and Snower (1984).
7These countries are Argentina, Chile, Colombia, Costa Rica, the Dominican Republic, Uruguay, Indonesia, Iran, Malaysia, the Philippines, and Sri Lanka.
8See Perotti (1996) for a survey of this literature.
9La Porta and others (1998) develop this point and provide supporting evidence.
10An extreme case is Italy, where the majority of the card-carrying union members are pensioners.
11In Italy, for example, per capita public employment is greater in the south than in the north. It is widely recognized that most public employment in the south is disguised compensation for permanent unemployment. See, in particular, Rostagno and Utili (1998).
12Interestingly, very similar results emerge from research focusing on U.S. states (Peltzman, 1992) and in Latin American countries (Kraemer, 1997).
14On this point, see Tanzi (1982); and Tanzi and Shome (1995).
16It turns out that the results are unaffected by the specific type of cyclical correction used.
17These unpublished preliminary results were obtained using the same data set in Gavin and Perotti (1997) but were not published in that paper.
18In Kenya, for example, half the labor force in the formal economy is employed by 93 government agencies (Alesina, Baqir, and Easterly, 1997).
19To be fair, donor countries share some responsibility for the failures of the foreign aid effort, as discussed in Alesina and Dollar (1998).
20The optimal reduction will vary among countries. For example, it is well known that the size of the welfare state is smaller in the United States than in most European countries.
21This is not a reference to the recent discussion on IMF intervention in East Asia; the point here is broader.
22In addition, several commentators have emphasized the use of creative accounting to achieve the targets.
23Argentina, for example, has been successful in its recent effort to increase tax collection in the context of the recent stabilization policies.
24See Johnson, Kaufman, and Shleifer (1997) for a discussion of the informal economy in Eastern Europe and the former Soviet Union.

    Other Resources Citing This Publication