7 Addressing Equity Issues in Policymaking: Lessons from the Chilean Experience

Ke-young Chu, Sanjeev Gupta, and Vito Tanzi
Published Date:
May 1999
  • ShareShare
Show Summary Details
Eduardo Aninat, Andreas Bauer and Kevin Cowan 


The democratic governments of the Chilean Concertación coalition that began in March 1990 have emphasized a development strategy of equitable and sustainable economic growth as a means of improving the quality of life of the population at large. In this paper, we illustrate how Chile has addressed the issue of equity through economic policymaking during the 1990s and present the lessons we have derived from that experience.

We begin our discussion with a brief review of the dimensions of the concept of equity. Next, we discuss the evolution of poverty and the distribution of income in Chile in recent years. As we show, Chile has achieved considerable reductions in poverty in a relatively short time. In contrast, however, the distribution of income has remained fairly stable, although it has not worsened as in a number of other countries in Latin America and elsewhere.1

In general, policymakers can address equity issues by using three main types of policy instruments: macroeconomic management; traditional fiscal policy, including public expenditure and revenue (tax) policies; and regulatory, or microeconomic, policy. The discussion of these instruments represents the core of this paper.

Regarding macroeconomic management, we stress the importance of macroeconomic stability for equity, because of its positive effect on economic growth, which in turn increases government income, generating room for targeted social spending. Other effects relate to the positive impact of increased price and labor stability on equity. In Chile, during the 1990s, prudent fiscal management fostered macroeconomic stability and helped spur economic growth. As we demonstrate, the strong economic growth recorded in that period generated an important component of the additional resources available for social spending.

Available evidence suggests that public revenue policies have only a limited impact on the distribution of income in Chile, where indirect taxes dominate the revenue structure and the government is relatively small. Under these conditions, even radical reforms of the tax system have only a marginal effect on the distribution of income. Because of the latter, public revenue policies have been formulated mainly to ensure macroeconomic stability and to generate the resources needed for targeted social spending in an efficient and sustainable way.

Public expenditure policies, on the other hand, go a long way toward achieving equity goals. As we demonstrate, Chile’s social spending has grown steadily during the 1990s and, in general, is well targeted toward the poorest 40 percent of the population. In addition, targeting of social policies has improved during the 1990s and therefore has contributed to increasing equity.

Social expenditure in Chile is concentrated on providing public goods and services, such as health care and education, which influence consumption levels but do not immediately affect distribution of income. This policy mix clearly favors investment in human capital over short-term income transfer policies, on the grounds that the former is a more sustainable path to improving equity in the long run. A number of studies have shown that differences in human capital are the most important explanation for income disparities in Chile. This evidence supports the special emphasis Chile has placed on education reform as a long-term instrument for improving equity.

Our discussion of the equity effects of regulatory, or microeconomic, policy as a policy instrument is limited to the labor market. We review the possible equity effects of Chile’s minimum wage policy and of its innovative system to protect unemployed workers (PROTRAC), which is currently under discussion in Chile’s Congress.

Policymakers may find it politically difficult to undertake equity-enhancing measures, because beneficiaries are poorly organized compared with other interest groups and/or because visible benefits appear only after a significant time lag. In the fifth section of this paper, we analyze some of the obstacles these policies have encountered and provide some strategies that have become politically viable in Chile. The last section summarizes the main conclusion of the paper.

Considerations Surrounding Equity and Economic Policy

In general, policymakers can address equity issues from two broad perspectives. The first is income or consumption levels and the quality of life for the population at large. The issue of poverty figures prominently in this perspective. Public policy must aim at ensuring that the entire population has access to a minimum level of income or consumption of basic needs. The level above this minimum is not necessarily a policy objective, however, and could be left to market forces.

The second perspective is distribution, which can be looked at from at least three different angles:

  • The distribution of individual monetary income;
  • The distribution of consumption, a proxy for individual welfare; and
  • The distribution of access to wealth-generating factors, or equality of opportunity.

The degree in which inequality in any (or all) of these three dimensions becomes a policy objective will depend on society’s perception of justice and/or implications of inequality for other economic objectives—mainly growth and poverty alleviation.2

Both the distribution of income and the distribution of consumption relate to final outcomes and thus depend on a mixture of endogenous characteristics (skill, effort, or even luck) and exogenous variables (e.g., access to education, distribution of wealth, and discriminatory practices). For policymakers, defining the concept of equity in terms of equal access to wealth-generating factors seems to be a more appropriate goal as the equality of access depends on variables external to the individual, which policymakers can influence. Current social policy in Chile has been inspired by the concept of equity as equality of opportunity. In his election platform, Chilean President Eduardo Frei said his social policy would move gradually toward addressing the issue of access, reinforcing the productive component of social spending, which implies the equalization of opportunities and the achievement of lasting forms of social equity.3

New challenges arise when the equal distribution of opportunities becomes the main objective of social policy. This third dimension of distribution is the hardest to measure. The distribution of current income can be objectively measured, but it is difficult to measure the increases in vertical mobility that result from opportunity-generating policies. Therefore, the true equity effect of a policy that promotes social mobility is often underestimated or ignored.

In addition, because the results of a more equitable distribution of opportunities are not immediately seen, these policies must have long time horizons. In Chile, this may have been easier to achieve in the early 1990s, when the probability of election turnover during the first two mandates of the Concertación government coalition after the end of the military regime was low.

In the long term, these three dimensions of distribution intertwine. For example, a more equitable distribution of education (i.e., improved access to and quality of education for poorer households) will improve the distribution of human capital and thus reduce income inequality.4 Vertical mobility is also likely to increase, especially if large differences in quality exist between the education accessed by the rich (i.e., private schools) and that by the poor (i.e., public schools).

An Overview of Poverty and Income Distribution in Chile

Although per capita income in Chile was, on average, 90 percent higher in 1996 than in 1987, the countrywide income distribution has remained virtually unchanged (as shown in Table 7.1), responding more to cyclical variations in output and employment than to any kind of long-term trend.5 However, a recent World Bank study revealed that at least in the distribution of labor income, the picture is slightly different.6 As can be seen from Table 7.1, although the Gini coefficient for the overall distribution of income remains stable, the trend in recent years is toward greater equality in labor income.

Table 7.1.Chile: Distribution of Monetary Income1
Income Decile19871990199219941996
Income share of top quintile to bottom quintile13.312.912.213.213.8
Income share of top decile to bottom decile27.526.124.527.729.5
Gini coefficient20.4870.4800.4750.4790.481
Gini distribution of labor income30.5760.5390.4740.4590.454
Sources: MIDEPLAN, CASEN surveys 1987 to 1996.

Monetary income = self-generated income + cash transfers from public sector.

World Bank (1997). Wage distribution is for Greater Santiago region.

Sources: MIDEPLAN, CASEN surveys 1987 to 1996.

Monetary income = self-generated income + cash transfers from public sector.

World Bank (1997). Wage distribution is for Greater Santiago region.

According to historical data drawn from standardized surveys of Greater Santiago (see Marcel and Solimano, 1994), there has been little change in the long-term income distribution either. The changes recorded from one presidential period to the next over the past thirty to forty years are minimal, except for the 1974–89 government period, when the income distribution deteriorated significantly. The lack of change is more striking when one considers the variety of economic orders in place over the entire period. International evidence (see Deininger and Squire, 1996) suggests that this “stickiness” in income distribution figures is common in most countries, and that aggregate measures of inequality stay more or less constant over time. Cross-country comparisons, although risky because of the quality and comparability of data, suggest that the income distribution in Chile is quite unequal (see Table 7.2).

Table 7.2.International Comparison of Inequality(Averaged Gini coefficients)
Eastern Europe.24.28
OECD member countries and other high-income.32.35
South Asia.35.31
East Asia/Pacific.35.36
Middle East and North Africa.40.40
Sub-Saharan Africa.41.49
Latin America.51.47

According to CASEN survey data,7 most of the difference in income distribution in Chile is explained by labor income differences, which depend heavily on the individual stock of human capital.8Beyer (1997) estimates that even if all capital income were redistributed toward the poorest 40 percent of the population, the income share of the richest 20 percent would remain above 50 percent.9 Therefore, more significant changes in Chile’s income distribution would be produced by a more equal distribution of human capital rather than a radical redistribution of property.

In stark contrast to the lack of progress in income redistribution, poverty has fallen drastically in Chile in the past 10 years.10 As shown in Figure 7.1, the percentage of the population living with per capita incomes below the poverty line fell from 45 percent in 1987 to less than 25 percent in 1996. Similar reductions occurred in extreme poverty. The considerable reduction in poverty achieved since 1987 occurred in a context of strong GDP growth, rising wages, and rapidly expanding employment (see Table 7.3).

Figure 7.1.Chile: Income and Poverty

Source: Mideplan, CASEN surveys 1987 to 1996, Central Bank of Chile.

Table 7.3.Chile: Selected Macroeconomic Indicators
19871988198919901991199219931994199519961997Average 1987–97Average 1994–97
Per capita GDP1987 = 100 (real)100.0105.5114.7117.0124.2137.1144.3150.1163.5173.0182.7
Annual percentage change4.
InflationDec.-Dec. percentage CPI21.512.721.427.318.712.712.
EmploymentIn thousands3,8964,1234,3524,4504,5184,7244,9925,0365,0955,1825,281
Annual percentage change3.
Wages1987 = 100 (real)100.0106.7108.6110.6116.0121.3125.2131.2136.9142.5146.0
Annual percentage change–
Wage bill11987 = 100 (real)100.0112.9121.4126.4134.5147.0160.5169.6179.1189.6197.9
Annual percentage change3.612.
Minimum wage1987 = 100 (real)100.0107.2119.2127.3138.7144.9152.1157.7164.8171.8177.8
Annual percentage change–
Average public wage1987 = 100 (real)100.099.497.388.189.991.692.795.498.8102.1105.9
Annual percentage change–7.5–0.6–2.1––0.23.4
Minimum pension1987 = 100 (real)100.0102.4102.0109.4122.6122.6124.4124.7129.7142.9147.0
Annual percentage change–3.62.4–
Source: Central Bank, National Institute of Statistics, and Ministry of Finance.

Average wage times employment.

Source: Central Bank, National Institute of Statistics, and Ministry of Finance.

Average wage times employment.

Studies by Contreras (1995) and the World Bank (1997) show that the reduction in poverty is substantial even when differences in household composition (equivalent adults) and regional prices are taken into account. Poverty has fallen in all regions and sectors, as can be seen in Anriquez, Cowan, and De Gregorio (1997). The World Bank study also reveals that poverty reductions have benefited those groups considered to be the most vulnerable in 1987.

Some studies define poverty in terms of basic needs rather than levels of per capita income. In this context, the United Nations Development Programme (UNDP) developed a Human Poverty Index (HPI) in 1997.11 According to the HPI in that year, Chile ranked third among developing countries in providing for the basic needs of its population. The average UNDP Human Development Index (HDI) ranked Chile fifth among developing countries and thirtieth among 175 countries in the world.

Policy Instruments to Address Equity Issues

Policymakers generally use three main types of instruments to address equity issues. The most obvious is traditional fiscal policy, which involves both expenditure and revenue policy. However, macroeconomic policy may also have important direct and indirect distributional consequences, and the same is true for efficient regulatory policy, especially that related to labor and financial market regulation.

Macroeconomic Management

Macroeconomic policy, especially overall fiscal management, has important direct and indirect equity effects, as shown in Figure 7.2. Macroeconomic policies that are aimed at providing price stability and sustained growth may provide a double dividend in terms of equity enhancement. First, increased price and job stability may increase the purchasing power of lower-income groups more than that of higher-income groups, and therefore directly enhance equity. Second, through its contribution to higher economic growth, macroeconomic stability may lead to increased public income, which may in turn improve equity if additional income is allocated to targeted social spending.

Figure 7.2.Chile: Chile: Impact of Macroeconomic Stability on Social Equity

Prudent fiscal management is fundamental to achieving overall macroeconomic stability; in practice, this means high fiscal savings or even overall fiscal surpluses. Fiscal surpluses are not easy to justify, however, especially in countries with high poverty and deficiencies in the supply of essential public goods. The pursuit of strict budgetary discipline should nevertheless be seen as an investment where cost is incurred in the short term to achieve equity enhancements in the long term. The best example of this issue is debt servicing: the repayment of public debt through continuous fiscal surpluses will free resources for redistributive spending and investment. Therefore, fiscal discipline in the short term will increase disposable government income in the medium and long term, and thus increase the government’s capacity to enhance equity.

In the past three decades, Chile has passed through two severe macroeconomic recessions. Unemployment did not drop below 10 percent until 1987, two years before the return to democracy. Inflation, currently at 5.3 percent, was close to 20 percent for most of the past decade. The present favorable macroeconomic setting is therefore characteristic of only the last 10 years. Chile’s experience has demonstrated the importance of adequate and responsible fiscal, monetary, and exchange rate policies. In periods of macroeconomic instability, fiscal policy revolves around issues of stabilization, and equity considerations are often put in second or third place.

A summary of recent macroeconomic indicators is presented in Table 7.3. From 1987 to 1997, the Chilean economy grew at an unprecedented rate, averaging 7.8 percent per year, or a 6.1 percent growth in per capita GDP.

Achieving and sustaining stability has been the guiding principle of macroeconomic policymaking in Chile. An autonomous central bank was established in 1989; it is responsible for the conduct of monetary and exchange rate policies and provides the institutional setting for price stability. In this context, the central bank has carried out a countercyclical monetary policy, pushing interest rates up whenever GDP growth or growth of domestic demand have reached levels deemed incompatible with a downward trend in inflation. Inflation has declined since 1991; since 1994, it has been in the single digits.

Fiscal policy has been relatively stable and sustainable during the 1990s, mostly as a result of the government’s long-term programs and objectives. Fiscal policy has also contributed to short-term macroeconomic stability, as can be seen in Figure 7.3, which shows the annualized growth rates of private and public aggregate demand. Because of the importance of several commodities, particularly copper, in Chile’s economy, terms of trade shocks produce strong expansionary cycles of aggregate demand, which in turn force the central bank to undertake contractionary policies to reduce inflationary pressures. The decomposition of aggregate demand into public and private shows, however, that public demand grows much more in line with GDP than private demand, which is proof that fiscal policy contributes to macroeconomic stability.

Figure 7.3.Chile: Growth of Private and Public Demand

(Annualized percentage change, four-quarter moving average)

Source: Budget Office, Central Bank of Chile.

1Includes consumption and investment.

2Excludes interest and financial investment.

Figure 7.4 shows the long-term fiscal policy trend in Chile: continuous fiscal overall surpluses for more than a decade. As mentioned, under the Concertación governments, fiscal policy has been aimed at (1) maintaining high levels of government saving and an overall fiscal surplus for macroeconomic stability and growth, and (2) maintaining a stable level of government expenditure to finance a sufficient number of well-designed, medium- and long-term social programs.

Figure 7.4.Chile: Fiscal Policy

Source: Chilean Ministry of Finance.

These goals have been met because every expenditure increase has been carefully matched by additional resources, and because the Copper Stabilization Fund has protected fiscal revenue from copper price volatility.12 The tax structure has been modified slightly in the past 10 years (Figure 7.5 shows the evolution of Chile’s main tax rates), leading to a moderate increase in tax revenue as a percent of GDP.

Figure 7.5.Chile: Tax System

(In percent)

Source: Chilean Ministry of Finance.

The fiscal “payback” of this investment in stability has been the significant growth in fiscal revenue due to GDP growth and the reduction of interest payments due to falling government debt. This has allowed fiscal spending to rise significantly without threatening the overall surplus and facilitated the shift from interest payments toward social functions.

Table 7.4 shows the result of a simple decomposition exercise for the growth of total social spending. With per capita government social spending as Gs, then:

Table 7.4.Chile: Decomposition of Sources of Growth of Real Social Expenditure
Average 1989–90 (1)Average 1996–97 (2)Percentage Change (1)–(2)Contribution to Change in Social Expenditure (In percent)
Social expenditure/total expenditure excluding interest payments656736
Total expenditure excluding interest payments/total expenditure9198817
Total expenditure/nominal GDP2221–5–11
Per capita GDP (deflated by CPI)1001393983
Per capita social expenditure (deflated by CPI)1001474711001
Source: Authors’ calculations based on data from the Central Bank and Ministry of Finance.

Figures in these columns do not add to these totals because of residual terms not considered in the decomposition formula.

Source: Authors’ calculations based on data from the Central Bank and Ministry of Finance.

Figures in these columns do not add to these totals because of residual terms not considered in the decomposition formula.

Gs = y · δ · α · β

and its percentage increase can be approximated as:


y = per capita GDP

δ = total government expenditure as a percent of GDP

α = percent of total expenditure that remains after interest payments

β = percent of expenditure (after interest payments) directed to social policy.

As shown in Table 7.4, per capita GDP growth accounts for more than 80 percent of the growth in social expenditure. Falling interest rate payments account for 17 percent, and 6 percent is explained by the reallocation of noninterest spending toward the social sector.

The lowest income groups also benefited from the 1990s’ trend of falling inflation because these income groups usually suffer the most from degenerating purchasing power.13 In Chile, the most direct redistributive effect of lower inflation is the resulting increase of the real value of public transfers to the poor. For example, a simple calculation shows that the reduction of annual inflation from 27.3 percent in 1990 to an estimated 4.5 percent by the end of 1998 would increase the annual purchasing power of the state minimum pension by as much as 9 percent.14

Fiscal Policy

Equity can be enhanced through fiscal policy in two ways. First, fiscal revenue policies, especially tax policy, can be structured so as to place a higher burden on upper-income groups. Second, government spending can be directed toward lower-income groups, to enhance equity.

Enhancing equity through fiscal revenue policies

In principle, equity objectives can be addressed directly through progressive (income) tax policies. Both theory and experience show, however, that fiscal revenue policies based on aggressive taxation are neither efficient nor effective in changing income distribution. High levels of taxation hamper economic growth rates and trigger socially inefficient tax avoidance by taxpayers. Moreover, in an integrated world, countries compete for tax bases, especially capital, thus limiting the scope for high income tax rates vis-à-vis neighboring countries.

For each of Chile’s democratic governments in the 1990s, fiscal stability has been the goal of fiscal revenue policies. Taxes have been adjusted to strengthen the overall fiscal position and to generate the resources needed for sustainable social spending.15 Efficiency, especially the achievement of more intersectoral neutrality, has also been taken into account in designing policy (Marfán, 1998, pp. 568–69). The redistribution of income, however, has not been a primary concern, because the authorities believe that equity can be better enhanced and sustained through targeted social spending. The tax system has generated the resources to fund equity-enhancing expenditure instead of modifying the distribution of disposable income. Indeed, public revenue policies have been formulated on the basis of joint incidence considerations, taking into account the use of public funds, rather than considering a partial analysis of tax incidence only.

Engel, Galetovic, and Raddatz (1997a) quantify the direct impact of taxes on income distribution at the household level in Chile. They also estimate the distributional effect of several tax system reforms, to assess the sensitivity of income distribution to changes in tax policy. Table 7.5 shows their main results: column 1 shows the Chilean income distribution in 1994 before taxes; column 2, the income distribution after taxes. These figures illustrate that the impact of the Chilean tax system on overall income distribution is slightly regressive.16

Table 7.5.Chile: Impact of Tax Reform on Income Distribution
Income DistributionAverage Tax Rate
DecileBefore tax (1)After tax (2)Raising VAT from 18 to 25 percent (3)Introducing 20 percent flat tax (4)Income tax (5)VAT (6)Tariffs and excise taxes (7)
Memorandum items:
Gini coefficient0.48990.49200.49630.4917
Income share of top quintile to bottom quintile13.5714.1314.4714.13

Columns 3 and 4 estimate the impact on income distribution of drastic modifications of the current tax system. The modifications being considered are an increase in the value-added tax—from 18 percent to 25 percent—and a 20 percent flat tax for the present progressive income tax. As can be seen in Table 7.5, the redistributive effect of these important tax changes on the final distribution is quite small.

The main reason for the limited redistributive potential of tax policy in Chile is the current tax structure, which is heavily concentrated on the regressive VAT and other indirect taxes, and the fact that taxes represent only a small fraction of GDP.17 Under these conditions, even large tax changes would have little impact on the income distribution, as De Gregorio and Landerretche (1997) show: income tax revenue in Chile is around 4 percent of GDP; if a radical tax reform were introduced—raising direct taxes by 50 percent from 4 to 6 percent of GDP—the share in total income of the poorest 20 percent would barely rise from 4.5 percent to 4.6 percent.

Despite the lack of explicit redistributive goals in the formulation of Chile’s tax policies, however, the tax reforms undertaken during the 1990s have not only generated more resources for social spending but also tilted tax income moderately toward a more favorable composition, from an equity perspective. Since 1990, and after five relevant adjustments in tax policy, revenue from income tax, which is by far the most progressive tax in Chile, has increased its share in overall tax revenue from 18 to 23 percent (Table 7.6).

Table 7.6.Chile: Structure of Tax Revenue
Revenue (As percent of GDP)Share (As percent)Revenue (As percent of GDP)Share (As percent)
Income tax2.6184.023
Excise taxes11.9132.112
Import taxes2.4171.911
Net tax revenue8.210017.6100
Source: Estadísticas de las Finanzas Públicas, 1988–97, Ministry of Finance.

On tobacco and fuels.

Includes stamp duties, inheritance tax, and lottery tax.

Source: Estadísticas de las Finanzas Públicas, 1988–97, Ministry of Finance.

On tobacco and fuels.

Includes stamp duties, inheritance tax, and lottery tax.

Two efficient and equitable ways to generate fiscal resources for targeted social spending are to intensify the fight against tax evasion and to eliminate distortionary tax exemptions and privileges.18 In Chile, the rate of VAT tax evasion has been reduced from 23.9 percent to 18.2 percent since 1990 by improving enforcement of taxes. Even in countries with low levels of evasion, the resources allocated to combating evasion promise high returns in terms of increasing tax revenue. A recent study by Engel, Galetovic, and Raddatz (1997b) shows that in Chile every dollar invested in tax administration resulted in an average increase of US$60 in tax income.

Enhancing equity through social expenditure

In principle, social expenditure is a more appropriate way to affect income distribution than taxation. Policymakers can influence social expenditure’s impact on equity through (1) the type and level of social spending—the amount of resources used and how they are distributed, for example, for health care and education; (2) the targeting of expenditure to certain beneficiary groups; and (3) the efficiency of services delivered.

In general, two types of equity-enhancing social expenditure affect inequality. The first is the provision of “essential” goods and services to all or part of the population, such as basic health care, sewerage, running water, basic housing, and survivor pensions. This kind of spending can enhance equity in terms of current consumption possibilities if properly targeted.

A second type of social expenditure affects inequality—in any of the three dimensions mentioned above—more dynamically, in the sense of investment, rather than ensuring a minimum threshold of welfare. A good example of this kind of spending is secondary and tertiary education.

As mentioned above, fiscal expenditure has indirect effects on income distribution and economic growth through its contribution to macroeconomic stability. However, fiscal expenditure also affects income, welfare, and opportunities directly. This is most obvious in the case of cash grants and transfers, which complement the recipient’s total income package. Likewise, social policies that provide goods and services (e.g., public health care, free education, and basic infrastructure investment) will supplement the consumption possibilities of those benefiting from such assistance. Therefore, carefully targeted social spending can help equalize consumption.

Guidelines for financing social expenditure

A suitable level of investment in health care, education, or housing requires considerable resources. Although in practice these resources come from all sectors of the economy, the public sector generally bears most of the burden; however, this does not imply that the private sector cannot contribute its own resources and independently offer these services. On the contrary, private sector spending should be treated as a welcome complement to public social expenditure.

Public policies on the financing of social expenditure must meet certain basic criteria if they are to be efficient and sustainable.

  • Ensure sustainability. Social expenditure can be broken down into operating expenses and investment expenses. Although investment can be financed by extraordinary funding, such as income from the sale of assets or borrowing, operating expenses must have stable, sustainable financing, to ensure the orderly development of the activities involved. The lack of a solid base for financing may create expectations of macroeconomic instability, with negative consequences for the country’s economic progress.
  • Target to reduce financing requirements. Social policy instruments are often poorly targeted and therefore increase the financing requirements of the government. Improving the targeting of expenditure should help generate more resources for the poor. It may be, however, difficult to cut poorly targeted transfers because of opposition from the beneficiaries. If this happens, ensuring targeting of incremental resources can lead, although more gradually, toward a more equitable outcome.
  • Mobilize private contributions. The public sector must take advantage of people’s willingness and capacity to contribute to the cost of social services, to free public funds for other uses. For example, because of the derived private benefits, creating a scheme of shared financing for secondary or higher education may be justified. Many individuals cannot contribute to education because they have no access to long-term loans with appropriate grace periods. The state must therefore take compensatory steps, such as establishing guaranteed funding or facilitating long-term refinancing for financial intermediaries.
  • Establish favorable regulations. Private social contributions reduce the need for public financing. Regulations should facilitate private contributions to such services as health care, education, and housing. In some countries, it may also be necessary to establish financial market regulations that stimulate the creation of saving and insurance instruments for higher education, catastrophic illness, or housing. Subsidies and tax breaks may also be needed to stimulate private contributions to education, provided they lead to an increase in marginal resources and not merely transfer revenue to the private sector at the cost of reduced fiscal savings.

Chile’s social expenditure policy in the 1990s

In Chile, total social expenditure has risen 47 percent in per capita terms since 1989. As discussed above (see Table 7.4), this increase has been the result of both economic growth and, to a lesser degree, expenditure’s reallocation. The evolution and composition of social expenditure is shown in Figure 7.6; this figure reveals the increasing importance of health care and education in total social expenditure, and the decreasing importance of pensions, due to the introduction of a private, fully funded pension system in Chile in the early 1980s.

Figure 7.6.Main Items of Social Expenditure

(Real per capita expenditure, 1990 = 100)

Source: Chilean Ministry of Finance.

However, the decreasing share of public pensions does not accurately reflect government action in this area. In fact, welfare pensions as well as minimum pensions have increased considerably more than the higher-level pensions since 1989 (see Figure 7.7). This policy has helped reduce the usual regressiveness of public pay-as-you-go pension schemes. In this context, it is interesting to note that the distributional impact of Chile’s pension system will dramatically improve once the transition to a private, fully funded system is completed. Currently, the state is serving as a bridge between the old and new systems. Under the new system, full public spending on pensions will become increasingly progressive, because the government will then finance welfare pensions and complementary resources for only those pensioners whose savings are not high enough to ensure by themselves the state-guaranteed minimum pension.

Figure 7.7.Chile: Growth of Public Pensions, March 1990–March 1998

(In percent)

Source: Chilean Ministry of Finance.

Regarding government spending on health care and housing, Chile seeks to provide all people with access to a minimum level of these services. The state does not provide these services on demand, but targets this expenditure. This targeting has improved the distribution of consumption.

The CASEN survey presents Chile’s income distribution for 1992–96 (see Table 7.7) but in a slightly different way than previously discussed: it splits the standard distribution of monetary income into self-earned income and cash transfers, and includes the monetary incidence of social expenditure. The resulting distribution of income is therefore an approximation of the distribution of consumption. Several interesting conclusions can be drawn from the data presented.

Table 7.7.Chile: Incidence of Social Spending on Income Distribution
Quintile of Self-Earned Income1
Income Category199219941996199219941996199219941996199219941996199219941996
Percent of total income per category
Self-earned income14.
Cash transfers and social programs35.035.243.426.125.525.420.018.319.313.
Cash transfers33.031.640.123.722.526.718.116.919.913.413.910.31.1.715.13.1
Social programs35.636.143.926.826.225.220.518.719.313.
Total income6.
Percent of total income per quintile
Self-earned income162.959.657.081.380.382.589.389.390.195.194.496.399.399.199.9
Cash transfers and social programs37.140.443.018.719.717.510.710.
Cash transfers7.
Social programs29.433.137.714.916.315.
Total income100.0100.0100.0100.0100.0100.0100.0100.0100.0100.0100.0100.0100.0100.0100.0
Source: Ministry of Finance.

Self-earned income consists of labor income, capital returns, pensions, and private sector transfers.

Source: Ministry of Finance.

Self-earned income consists of labor income, capital returns, pensions, and private sector transfers.

In Table 7.7, the second row shows that in 1996 more than 43 percent of cash transfers were received by the poorest quintile of Chile’s population. Between 1992 and 1996, the fraction of cash transfers received rose for the first three quintiles and fell for the highest two quintiles. Hence, the government succeeded in improving the targeting of these benefits. However, although this contributed to a more even distribution of income, these transfers accounted for only a small fraction of total income of the lower quintiles (as seen in the lower part of Table 7.7), and therefore had only a limited effect on the overall distribution of income.

Second, spending on health care and education had a much more pronounced effect on income distribution than cash transfers. In 1996, close to 75 percent of health care spending and 65 percent of all education spending were directed toward the poorest 40 percent of the population. These programs represented a much larger share of total consumption and therefore have a significant progressive effect on overall income distribution. As with cash transfers, considerable progress in targeting health care and education expenditure was achieved between 1992 and 1996.

Third, the combination of in-kind social spending provided through health care and education and cash transfers to the poorest households has clearly improved income distribution in Chile. Although the wealthiest 20 percent of the population receive over 15 times more “self-generated” income (labor income, capital returns, pensions, and private sector transfers) than the poorest 20 percent, this ratio falls to 8.5 when cash transfers and health care and education spending is included. Even so, overall income distribution remains very unequal in absolute terms: the richest 20 percent of the population still captures almost 54 percent of total income.

Fourth, the positive redistributive effect of social spending and cash transfers during 1992–96 compensated for a mild deterioration in the distribution of self-earned income, leaving overall income distribution virtually unchanged. This is important for policymaking, because it refutes the superficial conclusion—derived from the lack of progress in overall income distribution—that equity-enhancing public spending is ineffective.

The analysis of the distributional effect of public expenditure policy up to this point accounts for only the immediate effects of social spending on income distribution. In the long term, increased spending on education in the lower quintiles will lead to a more equitable distribution of human capital, which in turn will have a positive impact on the distribution of self-generated or autonomous income. In fact, as stated in the section “An Overview of Poverty and Income Distribution,” most of Chile’s unequal income distribution can be attributed to differences in labor income, which depend heavily on human capital. Therefore, a more equitable distribution of human capital is probably the most powerful long-term instrument for improving equity in Chile. However, a word of caution regarding the timing of these effects: no matter how efficient new and improved educational programs are, it will take considerable time for these changes to affect the labor market and income distribution.

In this section, the evidence presented on the incidence of public expenditure refers to the level of spending in the different social programs. However, these programs’ impact on equity will depend not only on the amount of resources spent but also on the efficiency with which resources are converted into quality goods and services and distributed to the respective social groups.19 Modernizing public administration will free resources to finance more equity-enhancing social programs. In some cases, it may be more practical—though not as attractive politically—to restructure existing programs than to increase spending.

The public health care system in Chile illustrates these problems. Although resources have nearly doubled in per capita terms over the past seven years, improvements in quality and coverage of services have lagged. Thus, although public health expenditure has a large (and increasing) role to play in improving equity, this role has been limited because of the lack of administrative improvements in the sector.

These experiences reinforced the government’s view that the public agencies and their services had to be modernized if the quality of public services and social spending was to improve. To this end, a joint ministerial committee—comprising officials from the Ministry of Finance, Ministry of the Interior, and the General Secretariat of the Presidency—was created to advise on and coordinate the modernization of public agencies.20

Box 7.1.Chilean School System at a Glance

Chile’s school system consists of eight grades of compulsory primary education (beginning at age 6) and four grades of noncompulsory secondary education. Preschool education up to age 6 is noncompulsory.

In terms of proprietorship and financing, three broad types of schools can be distinguished: municipal (public) education (which accounted for 57 percent of total enrollment in 1996); private but publicly subsidized education (33 percent of total enrollment); and private nonsubsidized education, which is fully financed by the children’s parents (8 percent of total enrollment).

Total enrollment in primary education was 2.18 million in 1996, or 95 percent of the relevant age group. Secondary school enrollment was 709,000, or 80 percent of the relevant age group. Preschool enrollment was 24 percent.

The public provision of funds is based on a per pupil grant system. Private grant-receiving schools and municipal secondary schools are allowed to charge families a fee, as a form of cofinancing.

This committee has sponsored the following activities:

  • Formulation of a modernization plan for the public sector (see Chile, Ministry of the Presidency, 1997);
  • Development of a methodology to measure the efficiency of public service (see Chile, Ministry of Finance, 1995);
  • Development of instruments to link pay to individual and institutional performance; and
  • Signing of modernization agreements between the central government and public agencies, which include commitments to fixed performance levels.

In 1997, 64 public services committed themselves to performance goals for 263 efficiency indicators. Fifty-four percent of the goals were completely met and 80 percent of the goals reported an achievement of more than 75 percent.

With a view to improving the efficiency of public spending, in 1997, the Chilean government hired external consultants to evaluate the quality and effectiveness of some spending programs. These evaluations are to provide important information for budget allocation. That year, 20 programs were selected by the government and the Congress for evaluation. The results were encouraging: the consultants endorsed all the programs but made suggestions on improving their quality. Because of this positive experience, in 1998 the number of programs selected for evaluation was doubled.

Enhancing equity through education policy

The Chilean educational system (Box 7.1) has two main characteristics that negatively affect the distribution of opportunities:

  • Poor quality of publicly financed education, reflected in very unsatisfactory levels of learning and high failure and repetition rates. According to the 1997 national SIMCE test, municipal schools, which cover most of the poor population, had achievement levels of only 59.5 percent in mathematics and 62.2 percent in Spanish. In contrast, private schools had levels of 80 percent in both subjects (see Table 7.8). Pupils in private schools need, on average, 13.2 years to complete secondary school (grade 12), and 89 percent of enrolled pupils graduate. In municipal schools, pupils need, on average, 15.8 years to complete secondary school, and only 53 percent graduate (see Table 7.9).
  • Large differences in preschool, secondary, and tertiary school enrollment rates between high- and low-income families. Although the primary education enrollment rate is above 96 percent for all income groups, the secondary school enrollment rate drops from over 95 percent for the upper two income quintiles to 75 percent in the lowest quintile. For tertiary education, the difference is even greater: almost 60 percent in the highest income quintile versus only 8.5 percent in the lowest income quintile (see Table 7.10).
Table 7.8.Chile: Results of National Test of Educational Achievement (SIMCE) by Type of School
MunicipalSubsidized Private SchoolsNonsubsidized Private Schools
19891997Change (In percentage points)19891997Change (In percentage points)19891997Change (In percentage points)
Source: Ministry of Education.Note: SIMCE is an annual standardized test that measures the degree of academic goals achieved by the fourth grade of elementary school.
Source: Ministry of Education.Note: SIMCE is an annual standardized test that measures the degree of academic goals achieved by the fourth grade of elementary school.
Table 7.9.Chile: Efficiency Indicators by Type of School, 1996
Rate of Graduation (In percent)Rate of Retention (In percent)Average Time to Graduate (In years)
Private nonsubsidized schools
Private subsidized schools
Municipal schools
Source: Compendio de Informatión Estadistica, 1996, Ministry of Education.Note: The indicators have been calculated on the cohort 1986–96 in primary school, and 1991–96 in secondary school.
Source: Compendio de Informatión Estadistica, 1996, Ministry of Education.Note: The indicators have been calculated on the cohort 1986–96 in primary school, and 1991–96 in secondary school.
Table 7.10.Chile: School Enrollment by Income Group(In percent)
Income Quintile
Higher education8.515.021.534.759.7
Source: MIDEPLAN, Departamento de Planificación y Estudios Sociales.
Source: MIDEPLAN, Departamento de Planificación y Estudios Sociales.

From an equity point of view, the poor quality of publicly financed education and lower enrollment rates of the poor in Chile are particularly alarming because studies show that in this country, educational differences are the most important cause of income disparities. Contreras (1998) finds that these differences accounted for between 30 percent and 40 percent of the wage disparity in Santiago in 1958–92. A study by the World Bank (1997) concludes that education “explains a much greater share of overall wage inequality than any other household attribute” (p. 13). Finally, as mentioned above, Beyer (1997) concludes that the main source of income inequality in Chile is labor income differences, which depend heavily on the individual stock of human capital.

Low-income earners are trapped in a situation of almost no vertical mobility. Poor households cannot afford private schooling and have low human capital accumulation because of the public school deficiencies. This in turn creates low labor productivity and therefore reduced labor income, which prevents the creation of resources necessary for access to private schooling for future generations.

The Chilean government has recognized that improving access to education is crucial to achieving a more equitable distribution of opportunities, and has thus made it a policy priority. Since 1990, public spending per capita has grown faster—90 percent in real terms—than any other social spending. Several studies have identified reasons for the low quality of public education in Chile: a school year of only 180 days, widespread deficiencies in curricula, and outdated methodologies, such as the domination of traditional lecturing.21 In response, the government has embarked on a historic project to reform public education. Now under way, it consists of four strategies:

  • Increase teaching hours. Public schools in Chile are switching from a two-or three-shift teaching day to a full school day, in order to increase substantially the number of teaching hours a year. The most visible feature of the reform package, this change is the most significant in terms of resource needs and qualitative implications.
  • Improve the quality and equity of schooling. Special programs to improve quality have been initiated, some of them targeted to the poorest schools. Significant efforts are being made to update school equipment and facilities. All high schools and 50 percent of the grade schools are being supplied with computers and connections to the new nationwide education network and Internet; in-class libraries, audiovisual equipment, and furniture are also being provided to many schools.
  • Modernize the educational content. Updated programs of study are being implemented in primary and secondary schools; for example, the recently launched Montegrande project has innovative educational programs that will benefit more than 40,000 students from so-called advanced schools (Escuelas de Anticipación).
  • Reinforce the teaching profession. Working conditions of public school teachers are being significantly improved, and teachers are being retrained. Scholarships for study abroad and prizes for teaching excellence are now awarded, and teachers have been given a substantial increase in real wages. Resources are being spent to improve the academic environment and to reduce the cost of teachers’ studies in an attempt to improve initial training of teachers.

The low secondary and tertiary school enrollment for lower income groups—the second most important deficiency of Chilean education—is the result of several factors. First, the high opportunity cost of attending poor-quality public schools deters secondary school enrollment. The calculations shown in Appendix I indicate that attending the publicly financed schooling system is costly, from both a private and a social perspective. Second, the poor quality of secondary public education leads to low achievement on the national aptitude test (Prueba de Aptitud), thereby restricting access to public universities. These factors plus the liquidity constraints and higher discount rates faced by lower income families help explain the low secondary and tertiary school enrollment of the poor.

The reform now under way should improve the quality of the publicly financed primary and secondary education, which in turn should increase the educational coverage of the poor. Moreover, the government has greatly increased public funding for student grants and tuition credit, to reduce the opportunity costs and liquidity constraints of the poor. Tables 7.11 and 7.12 show that public spending for grants and credits has increased almost 67 percent since 1990. Presidential grants, which cover the living expenses of poor secondary and higher education scholars, have increased even more—132 percent in real spending—which has facilitated a rise in the individual grant to a more realistic amount and the expansion of enrollment from 8,500 to about 12,500 beneficiaries.

Table 7.11.Chile: Evolution of Public Spending on University Credit and Grants(In millions of 1998 pesos)
Tuition CreditTuition GrantsPresidential Grants1Total Spending
Source: Budget Office, Ministry of Finance.

Grants that cover living expenses: beneficiaries are scholars from secondary and tertiary education.

Source: Budget Office, Ministry of Finance.

Grants that cover living expenses: beneficiaries are scholars from secondary and tertiary education.

Table 7.12.Chile: Estimated Number of Presidential Grant Beneficiaries(In millions of 1998 pesos)
Type of Education199319941995199619971998
Source: Budget Office, Ministry of Finance.
Source: Budget Office, Ministry of Finance.

Over the next several years, this education reform will require financing efforts of unprecedented size. The extension of the school day alone implies an annual cost of 0.6 percent of GDP. Given the magnitude of this reform, the government has insisted that its execution be accompanied by a system of permanent financing. To accomplish this, a consensus was reached to maintain indefinitely an additional percentage point of the value-added tax, which, according to previous tax agreements, was to have been lowered in 1998.

The government has also introduced a number of measures aimed at increasing and diversifying the ways to finance higher education. Last year, a new private credit facility was established, to complement soft state credit and grants in the financing of undergraduate university studies. This instrument is now available to students, especially to those from middle-class families who have no access to soft public university credit as a long-term financing alternative for their enrollment and tuition expenses.22

Also, now in the final stage of development is a massive savings instrument for university education. This would be an easily accessed savings account; it would motivate Chilean families to contribute voluntarily to the financing of their children’s university education.

Regulatory Policy

Market imperfections can be a source of inequity. Consider, for example, the financial market. If imperfections, such as incomplete information, exist, financial markets may fail to provide consumption-smoothing opportunities or educational access to lower income groups, or some agents may incur socially undesirable risks unless proper regulation is in place.

A well-functioning labor market is crucial to improving equity, because labor income represents a large part of total income, especially in the lower quintiles. As an extensive discussion of equity-enhancing regulatory issues is beyond the scope of this paper, we will focus on some aspects of labor market policy.

Enhancing equity through labor market regulation

Labor market regulation can influence equity through its effects on the overall employment level and/or its effects on wages. In a simplistic world of competitive markets, factor mobility, and full information, active labor market policies can distort and thus negatively affect employment, overall efficiency, and, eventually, the long-term distribution of income. However, in a more realistic setting, especially in developing countries where workers are not fully informed about job opportunities and where searching for employment is not an option for most of the population because of liquidity constraints, appropriate regulations may actually enhance equity and the overall efficiency of the labor market.

Minimum wage policy is a prominent instrument of active labor market policies in most countries. Sometimes a minimum wage policy is viewed as an instrument to guarantee a minimum subsistence level and thus help reduce poverty. From an economic point of view, however, it is more appropriate to think about potential market failures as the main rationale for minimum wage regulations. In an unregulated market, low-productivity workers may be paid less than their marginal productivity, because information costs and precarious income situations severely restrict job searches and wage negotiations. But, under some circumstances, a minimum wage policy can improve income distribution, if it is enforced and if wage levels are not pushed above worker productivity.23

Unemployment protection is also important in labor market regulation. Low-income workers are particularly vulnerable to layoffs, because they are usually the least skilled. Low-income workers also have the most difficulty monitoring the job market for better work opportunities, because of their liquidity constraints. Thus, a case can be made for unemployment protection on equity grounds.

Chile’s minimum wage policy

Minimum wages in Chile grew at an average annual rate of 5.4 percent in real terms during 1990–96—more than the average wage of the economy, which grew at 4 percent in real terms, and the average labor productivity, which increased 3.5 percent a year. Despite strong growth in the minimum wage, unemployment and the extent of informal labor markets decreased during the same period.24

The Chilean government sends a bill to the Congress each year proposing a nominal increase in the minimum wage.25 These proposals are based on three factors: expected inflation, estimated relevant productivity growth, and a catch-up component to compensate for the more than 35 percent loss of minimum wage purchasing power during the 1980s.

The strong increase in the minimum wage during 1990–96 coincides with increased wage income equality as stated in the section “Considerations Surrounding Equity and Economic Policy.” However, there is no clear evidence that the former had a significant impact on the latter. This is not surprising because the minimum wage is earned by only about 5 percent of the labor force and because minimum wage coverage actually decreased during that period. Nevertheless, six policy lessons can be found in the Chilean experience:

  • Tie increases to objective factors. Increases in the minimum wage must be tied to such factors as productivity growth; however, the appropriate productivity measure may not be easily observed. For example, Anriquez, Cowan, and De Gregorio (1997) argue that the relevant rate of productivity growth is not necessarily the economywide average but that of labor with minimum skills. For the past six years in Chile, the latter has probably amounted only to half of the average productivity growth.
  • Use expected inflation to adjust. Using expected rather than past inflation to adjust the nominal minimum wage contributes to the deindexation of the economy and therefore helps to reduce inflation, which in itself is beneficial from an equity point of view, as mentioned above. It also helps to ensure that nominal increases translate into more purchasing power for the minimum wage.
  • Promote flexibility. As seen in Chile, under specific labor market conditions, minimum wages may be increased above the rate of productivity growth and expected inflation without major harm to employment and/or inflation; however, this result cannot be generalized.26 Even in Chile, past experience does not ensure that maintaining the pace of minimum wage increases will not have detrimental effects on employment, inflation, poverty, and the distribution of labor income. For this reason, minimum wage adjustments must be evaluated on a case-by-case basis.
  • Differentiate among minimum wage levels. The minimum wage should be regarded as an entry-level wage rather than a guaranteed subsistence minimum. Because entry-level productivity may vary according to a worker’s attributes, a differentiation of minimum wage levels may be appropriate. In Chile, for example, a reduced minimum wage applies to workers under the legal age of 18, to account for the lower productivity of youngsters. This also helps prevent early school leaving, which, from an equity point of view, is probably the most undesirable side effect of excessively high minimum wages.
  • Target complementary policies. Complementary labor policies can be targeted toward minimum wage earners, to increase their productivity levels and help promote their takeoff from the bottom of the wage scale. For example, a special training program for minimum wage earners was recently started in Chile, and is scheduled to cover 20,000 workers in three years.
  • Enforce minimum wage laws. Important gains can be made in the income levels of low-productivity workers by strengthening the enforcement of minimum wage laws, and by making a strong effort to attract micro and small enterprises to the formal economy.

Chile’s proposed worker protection system

Most countries have developed some sort of insurance system to protect unemployed workers. In Chile, workers are protected against involuntary dismissal by a monetary compensation system that is based on job seniority. When dismissed for business reasons, the worker must be given one month’s salary for every year employed by the company, up to a total of 11 months’ salary.

Unemployed workers who have been dismissed involuntarily also receive a public transfer called Subsidio de Cesantía (unemployment subsidy) for up to 12 months. To qualify for this subsidy, which is equivalent to 11–22 percent of the minimum wage,27 a worker must have made social security (pension fund) contributions to the private or public system for at least 12 months.

Chile’s worker protection system has a number of shortcomings:

  • Low coverage—only 26 percent of the eligible beneficiaries use the unemployment subsidy, partly because of its low value;
  • Worker uncertainty—workers often must go to court to oblige firms to pay, and workers lose their benefits if the employer files for bankruptcy;
  • Job rigidity—workers expecting large compensation resist job turnover; and
  • Negative workplace effects—workers with large accumulated compensation may force their own dismissal, or employers may try to force them to quit voluntarily.

These shortcomings weaken the worker protection system and the job market efficiency.

To increase worker protection and mobility, the Chilean government has developed a novel unemployed-worker protection system, called PROTRAC (Sistema de Protección at Trabajador Cesante). PROTRAC has some innovative characteristics and sidesteps the problems associated with most traditional unemployment insurance schemes. In brief, PROTRAC is based on the Individual Savings Account for Unemployment (ISA), which is property of the worker and is privately administered. The ISA is fed by monthly contributions from both worker and employer.28 If involuntarily dismissed, the worker can withdraw the funds accumulated in the ISA. The employer will be obliged to make a matching payment, which covers the difference between the funds accumulated in the ISA (including interest) and the equivalent of one month’s salary for each year of company employment, to a total of 11 months.

If the worker voluntarily quits his or her employment, dies, or becomes an invalid, he or she (or the surviving next of kin) is entitled to use the funds accumulated in the ISA to cover expenses during unemployment but receives no additional payment from the employer.

The proposed unemployed-worker protection system would resolve most of the problems inherent in the present system, and offers some additional benefits. Worker mobility would be greatly improved because workers would be protected against all types of unemployment, even that caused by voluntary resignation. The system would also increase worker security in the event of involuntary dismissal—the present system does not—because most of the employers’ obligations would be accumulated in advance in the ISA, which would be property of the worker. The working climate would likely also improve, as perverse behavioral incentives for workers and employers lessen.

In addition, the new system would probably increase overall savings, as it would have elements of forced savings for both employers and workers. From a fiscal point of view, PROTRAC would be neutral since it would not involve any contribution from the state.

Political Economy of Equity-Enhancing Policies

Equity-driven policies, despite their being undeniably beneficial, can easily get trapped in political and social conflicts that overshadow their rationale and merit and hinder implementation. This is likely to occur for four main reasons:

  • First, these policies mainly benefit people who, because of their poverty, are likely to be poorly organized, heterogeneous, and dispersed and, therefore, wield less political power than those who provide essential services, such as schoolteachers, doctors, and building contractors.
  • Second, improving health care, education, and housing often requires major reforms in their delivery systems, to ensure that scarce public funds are used effectively. Reform opponents may protest more loudly than the potential reform beneficiaries, thereby shaping the public debate on the subject.
  • Third, as the benefits of most of these policies will not be seen in the short run, policymakers must be persistent and consistent. This lag in the effect of equity-driven policies, however, will give opponents room to claim the failure of the policies and lobby for reversal.
  • Fourth, in developing countries, the ministries in charge of social policies are often disorganized, highly bureaucratic, overstaffed, and lacking in the technical expertise to deal with all the political, economic, and institutional issues involved in a reform. An indicator of the weakness of Latin America’s social ministries is the average tenure of ministers: about one year.

Given these facts, how can policymakers foster social reform? Some of the strategies that have been successful in Chile include the following:

  • Strive for objectivity. Equity-enhancing measures will be more successful if emotions and ideological issues are resolved early on. To this end, consultations, studies, and investigations with apolitical experts or a board of widely respected individuals of a different political couleur is helpful. In Chile, the Brunner Commission made a detailed assessment of the quality of the educational system and proposals for modernization.
  • Create public concern over priority issues. Once the realities, based on objective indicators and evidence, are revealed, social and political actors should be informed and public opinion mobilized. New evidence can often turn an issue into a public concern, then a political priority. In Chile, studies on domestic violence focused public interest on that issue, creating pressure to modernize existing legislation. And public awareness of the social cost of pollution in Santiago has made urban pollution a top priority on the political agenda.
  • Build political consensus. Growing public concern about a particular social issue usually attracts the interest of politicians, who then become more open to discussing solutions. Agreements that cross party lines are more likely to be reached in such an environment. These broad agreements are particularly important for long-term policies because they ensure sustainability in the event of political change; for example, a path-breaking judicial reform was recently approved by all political parties in Chile, and is scheduled to be gradually implemented well into the next century.
  • Build and preserve the loyalty of crucial social and political actors. Consensus may be difficult to achieve, even among political actors who usually support government action. The viability of controversial equity-enhancing policies can be improved if loyalty links are established in advance between the social and political actors. These links create pressure to share, at least partly, political costs.
  • Build checks and balances. If the potential beneficiaries are less well organized than the special interest groups, policymakers should help organize the former. If organization is difficult, then surveys should be developed to help beneficiaries voice their concerns and needs. Because urban homeless families (allegados) in Chile had traditionally been difficult to target with conventional social policy, government promoted their organizing into local committees to represent themselves. These committees soon developed into forceful organizations that now interact directly with government institutions, helping to improve delivery and targeting of social services.
  • Ensure consistency of partial steps. Political and/or economic factors may require a gradual implementation of reforms. If so, consistency in these programs, initiatives, and measures is crucial to providing a sense of direction for the whole reform effort. Even if progress is slow, the involved parties—including opponents—will know where the effort is heading. This will ensure that the reforms’ objectives are pursued.
  • Improve measurement of results. Individual perceptions of progress may differ from objective results. Therefore, it is important to develop objective indicators to measure—and control—the performance of equity-enhancing policies. Chile’s national educational achievement test (SIMCE), for example, has revealed a catching-up of publicly financed schools since 1989 (see Table 7.8).
  • Improve implementation. In many cases, social programs are not carefully implemented. The widespread belief seems to be that if a need exists, people will benefit from any program that addresses it. But social program beneficiaries should be treated more as customers: their needs should be well researched, the quality of existing programs tested, the matching of programs with needs assessed, and the programs adjusted, if necessary.
  • Strengthen and build institutions. Effective social programs require effective management. Weak institutions can destroy the best-designed programs. External management units can help temporarily, but the expertise, management skills, capabilities, and knowledge must be transferred to permanent institutions.


Equity-enhancing policies undertaken in Chile during the 1990s have been based on several key elements. Prudent fiscal and macroeconomic management has been understood as a means of improving equity. Past experience in Chile has shown that in periods of macroeconomic instability, fiscal policy revolves around issues of stabilization; in this context, equity considerations are often put in second or third place. Macroeconomic stability therefore fosters economic growth, which increases sustainable fiscal resources available for redistributive public expenditure policy. Macroeconomic stability also has positive direct effects through increased job stability and creation and lower inflation, which tend to benefit low-income groups more and therefore improve equity.

The Chilean experience shows that fiscal policy through public expenditure can be equity enhancing. During 1992–96, additional resources, together with steps taken to improve the targeting and efficiency of social spending, had a positive impact on equity—in fact, the positive redistributive effect of social spending and cash transfers compensated for a mild deterioration in the distribution of self-earned income, leaving the overall income distribution virtually unchanged.

However, most social spending, particularly transfer policy, provides only a short-term instrument for improving equity. In the long term, access to wealth-generating factors for low-income groups must be improved in order to achieve a more equitable society. Because studies show that education is the single most important factor in explaining income disparities in Chile, the government has focused its long-term equity strategy on improving the quality of education. To achieve this goal, a radical reform of the public education system is currently under way, aimed at enhancing the quality and access to better education for the poor. Therefore, significant changes in the distribution of income and improved vertical mobility can be expected once these reforms fully develop and equality in educational opportunities is achieved.

Appendix I: Assessing the Impact of Low-Quality Education

Educational quality varies greatly depending on the type of school, as shown in Table 7.13. In general, nonsubsidized private schools show the highest educational achievement, followed by publicly subsidized private schools. Municipal education, which is fully paid for by the state, shows the lowest achievement as seen in the SIMCE scores and in the time required for graduation. Public school deficiencies, such as large class size and poor quality of instruction, lead to higher repetition rates because students do not meet the minimum standards.

Table 7.13.Chile: Quality Indicators by Type of School, 1994
MunicipalPrivateNonsubsidized Private
Students enrolled in first year of primary school180,289101,35422,596
Average rate of graduation (in percent)
Primary school70.8472.9985.16
Secondary school57.7375.0388.82
Average time to graduate (in years)
Primary (8 grades)10.259.868.79
Primary and secondary (12 grades)16.0814.8513.19
SIMCE scores 1995 (in percent)
Public subsidy per student (in US$ per year)
Primary school324.10324.100
Secondary school385.80385.800
Source: Ministry of Education.
Source: Ministry of Education.

Because of the lower quality of municipal and subsidized private schools, students from low-income families who cannot afford a nonsubsidized private school face a double opportunity cost: first, lower labor productivity and thus lower wages, because of lower SIMCE scores; and second, forgone income, due to the longer time needed to complete their secondary education.

Although the former cost is more difficult to measure, the latter can be easily calculated using the information in the table above. Using as a benchmark the average time to graduate from a nonsubsidized private school, the individual opportunity cost can be approximated as follows:29


Om = average opportunity cost of student graduated from municipal secondary school

Tm = average time to graduate from municipal secondary school

Tpn = average time to graduate from nonsubsidized private secondary school

w = salary forgone.

If we assume the minimum wage as the relevant wage rate, then, using 1994 data,30 a graduate of a municipal secondary school bears an opportunity cost of US$5,020, equivalent to almost 22 months of the average household income of the poorest 40 percent of the population.31 The opportunity cost for students enrolled in a publicly subsidized private school is US$2,883, or 12.4 average months of the household incomes.

However, there is an additional social burden that results from the poor quality of public and publicly financed education. The longer time needed for graduation means a higher cost in terms of infrastructural needs and state subsidies. This cost can be calculated by multiplying the delay in graduation by the annual value of the per student subsidy:


ESm = excessive subsidies paid because of graduation delay of municipal school students

Tpm = average time to graduate from municipal primary school

Tppn = average time to graduate from nonsubsidized private primary school

Sb = public subsidy for primary school education

Tsm = average time to graduate from municipal secondary school

Tspn = average time to graduate from nonsubsidized private secondary school

Ss = public subsidy for secondary school education.

Using this methodology, the additional public cost per municipal school student is, on average, US$1,024.90, and the cost per student in a publicly subsidized private school is US$574.40.

We can also simulate how much excess subsidy the government will pay due to excessive duration of schooling for an entire cohort of students. This cost is given by


TESm = total excessive subsidies paid because of graduation delay of municipal school cohort

epm = rate of success in primary municipal schooling (1-dropout rate)

esm = rate of success in secondary municipal schooling (1-dropout rate)

nm = number of students enrolled in first year of municipal primary education.

This exercise indicates that the government will pay an excess subsidy of almost US$118 million for the cohort of students enrolled in 1994 in the first year of municipal primary education, and US$43 million for private publicly subsidized education. The sum of the two is equivalent to 10 percent of Chile’s total annual central government spending on education in 1994.


    AnriquezG.K.Cowan and J.De Gregorio1997“Poverty and Macroeconomic Policies: Chile 1987–94,”Documentos de Trabajo No. 27 Department of Industrial Engineering Universidad de Chile (unpublished; Santiago: Department of Industrial Engineering, Universidad de Chile).

    • Search Google Scholar
    • Export Citation

    AtkinsonA.B.1997“Bringing Income Distribution in from the Cold,”Economic JournalVol. 107 (March) pp. 297321.

    BénabouRonald1996“Inequality and Growth,”NBER Working Paper Series No. 1450 (Cambridge: National Bureau of Economic Research).

    • Search Google Scholar
    • Export Citation

    BeyerH.1997“Distribución del Ingreso: Antecedentes para la Discusión,”Estudios PúblicosNo. 65.

    BonifázR. and DavidBravo1998“Mercado del Trabajo e Institucionalidad Laboral en Chile durante los Gobiernos de la Concertación,” in Construyendo Opciones-Propuestas Económicas y Sociales Para el Cambio de Sighed. by RenéCortázar and JoaquínVial (Santiago: CIEPLAN, Dolmen Ediciones) pp. 31760.

    • Search Google Scholar
    • Export Citation

    BravoDavid and D.Contreras1998“Comentarios sobre el reciente reajuste del salario mínimo,”Serie Opinión No. 4 (Santiago: Departamento de Economía, Universidad de Chile).

    • Search Google Scholar
    • Export Citation

    BravoDavid and JoaquínVial1997“La Fijación del Salario Mínimo en Chile: Elementos para una Discusión,”Colección Estudios CIEPLAN 45 (June) pp. 11751.

    • Search Google Scholar
    • Export Citation

    BrunnerJosé and others1995“Los Desafíos de la Educación Chilena Frente al Siglo XXI,”Informe del Comité Técnico Asesor del Diálogo Nacional sobre la Modernización de la Educación Chilena Designado por S.E. el Presidente de la República (unpublished; Santiago).

    • Search Google Scholar
    • Export Citation

    BulírAles1998“Income Inequality: Does Inflation Matter?”IMF Working Paper 98/7 (Washington: International Monetary Fund).

    Chile Ministry of Finance1995Guia para la Construcción de Indicadores de Desempeño en los Servicios Públicos (Santiago).

    Chile Ministry of Finance Ministry of the Presidency1997Plan Estratégico de Modernizatión de la Gestión Pública: El Estado al Servicio de la Gente (Santiago).

    • Search Google Scholar
    • Export Citation

    ContrerasD.1995“Poverty Measures, Robustness of the Poverty Profiles, Welfare and Targeting: Evidence from Chile” (unpublished; Los AngelesUCLA).

    • Search Google Scholar
    • Export Citation

    ContrerasD.1998“Explaining Wage Inequality in Chile: Does Education Matter?” (unpublished; Santiago: Department of Economics, Universidad de Chile).

    • Search Google Scholar
    • Export Citation

    CowanK. and J.De Gregorio1996“Distribución y Pobreza en Chile: Estamos mal? Ha habido progresos? Hemos retrocedido?”Estudios PúblicosVol. 64.

    • Search Google Scholar
    • Export Citation

    CoxCristián and PabloGonzález1998“Educación: De Programas de Mejoramiento a Reforma,” in Construyendo Opciones-Propuestas Económicas y Sociales Para el Cambio de Sigloed. by RenéCortázar and JoaquínVial (Santiago: CIEPLAN, Dolmen Ediciones) pp. 23378.

    • Search Google Scholar
    • Export Citation

    De GregorioJosé and OscarLanderretche1997“Equidad, Distribución y Desarrollo Integrador,” in Construyendo Opciones-Propuestas Económicas y Sociales Para el Cambio de Sigloed. by RenéCortázar and JoaquínVial (Santiago: CIEPLAN, Dolmen Ediciones) pp. 15190.

    • Search Google Scholar
    • Export Citation

    De GregorioJ. and J.W.Lee1998“Education and Income Distribution: New Evidence from Cross-Country Data” (unpublished).

    DeiningerKlaus and LynSquire1996“A New Data Set Measuring Income Inequality,”World Bank Economic ReviewVol. 10 (September) pp. 56591.

    • Search Google Scholar
    • Export Citation

    EngelE.A.Galetovic and C.Raddatz1997a“Taxes and Income Distribution in Chile: Some Unpleasant Redistributive Arithmetic” (unpublished; Santiago: Department of Industrial Engineering, Universidad de Chile).

    • Search Google Scholar
    • Export Citation

    EngelE.A.Galetovic and C.Raddatz1997b“Efectos de la acción del servicio de impuestos internos sobre la recaudación de IVA y renta” (unpublished; Santiago: Department of Industrial Engineering, Universidad de Chile).

    • Search Google Scholar
    • Export Citation

    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

    LópezR.1996“Poverty in Chile: Analyzing the Role of Minimum Wages and Other Factors Using Regional Evidence” (unpublished;University of Maryland).

    • Search Google Scholar
    • Export Citation

    MarcelMario1997“Effectiveness of the State and Development: Lessons from the Chilean Experience,”paper presented at the seminar of the World Bank’s Economic Development Institute, “Chile: Development Lessons and Challenges,”WashingtonDecember.

    • Search Google Scholar
    • Export Citation

    MarcelMario and AndrésSolimano1994“The Distribution of Income and Economic Adjustment,” in The Chilean Economy: Policy Lessons and Challengesed. by B.BosworthR.Dornbusch and R.Labán (Washington: Brookings Institution).

    • Search Google Scholar
    • Export Citation

    MarfánManuel1998“El Financiamiento Fiscal en los Años 90,” in Construyendo Opciones-Propuestas Económicas y Sociales Para el Cambio de Sigloed. by RenéCortázar and JoaquínVial (Santiago: CIEPLAN, Dolmen Ediciones) pp. 54574.

    • Search Google Scholar
    • Export Citation

    PerottiR.1995“Growth, Income Distribution and Democracy: What the Data Say,”Columbia University Department of Economics Discussion Paper Series No. 757 (New York).

    • Search Google Scholar
    • Export Citation

    SchkolnikMariana1996“Estudio de Incidencia Presupuestaria: El Caso de Chile,”Proyecto Regional de Política Fiscal (Santiago: CEPAL).

    • Search Google Scholar
    • Export Citation

    SolimanoA.1996“Economic Growth and Social Equity: An Overview,”paper presented at the conference on “Economic Growth and Equity: International Experiences and Policies,”SantiagoJuly.

    • Search Google Scholar
    • Export Citation

    TanziVito1996“Fiscal Policy and Income Distribution,”paper presented at the conference on “Economic Growth and Equity: International Experiences and Policies,”SantiagoJuly.

    • Search Google Scholar
    • Export Citation

    United Nations Development Programme (UNDP)1997Human Development Report (New York: UNDP).

    United Nations Economic Commission for Latin América and the Caribbean1997a“Panorama Social de America Latina 1996” (Santiago: CEPAL).

    • Search Google Scholar
    • Export Citation

    United Nations Economic Commission for Latin América and the Caribbean1997b“Evolucíon Reciente de la Pobreza en Chile” (Santiago: CEPAL).

    • Search Google Scholar
    • Export Citation

    World Bank1997Chile—Poverty and Income Distribution in a High-Growth Economy: 1987–95 Report No. 16377-CH (Washington: World Bank).

    • Search Google Scholar
    • Export Citation

Note: Mr. Aninat, who was a conference participant, is the Minister of Finance, Chile; Messrs. Bauer and Cowan are in the Ministry of Finance, Chile. The section entitled “Political Economy of Equity-Enhancing Policies” is drawn mainly from a contribution made by Mario Marcel, who is with the Inter-American Development Bank.


For a review of the theory of distributive justice, see Solimano (1996). Bénabou (1996) offers recent reviews of the literature relating to income distribution and growth.


De Gregorio and Lee (1998) present recent international evidence on the relationship between income distribution and education.


For further details, see Cowan and De Gregorio (1996).


A nationwide socioeconomic survey carried out every two years in Chile since 1985.


Beyer (1997), based on CASEN survey data, estimates that in Chile nonlabor income accounts for less than 20 percent of total income.


De Gregorio and Landerretche (1997) show that the per capita income in 1994 PPP (purchasing power parity) dollars of the poorest 20 percent of Chile’s population is one of the highest in Latin America.


United Nations Development Programme (1997). The HPI focuses on three aspects of human development: longevity, access to basic knowledge, and a minimum consumption level. These three dimensions are measured by the probability of premature death, illiteracy rates, and the proportion of people with disposable personal income less than 50 percent of the national median.


The Copper Stabilization Fund operates by saving fractions of copper revenue obtained over and above a predetermined reference price. When the price of copper falls below this price, fiscal revenue is compensated with previously accumulated savings.


Wages of high-income workers, most of whom are employed in the formal (and, eventually, more unionized) sector, usually exhibit a higher degree of indexation than those of low-income workers. In addition, the holding of nominal assets is likely to be proportionally lower for the wealthier. Both of these factors explain why the less wealthy suffer more from inflation. See Bulír (1998) for recent cross-country evidence on the positive impact of reduced inflation on the distribution of income.


This calculation assumes constant inflation every month (geometric average rate). The state minimum pension is currently at US$135.


During the 1990s, five tax reforms were undertaken in Chile. In 1990, authorities reformed income taxation to provide the necessary resources for financing the new government’s social program without jeopardizing macroeconomic stability. The reduction of the uniform import tariff rate to 11 percent from 15 percent, and compensation for the resulting revenue loss by the increase of several indirect taxes, were undertaken in 1990 to prevent an unsustainable rhythm of real appreciation of the peso and to ensure current account stability. Because the tax increases enacted in 1990 were limited to a four-year period, a permanent solution to the government’s resource needs had to be negotiated in 1993 to ensure long-run fiscal equilibrium. Finally, in 1995 and 1997, policymakers increased the value-added and excise taxes on gasoline and tobacco, to finance two redistributive spending programs: a 10 percent real increase of the lowest public pensions and an education reform (explained in more detail below).


A World Bank (1997) study reaches a similar conclusion, although that is not surprising. Tanzi (1996) shows that income distribution in Latin America hardly changes after taxes because of that region’s low levels of direct taxes.


The regressiveness of these taxes is shown in Table 7.5, columns 5–7.


Engel, Galetovic and Raddatz (1997b) simulate the positive distributive impact of such policy in Chile.


Corruption is also likely to have a negative impact on equity. On this issue, see Gupta, Davoodi, and Alonso-Terme (1998).


On the modernization of the public sector management during the 1990s, see Marcel (1997).


See, for instance, Brunner and others (1995).


Through this credit facility, which has already been used for graduate studies, commercial banks open a credit line with the state-owned Corporation for the Development of Production (CORFO) at a relatively low real interest rate of 5 percent for the refinancing of credits up to 15 years for university study and technical training in public and private establishments with full academic autonomy. CORFO guarantees against default, allowing banks to reduce their guarantee requirements for students from low-income families; if the borrower does not pay, CORFO will reimburse the banks 75 percent of the final net loss for nonpayment of principal and regular interest. This credit facility is to help the government meet its objective of offering long-term funding through the financial system to facilitate the training of highly skilled professionals.


The impact of minimum wage regulations on income distribution is not clear a priori, because workers subject to such regulations are not necessarily the poorest. Workers at the low end of the income distribution are often employed in the informal sector or self-employed. Minimum wage regulations may affect income distribution in countries where informality and self-employment are low, or where the setting of the minimum wage is used by the informal sector to signal productivity increases and therefore tends to increase wages for these workers too.


On the data see Bonifáz and Bravo (1998), p. 328. According to Bravo and Contreras (1998), there is no evidence that a minimum wage policy has affected unemployment in Chile.


In Chile, no automatic indexation applies to the minimum wage; to increase the minimum wage, the current law would have to be amended. The Congress can only approve or reject—not modify—a government-proposed increase.


As mentioned, in Chile, the excessive depression of real minimum wages during the 1980s created room for a catch-up during the 1990s.


The amount of the transfer decreases each month.


A government project currently being discussed by the Congress establishes an employer contribution of 3.6 percent of the worker’s salary and a contribution of 0.8 percent from the worker during the first 11 years of employment in the company (after this time, no further contributions are to be made).


Obviously, this is a very simplistic exercise and has its shortcomings. First, the longer time to graduate from public and publicly subsidized schools cannot be attributed exclusively to differences in educational quality. As these schools serve mostly the lower-income groups of society, part of the delay in graduating may he caused by special social conditions—for example, children may be obliged to help at home or to work temporarily. Although the opportunity cost might be overestimated, available data suggest that the influence of this factor is limited.


In 1994, the minimum wage in Chile was Ch$358,900, or about US$145 per month.


According to CASEN, the average monthly household income of the first two quintiles of Chile’s population in 1994 amounted to US$232.20.

    Other Resources Citing This Publication