Comments: Guillermo Perry

Ke-young Chu, Sanjeev Gupta, and Vito Tanzi
Published Date:
May 1999
  • ShareShare
Show Summary Details

We have on the table two excellent papers, rich in policy implications. Let me first make a few comments on Alberto Alesina’s paper. It is a very well argued paper, and his model of the two different political and economic equilibria—of governments that “fail” because they are “too large,” and governments that “fail” because they are too “small and inefficient”—is powerful. The possibility of a “bad” equilibrium, with a small government and a large informal sector, has already been studied by Norman Loayza (1997), both theoretically and empirically, with special reference to Latin America. Also, as shown by Johnson, Kaufmann, and Shleifer (1997), the events following the dissolution of the former Soviet Union have lent a lot of credibility to the possibility of this type of equilibrium.

However, in Alesina’s model it is not easy to see what can produce an abrupt change in the policy regime. How do you get out of the trap in the “small government” equilibrium? It seems possible to escape only very gradually, as the state accomplishes two goals: increases tax revenue through better administration and substantially improves the quality of its expenditure and its policy interventions. Neither of these can be done overnight. Authorities cannot increase tax revenue simply by raising rates, because the model itself suggests a rate hike will probably increase informal sector activity. So, the way to increase revenue is basically through controlling tax evasion, and that takes time and a lot of effort, as we know.

Also, the model shows that the issue of improving the quality of expenditure is much more important in a country with a small state, precisely because the government needs to show that it can deliver really good-quality services and, in particular, effective protection of property rights that will encourage enterprises to move from the informal to the formal economy.

A footnote on the importance of institutional development. The recent literature has again brought to our attention that institutional development is critical to growth. The recent crisis in Asia has shown that it is also critical to stability (Perry and Lederman, 1998).

There are three papers by Alberto Chong and César Calderón (1997a; 1997b; 1998), not yet published, that explore the relationship between institutional development, poverty reduction, and inequality. Improved institutions clearly have the effect of reducing poverty, both directly and indirectly, through their effect on growth. However, evidence suggests that in developing countries, institutional reform may initially lead to an increase in inequality, with positive effects arising only after some threshold of institutional development has been reached. This is a provocative result, and its interpretation is not completely clear. It may be that improved institutions basically lead to a better enforcement of formal contracts and property rights and reduce some basic transaction costs in the formal sector. Although this initially increases the size of the formal sector, it benefits mainly those who are already in it; not until a higher level of institutional development has been reached does further institutional reform begin to benefit the poor. As I just mentioned, this is a provocative issue that one may wish to explore.

Let me turn to Aníbal Cavaco Silva’s paper. I fully agree with his proposition that policy should focus on the lower-income groups, and that the key policies are those that enhance human capital, especially of the poor, and particularly by improving the quality of education and by implementing pro-employment policies in general. Let me mention that, especially in Latin America, the issue of education is absolutely critical. Latin American countries are lagging significantly in the quality of education with respect to their competitors in the world economy. Michael E. Porter, in remarks he made on the Global Competitiveness Report 198 at a World Bank seminar in June 1998, said this lag in the quality of education is clearly the major current restriction to competitiveness in Latin America (Porter, 1998a; see also Porter, 1998b and 1998c). At the World Bank and the Inter-American Development Bank (IDB), we have come to similar conclusions. The problems are not only related to the lack of quality in the education provided, but also to the huge inequality in access to quality education. In Latin America, although most children enter primary schools, especially in urban areas, the dropout rate, especially of children who come from low-income families, is very high. On average, only half of the children complete primary school in Latin America, and most of them come from high-income and medium-income families; very few come from low-income families. So the system is clearly perpetuating the structure of inequality and poverty in a very serious way.

When you look at figures on entry-level enrollment in primary schools, Latin American countries appear to be similar to East Asian countries. When you look at the figures for the end of primary school, Latin America looks more like West Africa: the percentage of children from low-income families who receive further schooling is very low. This is a dramatic issue in Latin America. We welcome the emphasis that this conference has placed on education, and we think that the multilateral institutions must make it a critical priority. Both the World Bank and the IDB are delivering this message in Latin America.

Let me illuminate, however, a few differences between the European Union and Latin America. First, Latin America has much higher poverty rates. The share of income of the lowest quintiles is the lowest in the world because the region has an income distribution that is very skewed against the poor. Thus, the focus, in Latin America much more than in any other region, clearly has to be on reducing poverty and distributing income in favor of the groups with the lowest income. This is not the same as a general concern with inequality. In Latin America, the political discussion on inequality often centers on the amount of income that goes to the middle and top quintiles of the income distribution, rather than the lowest quintile groups. Personally, when I see that it is inequality and not poverty reduction that is high on the political agenda of Latin America, I worry a little.

One reason for this focus of the political discussion may be that, to some extent, poverty levels were reduced at the beginning of the 1990s owing mainly to the effects of stabilization—especially in countries like Brazil—and of increased growth. But income distribution has either remained stagnant, very unequal, or has worsened. The opening of the economies to trade and foreign investment may have produced, in some countries, a widening of wage disparities and some concentration of income in the top quintiles of the income distribution.

This outcome is contrary to what we expected. Export growth and capital inflows have not been concentrated on unskilled-labor-intensive sectors, but rather on skilled-labor-intensive sectors and natural-resource-intensive sectors—mining, agriculture, and industries based on natural resources. So the path of integration of Latin America into the world economy seems to be—with the exception, perhaps, of Mexico—very different from the pattern of East Asian countries. Thus, it is of great importance that these issues—and the corresponding policy responses—be looked into further.

I suggest that, first, the international community try to keep the focus of Latin American policymakers on poverty reduction. In that region, the discussion of income redistribution has too often been used, by both populist and right-wing regimes, to justify subsidies to very vocal and electorally important medium- and high-income groups. Thus, we have, for example, huge subsidies for electricity consumption and for higher education, which are not helpful to the poor and in fact crowd out expenditures that could help the poor.

Second, some of the labor market differences should be taken into account. Open unemployment is lower in Latin American countries than in the European Union, but there is much higher underemployment associated partially with large informal sectors. Besides, in Latin America, the unemployed are not always the poorest. The poorest cannot afford to be unemployed: they have very low family incomes and assets, and no unemployment insurance.

In practice, the flexibility of labor markets is higher in Latin America than in the European Union, not because the labor legislation or institutions are better, but because in many countries regulations are not well enforced. Thus, although in some Latin American countries reforms aimed at increasing labor market flexibility are important for reducing poverty—Argentina is a case in point—we have found that in some other countries, like Mexico, these reforms would have a much smaller impact. However, reducing taxes on labor, if we can compensate with increased revenues elsewhere, would help increase employment in every country.

Given the present importance of the informal sector, any policy that can improve its productivity and move it toward formality is even more important. For example, financial reform—including financial deepening—is critical in Latin America. The lack of access to credit for small producers, especially microenterprises, tends to limit growth significantly and, as mentioned, tends to reduce the growth of the more labor-intensive segment of the economy. This is an institution-building task. We do not want to create public banks for microenterprises, because that has clearly not worked in Latin America. The tasks include the improvement of creditor rights, the reduction of transaction costs, and so on.

Also, the problem of access to land remains significant in Latin America. Land reform and titling are unresolved issues. Command-and-control land reform did not work well, but market-based land reform is something that we are experimenting with, and the World Bank is supporting experiments in Brazil, Colombia, and elsewhere.

Although the large informal sector makes targeting more difficult, we must look for programs that do self-select. A good example is a program that the World Bank is supporting in Argentina, called Trabajar. This is a public works program that pays a below-market wage, so that it induces self-selection. It is really the poor, those who do not have access to jobs, who get into the program.

Let me conclude by saying that economic insecurity, especially for low-income groups, is an important new equity issue in Latin America. The previous development model, with all its inefficiencies deriving from protectionism of industry and patronage employment in the public sector, gave a measure of job security to some groups. With deregulation and trade opening, income insecurity has increased.

This economic insecurity is something that we must put on the agenda. Dani Rodrik, at Harvard, has voiced this concern, and I fully concur: we must examine more carefully how to build efficient safety nets and social security systems in Latin America. This is also an important issue for the Eastern Europe and Central Asia region and now for East Asia.


    ChongAlberto and CésarCalderón1997a“Empirical Tests on the Causality and Feedback Between Institutional Measures and Economic Growth” (unpublished; Washington: World Bank).

    ChongAlberto and CésarCalderón1997b“Institutional Efficiency and Income Inequality: Cross-Country Empirical Evidence” (unpublished; Washington: World Bank).

    ChongAlberto and CésarCalderón1998“Institutional Quality and Poverty Measures in a Cross Section of Countries” (unpublished; Washington: World Bank).

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

    LoayzaNorman A.1997“The Economics of the Informal Sector: A Simple Model and Some Empirical Evidence from Latin America,”World Bank Policy Research Working Paper 1727 (Washington: World Bank).

    PerryGuillermo and DanielLederman1998“Financial Vulnerability, Spillover Effects, and Contagion: Lessons from the Asian Crises for Latin America,”World Bank Latin American and Caribbean Studies: Viewpoints (Washington: World Bank).

    PorterMichael E.1998aWorld Bank Seminar on “Building Latin America’s Future: Public/Private Partnerships for Education”WashingtonJune.

    PorterMichael E.1998b“The Macroeconomic Foundations of Economic Development,” and “Measuring the Microeconomic Foundations of Economic Development,” in The Global Competitiveness Report 1998 (Geneva: World Economic Forum).

    PorterMichael E.1998cOn Competition (Boston: Harvard Business School Press).

    Other Resources Citing This Publication