Comments
[Crises and the Poor: Socially Responsible Macroeconomics]
Albert Fishlow: This paper is a nonpopulist manifesto. It represents an alternative to the literature of the 1970s in that it suggests a very different possibility of dealing with the distribution problem. Then, the emphasis was on direct intervention: increases in the minimum wage, direct intervention to improve consumption of essential services like health and housing, and greater public engagement were seen to be positive policies. Now, a more subtle series of efforts has been undertaken to improve the distribution of income and, in particular, to assist the still large--and recently increasing--share of the population that lives in poverty. Efforts taken by the multilateral institutions, and by individual governments, have been impressive. Although, alas, the distribution of income remains highly unequal, policies to ameliorate the larger number of poor have become more common.
In this respect, Lustig offers a welcome parallel to Dani Rodrik's paper also presented at this conference, which focused more exclusively on the middle class. Her analysis is concerned exclusively with the poor. But are some of the poor formerly of the middle class, having descended the income ladder? The middle class in Latin America is in the process of rapid change. The old middle class was defined by labor unionization, concentration in a manufacturing sector upheld by import substitution, and limited entrepreneurial participation. Today, Latin America enjoys much freer international trade, and the old middle class faces much greater competition in the labor market. A new, and larger, middle group is beginning to emerge. In this sense, the two papers interestingly overlap.
One of the paper's few disappointments is its failure to emphasize that Latin America today is really very different than it was ten or fifteen years ago. That difference has implications not only for growth strategies but also for poverty. One of the important advances that is completely left out of Lustig's analysis is the end of inflation in Latin America. That simple fact has had a greater impact on the distribution of income of the [End Page 20] region's poorest households than virtually any other policy implemented over the last few years. For it directly ended the daily erosion of wealth the poor suffered; the rich could rely on daily deposit accounts that yielded a rate of return.
Additionally, the region has undergone fundamental change. It is now divided into three groups. The first group comprises the successes of new capitalism, namely, Brazil, Chile, and Mexico, as well as some of the countries in Central America and the Caribbean. Second is the case of Argentina, where commitment to a fixed exchange rate has led to dramatic change, but only limited trade growth. The Argentine experience is a key test, for if the currency board were to fail, the effects would spill over to the successful economies. Finally, the Andean subregion is in serious and suddenly explosive trouble. In addition to political difficulties, the economic reforms implemented in much of Latin America over the past decade have been pursued far less extensively in these countries, with the exception of Peru. Thus, the economies have not advanced nearly as far as the first two groups in areas such as privatizing state-owned companies, curbing inflation, or instituting a flexible exchange rate policy (with the exception of Colombia's recent move to a more floating rate). Politically, the rise of new leaders such as Fujimori and Chavas has meant the end of the earlier party structure, but this has not necessarily been accompanied by the creation of a new political order. This stands in sharp contrast to the rehabilitation of political structures in Argentina, Brazil, and Chile. Recent elections in Peru and Venezuela underline this difference.
The real focus of the paper is on the possibility of establishing safety nets to avoid further growth of poverty in the future. François Bourguignon deals with that topic extensively in his comments. In terms of comparative advantage, I choose to emphasize the aggregate side.
Macroeconomics enters the debate directly, even in the title. Lustig tries to follow an intermediate position, against both inaction and overreaction. She seems, however, to be too inclined to accept the claims against the International Monetary Fund (IMF) that have recently become popular, ironically uniting such diverse critics as Allan Meltzer and Jeffrey Sachs. In brief, from December 1994 to January 1999 the IMF intervened extensively in response to the Mexican crisis, the Asian crisis, the Russian failure, and the Brazilian devaluation. Yet this series of recent setbacks was very soon overcome. Contrast the recent situation with the tragedy of the 1980s. In 2000 growth has been restored virtually everywhere. [End Page 21]
Lustig nowhere considers the central role of privatization revenues, which have been a major component of the fiscal improvement of virtually all Latin American countries as they combat inflation. Privatization has not only added to the inflow of revenues, but also led to the sudden necessity of effective regulation. Earlier, when the government owned the communications, railway, and power sectors, regulation was largely irrelevant. Now decisions must be made on what rates to accept and what incentives are efficient. This new area of institutional reform is a central part of the evolution of the state in Latin America.
Lustig argues in favor of government adjustment programs that will accomplish an efficient recovery from crisis while at the same time paying attention to the poor. That is very much a current consensus, not merely of academics but of governments themselves. She cites the cases of numerous Latin American countries that have moved in this direction. She fails to note, however, that in the cases of Argentina, Mexico, and Venezuela--all of which did do so--the government lost subsequent elections. Did this happen because the poor did not vote in as high percentages as the rich? Or because they included other measures in evaluating the adequacy of the government?
The ability to establish and operate safety nets depends very centrally upon politics. Yet this issue is somewhat understated in the paper. Legislating and implementing policies in Latin America is subject to important national differences. Furthermore, changing rates of population growth greatly influence not only pension policy, but also electoral results. For example, young voters currently are dominant in much of the region. In raising the question of safety nets, it is important to indicate whom they are designed to protect. If created simply to protect the poor without regard to age, they may wind up serving an elderly population with little formal education, little geographic mobility, and little opportunity to take advantage of subsequent increases in income. This issue is highly relevant.
Finally, one must also consider the future growth prospects of the region in evaluating the adequacy of policy. If Latin America is to plod along at its recent rate of expansion, safety nets will hardly be enough. Many analysts presume that higher expansion rates will occur, leaving only the cyclical crises to be overcome. But accomplishing that happy situation will require fundamental changes. I mention here in closing only one: the degree of education. In much of the region, public universities are essentially free. Consequently, significant public resources are spent on higher education, which diverts funds away from the expansion of secondary [End Page 22] education. If one is to compare Latin America with Asia, that--and not only the differential rate of savings and the growth of exports--is an area worthy of greater emphasis.
Nora Lustig has provided us with a rich paper. Hopefully, Latin American expansion in subsequent years will make its subject less relevant than it presently is.
François Bourguignon: This paper has a very strong message: poverty, or more exactly the fate of the poor, must be one of the objectives of responsible macroeconomic policy. This message would be without consequence if, as most often assumed, there were full symmetry in the way various social groups in an economy are affected by macroeconomic shocks and policies. It would also be unimportant if the fate of the poorest were already among the priorities of policymakers. What Nora Lustig's paper shows, in the first place, is that this is not the case and that dynamic efficiencies or inconsistencies may therefore characterize the way macroeconomic shocks are dealt with in Latin America. After giving evidence of these basic facts, she asks three main questions. How could and should macroeconomic policy explicitly account for poverty? How can poverty-oriented public spending be protected from macroeconomic-induced variations? What kind of safety nets could and should be implemented?
I am personally very sympathetic to the general argument in the paper and the questions asked, even though I do not always agree with the author's recommendations. However, I feel that a few general issues should have received more emphasis in the paper. They all have to do with safety nets, which are presented in the paper as one of the possible instruments for alleviating the burden of macroeconomic shocks on the poor. I believe their role is indeed much more fundamental and has important implications for the conduct of macroeconomic policy.
Macroeconomic stabilization generally is a matter of urgency. For good or for ill, those in charge of stabilizing an economy in times of shocks do not generally worry too much about what will happen to the poor. The implicit argument for ignoring this objective is that the consequences of failing to stabilize the economy might be worse for the poor than for other people. Somehow, the responsible macroeconomics that Lustig calls for in this paper goes precisely against this view that there should be some dichotomy between macroeconomic policy and distribution issues, at least in periods of crises.
Whether it is possible to reconcile efficient stabilization policy and the protection of the poor essentially depends on the number of policy instruments [End Page 23] that are available and their function. Preventing the increase of poverty may well be among the objectives of policymakers. However, they would not have to consider the issue in the midst of stabilizing an economy if the appropriate instruments existed to deal with it, independently of aggregate monetary and fiscal policy. In the presence of well-functioning safety nets, policymakers would not need to ask whether a specific macroeconomic policy or exchange rate regime were pro-poor or the opposite: the existing safety nets would cushion the poor from whatever happened on the macroeconomic front. At the same time, macroeconomic policy would be more efficient in dealing with the truly macroeconomic economic imbalances because it would be freed from consideration about its possible impact on poverty.
Rather than presenting safety nets as one of the various dimensions of socially responsible macroeconomics, as is done in the present paper, I would consider instead that it is a crucial condition for efficient macroeconomic policymaking. If such safety nets were in place in Latin America, many of the questions in the first part of the paper would not arise, and many of the debates on the effects of macroeconomic stabilizing policies would simply have no ground. Building those safety nets is thus like providing new instruments and contributing to more efficient policymaking. Beyond making macroeconomic policy socially responsible, it makes the state socially responsible and at the same time frees the state--at least partially--from distribution constraints in periods of negative shocks.
Building safety nets should thus be a priority in the Latin American region. Unlike Lustig, I believe most countries are still very far from that objective. Progresa in Mexico--and similar programs in Brazil and Ecuador--and Trabajar in Argentina show that things are moving in the right direction, but this is still a long way from anything resembling the income guarantee schemes found in industrial countries, namely, unemployment insurance and minimum income programs. Progresa currently provides cash transfers that are conditional on children going to school and receiving regular medical examinations. Beneficiaries are identified as being poor on the basis of permanent attributes like household composition or housing characteristics; this is very different from more or less instantaneous means testing, that is, cash transfers that are conditional on current income. The insurance part of Progresa lies in the fact that it is a permanent transfer that does not depend on the level of local or national economic activity. On the other hand, conditioning the payments on the children's schooling means that it is education rather than household [End Page 24] income that is insured. Such insurance is certainly very important, as schooling disruption among the poor may be the cause of the apparently irreversible surges in inequality seen in many Latin American countries during and after crises. But again, this is only a partial solution. In particular, a program like Progresa has no effect on households that fall into poverty because of a macroeconomic shock.
Public works schemes may play that role, but mechanisms must be implemented to ensure that they will indeed be able to expand in times of crises so as to cover rapidly increasing needs. As discussed in the paper, this could be obtained through an unemployment insurance contribution that accumulates in a special fund. But then, how should this be articulated with more standard unemployment insurance in the formal sector? Alternatively, is it conceivable to extend programs like Progresa to urban areas and to have them means-tested with a shorter horizon? Such questions should be seriously analyzed if macroeconomic policy in Latin America is to be partially freed from acute distributive considerations and if the socially expensive indirect costs of recessions are to be avoided. At the same time, the extension of these insurance mechanisms has macroeconomic implications that cannot be ignored.
I would like to add two sets of remarks to these general comments on the macroeconomic importance of safety nets. First, it must be kept in mind that building safety nets is costly. In particular, there are political economy aspects that must be taken into account--aspects that are surprisingly absent from the paper. In this respect, it is often said that a crisis is not the right time to innovate in the social field, but that there is no incentive to do so in good times. Thus, safety nets might never be built. This may be so, yet it is worth stressing that many social security systems were born more or less directly from the experience of the middle class with poverty during major economic crises. For instance, the crisis of the 1930s did very much for the progress of social insurance in pre- and post-war Europe. In normal times, the median voter might not be interested in the fate of the poorest people in society. After a crisis, however, he/she may have experienced poverty directly or may know people who did. This may be precisely the time to build permanent safety nets. If, on the other hand, the decisive voter is much above the median, a positive decision on safety nets may be difficult to obtain because the decisive voter will be too far from the poor, or from the risk of becoming poor during a crisis. Is this the situation observed in Latin America? [End Page 25]
My second remark has to do with the issue of discretion and rules in macroeconomic policymaking. The existence of safety nets is not independent of the conduct of macroeconomic policy. It may free policymakers from distribution concerns (at least partially), and thus give more credibility to some of their commitments. For instance, market actors may be more willing to believe in the government's commitment to a strict monetary or fiscal policy if they know there is no risk that distribution issues may eventually force the government to loosen its policy. As a result, there is more need for discretion than for rules in macroeconomic policy.
It strikes me that the paper is advocating very strongly in favor of all sorts of rules that might progressively make policymakers useless: stabilization funds, priority ranking of public spending, possibly dollarization. What instrument will be left? If distribution is partly taken care of through well-functioning safety nets, incentives for populist policies and therefore for imposing rules might be reduced, and policymakers might securely be given more discretion. In the long run, it is thus probably more efficient to build effective safety nets than to establish rules, which at some stage or in some circumstances will inevitably prove inadequate.
