12 Lessons for Countries and the IMF

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

Alassane Ouattara

Let me welcome you to this roundtable discussion. This session aims to draw lessons from the previous sessions on the design and implementation of equitable policies for countries and for the IMF.

Amartya Sen, and to some extent Minister Aninat, argued that equity is a multidimensional concept, encompassing more than the distribution of income. This suggests that countries and the IMF need to monitor and target improvements not only in the distribution of income but also in a number of social indicators that measure the degree to which the poor are empowered to participate in society. Empowering the poor requires policies that improve their human capital through well-targeted primary education and health spending.

However, as noted by Grzegorz Kolodko, policymakers are sometimes prone to making errors while pursuing the objective of equity. A similar thought was expressed by Aníbal Cavaco Silva, who said that governments tend to prefer high-visibility expenditure programs geared to the groups of voters most beneficial to the incumbent government. Governments may also opt for regulatory measures—such as high minimum wages—ostensively to promote equity. Such measures could be detrimental to an important dimension of equity—the ability of the economy to employ all members of society productively. Alberto Alesina argued that in both industrialized and developing countries, governments have failed to promote equity effectively. For most industrialized countries, governments have tried to do too much, and their excessive provision of social insurance has become counterproductive. On the other hand, in developing countries, governments have not done enough, either to provide infrastructure or to reduce inequality.

Prudent fiscal policy can be instrumental to achieving equitable growth. Minister Aninat noted that fiscal restraint allowed Chile to repay its debt and reduce interest payments, providing space for equity-oriented social expenditures. In Chile, good macroeconomic policy and equity have gone hand in hand.

Globalization has brought to the fore new issues and challenges for policymakers. We all recognize the benefits of globalization. One view, expressed by Anthony Atkinson, is that the constraints on government’s redistribution of income in a globalized environment are exaggerated and are largely political in origin. Pressures to reduce tax rates may constrain governments—particularly those of OECD member countries—from pursuing a more active role in income redistribution. I would be interested in hearing views on whether this political constraint has led to better or worse economic policies.

Against this background, I would like to ask what more the IMF can do in light of its essentially macroeconomic mandate. For example, Alesina argued that to provide the right incentives for countries to implement equitable policies more effectively, the IMF and World Bank should consider withdrawing technical and financial assistance to governments that do not satisfy minimum standards of efficiency and are excessively corrupt. Is this position realistic? What are the other implications for countries and IMF policies and operations?

In promoting growth with equity in its member countries, the IMF has concentrated on promoting macroeconomic stability, structural reforms, good governance, implementation of fair and efficient tax systems, growth, human-capital-enhancing changes in the composition of expenditures, and well-targeted social safety nets. How can the treatment of equity issues be further strengthened in IMF-supported programs? Is it feasible for countries and the IMF, with the help of other international organizations, to monitor a wider range of equity indicators? How should the results be used? How should IMF programs respond to variations among countries in perceptions of equity? And, can an international consensus be forged on a minimum set of equity conditions that should be met? I hope that this roundtable discussion will address some of these questions.

Alberto Alesina

I learned a lot from this conference. I would like to share with you some of the thoughts I had while listening to the various presentations. In economics there are always trade-offs, and several important ones have been highlighted in this conference. I would like to mention several of those trade-offs and their implications for policies.

The first trade-off is small government versus big government: which is better for growth? I want to clarify one thing, using a model from economic theory. Essentially, there is an inverted U-curve relating the size of government to growth: the maximum level of growth is achieved with an intermediate size of government. If a government is too small, it will not achieve the maximum growth because it will not have enough infrastructure, telecommunications, and so on. But, beyond a certain point, when the government becomes too large—for example, because it is too involved in redistribution—a point is reached where the burden of taxation becomes predominant and growth begins to slow down.

Thus, the government should be neither too small nor too big. The problem that I tried to highlight in my paper is that many countries, particularly OECD countries, are on the right, downward-sloping side of the inverted U-curve. As a result, reducing the size of government by cutting government spending in these countries would actually boost growth. By contrast, many developing countries are on the left, upward-sloping side of the inverted U-curve, and may actually boost growth by increasing government spending. The problem, of course, is that government spending in developing countries can be directed at either productive or nonproductive expenditure, and wasteful spending will not fuel growth.

The second trade-off also concerns big versus small government: not in relation to growth but as a function of income inequality. For example, several members of the audience contended that even if big government does reduce growth, pursuing the policy of having a large government may be worthwhile because it certainly would reduce income inequality. Arguably, it is worthwhile to trade off some growth for some reduction in inequality. I disagree. I simply do not believe that larger government does indeed reduce income inequality. And I am not sure that strong evidence to that effect exists. However, even if it were true that bigger government somewhat narrowed income inequality, and that bigger government hurt growth, I would be cautious about trading off even a small amount of steady-state growth. Just a simple computation shows that even a small reduction of growth each year can lead to tremendous difference in the level of income over 20 years, and we know that growth results in reduction of poverty, if not income inequality. If the goal is to reduce poverty, then no policy that lessens per capita growth in income should be followed except under very special circumstances. A trade-off in which you forgo even a fraction of 1 percent of growth in the steady state becomes an extremely costly policy over time. My point is that if it is true many developing countries are on the upward-sloping side of that inverted U-curve, increasing a certain type of government spending—for example, on infrastructure and education—may be a win-win policy: more and better-targeted government spending may actually enhance growth.

This leads me to the third trade-off, that between government size and government quality, which we discussed earlier. The size of government may not mean much, but quality of government does. Of course, nobody can argue that a good government is not better than a bad government. But, empirically, it seems that large governments in OECD countries are, on average, more efficient than small governments in developing countries.

The empirical discussion may be somewhat confused by the fact that as the size of government grows, efficiency also grows—that is, empirically, small governments tend to be rather inefficient and large governments tend to be relatively efficient. Moreover, perhaps the optimal size of government is much smaller than the OECD average, but OECD governments are more efficient than what we see in many developing countries with small governments. So this issue of size and efficiency becomes complex.

Now, what to do about all these trade-offs? Let me pick up on just two of the issues I find particularly difficult and intriguing: one concerns developing countries; the other, OECD countries.

In developing countries, the single most difficult problem is the composition, efficiency, and targeting of spending. Even though we all agree that many of these countries’ governments are too small and efforts should be made to increase revenue and spend more on social programs and education, simply throwing more money at existing programs is not likely to work. There is a large amount of evidence, some of which I have discussed, that social spending in developing countries is mistargeted and does not reach the very poor. Thus, throwing more money at existing programs handled by bureaucracies, which are often inefficient and corrupt, is not going to solve the problem of poverty and income inequality.

This is a formidable problem: expanding the size of government while increasing efficiency. And that is why I support the view to which Mr. Ouattara referred: perhaps we should use institutional conditionality and stop lending to countries that clearly do not satisfy the minimum standard of efficiency. This morning, someone suggested that the IMF already does this, but this afternoon I hear that, in fact, this institutional conditionality is perhaps not effective enough. These opposing views suggest that this difficult and unresolved issue needs further discussion. In my opinion, not only the IMF but the other international organizations as well should use more “institutional conditionality.”

One final point concerning OECD countries. Ravi Kanbur gave a splendid discussion of my paper, and he ended with an intriguing example. He agreed that the welfare state in OECD countries had to be retrenched, but he worried about the 50-year-old steelworker who is displaced by the technological revolution and must be retrained to become a computer programmer. He said this would be difficult, and suggested that we might have to provide a social transfer to this steelworker for the next 30 years of his life because he could not be productive. This is how Europe generally approaches every issue: when there is a problem, a transfer is provided.

I suggest another solution, however. I suggest cutting taxes and regulations, and reducing labor costs, so this 50-year-old person can be reemployed—if not as a computer programmer, then perhaps doing something without computers. Let me give you an example. A few years ago I was in Stockholm, and every day I took the subway to the university about 8:00 a.m. One day I overslept and had to take the subway at 11:00 a.m., a nonpeak hour. I could not figure out how to buy my ticket, so a kind Swede told me that at 11:00 no one collects tickets, because the cost of paying someone to collect them is higher than the revenue forgone from giving people free use of the subway. My point is that the 50-year-old steelworker who cannot feasibly be trained as a computer programmer could perhaps work as a ticket collector if taxes were cut and unit labor costs were not high.

Karin Lissakers

I would like to focus on what lessons the IMF can draw from the preceding discussions. Clearly, the first lesson is that we at the IMF have to think about the issues addressed at this conference. And I was struck by Grzegorz Kolodko’s comment that in his nine years as a policymaker, he was never asked about equity or inequality during his discussions with the IMF. This attitude—this failure—has changed, and we have begun to address equity issues, both in our annual economic consultations as part of our surveillance exercises and in program design. But the question remains, How do we think about equity in a context that makes sense for the IMF and for its members?

One of the benefits of this conference is that we will take with us from the papers and discussions a number of ideas on how to improve the IMF’s performance in this area. Some of the early speakers pointed to the definitional debates and to the difficulty of numerically measuring equity and equity performance. As the IMF is an institution that historically has dealt with numbers and based its program design and, to some extent, its surveillance activity on numbers, this presents a practical problem. But anyone who looks at what the IMF has been doing recently in Asia sees that we have moved beyond numbers and have looked at institutional structures and governmental organization. This is necessary in order to address equity issues effectively.

The IMF has addressed many of the concerns and issues related to equity quite concretely in recent years, in its pioneering efforts in the area of governance—which is absolutely central to the empowerment issue that several speakers addressed effectively—and transparency, which have become watchwords and performance criteria in programs of both the IMF and the World Bank.

Again, observers who have examined what the IMF has done in Korea, Indonesia, and Thailand see very specific, concrete manifestations of that concern for equity, and the institution’s recognition that good governance and good economic performance go hand in hand and, most particularly, the institution’s focus on spreading the benefits of improved economic performance to the wider population and not to just one privileged segment of society.

Archbishop Ndungane gave a very impassioned plea for debt relief and debt forgiveness. This is another equity issue between rich and poor nations, and one that the IMF—again, with the World Bank—is confronting. The IMF has designed a somewhat complex debt-relief structure, as you are probably aware. One of the complexities, and a target of criticism, is the conditionality that accompanies debt relief and the delay that is sometimes required in its delivery. But I think that Archbishop Ndungane himself spoke of this issue when he said we need strict monitoring and control of policies to ensure that the debt relief, which is clearly essential for many countries to achieve sustainable growth, is not wasted because the underlying policy corrections have not been undertaken.

It is clear from the IMF’s preliminary experience in this area that the prospect of debt relief is a powerful incentive for governments to stay the course on essential economic reforms; this is why most IMF members, including the United States, have been insistent, as has management, that debt relief be linked directly to meaningful and lasting policy reforms.

In the end, the IMF’s largest potential contribution is probably in its traditional area, which is to help member governments lay the basis for sustained high levels of growth because, as many speakers noted, high growth is the quickest path to poverty reduction even if it does not solve all of the equity issues. And although many equity issues will still need to be addressed after high growth has been achieved, high growth will make the process easier and improve life for people on the bottom half of the economic ladder as well as those at the top. I came away from the conference sessions with the sense that there is no real controversy about the need for macroeconomic discipline as a basis for improved growth and equity. In this context, the IMF has paid increasing attention to the quality of fiscal adjustment in recent years.

The IMF now specifically encourages governments to reduce nonproductive expenditures, such as military expenditures, and to shift resources to primary education and primary health care. These areas are critical, as many here have said, to improved economic performance and, above all, to improved equity and to the distribution of the gains of growth and development.

The IMF, however, has no magic formula. Even the countries that are making excellent, broad, economic policy efforts, as Minister Aninat said, have not found the key to closing the gap quickly between those at the lower and higher ends of the socioeconomic ladder. The IMF can help greatly in these areas, but it probably cannot carry the weight that other institutions, like the World Bank and the multilateral development banks, can. The governments—the leadership—in each country must work in an open and transparent fashion with civil society to make a difference. The IMF has advised many countries on labor market flexibility and has recognized that it is not enough to just talk about flexibility and employers’ ability to shed labor: the IMF must look at whether workers have a fair chance of finding good jobs, and when there is unemployment, whether social safety nets, unemployment benefits, and so on, are sufficient to cushion job transitions.

The IMF has not done enough systematic thinking or program design in this area. But, again, I would point out the beginnings of efforts in this area in the Asian programs. The IMF has played an important catalytic role in getting the tripartite dialogue going in Korea. Less visibly but, I think, as effectively, we played a role in Indonesia and Thailand in giving workers a voice in policy formulation and execution. And it is in that nexus between macroeconomic management and the relationship with civil society where the IMF is beginning to make a new and valuable contribution.

Jean-Claude Milleron

What I learned at this conference is related to measurement, values, and policies. The first lesson is that there is broad agreement on measurement, especially after Amartya Sen’s enlightening presentation. Equity is not just the fair distribution of income or even wealth; it is also opportunities available to human beings. It is not only a static but also a dynamic concept, which must be analyzed in a world of uncertainty—when uncertainty itself is a potential source of inequality.

I would like to make three observations on the issue of measurement. First, we must be aware that in many cases, the data simply do not exist or were “invented” some years ago by good people who, perhaps, were not fully aware of the standards required of the professionals in the discipline.

Second, we should fight the temptation to sensationalize equity issues in the press by disseminating simplistic indicators. Several speakers have warned at this conference that working with just one index is illusory and, moreover, may be dangerous.

I was impressed by Alicia Munnell’s remarks on social mobility. Clearly, the special problem of people who are likely to remain poor throughout their lives deserves attention. This is certainly true at the national level. Furthermore, it may be especially relevant in development economics, because we now know that many changes can take place in just one generation.

My third observation relates to the value judgment or social welfare function assumptions implicit in the indicators we use. This approach was promoted some years ago by various economists in the United States and in Europe; and Anthony Atkinson himself contributed to this literature. For instance, we all know the limits of Gini coefficients and have debated them. But we could enrich this debate by keeping in mind that the use (under various constraints) of the Gini coefficient to minimize inequity is compatible with some implicit value judgments in terms of the so-called social welfare function. And this issue brings me to my next point, values.

On values, it seems to me that the old debate on Rawlsian versus utilitarian social welfare remains relevant. From that point of view, our international organizations may diverge from the national authorities. Here, I dare to say that the international financial institutions may be more Rawlsian—truly keen to help the worst-off of the world. Conversely, as Grzegorz Kolodko put it so well, many governments tend to be more utilitarian because their responsiveness is constrained by electoral pressures.

In our role as members of international institutions, we should examine whether we are sufficiently consistent in the policy recommendations we make over time and in different contexts. For instance, in the case of a negative shock or when proposing some necessary adjustment policy to the member states, we are clearly Rawlsian. As Stanley Fischer put it, we try to minimize as much as possible the effects of our policies on the poor. This is the general idea behind safety nets. It is also, for example, what the World Bank does in Indonesia in providing temporary subsidies to help low-income groups.

In the converse case of possible windfall profits versus sharing the fruits of growth, perhaps the international financial institutions are not fully consistent; moreover, I would be willing to argue that the international financial institutions are tempted to be more utilitarian in such cases. Certainly, we have to be aware that the utilitarian approach is not inconsistent with decreasing marginal utility of income, but it leads to less radical conclusions about the help we provide to the worse-off of the world.

I would also like to note an interesting debate on the foundations of the concept of solidarity. I do not think that an individualistic approach necessarily means less solidarity, but the coexistence of the two does require a categorical imperative. It is true that solidarity comes also from a feeling of belonging to something much broader than ourselves: the family, the village, the country, and so on—something that is, in a sense, the opposite of exclusion, to choose a word often used in our debate. Here, I would also like to note the profound interaction between values and policies: inequality leads to exclusion, equity to better social integration.

The international financial institutions can no longer avoid using the word “moral.” Although it is often associated with “hazard” by members of the U.S. Congress, I prefer to associate it with philosophy. The values we are talking about are an essential part of today’s moral philosophy. In this context, we must recognize that corruption is a kind of moral violence, as it is one of the most burdensome inequities of our times.

Finally, let me turn to policy issues. In the interest of brevity, I will mention just seven points. First, as Alassane Ouattara reminded us, we must never lose sight of the fact that there is no substitute for sound macroeconomic policy. Also, as speakers have said generally and about India and continental Europe—unemployment remains a major component of inequality.

Second, to reiterate an important point made by Atkinson: in terms of policy mix, we should avoid blanket prescriptions of standard medication. Because history and social norms vary among countries, economic policy must be custom-made. It could be that the IMF prescriptions sometimes too closely resemble over-the-counter medicine rather than therapy designed for the individual patient; however, this is not to say that tough decisions will not have to be made.

Third, on the issue of a possible trade-off between equity and efficiency, I persist in thinking that the answer must be “maybe so/maybe no.” We could come up with very different responses, depending on which stage of development an economy has reached.

Fourth, at the microeconomic level, I am convinced that in the years to come, the question of the geographical mobility of people and activities will become a major component of the Fiscal debate. In Europe, “competitive regulation” has long been considered critical, and it is interesting to note that the OECD has initiated some work in this field.

Fifth, we must draw a distinction between trends in inequality as measured by a set of indicators and analysis of the distributive consequences of a given project or policy.

Clearly, as stated by Monsignor Martin, unsustainable economic patterns of growth and inequities are causally linked. However, we now know that within one generation, many things can be changed for the worse or the better. The difficulty is the wide discrepancy in time horizons: most politicians have a horizon closer to that of a three-month treasury bill than that of demographers. It is something that we have to live with.

I would like also to emphasize that even if it is difficult to make a precise assessment, the international financial institutions do and must continue to try to be explicit about distributive consequences of their decisions. This was the focus of discussion of the 1970s, mainly in the World Bank, on investment criteria. And it underlies the discussion of today on adjustment policy. There is no doubt that such an issue calls for strengthened cooperation within the Bretton Woods family but also with the UN agencies involved in field activities.

Sixth—and this point I emphasize—in making judgments about inequality, we should consider not only what people say but also what they do. For instance, I have been impressed by the enormous efforts made in most parts of the United States to make buildings accessible to the handicapped. I know the difficulties that remain in my country, France, for people in wheelchairs to gain access to public transportation, public buildings, or even a movie theater. In these cases, my advice is, “Don’t talk too much, just do it.” This may be a European reaction, but it may also be a part of what categorical imperative really is or should be.

Seventh, a consequence of all these points is that policymaking is becoming increasingly political—in the best sense of the term—which explains why the issue of political accountability is becoming increasingly important in the IMF and should be addressed as such. Someone in our discussions mentioned the question of sanctions; in this context, political accountability is very important indeed.

In conclusion, I would like to share three thoughts I will take with me from this conference.

  • Data matters. Investing in data collection and knowledge is certainly a prerequisite for increasing equity in the development process.

  • Our international financial institutions must focus to some extent on the worst-off of the world.

  • In preparing our policy recommendations, we must never shirk our duty to take into consideration the specific history and social norms of the country we are examining. And, some good advice to economists: Don’t be overcautious. Political issues should be taken into account—they are not necessarily dirty.

David Smith

One of the pleasures of speaking at the end of such an informative and provocative conference is that one is wiser after listening to all of the preceding speakers. The corresponding difficulty is, of course, that much of what I had hoped to say has been said. Our panel’s task is to lead off a discussion of lessons, and I want to suggest six, most of which have been discussed or raised over the last few days.

At lunch yesterday Stanley Fischer noted that IMF Managing Director, Michel Camdessus, as he travels the world, makes time to meet with religious leaders. And Lawrence Summers noted the odd-bedfellow combination of the IMF and the American Federation of Labor-Congress of Industrial Organizations (AFL-CIO). One lesson is that such meetings between financial, religious, labor, and other leaders ought to become commonplace. They ought to become so commonplace that no future speaker will think them worthy of comment.

Trade unions and the other institutions of civil society need to be engaged early and often as governments and the international financial institutions consider stabilization programs and institutional reform. There are multiple reasons for this, most of which have been discussed. It is simply not possible to build the broad social consensus that is necessary to support the very difficult—often wrenching—measures that must be undertaken unless the broadest possible coalition is engaged in developing them and exploring the alternatives. The concerns of working people must be explicitly on the table and part of the policy mix in the development process, not simply included as afterthoughts. Moreover, as several of our colleagues have pointed out, finance ministers’ and their advisors’ understandings of reality, both domestic and international, is incomplete. Just as they must consult with their colleagues in the education, labor, and social ministries, they must consult with representatives of civil society and trade unions. When they do that, they make better decisions. As Grzegorz Kolodko reminded us, policymakers need to make decisions that can withstand the test of political viability.

A corollary lesson is that popular and democratic institutions must be allowed to exist: the rights of freedom of association, of political participation, of religious freedom, and respect for the core conventions of the International Labor Organization (ILO) must be honored in fact as well as in theory. Broad consultation with popular groups cannot occur if those groups are repressed, if voices are hushed, if trade unionists and political opponents are locked up. The IMF and the other international financial institutions must insist that these rights be protected and enforced as an integral part of any and all stabilization programs.

Some will claim that this is too difficult, that stabilization must come first. While we are certainly not done learning the lessons from East Asia, one important lesson we have already learned is that economic stabilization programs, with all of their political and social difficulties, cannot be sustained when basic rights are denied, when voices of citizens are muffled. In addition, important evidence indicates there is a consistent relationship between improvements in the right of free association—most importantly, but not only, trade union rights—and economic performance. A new paper by my colleague, Thomas Palley,1 examines some of this evidence in more detail.

In stating the third lesson, I want to echo what my panel colleagues have said but draw a slightly different conclusion about the importance of growth than our colleagues from Chile and Poland may have intended. The lesson is that the IMF must be extremely cautious about requiring too much austerity, too much attention to fiscal surpluses, too much reliance on high interest rate regimes, and, even, too little inflation. We cannot all export our way to prosperity. Growth led by domestic demand must continue to be our objective, and rigid deflationary prescriptions—the kind of one-size-fits-all prescriptions that Jean-Claude Milleron described—too often work against this objective.

Hand-crafted, tailored prescriptions that leave as much room as possible for an early and robust resumption of growth in domestic demand need to be our objective. Sustainable growth, on a global basis, cannot be achieved if we leave nation after nation little choice but to engage in competitive devaluations in order to hold on to export markets. Export markets are simply not ample enough, and the political reaction in the European Union (EU) and the United States is also an enormous threat to a regime that depends on export-led growth.

The fourth lesson is that government matters. The question is not whether government is too big or too small. It is a question of building—at both national and international levels—governmental capacity that can cope with a new set of questions. We have all described this world in terms of new architecture. The architecture of our public institutions has to adapt, improve, and adjust to match the changes in markets. Markets are powerful, often wonderful things, but they are not enough, and they are not always right. They overshoot. They overcorrect. They misprice public goods. They underinvest in human beings. And when they are wrong, they leave enormous costs in their wake.

In the last couple of days, we have talked a lot about competent government. Surely no one here would argue that incompetence is better, but we must underscore that competence is only part of what is necessary. Governments must have the capacity, they must have the authority, and they must have the tools to set and enforce rules for the global marketplace; rules that enforce workplace standards, protect the environment, assure financial transparency and integrity, and allow intervention when markets—especially financial markets—threaten political and even national stability. To insist that governments should always, as a matter of rule, eschew transaction taxes, forbear from using capital controls, and refrain from currency market interventions is too simple and wrong. In general, capital account liberalization is desirable. But the world does not operate “in general,” and the IMF should not change either its practice or its charter to insist that in each and every instance capital account liberalization should be required of member countries.

On a similar note, Maria Ramos and Santiago Levy made important points about timing and institutional competence. These points are even more significant when they are joined. Deregulating financial markets before the institutional capacity exists to oversee those institutions is a moral hazard of the first order. Encouraging and seeking direct investment before the capacity exists to ensure that health, safety, and environmental protections will be maintained, and workers’ rights protected, invites the very race to the bottom that Larry Summers cautioned against yesterday.

For the fifth lesson, let me return to the concerns of working people in the industrialized world. They fear that they, too, will become globalization’s victims as investors seek the most exploitable worker or the most degradable environment and, consequently, threaten to or appear ready to move work and wages away from high-standard countries. If we do not take U.S. President Bill Clinton’s admonition to the World Trade Organization seriously and seek in all available forums—domestic and multinational—to begin the difficult process of harmonizing standards upwards, I fear that the backlash we have seen in the United States and in the EU will intensify and potentially choke off this generation’s opportunity to make the global economy work for workers everywhere.

In a recent speech at the Council on Foreign Relations, AFL-CIO President John Sweeney said that across the world people understand that the global economy is a reality, but they are no longer passively accepting global rules that lead to the return of inhuman conditions, the erosion of opportunity, and the instability of untrammeled markets. The potential of the global market can be tapped only if its blessings are shared, its excesses are curbed, and its brutalities outlawed.

And, of course, that is the sixth and most important lesson. If we want equity, we need to insist on rules that promote it just as vigorously as we insist on rules that protect property. If we want social justice, we must be as ready to insist that tyrants open up the political process as we are to insist that they open up their financial marketplaces.

Vito Tanzi

I thought about what I could say that would be complementary to what was said earlier. I was intrigued by Anthony Atkinson’s statement about social norms and their impact on income distribution. Incidentally, the first time I saw some reference to social norms was not in Atkinson’s paper but in a paper presented at the last IMF conference on income distribution by Nicola Rossi and Andrea Brandolini.2 In that paper, the authors made the point that some institutions may determine, to a large extent, the income distribution of countries.

As I thought about it, I realized that many everyday institutions outside the government and outside the economy may have a major impact on income distribution. In the country where I was born, Italy, there is still, today, an institution called Mezzadria, originating from Roman times, which stipulates that the owner of the land keeps 50 percent of its output and the one who cultivates it receives 50 percent. There is no reason for this 50/50 split, it is just a practice inherited from the Romans that still prevails in some parts of Italy.

Other attitudes or customs affect, for example, employment. We are told that in Korea and Japan, employment for life is an institution. What about inherited jobs? I once thought that jobs were inherited only during the European Middle Ages, but that is not so. I learned a few minutes ago from Aníbal Cavaco Silva that if you own a pharmacy in Portugal, you simply pass it on to your children, and that a pharmacy generates tremendous income. In Italy, it goes further: if you are a pharmacist and die, and your child does not have a pharmacy degree, he or she has 10 years to complete a degree, and can then run the pharmacy and earn a substantial income. The same story applies to Italian notaries, who are settlement lawyers. These positions are extremely difficult to obtain. Often, a son or daughter inherits a notary’s post and its lucrative income from the father. And, in fact the last time I was in Italy, a friend of mine, who is not an economist, talked about rendite di posizione, that is, positional rents. I have never encountered this term in economics. These are rents that come with the position that you have somehow acquired or into which you have been born. As for rules on promotion, consider that, in some countries, seniority rules prevail, whereas in other countries, merit prevails. Again, these rules on seniority tremendously influence income distribution. What about the role of marriage? In some countries, the family of the bride must pay a dowry, and dowries clearly play an important role in income distribution and in the standard of living of those who marry. What about the choice of partners? In the town where I was born, people say that you should never marry somebody simply because she or he is rich, but you should always choose from the rich ones! If the rich marry the rich, and the poor marry the poor, the existing income distribution will tend to perpetuate itself.

Speaking of family matters, what about lending? In most traditional societies, lending within the extended family at zero interest is normal; it is unthinkable to charge interest to a close friend or to a relative. What about the decision on who goes to school? We know that in some societies boys go to school and girls stay home. What about birth order? We know that in some societies the first child is pushed forward, the second is not.

These are all examples of social norms that must have an impact on income distribution. If the economic environment were neutral and the government were neutral, these norms would, to a large extent, determine income distribution.

Social norms are stable in a traditional society, whereas in societies undergoing change, they are not. A major revolution or globalization can alter social norms. For example, immigrants venture into marriages with less consideration of the social, or even the economic, background of the persons they marry. One is much freer to choose a partner in the new country than in the country of origin. And when social norms begin to lose importance, changes in income distribution follow. Perhaps the income distribution that was acceptable earlier is no longer acceptable. At this point, the government often steps in.

As we have discussed, the role of government varies. Paradoxically, countries that need good government the most—the poor and/or the developing countries—are the ones that have the least ability to promote efficiency. Instead, the countries that need government the least—because, for example, their markets are well developed, information is more available, credit markets are good—are the ones that normally end up with the larger government, at least as measured in terms of spending and taxing.

I have written several papers on these different points, arguing that one cannot measure the size of government simply in terms of taxes and public spending. Analysts must take into account the regulatory role of the government and also its macroeconomic role. Minister Aninat spelled out the fact that macroeconomic stability is essential. Without macroeconomic stability, it is difficult to pursue economic policies. Grzegorz Kolodko also mentioned that in a crisis, what deteriorates first is macroeconomic stability. So stability is key. The size of the public sector measured in terms of spending and taxing is important. Alberto Alesina put out a graph with growth on one axis and the size of government on the other, and said that the government might be too small at a certain level and too large at the other level. You can have stability at 15 percent of spending, 25 percent of spending, or 50 percent of spending; thus, the role of government in that sense is very important. But so are the structural role and the regulatory role.

I would like to give you an example using the U.S. law on disability mentioned by Jean-Claude Milleron a few minutes ago. The government could protect the disabled by directly spending money. It could spend money and build ramps to give access to those in wheelchairs, or it could simply pass a law saying that all private activities, universities, and so on, must have ramps and elevators that can be accessed by the handicapped. By passing this regulation, the government is effectively shifting expenditures to the private sector. It is achieving the same purpose, the same objective, but it is doing it in an indirect way. So, even though all these factors that we have discussed are important, the role of government that attracts the most attention of economists is the taxing and spending role.

Economists now seem to feel that high tax and expenditure levels do not do much. It was not so in the past, when, for example, the marginal tax rate on capital income was 98 percent in the United Kingdom and around 90 percent in the United States. The view now is that it is better to have an efficient tax system with broad bases and reasonable rates.

We all agree that what we want to avoid through taxation is inequity. Sometimes tax systems single out particular groups or individuals for preferences, but equity issues emerge more often in the area of spending. In theory, one could spend money and lift the income of the very poor; in practice, this policy soon runs into problems. Over the years, my own analyses have convinced me that many of the greatest economic crimes were committed in the name of income redistribution. Many programs, often unjustified on almost any basis, were promoted as long as the government could claim that these programs were creating employment or were redistributing income in some undefined way. Very bad policies were enacted on this basis.

When major public spending is undertaken, two large problems frequently arise. One problem, spelled out by Alesina in his paper and mentioned by me in a paper I wrote in 1973, is that very often public spending is “hijacked” by powerful groups—that is, interest groups with great power over politicians manage to divert spending toward themselves. Thus, a program that started as a subsidy for the poor becomes a subsidy for the middle class or for everyone. Powerful groups may direct spending in ways that are not the most equity-enhancing. For example, secondary and tertiary education may enjoy much more attention than primary education. When I was working on Brazil about two decades ago, I remember reading that spending for each university student was about fifty times that for the average primary school student. Remembering that those who go to university normally are not the poor, it is easy to see why the education budget is misdirected.

Hijacking may be practiced by the providers of the services; in other words, school teachers who appropriate many of the benefits of educational spending, and doctors and nurses who do the same for health spending. Yesterday, Minister Aninat referred to the health care experience in Chile. There, about four or five years ago, the government became worried that the health system was not delivering what it should. So it decided to triple expenditures on health. After three or four years, an evaluation of the policy concluded that the only things that had changed were the salaries and other benefits of the doctors, nurses, and other hospital workers. Nothing else had changed: not the number of operations, not the output of health services.

This leads me to the final point that I want to make. We cannot judge whether a government is socially minded only on the basis of how much it is spending in a particular category. We need to know how the money is being spent. Unless we evaluate the institutional side of education, health, and so forth, we cannot reach a conclusion. This point is relevant to the accusation that is often directed at the IMF: that its programs have led to less spending for education and health in Africa. In fact, evidence shows that these expenditures have increased, not decreased. As discussed, however, the expenditure statistics do not tell us all that much. What we really need to focus on, in order to achieve change, is the political reform that will allow a certain level of scrutiny of a particular sector and the institutional reform. This is very difficult. It is easy to say we have to spend 5 percent more or 10 percent more for education and health; it is much harder to design a new curriculum and make sure that, if more money is spent, this increased spending will not be of benefit to the teacher but to the student, in terms of better performance and so forth. This is the real challenge.

The Beneficial Effect of Core Labor Standards on Economic Growth” (unpublished; Washington: AFL-CIO, 1998).

Rossi and Brandolini, “Income Distribution and Growth in Industrial Countries,” in Income Distribution and High-Quality Growth, ed. by Vito Tanzi and Ke-young Chu (Cambridge, Massachusetts: MIT Press, 1998).

    Other Resources Citing This Publication