A quarterly magazine of the IMF

December 2000, Volume 37, Number 4

Dani Rodrik is Professor of International Economy at the John F. Kennedy School of Government, Harvard University, and the research coordinator for the Group of 24. His research covers international economics, economic development, and political economy. He has published numerous articles and books, including Has Globalization Gone Too Far? (1997) and The New Global Economy and Developing Countries: Making Openness Work (1999).

Should governments pursue economic growth first and foremost, or should they focus on poverty reduction?

Recent debate on this question has generated more heat than light because it has become embroiled in broader political controversies on globalization and the impact of World Bank and IMF programs on developing economies. Experience has shown, however, that growth and poverty reduction go largely hand in hand. The questions we should be asking are (1) what kinds of policies lead to both growth and poverty reduction? and (2) would a poverty focus facilitate the adoption of such policies?

Before attempting to answer these difficult questions, we need to ask some easier ones.

Does growth benefit the poor?

Yes, in general. The absolute number of people living in poverty has dropped in all of the developing countries that have sustained rapid growth over the past few decades.

In theory, a country could enjoy a high average growth rate without any benefit to its poorest households if income disparities grew significantly—in other words, if the rich got richer while the incomes of the poor stagnated or declined. This outcome is rare, however; income distribution (for example, as measured by the Gini coefficient, which ranges from 0 for absolute equality to 1 for absolute inequality) tends to be stable over time within countries.

Moreover, to the extent that income distribution changes, its relationship to economic growth varies from country to country. Growth has been accompanied by greater equality of income in Bangladesh, Egypt, and Taiwan Province of China, for example, but by greater inequality in Chile, China, and Poland. This suggests that the magnitude of the poverty reduction payoff from growth depends, in part, on a country's specific circumstances and policies.

Is poverty reduction good for growth?

Again, yes, in general. It is hard to think of countries where a large decrease in the absolute number of people living in poverty has not been accompanied by faster growth.

Just as we can imagine growth occurring without any reduction of poverty, we can also imagine a strategy of poverty reduction that relies exclusively on redistributing wealth from the rich and the middle classes to the poor. In principle, a country pursuing redistributive policies could reduce poverty even if its total income did not grow. But we would be hard pressed to find real-world examples. Policies that increase the incomes of the poor—such as investments in primary education, rural infrastructure, health, and nutrition—tend to enhance the productive capacity of the whole economy, boosting the incomes of all groups.

What does a high correlation between growth and the incomes of the poor tell us?

Practically nothing, for the reasons outlined above. All it shows is that income distribution tends to be stable and fairly unresponsive to policy changes.

Moreover, a strong correlation between economic growth and poverty reduction is compatible with both of the following arguments: (1) only policies that target growth can reduce poverty; and (2) only policies that reduce poverty can boost overall economic growth. Therefore, the observed correlation between growth and poverty reduction tells us little of interest as far as policy choices and priorities are concerned.

The real question is not whether growth is good for poverty reduction or vice versa, but whether the well-being of the poor should enter into the equation as an independent determinant of policy choices, in addition to the usual focus on macroeconomic stability, microeconomic efficiency, and institutional quality.

Should economic reform strategies have a poverty focus?

Yes, for at least three reasons.

First, in considering social welfare, most people, and democratically elected governments in particular, would give more weight to the well-being of the poor than to that of the rich. The economy's growth rate is not a sufficient statistic for making welfare evaluations because it ignores not only the level of income but also its distribution. A policy that increases the income of the poor by one rupee can be worthwhile at the margin, even if it costs the rest of society more than a rupee. From this perspective, it may be entirely rational and proper for a government considering two competing growth strategies to choose the one that has a greater potential payoff for the poor, even if its impact on overall growth is less assured.

Second, even if the welfare of the poor does not receive extra weight, interventions aimed at helping the poor may still be the most effective way to raise average incomes. Poverty is naturally associated with market imperfections and incompleteness. The poor remain poor because they cannot borrow against future earnings to invest in education, skills, new crops, and entrepreneurial activities. They are cut off from economic activity because they are deprived of many collective goods (such as property rights, public safety, and infrastructure) and lack information about market opportunities. It is a standard tenet of economic theory that raising real average incomes requires interventions designed to close gaps between private and social costs. There will be a preponderance of such opportunities where there is a preponderance of poverty.

Third, focusing on poverty is also warranted from the perspective of a broader, capabilities-oriented approach to development. An exclusive focus on consumption or income levels constitutes too narrow an approach to development. As Nobel Laureate Amartya Sen has emphasized, the overarching goal of development is to maximize people's ability to lead the kind of life they value. The poor face the greatest hurdles in this area and are therefore the most deserving of urgent policy attention.

Do priorities matter?

Yes, a lot. Policymakers make choices all the time. The lens through which they perceive development will profoundly affect the outcomes. Keeping poverty in sight ensures that their priorities are not distorted. Consider some illustrative trade-offs.

Fiscal policy. How should a government resolve the trade-off between higher spending on poverty-related projects (rural infrastructure, say) and the need for tight fiscal policies? Should it view incurring the disapproval of financial markets as the price it must pay for better irrigation? How should it allocate its educational budget? Should more be spent on building primary schools in rural areas or on training bank auditors and accountants?

Market liberalization. Should the government maintain price controls on food crops, even if such controls distort resource allocation in the economy? Should it remove capital controls on the balance of payments, even if that means fiscal resources will be tied up in holding additional foreign reserves—resources that could otherwise be used to finance a social fund?

Institutional reform. How should the government design its anticorruption strategy? Should it target the large-scale corruption that foreign investors complain about or the petty corruption in the police and judicial systems that affects ordinary citizens? Should legal reform focus on trade and foreign investment or on domestic problems? Whose property rights should receive priority, those of peasants or of foreign patent holders? Should the government pursue land reform, even if it threatens politically powerful groups?

As these examples illustrate, in practice, even the standard, growth-oriented desiderata of macroeconomic stability, microeconomic efficiency, and institutional reform leave considerable room for maneuver. Governments can use this room to better or worse effect. A poverty focus helps ensure that the relevant trade-offs are considered explicitly.

How much do we know about policy impacts?

Not nearly enough. We have evidence that land reforms, appropriately targeted price reforms, and certain types of health and education expenditures benefit the poor, but we are uncertain about many things. It is one thing to say that development strategies should have a poverty focus, another to identify the relevant policies.

But this is not a strike against poverty-oriented programs, since we are equally uncertain about growth-oriented programs. The uncomfortable reality is that our knowledge about the kinds of policies that stimulate growth remains limited. We know that large fiscal and macroeconomic imbalances are bad for growth. We know that "good" institutions are important, even though we have very little idea about how countries can acquire them. And, despite a voluminous literature on the subject, we know next to nothing about the kinds of trade policies that are most conducive to growth.

The policies that promote growth are probably not that different from those that target the poor directly, for the reasons just discussed. These policies are likely to vary considerably depending on institutional context, making it difficult to generalize. The debate on growth versus poverty reduction is a meaningless debate that diverts attention from the questions that should be our real focus: what works, how, and under what circumstances?