Read an Excerpt

Brookings Trade Forum 2004

Brookings Institution Press

Chapter One

Globalization and its linkages to global poverty and inequality are the subjects of heated debates. These topics are also the focus of large and growing literatures, in economics and other disciplines. What have we learned from all of this research? Or put another way, what do we know, and what do we still not know? This broad topic clearly involves a myriad of more specific issues, and most, if not all of them, are quite complex. To make sense of the interwoven strands, a series of papers by researchers working actively on a variety of subtopics were commissioned for this project. These experts, as well as the formal commentators, were explicitly selected so as to cover a range of perspectives and views.

All of the contributors to the project grappled with difficult issues of definition. As discussed below, there are no agreed-upon definitions of the terms globalization, poverty, or inequality. However, clarifying the way such terms are used is critical, as studies based on different definitions of these terms yield starkly contrasting conclusions.

A number of other themes emerged from the analyses. Perhaps the most consistent was a note of caution against making definitive, across-the-board statements about the relationships between globalization, poverty, and inequality. The extensive discussions highlighted tremendous heterogeneity in outcomes as countries integrate into the global economy. This heterogeneity is driven as much by differences in countries' initial endowments and institutional capacity as it is by the globalization process per se. Another theme is the extent to which implicit value judgments are embedded in the measures and methods that are used to evaluate these relationships. For example, economists that study inequality typically focus on proportionate differences in income across countries and individuals, while critics of globalization are more concerned about absolute differences between individuals and countries-for example, the income gap between the rich and the poor. This results in starkly different visions of the relationship between globalization and inequality.

Four of the papers focus on measures of global welfare. These contributors examine traditional indicators of poverty and inequality, as well as explore broader indicators including measures of health, subjective well-being, and civil conflict. Their papers highlight that there are difficult issues of measurement inherent in each of these concepts. Reasonable people may disagree about which one (or ones) is most appropriate, often because of the value judgments embedded in the methodology used to construct them and, furthermore, because each of the measures appears to behave quite differently.

Three papers focus on the channels through which globalization may affect the world's poor. One of these does so using the lens of analytic trade models to explore the importance of cross-country differences in factor supplies versus productivities. Another provides an extensive review of the empirical literature on the implications of trade liberalization using microdata from country case studies. The third tackles a broad set of issues, including the influences of both domestic and international factors on the poor in low-income countries.

Authors of the final two essays were asked to comment broadly on what they had heard and to look forward. One of these emphasizes why inequality, and not just poverty, matters. The other takes a very long-term view, discussing implications of growth on poverty and the risks to that growth scenario.

HOW MUCH ARE THE WORLD'S POOR sharing in the gains that arise from economic growth, fueled by greater economic integration? In the second paper of the volume, Martin Ravallion asks why different sides of the ongoing debate about globalization and inequality give seemingly conflicting answers. Members of the antiglobalization movement claim that inequality between and within nations has been increased by globalization policies. In stark contrast, those who support globalization claim that it raises nations' incomes and that the poor benefit substantially. Ravallion's main thesis is that the different sides in this debate do not share the same values about what constitutes a just distribution of the gains from globalization. But value judgments are embedded in the standard measurement practices used to construct the factual claims made about what is happening to global inequality. He explores several differences in the value judgments made about distributive justice and examines their implications for the globalization debate. While recognizing that knowledgeable individuals might reasonably make different choices, he offers some tentative conclusions about preferred measures of inequality. More generally, he argues that explicit attention to the values that underlie the seemingly conflicting evidence on what has happened to poverty and inequality is essential for moving the debate forward.

The paper begins with an overview of some of the main empirical findings about what has happened to inequality and poverty, relevant for the globalization debate. Inequality can be measured in a variety of ways. One might focus on inequality among people in the world. Noting that the results here depend on the study and time period, Ravallion concludes that there has been no clear trend toward either increasing or decreasing global inequality. One might focus on inequality within countries. Here, Ravallion discusses the common finding that growth is distributively neutral on average. In other words, among growing economies, inequality (measured by the usual Gini index) rises about half of the time and falls about half of the time. This is true for poor as well as for rich countries. Thus the well-known Kuznets hypothesis has not been borne out. However, this finding is based on measures of relative inequality, as discussed further on. It is not true if inequality is measured so as to depend on absolute differences in living standards instead of income ratios relative to the mean. The absolute gap between the rich and the poor tends to rise in growing countries.

Perhaps most controversial has been the question of what has happened to inequality between countries. It is true that, on average, poor countries are not catching up with rich countries. Over the past 100 years, countries that were initially poor have grown relatively slowly. In addition, the average income of the world's richest country was roughly ten times that of the world's poorest country in the late nineteenth century, but today that income is closer to sixty times greater. Taking each country as one observation, critics of globalization have pointed to a sharp increase in inequality between countries beginning in the 1980s. However, if countries are weighted by population, the data show a trend toward a decline in the between-country component of inequality since the mid-1970s. This is the approach preferred by globalization proponents. The exchange about which approach should be used has often been heated. As discussed below, Ravallion argues that this is one area in which value judgments are implicit in the arguments on both sides.

The paper also surveys some of the empirical findings about global poverty. One frequently used indicator of absolute poverty is the share of the population with income below a fixed real value (such as one dollar a day). By this measure, poverty has clearly fallen. In developing countries the percentage of the population classified as poor fell from 40 percent in 1981 to 21 percent in 2001, and the number of poor fell from 1.5 billion to 1.1 billion. But it is far from clear that this decline in the number of poor is attributable to globalization. If one excludes China, the number of poor has remained roughly stable over this period. And the sharp decline in China's poverty incidence appears more closely linked to the post-1978 decollectivization of agriculture since it occurred during the early 1980s-before China's external trade liberalization.

Ravallion stresses that it is extremely difficult to assess what might have happened in the absence of globalization, raising questions about the claims (by either side) about precisely what role globalization played. Furthermore, he cautions against the types of broad generalizations that too frequently are made. For instance, the finding that growth tends to be associated with reductions in poverty does not imply that all growth-promoting policies are necessarily good for the world's poor.

The paper highlights three differences in the value judgments made about distributive justice that underlie the globalization debate. The first concerns the dispute over what has happened to inequality between countries. The value judgment here relates to whether one should weight countries equally or people equally when assessing distributional outcomes. Weighting countries equally makes sense for the large macroeconomic literature that studies the empirics of growth and distribution, because each country is considered a draw from the urn of possible combinations of country policy, shocks, initial conditions, and outcomes. But if it is the impact of policies and performance that economists and policymakers care about, then why use equal country weights, thereby giving smaller weight to those living in populous countries? And it seems inconsistent to use population weights in measuring within-country but not between-country inequality. However, population-weighted measures ignore that individual welfare is likely to reflect country of residence as well as own income. Furthermore, population weights make the inequality indicator very sensitive to outcomes and to possible measurement errors in the most populous countries (China and India). Ravallion concludes that neither weighting method is ideal but that it is hard to defend weighting countries equally when the objective is to make normative judgments about inequality.

The second difference in concepts of inequality relates to how much weight one should attach to the way average gains from reform vary with income (horizontal impact) versus the differences in impacts found at a given level of income (vertical impact). The pro-globalization side has tended to focus on the former aggregate measure. The antiglobalization side has pointed to the losers among the poor and those vulnerable to poverty-often to the point of ignoring the aggregate outcomes. Although conventional measures incorporate both vertical and horizontal inequality, Ravallion discusses a number of conceptual and measurement issues and concludes that standard indicators may indeed underweight the horizontal component. Measured income may not adequately capture horizontal inequities. Differential impacts on initially similar households may be viewed as unfair, fueling social conflict. And conventional indicators do not reflect "churning" as some people gain while others lose, swapping places in the income distribution.

The third issue concerns the distinction between relative inequality, which depends solely on proportionate differences in incomes, versus absolute inequality, which depends on the absolute differences-the "income gap between rich and poor." Ravallion notes that virtually all the research by economists on world inequality has used the former concept, and this has become embedded in popular writings supporting globalization. In contrast, critics of globalization often seem more concerned about absolute inequality. Arguably it is absolute inequality that most people see in their daily lives and that motivates their concerns about distributive justice. One's preference between the two concepts reflects one's (implicit) value judgment about what is a fair division of the gains from growth. Ravallion argues that both sides need to be clearer about which concept they are using and more aware of the underlying value differences.

Thus Ravallion argues that the empirical "facts" in contention do not stem solely from objective data on incomes, prices, and so on. They also depend on value judgments that have become embedded in the methodological details underlying statements about what is happening to inequality in the world. Bringing the value differences to the surface will be one critical step in moving the globalization debate forward.

Both discussants agreed with the main points made in the paper. Erik Thorbecke extended the analysis in a number of directions. In particular, he argued that "churning" may be even more important than Ravallion's paper suggests since the welfare losses of those who are hurt will exceed the gains of those whose income rise. He argued that we need to better understand how sensitive inequality measures such as the Gini index are to measurement errors in the data and to the underlying assumptions used to construct them. Finally, he discussed a variety of difficulties with using income as a measure of welfare, suggesting that a range of indicators should be considered, including those based on health and other outcomes.

Lant Pritchett argued against focusing on how one concept he sees as poorly defined (globalization) affects another (poverty). First, he disagreed that the central driving phenomenon of the period in which we live should be labeled "globalization." He argued that the past fifty years have seen a proliferation of sovereign states, significant technological improvements reducing transport costs, liberalization of restrictions on the cross-border movements of merchandise in some countries, and binding restrictions on the ability of labor to move across borders. In his view, this is not globalization. Furthermore, he sees no reason for strong preconceptions about how these developments should affect global inequality. Finally, he took issue with the usage of a single, low poverty line (for example, a dollar a day) because we do care about income gains above that line. He argued instead for use of a higher poverty line or, better yet, for focus on the distribution.

IN THE SECOND PAPER, William Easterly examines the channels through which globalization affects inequality between and within countries. He defines globalization as the movement across international borders of goods and factors of production. Conventional analyses of the consequences of globalization for inequality examine the effects of trade and factor flows on the returns to factors (labor, capital, and land), on accumulation of factors, and on national income. However, the central point of his paper is that different economic models give very different predictions about exactly what the effects of these flows are likely to be.

Easterly focuses on the two main alternative models.

Continues...

---

Excerpted from Brookings Trade Forum 2004 Copyright © 2004 by Brookings Institution Press . Excerpted by permission.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.

---