Methodological and Data Challenges to Identifying the Impacts of Globalization and Liberalization on Inequality

Globalization (the increasing degree of economic interaction among countries) and liberalization (reductions in government intervention in markets, partly with respect to international interaction but also more generally) are two of the defining features of the last couple of decades. Both have given rise to contentious debate, with views ranging from the very optimistic to the very sceptical. In this paper, Albert Berry reviews the evidence on how the two trends have affected inequality (and hence poverty) at the world level and within countries. The notable absence of consensus on these effects reflects both the dearth of adequate quantitative information and the lack of and difficulty in the analysis of the causal links among the phenomena. These weaknesses have interacted with the strongly held prejudices of many advocates and critics of globalization and liberalization (G&L) to produce many weak analyses and the extreme range of views. Nearly all of the strong policy recommendations of the last couple of decades have been based on analyses that were at worst weak and at best needful of considerable refinement.

Different groups of analysts and different organizations, especially during the earlier years of this phase of development, tended to reach systematically different conclusions. Within the United Nations community, broadly defined, the International Monetary Fund and the World Bank have been most inclined to view G&L as desirable phenomena, to be encouraged (which they have vigorously done). The other UN agencies have been less enthusiastic. Recent work carried out at the Economic Commission for Latin America and the Caribbean has been among the more balanced and higher quality work done thus far.

For the period since around 1980, which roughly corresponds to the new era of globalization and to a wave of liberalization, some conclusions on inequality and poverty trends seem more or less secure. First, this period, and especially the 1990s, has been characterized by a within-country tendency toward increasing inequality of income distribution among persons. Second, in spite of that intracountry pattern, the world distribution of income among persons has been relatively stable (as is true for the last half-century, in fact), probably declining slightly. Third, since average income growth for the poorest deciles of the world distribution was substantial, poverty levels have continued to decline, though the improvement of this variable was small in the 1990s when a low poverty line (for example, $500 in 1985 US dollars, or less) is used. The surprising fact that world inequality has fallen, even with a general tendency for intracountry inequality to rise, is explained by the strong growth of the world’s two most populous low-income nations, China and India. Excluding these two countries, there has been a marked increase in inequality worldwide and no poverty reduction. Other worrisome aspects of the recent record are that the 1990s were less good than the 1980s, and growth continued to be slow in sub-Saharan Africa, increasingly the locus of the world’s poor people.

Analysis of the possible links between globalization, liberalization and inequality/poverty is far too limited to provide any reliable conclusions at this point. Such inquiry is farther advanced for Latin America than elsewhere. The range of likely average impact of G&L on country-level inequality goes from mildly negative to strongly so over the short to medium run. No studies have thrown much light on the long-run impact due to the recentness of the liberalizing reforms and the current wave of globalization. A high multiple of the amount of research carried out thus far would be necessary to provide reliable interpretations of these links.

While the effects of G&L on inequality and poverty remain murky, a reasonable guess would be that these phenomena cannot be credited either with a major positive contribution or a major negative one at the world level, though their impacts in certain specific countries are probably much greater. Over recent decades, growth of per capita income in the low-income countries has been the key determinant of changes in world inequality and poverty; thus the most important effects of G&L could be on the growth rates of those countries. If G&L have in fact contributed significantly to declining world poverty over the last couple of decades, the only likely route through which they will have done so would be a positive impact on growth in China and India, a matter which requires detailed analysis of the experiences of those two countries.

Future impacts of G&L could be either more or less positive than past ones. Possibly the “convergence” often postulated as an effect of globalization will eventually occur. Two current causes for doubt about its early arrival are the poor performance of sub-Saharan Africa, and the frequent observation (especially in Latin America) that globalization has widened the technology and productivity gaps between those operating in world markets and everyone else, without raising the employment share of the former.