The new geography of global income inequality.

Anderson, Kathryn

The New Geography of Global Income Inequality Glenn Firebaugh Harvard University Press: Cambridge, MA, 2003, 257pp.

Comparative Economic Studies (2007) 49, 326-329. doi:10.1057/palgrave.ces.8100197

Globalisation and technological innovations in communication and transportation promoted rapid growth and economic opportunity for many people in the world. Some countries that stagnated in the 1970s and 1980s flourished in the 1990s, with economic growth at unprecedented levels. The economies of many other countries stagnated in the 1990s, and political discontent increased. The collapse of the Doha Round, protests in the United Nations, calls for military interventions, and the increase in support for socialist Latin American politicians are all indicators of the economic bifurcation that we see today. In his very careful analysis, Firebaugh addresses these sensitive issues and brings together the most recent data, methodologies, and literature to measure and explain the recent development of the global distribution of income.

Firebaugh's book focuses on the measurement of global income inequality and the change in the between- and within-country components of income inequality during the last 20 years. His work complements recent empirical work on global income inequality by Sala-i-Martin (2002a), Li et al. (1998), and others and the new theory of economic geography (Fujita et al., 1999).

After reviewing the literature on global inequality in the 19th and 20th centuries, he presents new evidence on the change in inequality since 1990. Over the last two centuries, the world experienced a large increase in real average per capita income; on an average, people in all regions of the world are economically better off today than they were in the 19th and early 20th centuries. At the same time, the distribution of income changed dramatically. Rich nations disproportionately benefited from real income growth, and overall global inequality worsened until the middle of the 20th century. Between-nation inequality increased and within-nation inequality decreased during this period of time. In the second half of the 20th century, inequality between countries stabilized until about 1990 and declined thereafter. Within-nation inequality increased, but Firebaugh estimates an overall improvement in global inequality in the late 1990s. He argues that the causes of these changes in inequality were industrialisation, the movement out of low-income agriculture in the largest lower income nations such as China, and the rise of the information-based economy.

Firebaugh's conclusions are dependent on whether the measure of between-nation inequality is unweighted or weighted by population share. The unweighted Theil index of between-nation inequality indicates that income inequality between nations in the 1990s increased significantly. More states of all sizes fell behind the global average during the 1990s than rose above it. Many of the economies that did not benefit from globalisation were in sub-Saharan Africa. If the measure of between-national inequality is calculated using population weights--so that inequality is estimated among persons--then he finds between-country inequality fell in the 1990s. Asia--especially China and India--is the region responsible for this improvement in between-country inequality. Because of the Asia turnaround, relatively more people in the world apparently gained than lost from globalisation. Firebaugh estimates that if Asia had experienced the same economic trend in the 1990s as Africa, income inequality between nations would have increased using weighted inequality measures. Because of the disparity in the economic progress of large and small low-income countries, it is not surprising that we see so much dissension among nations in trade negotiations, global economic policy, and foreign affairs.

Firebaugh takes great care in the measurement of between-nation inequality over time, and he presents a variety of robustness checks that make his results even more compelling. He compares the trends in three inequality measures: Theil, mean logarithmic deviation or entropy measure, and the Gini coefficient. He uses the best and most complete macroeconomic data available to date and presents the reader with a clear discussion of the problems inherent in cross-country comparisons. His preferred estimate of income is adjusted for purchasing power parity (PPP) or world prices, but for comparison he presents results using the United Nations' Human Development Index. His basic conclusions about the patterns of weighted and unweighted between-nation inequality are robust to most of these measurement issues, but the sizes of the effects (point and interval) are quite sensitive to measurement. Three excellent chapters comprise the best, non-technical discussion I have read on inequality measurement and are a must-read for those unfamiliar with the concepts.

Firebaugh includes a thorough, non-technical discussion of the possible causes of the inequality changes he has measured. His conclusions about the role of manufacturing growth in income inequality are supported by reason and data on relative sectoral growth between and within countries. Although he does not model income growth formally or use multivariate statistical analysis to measure the relative importance of manufacturing expansion for income inequality, his descriptive evidence is convincing. In the future, a more controlled evaluation of the causal paths should be pursued.

The weakest part of the analysis in this book is the measurement of within-country inequality. Firebaugh uses quintile data on income to measure the within-country variation in income, but this approach assumes that all persons within the quintile have the same level of income. In a recent paper, Sala-i-Martin (2002b) estimates country-specific income distributions using kernel density estimation, which allows for variation in income within quintiles, and he finds larger within-country inequality when variation within quintiles is allowed. The general patterns over time of declining between-country inequality and increasing within-country inequality are evident in the quintile and kernel density estimates of inequality, but the measure of within-country inequality is smaller in the 1990s when the quintile estimate is used. This comparison suggests that Firebaugh's within-country inequality and overall inequality measures may be biased downward.

Firebaugh's estimates are also sensitive to price assumptions--foreign exchange rates (FX) or PPP. He rejects the exchange rate measure of national income because exchange rates value only goods that are traded across countries and ignore the purchasing power of untraded goods. However, PPP-adjusted inequality measures are also biased because the standard PPP conversion as in the Penn World Tables ignores the ability of poor consumers to substitute towards cheaper goods and services in the local economy. Dowrick and Akmal (2005) compare estimates of global inequality on a variety of measures in the 1990s using FX, PPP as in the Penn World Tables, and an intermediate (Afriat) method that they label 'true income'. In comparison to the 'true income' inequality, they find a significant downward bias in the measures of overall inequality using PPP-adjusted income and an upward bias in FX-adjusted inequality. They also find that the magnitude of the substitution bias inherent in PPP increased over time as the price structures across countries became less similar. The Dowrick and Akmal results are not definitive but do illustrate the problems inherent in cross-country comparisons of economic well being.

In summary, The New Geography of Global Income Inequality is an excellent study of the global trends in income inequality. The results are provocative and raise many questions about future economic development. The book should be required reading for anyone interested in the measurement of global inequality; it complements the new theories of economic geography and recent empirical research on spatial inequality and growth. It should be on reading lists for undergraduate and graduate economics courses in labour and human resource economics and economic development. I recommend it to all of my students who are interested in the study of global inequality.

Kathryn Anderson

Vanderbilt University, Nashville, Tennessee, USA

REFERENCES

Dowrick, S and Akmal, M. 2005: Contradictory trends in global income inequality: A tale of two biases. Review of Income and Wealth 51(2): 201-229.

Fujita, M, Krugman, P and Venables, Anthony J. 1999: The Spatial Economy: Cities, Regions, and International Trade. The MIT Press: Cambridge, MA.

Li, H, Squire, L and Zou, H-f. 1998: Explaining international and intertemporal variations in income inequality. Economic Journal 108(446): 26-43.

Sala-i-Martin, X. 2002a: The "disturbing' rise off global income inequality. National Bureau of Economic Research Working Paper #8904: Cambridge, MA.

Sala-i-Martin, X. 2002b: The world distribution of income (estimated from individual country distributions). National Bureau of Economic Research Working Paper #8933: Cambridge, MA.