New World Math

European Economic Review, Vol. 51, Issue 5, July 2007

Not long ago, the tools of the trade for economists were fairly basic. They had little more than hand-held calculators at their disposal to calculate the most complicated sets of variables. That meant they had to be fairly certain about the relevance and quality of the data they were analyzing. Fast-forward to today, and every economics graduate student has massive computing power available at the push of a button. Combined with the availability of a whole host of dubious databases, there has been a massive proliferation of supposedly scientific research on nearly every aspect of countries’ political, economic, and social standings.

The pitfalls of this burgeoning academic research are illustrated by a recent article in the European Economic Review. In an article titled "Income Inequality and Colonialism," economist Luis Angeles of the University of Glasgow asserts that the degree of colonialism is the major explanation behind the differing levels of inequality within countries.

He divides the New World into three categories: "New Europes" are Australia, Canada, the United States, and New Zealand, where European immigrants eventually became the majority of the population; "settler colonies" are places where those of European descent made up 10 to 30 percent of the population, such as Latin America, the Caribbean, and parts of southern and northern Africa; and "peasant colonies" are places like West and East Africa and Southeast Asia, where Caucasians were only a small minority, usually less than 1 percent. Angeles then analyzes the Gini coefficient of different countries — economists’ favorite tool for measuring how evenly income is dispersed across a society (the higher the number, the more unequal) — and observes that income inequality has historically been much higher in settler colonies than in New Europe or in peasant colonies.

Angeles’s thesis has an elegantly tidy appeal, its implication being that racial homogeneity leads to more equality. But his premise is shaky from the outset. For starters, his classification of countries by degree of colonialism is sloppy, at best. Angeles’s New Europe consists mainly of countries where white colonists arrived in numbers large enough to exterminate or expropriate the indigenous and aboriginal populations so that those countries became primarily white. But why exclude Argentina and Chile from this category, since their populations are dominated by Spanish descendants? Angeles also classifies countries such as Kenya and Zimbabwe as peasant colonies, despite the fact that most Africans would consider them to be settler colonies.

Angeles’s choice of statistics is equally questionable. Gini coefficients are not truly comparable across countries, as the World Bank itself acknowledges. Some are based on the distribution of income, while others are based on consumption expenditure. Some use individuals as a unit of observation, while others look at households. In short, the statistics that are available simply make it impossible for an apples-to-apples comparison.

Not only does Angeles’s thesis have a weak foundation, but it is also not particularly persuasive. From a historical perspective, other sources of inequality besides European colonialism are clearly relevant. Take, for example, the agrarian civilizations of Eurasia: China, India, and Mesopotamia. These regions were sandwiched between two vast areas of nomadic pastoralism: the steppes of Central Asia and the Arabian Desert. To protect themselves against the periodic raids that nomads mounted against them, these rich sedentary civilizations required specialists in warfare. That necessarily led to social stratification, with the warrior class distinct from the peasants tilling the soil. Another source of inequality in these civilizations was that labor was scarce. That meant that peasants would have to be coerced to hand over the rice and wheat they cultivated to feed the warriors and priests who lived in the towns. Various unequal social institutions like serfdom, slavery, and the caste system were descendants of these trends.

By contrast, countries of New Europe did not face this perpetual military threat; Australia, New Zealand, and North America were all protected by vast oceans. Moreover, the temperate agriculture they practiced allowed white settlers and their descendants to expand their family farms with relatively little social stratification. With the notable exception of the American South, these countries did not rely on coerced labor to survive and could maintain more equal societies.

But never mind these historical lapses. Let’s assume that Angeles’s primary data are robust and that his conclusions are correct. So what? Does that mean colonialism was bad for the developing world? The evidence suggests that this is not the case. As Angeles himself recognizes, even if colonialism resulted in highly unequal societies in settler colonies, that does not imply that they had poorer economic growth records than places that were more homogeneous. Nor does it mean that the impoverished majority of the population in a settler colony would be poorer in absolute terms than the population of a peasant economy. Take Brazil and Mexico, archetypical settler economies by Angeles’s definition, where the white settler population ranged from 10 to 30 percent. Their economies grew at an annual rate of more than 4 and 6 percent, respectively, from 1950 to 1980, with substantial reductions in absolute poverty. Compare that with the peasant colony of India, where the white population was less than 1 percent. Its economy grew at only 3.5 percent a year during the same period, with no decrease in absolute poverty.

Certainly, there are some dogmatic egalitarians in the world who would cheerfully trade growth or the reduction of absolute poverty for equality. But many people would be happy to look upon colonialism favorably — if it raised growth rates and reduced absolute poverty — irrespective of how it distributed wealth across society. Angeles provides no reasons for taking a jaundiced view of the economic record of empires and colonialism in general. He ignores the positive effects of empires on the well-being of the majority of the population; dismisses the order that they promoted in disorderly societies; and chooses not to tackle how they might have contributed to the integration of the developing world into the global economy.

There is little danger that Angeles’s arguments will have any effect on policy. But the fact that such studies can masquerade as research should raise concern. The problem with such supposedly scientific studies is that they begin by offering dubious suppositions, build them further with suspect data, and derive misleading and ill-informed conclusions. But they may prove one thing: Sometimes it is better if economists just work with calculators.