iOS app Android app More

No Need to Oversimplify Poverty


Bill Easterly takes a complex problem, African poverty, and tries to reduce it to a single factor: "the consensus among most academic economists is that destructive governments rather than destructive geography explain the poverty of nations." This is a strange assertion. Geography and government policies both matter.

The idea that geography affects economic performance is an old one. Easterly and some other economists have taken a particular position about the relationship between geography and development. They too have recognized the high correlation of a country's poverty with being in a malaria-transmission region, or being landlocked, or being in an ecological zone leading to low food productivity. Those correlations after all are powerful, as recently shown again by Prof. William Nordhaus of Yale , in the Proceedings of the National Academy of Sciences.

Easterly and the others, however, have made a very unusual argument: yes, the correlations are there, but only for historical reasons. Bad geography two centuries ago led colonial powers to adopt exploitative political and economic institutions in the adversely affected regions. The adverse geography itself is no longer important, Easterly and his colleagues have claimed, but the adverse political and economic institutions live on nonetheless.

Specialists in many fields, inside economics and beyond, disagree strongly with this way of thinking. They believe that various dimensions of geography were important in the past, and are still important directly today. A region that suffers from malaria today, whether because of its tropical climate or the species of its mosquitoes, is hindered in development not only because it has poor institutions inherited from 1820, but because it has malaria, which kills and disables children, discourages public and private investments, and hinders economies in many other ways. A recent academic study by Kai Carstensen and Erich Gundlach, published in the World Bank Economic Review in 2006, made this point powerfully and directly: "After controlling for institutional quality, malaria prevalence is found to cause quantitatively negative effects on income."

Adam Smith, the pioneer of market economics, knew about the direct role of geography in affecting transport and trade all the way back in 1776. Even though the main purpose of the Wealth of Nations was to discuss the implications of economic policy and the division of labor on economic wealth, Smith also emphasized the role of geography in affecting national wealth. He cited Africa as a region suffering from especially high transport costs and therefore poor economic development:

There are in Africa none of those great inlets, such as the Baltic and Adriatic seas in Europe, the Mediterranean and Euxine [Black] seas in both Europe and Asia, and the gulfs of Arabia, Persia, India, Bengal, and Siam, in Asia, to carry maritime commerce into the interior parts of that great continent: and the great rivers of Africa are at too great a distance from one another to give occasion to any considerable inland navigation.

More recently, Paul Collier in The Bottom Billion, as well as my colleagues and I in several studies, have shown that being landlocked and far from the coast continues to be a major hindrance to participating in certain kinds of international trade, especially for manufacturing exports. This is seen powerfully in sectors like apparel and other assembly exports, where semi-processed goods are imported from abroad, further processed, and then re-exported. It is very disadvantageous in general to set up such export operations in landlocked countries or far from ports or major markets. Collier emphasizes the special difficulty for a poor landlocked country that is surrounded by poor neighbors. The nearby markets are insufficient to generate much trade, and participation in globe trade is very hard. As Collier puts it, landlocked countries are "hostages to their neighbors."

When it comes to agriculture, the geography-based problems of water are often paramount. The Green Revolution in Asia, which helped to trigger long-term economic growth in India and other countries, depended heavily on irrigation from the massive rivers systems of the region. In dry land areas with much higher costs of irrigation, the Green Revolution has been much harder to achieve. It may not be impossible, but the costs of entry for poor dry land countries are very high. The result can be low farm productivity, chronic rural poverty, and often no escape from extreme deprivation.

Bill Easterly seems not to want to be bothered by the details of irrigation-based versus rain-fed agriculture, or the types of mosquito species in Africa, or the implications of being landlocked for international trade, or the effects of a dry climate and high costs of irrigation on food production in poor rural areas. He seems to prefer a one-factor solution. He is not alone among economists in ignoring these "details," but this is nonetheless an odd approach scientifically, especially when dealing with a complex system like an economy. When biologists deal with a complex system like the human body, they know that thousands of particular causes - even one single change of base pair in the genetic code - can cause crippling diseases or deaths. They are attentive to the large number of possible causes and their interactions, rather than claiming as the ancients once did that disease is the result of an imbalance of the four bodily humours. The key is to use a "differential diagnosis" to ascertain the causes underlying a specific situation, rather than assuming that a problem like poverty has a single cause.

Sub-Saharan Africa faces a constellation of special challenges, with greater or lesser impact in different parts of the region, including: a climate and ecology especially burdened by infectious diseases such as p. falciparum malaria and other vector-borne diseases; a rain-fed agriculture, much of it in sub-humid or arid zones that are prone to drought; high overland transport costs, including the greatest number of landlocked countries of any continent and a relative paucity of ocean-navigable rivers; low population densities in rural areas, characteristic of many rainforest and dry ecosystems, which make rural infrastructure relatively expensive; a historical legacy of colonial rule in which the colonial powers left behind relatively little infrastructure; and of course challenges of bad governance like Mr. Mugabe of Zimbabwe. These challenges should be addressed forthrightly and in an integrated manner.

It's just bad science, and offensive, to read in Easterly's blog that I offer "a bizarre geographic theory of Africa's poverty and [am] oblivious to the bad governments that many courageous dissenters have fought at great sacrifice." The geographic factors are not "bizarre," and I have never been oblivious to the tremendous costs that can be caused by bad governments. In the End of Poverty (p. 194) I wrote "I visited Zimbabwe several times, and saw Robert Mugabe's depredations firsthand. Zimbabwe is a case where the traditional explanation of miserable rule is a sufficient explanation for a country's ills (although the nation no doubt suffers from other serious problems as well)." I have always pointed to geography and good governance as complementary factors, not a choice of one or the other.

Complex systems, in summary, require explanations that acknowledge complexity. An economy is affected by many factors: its proximity to trade, resource base, climate, history, social divisions, as well as government policies. A true economic science treats the economy with the care and sophistication that biologists treat an organism or that ecologists treat an ecosystem. Single-factor explanations for poverty take us back to pre-scientific realms and ways of thinking that are counterproductive for solving real problems.