This weekend, Gary Becker and Richard Posner — two of the smartest and most prolific academics in the nation — wrote about food prices in developing countries on their blog. As always, their analysis is excellent and their writing is top-notch, so what I discuss in this post is really a quibble.
Both Becker and Posner appear to believe that rising food prices have caused political unrest in the Middle East and North Africa (MENA). Becker first writes:
“Higher prices of foods mainly hurt the poor since poor countries and poorer families within a given country spend a much larger fraction of their incomes on foods than do rich countries, and then richer families within a country. (…)
This simple arithmetic explains why the current rapid price increase in foods and other commodities, and past large increases in these prices, often caused great distress among poor families of Africa, Asia, and elsewhere in the world. This distress led to food riots in many countries, and other protests by the poor against the large rise in their cost of living. Governments responded to these protests in various ways that often lowered the cost of food to consumers, but usually at the expense of inducing inefficient behavior by farmers and consumers, and often even at the expense of the poor.”
While Becker only alludes to food prices causing political unrest in the MENA, Posner establishes a more direct causal link when he writes:
“The concentration of population in a city, especially in a nation’s capital, makes urban residents a potential threat to political stability. It does this by facilitating large-scale demonstrations, riots, and other mob action (not only because there are many people to fill the streets, but also because information spreads very rapidly in a city, facilitating the coordination of a large group of people), with consequences—which can include bringing down the government—that we are seeing in the Middle East and North Africa today. Hence it makes political sense for a government to provide more generous food subsidies to urban than to rural residents even though the latter are needier.
It appears that rapidly rising food prices have been a major factor in the recent and continuing unrest in the Arab countries. As Becker emphasizes, food prices are a big part of the budget of families in poor countries, even of urban residents, with their higher incomes. In fact the demonstrators who brought down the Tunisian and Egyptian governments were complaining vociferously about surging food prices, though they had other serious complaints as well.”
As much as I suspect that there is a causal link between food prices and political instability (which is not the same as believing that food prices are the only cause of political unrest in the MENA), I would never be able to claim that the relationship is causal. This is in line with my post yesterday about confirmation bias. The problem with the relationship between food prices and political unrest is that (i) we do not observe the counterfactual, i.e., a world in which everything else would have remained the same but in which food prices would not have risen sharply in the second half of 2010; and (ii) there are just too many factors that confound the relationship between food prices and political instability.
This does not mean that the question is not worthy of being investigated with the best empirical tools available. Indeed, abandoning that question altogether would be akin to the drunk who lost his car keys and is searching for them under the streetlight because that is the only place he can see. But one must be very careful when making causal statements, since there is little we do know with certainty.
(To take a completely and extremely agnostic position on causality, I don’t really “know” that I am not going to fall into the sky when I walk out of the house tomorrow morning. Somewhere along the way, I had to take a leap of faith in deciding to “know” about gravity. Interested readers should read David Hume’s A Treatise of Human Nature.)
To their credit — and I wish to state that Becker and Posner’s academic writings have been very influential on my thinking — neither Becker nor Posner is an empiricist, so they can both be forgiven for not approaching this like someone who came of age during the “credibility revolution” in economics.