Perils of Foreign Aid

In spite of the best efforts of Columbia University’s Jeffrey Sachs and U2’s Bono, the idea of using foreign aid to help poor countries keeps taking it on the chin. And it is not just the pull-yourself-up-by-your-bootstraps Republicans either; academic economists—other than Sachs—haven’t lately found anything nice to say about foreign aid. (Academics can be pretty snarky; which is why I should perhaps have begun this post with “because of” instead of “in spite of.”) In the near future, I hope to discuss recent work on foreign aid by two IMF economists. In today’s post, I will discuss a recent paper by Francesco Caselli of LSE and James Feyrer of Dartmouth: “The Marginal Product of Capital,” National Bureau of Economic Research, August 2005, http://www.nber.org/papers/w11551.

Caselli and Feyrer find that you can get a lot more output out of one unit of physical capital in a typical poor country than in a typical rich country. (Using the juicy piece of jargon in the paper’s title, the marginal product of capital is higher in poor countries than in rich countries.) One would have thought that this would have provoked a large flow of capital from rich to poor countries. After all, people in rich countries who have savings could take their dollars, euros, and yens to the poor countries and buy machines, factory buildings, land, etc.—collectively called physical capital—which would then yield huge outputs, as one would expect from the (genuinely) high marginal products of capital in the poor countries.

Unfortunately, it doesn’t work out that way. We don’t see a large flow of private capital from rich to poor countries despite the latter’s high marginal products of capital. The question then is: Why? Why are the poor countries so starved of capital despite their high marginal products of capital?

Caselli and Feyrer claim to have found an answer: capital goods—machines, factory buildings, etc.—are way too expensive in poor countries. This prevents the financial return to capital in poor countries from rising above rich-country levels despite the high marginal product of capital in poor countries. Consequently, you don’t see private capital flowing from rich to poor countries.

Let’s consider a simple made-up example. Suppose $10 buys you one unit of physical capital—say, a machine of some sort—in rich countries and this additional unit of capital yields $2 of additional output—this, recall, is the marginal product of capital—in every subsequent year.

Now, suppose a unit of physical capital in a poor country yields $4 of additional output annually—that is, the poor-country marginal product of capital is twice the rich-country marginal product of capital. But suppose—along the lines of what Caselli and Feyrer found—that you need $20—not $10—to buy one unit of physical capital in a poor country. Then, the poor countries’ advantage of high marginal products of capital is neutralized. A rich-country investor can get the same annual payoff from $20 irrespective of where the money is invested. Therefore, he or she has no particular reason to invest in poor countries.

Now, what does all this have to do with the effectiveness of foreign aid in helping poor nations out of poverty? Caselli and Feyrer have explained why private money does not flow from rich to poor countries. But what does that have to do with foreign aid?

We all know one important lesson of supply and demand: whatever’s abundant is also cheap. If there is a sudden influx of, say, immigrant labor, wages will fall. And if there is an inflow of foreign capital, the return on capital will fall. Going back to my example in which a $20 investment generated $4 of additional annual output in both the rich and the poor countries, let us consider how foreign aid would affect the situation.

The greater availability of capital in the poor countries would reduce the return to capital: $20 would now generate, say, a mere $3.95 of additional output every year. On the other hand, the (slightly) reduced availability of capital in the rich donor countries would leave the rich-country return on $20 either unchanged at $4 a year or maybe even a bit higher, say, at $4.01 a year. As a result, investors in poor countries—these could be rich-country citizens or poor-country citizens—would take their money out of poor countries and send it to rich-countries.

In fact, this reverse flow of capital from poor to rich countries would continue till the earlier parity of the return to capital is restored. The effect of foreign aid would be totally neutralized.

Think of it in terms of hydraulics. Think of it in terms of water seeking its own level. If there are several connected bodies of water, pumping water out of one and into another in an effort at raising the latter’s level would, quite naturally, be a complete waste of time.

Bummer!