Eric Werker, an assistant professor here at HBS, co-authored a rather provocative piece in the New York Times a week ago. In it, he argued that the U.S. should replace its foreign aid program with a 39-percent tax credit for foreign investment in targeted countries.
Now, let me start by saying that Eric is a brilliant colleague. His work is clever, interesting, and iconoclastic. He hasn’t written a paper that I didn’t wish I’d thought of first. He’s also got a ton of real world experience in development work in the hairier places on the planet. I think I’ve probably seen more dead bodies than he has, but I’m not sure. Other than that, Eric’s got me beat cold on the travel-to-serious-places and do-serious-things tip. In fact, to be brutally frank, he’s got me beat cold in pretty much everything. Eric is smarter than me, taller than me, better-looking than me, younger than me, and he dresses hella better than I do.
Nevertheless, this proposal seems, well, wrong.
The first assumption is that the welfare gain from a dollar of foreign investment is more than the welfare gain from 39¢ of foreign aid. Is that true? I accept that only 40% of aid dollars actually go to “economic development,” which today means providing public goods of some sort. So that 39¢ becomes 16¢. We’ll further assume that the 16¢ of spending only produces 16¢ of welfare gains: that is, there are no benefits to aid spending other than the value of the spending itself.
So, in order to match the welfare gains of the one dollar of aid, our hypothetical dollar of investment spending would also have to produce 16¢ in welfare, a social rate-of-return of 16%. That’s not low, I don’t think. It’s higher than the social returns that Carlos and I calculated for the Panama Canal, and I’m just not sure I believe that a textile factory in Madagascar will generate social returns twice as high as the Big Ditch.
In fact, another colleague of mine — the equally-brilliant Laura Álfaro — has some papers showing that foreign direct investment (FDI) produces no spillovers in very poor countries. Their institutions — what an MBA student would call the “business environment” — are just tan jodido. (That last is a technical term.) Of course, there are plenty of places where FDI generates lots of spillovers — Trinidad comes to mind — but Trinidad and countries like it aren’t that poor and aren’t the kind of place that this program would target.
The proposal also assumes that the big problem facing poor countries is a lack of investment. That’s certainly possible, but I have some doubts. First off, real interest rates are rather low in a broad swathe of underdeveloped nations, which is not consistent with the idea that there are all these investment opportunities going unseized for lack of capital. Second, while a 39% rise in your return on capital is very hefty, you have to wonder about the elasticity of investment with respect to returns in very poor — and hence pretty risky — nations. Are there that many marginal projects sitting around out there? I’m remembering my own flirtation with trying to encourage economic development in the Tzotzil-speaking parts of Chiapas: I’ll be damned if I can think of any projects where we muttered to ourselves, “If only our effective hurdle rate could be reduced from 9.8 percent to six!” (Which is not to say that more profit isn’t preferable to less.) Third, if public goods aren’t universally-provided, then subsidizing private returns is going to exacerbate whatever distorted investment patterns already exist: e.g., investment goes to the places in the poor country that already have decent roads, electricity, and public health. And that, of course, reduces the social rate of return on the investment.
Of course, there’s another problem: how do you know that companies won’t just use the tax credit to build factories that they would have anyway? In fact, even if you’re pretty sure that the project would not have been built in that country without the credit, it’s going to be really hard to be sure that it would not have been built at all without the tax credit. Now, if countries were really good proxies for economic deprivation, this wouldn’t be a problem — but do you really want to subsidize Nicaraguan textile factories at the expense of, say, Mexican ones? Mexicans are a lot richer than Nicaraguans; it isn’t clear that Mexican textile workers are a lot richer than Nicaraguan textile workers. And it’s even less clear how much subsidizing more textile factories will help peasants in either place.
I’d rather American taxpayers build schools and roads and immunize children and maybe even pay for (or provide directly) teachers and police, than make private goods cheaper. If aid programs aren’t providing those public goods, and most of them aren’t, better to change the programs than junk them for a tax credit. And if they can’t provide the goods — a result that would make the do-gooder in me very unhappy, but if that turns out to be the case, then I’ll just have to accept it — it would probably be better to get rid of aid altogether, except where we openly wanted to use our money as a political weapon.
(Eric, in fact, has an absolutely brilliant paper showing how aid inflows go up when countries get elected to a seat on the Security Council, a perfect example of using money as a political weapon. And the paper really is brilliant, systematically demolishing any other possible explanation for the temporary increase in aid. As a left-wing imperialist and a patriotic peace-loving American, I highly approve of bribing other governments to get what we want.)
The op-ed uses the New Market Tax Credit passed in the waning days of the Clinton Administration as an example of how the tax credit approach has been “a resounding success.” Well, maybe it has, but I’m not so sure. The problem is that the only metric of success that they give is that, “Seven years later, so many businesses want to invest in poor areas that only a quarter of the companies that applied for tax credits in 2006 received them.” That’s kinda weak gruel on the success metric tip. We’d need to know whether the companies that didn’t get credits abandoned their investments. We’d also need to know how the investments affected employment, wages, poverty, educational attainments, public health, and crime in the areas they invested in.
You can see how the New Market credits are used in Carlos’s motherland here. Another report is here. If you read through them, the projects financed by New Market money don’t sound like stuff that wouldn’t have happened without it. Even the frozen-pizza plant in the Menomonee River Valley, which quite possibly would not have been built there without the program, just moved 85 jobs from Illinois.
I should also point out that New Market is far from a free-market initiative. The companies that get the credits need to have local community members directly involved in decision-making, which means that they tend to go to public-private partnerships and the like. I don’t think that the Global Tax Credit proposal envisions anything similar. Now, that’s probably a good thing. (If you’re going to give tax credits to FDI, it would probably be a bad idea to put local government officials on the project’s board.) But it does make the domestic and foreign programs even less comparable.
En fin, I have very few doubts that ending the New Market program and spending the money on expanding the Earned Income Tax Credit or State Children’s Health Insurance Program would do quite a bit to reduce poverty in the United States. In other words, the domestic programs lauded in the op-ed aren’t the best use of American tax dollars to fight poverty within our country, and they don’t provide much of a model for how we should use our tax dollars outside it.
Eric tells me that the proposal is serious. I’m not so sure, because Eric is beginning to make a habit of making outrageous proposals in the popular press — in one astoundingly entertaining piece, he suggested that we let corporations run for mayor. But regardless of Eric’s intentions, there a lot of insane people — also known as “Republican congresspeople” — who will be all over this proposal like foul odors on excrement. Since I for one happen to be a bit more agnostic about the idea that old-fashioned aid is useless — and this after getting down close to the dirt in Afghanistan, which should be enough to make even the most fervent aid proponent doubt that old-time religion — I’d like to hope that traditional aid isn’t replaced by tax credits that seem very unlikely to have any effect on growth or welfare in the countries that they’re intended to help.
Plus, I’ve got my own crazy proposals to fund. More on them some other time. Comments, criticisms, counter-arguments? All very welcome.