It's not easy to save the world

I wrote a post here recently about charity (specifically, on incentives vs. grants), and in the wake of the Super Bowl, I've come across another interesting item that falls in the same basic category. Once again, there is some question as to what might be the best (or at least, most efficient) way of donating to charitable causes.

Every year before the Super Bowl (and other large sporting events, I'd imagine), officially licensed t-shirt producers print up two batches of "Super Bowl Champions" shirts, one for each participant in the game. The winning team's shirts are sold and/or given to the winning players, while the losing team's unfortunate shirts are shipped to Africa by an intermediary. As a Patriots fan, it's a process I'm all too familiar with:

This donation seems like a nice and charitable use for what would otherwise be wasted textiles, but according to guest poster Dean Karlan at the Freakonomics Blog, it's not quite that simple.

The Super Bowl stirred up an old controversy in the international aid community.  What happens to all those preprinted “Pittsburgh Steelers 2011 Super Bowl Champion” t-shirts?  Apparently, each year the NFL gives them to the international relief and development organization World Vision, who then ships them to Africa.

Is this good or bad? And why should anyone care?

This is not the first time these questions have been asked. Less than a year ago, [web entrepreneur] Jason Sadler planned to send a million t-shirts to Africa, only to be bombarded by scathing criticism from the aid blogosphere...

Opponents argue that sending shirts destroys local textile economies by flooding the market with free goods and undercutting local t-shirt producers. World Vision responds by saying something to the tune of, “but we spread it out.” That is kind of like arguing that something bad is okay if you do it in small enough doses to lots of people (rather than a large dose to a few people). Of course, such arguments don’t hold water: bad is bad, even if marginally so.

I think World Vision might have a better defense. They could argue that critics of the annual t-shirt migration (or at least all the critics I’ve heard) are thinking about the wrong counterfactual. The choice is not between (a) doing nothing — which, critics infer, would leave Africans to produce and sell 100,000 new t-shirts — and (b) shipping 100,000 t-shirts to Africa. Rather, the choice is between (a) selling the t-shirts in the U.S. as rags (or novelty souvenirs for delusional Steelers fans) and then sending to Africa the proceeds plus the money that would have been spent on shipping, or (b) shipping 100,000 t-shirts to Africa.

In other words, the NFL surely isn’t going to pay local producers to make 100,000 t-shirts after the Super Bowl. That option is not on the table. So in the end, the t-shirt migration has one pro and two cons, and we have no real data to tell us what to do. The pro: some people in Africa get some t-shirts, and hopefully those people extract some value from the t-shirts (either by wearing them or by selling them). The first con: market prices for t-shirts may go lower in Africa, and this adversely affects some. The second con: there may simply be a better way, such as selling the t-shirts in the US and sending the profits, as in (a) above.

Because of a lack of reliable data, Karlan is unable to go much farther in his analysis, leaving us to grapple with an open-ended question. As in my previous case of incentives vs. grants, it's not completely clear whether our "charity" is in fact doing more good than harm.

In attempting to devise a proper charitable program, these are exactly the kinds of questions that must be answered, and yet they're the most difficult ones to decipher. It is indeed difficult to save the world, even if our hearts are in the right place. So, as Dean Karlan did before me, I'll ask you: what do you think? Is it better to try and do harm than never to have tried at all?

[Freakonomics Blog]