Search the Site

How Pure Is Your Altruism?

We’ve had a lot of conversations on this blog about charitable contributions. For instance: where people like to give, and why; how a young philanthropist should disburse $70 million; whether to give to a street beggar, a hot dog vendor, or neither. So let’s start one more conversation on the subject.

There have been a pair of huge natural disasters in the past two weeks: a cyclone in Myanmar and and an earthquake in China, each of which have killed tens of thousands of people.

Have you written a check yet to donate to either cause? I seriously doubt it.

Why do I say that? Before looking at these recent tragedies, first consider the following three natural disasters from a few years ago, listed along with number of fatalities and amount of U.S. individual charitable donations (according to Giving U.S.A.):

1. Asian Tsunami (Dec. 2004)
220,000 deaths
$1.92 billion

2. Hurricane Katrina (Aug. 2005)
1,577 deaths
$5.3 billion

3. Pakistan Earthquake (Oct. 2005)
73,000 deaths
$0.15 billion ($150 million)

Americans gave nearly three times as much money after Hurricane Katrina as they did after the Asian tsunami, even though the tsunami killed many, many more people. But this makes sense, right? Katrina was an American disaster.

Then along comes a terrible earthquake in Pakistan, killing 73,000 people, and U.S. contributions are only $150 million, making the $1.92 billion given after the tsunami look very, very generous. That’s only about $2,054 per fatality in Pakistan, versus an approximate $8,727 per fatality for the tsunami. Two far-away disasters both with huge loss of life — but with a huge disparity in U.S. giving. Why?

There are probably a lot of explanations, among them:

1. Disaster fatigue caused by Katrina and the tsunami; and

2. Lack of media coverage.

Do you remember coverage of the Asian tsunami? I am guessing you do, especially because in addition to hitting poor areas, it also struck high-profile resorts like Phuket. Do you remember coverage of Hurricane Katrina? Of course. But what about the Pakistan earthquake? Personally, I remember reading a couple of brief newspaper items but I didn’t happen to see any coverage on TV.

Consider the recent paper “Media Coverage and Charitable Giving After the 2004 Tsunami,” by Philip H. Brown and Jessica H. Minty. Here’s their rather startling — if sensible — conclusion:

Using Internet donations after the 2004 tsunami as a case study, we show that media coverage of disasters has a dramatic impact on donations to relief agencies, with an additional minute of nightly news coverage increasing donations by 0.036 standard deviations from the mean, or 13.2 percent of the average daily donation for the typical relief agency. Similarly, an additional 700-word story in The New York Times or Wall Street Journal raises donations by 18.2 percent of the daily average. These results are robust to controls for the timing of news coverage and tax considerations.

And what causes one disaster to get a lot of coverage while another doesn’t? Again, there are probably a lot of factors, foremost among them the nature of the disaster (i.e., how dramatic/telegenic is it?) and location. Getting back to the recent disasters in Myanmar and China, I’d say there are a few other things worth considering:

1. We are in a season of heavy political coverage in the U.S., which is hard to dislodge from the airwaves.

2. Covering far-away disasters is time-consuming and expensive, which becomes doubly prohibitive when media outlets are in cost-cutting mode.

3. Neither Myanmar nor China (nor Pakistan) have what one would consider a very high Q Score among Americans. I am guessing that most Americans couldn’t find Myanmar on a map, and if they have any impressions about the country at all, they are not good impressions (think “military junta”).

Indeed, donations to Myanmar so far are very low. Considering how unevenly disaster aid is often distributed, maybe this isn’t so terrible. But still: if you are the kind of person who donates money to people in need, isn’t the family of a cyclone victim in Myanmar as worthy of your charity as anyone else? The political or narrative forces of a disaster shouldn’t change our response to the need, should they?

We might like to think that we donate almost blindly, depending on need rather than our own response to the particulars of a disaster. But the growing economics literature on charitable donations shows that isn’t the case.

In a narrow but very compelling piece of research, John List argued that if you are trying to solicit donations door-to-door, the single best thing you can do to get large donations is to be an attractive blond woman.

I thought of this research when the N.F.L. was raising money in a weekend telethon after Hurricane Katrina. Between games and during halftimes, the league had star players manning the phones; in the end, the money the league raised was relatively very, very low. They probably would have done a lot better if they had used cheerleaders to solicit donations instead of the players.

So given the particulars of the disasters in Myanmar and China, as tragic as they are, I feel pretty confident in predicting that U.S. charitable contributions in each case won’t be very large. (One surprising upside may, however, emerge for China: activists from the U.S. and elsewhere who’ve been urging an Olympic boycott may find it harder to stir up anger against a country that will still be in mass mourning.)

It may be that the only kind of altruism that truly exists is what economists like to call “impure altruism.” (This is a subject we’ll be writing about at some length in SuperFreakonomics.) Does this mean that human beings are shallow and selfish — that they only give to a cause when it is attractive to them on some level? Will the future produce some sort of “disaster marketing” movement in which aid agencies learn to appeal to potential contributors?

[Note: I’ll be speaking more about this subject tomorrow (Wed., May 14), at about 6:15 a.m. EDT on the new public-radio show The Takeaway.]


Comments