Behavioral economics and poverty

Mullainathan worked with a bank in South Africa that wanted to make more loans. A neoclassical economist would have offered simple counsel: lower the interest rate, and people will borrow more. Instead, the bank chose to investigate some contextual factors in the process of making its offer. It mailed letters to 70,000 previous borrowers saying, “Congratulations! You’re eligible for a special interest rate on a new loan.” But the interest rate was randomized on the letters: some got a low rate, others a high one. “It was done like a randomized clinical trial of a drug,” Mullainathan explains.

The bank also randomized several aspects of the letter. In one corner there was a photo—varied by gender and race—of a bank employee. Different types of tables, some simple, others complex, showed examples of loans. Some letters offered a chance to win a cell phone in a lottery if the customer came in to inquire about a loan. Some had deadlines. Randomizing these elements allowed Mullainathan to evaluate the effect of psychological factors as opposed to the things that economists care about—i.e., interest rates—and to quantify their effect on response in basis points.

“What we found stunned me,” he says. “We found that any one of these things had an effect equal to one to five percentage points of interest! A woman’s photo instead of a man’s increased demand among men by as much as dropping the interest rate five points! These things are not small. And this is very much an economic problem. We are talking about big loans here; customers would end up with monthly loan payments of around 10 percent of their annual income. You’d think that if you really needed the money enough to pay this interest rate, you’re not going to be affected by a photo. The photo, cell phone lottery, simple or complicated table, and deadline all had effects on loan applications comparable to interest. Interest rate may not even be the third most important factor. As an economist, even when you think psychology is important, you don’t think it’s this important. And changing interest rates is expensive, but these psychological elements cost nothing.”

Mullainathan is helping design programs in developing countries, doing things like getting farmers to adopt better feed for cows to increase their milk production by as much as 50 percent. Back in the United States, behavioral economics might be able to raise compliance rates of diabetes patients, who don’t always take prescribed drugs, he says. Poor families are often deterred from applying to colleges for financial aid because the forms are too complicated. “An economist would say, ‘With $50,000 at stake, the forms can’t be the obstacle,’” he says. “But they can.” (A traditional explanation would say that the payoff clearly outweighs the cost in time and effort, so people won’t be deterred by complex forms.)

Thanks to a reader — his name mistakenly deleted from my email — for the pointer.

Comments

Couldn't help noticing the word 'pathological', and its implication that if people behave different than the neo-classical model, they are wrong, not the model. Reminds me of Brecht's poem that after some riots suggested that the DDR government might wish to depose of their people and choose a new one.

In fairness to "the model," the more educated and intelligent people are, the closer they come to being "right."

From the economist (Caplan link):

"Putting the fuel filler doors on different sides of different cars thus means that some cars can access pumps from the left. And this makes it less likely that drivers will have to wait in line for gas."

Looks like a nice try at explaining why gas caps are on both sides, but the example is poor. I've never seen a gas station with only one entrance. If one passes the first entrance, and enters the second, you are able to use the other sides of the pump.

If all the caps were on the driver side, you would at least have the benefit of not circling you car to fill up.

I've never seen a gas station with only one entrance.

Strange, there's three like that right by where I live in Centreville, VA, one in the Centreville Square shopping center, another (a Shell station) where Union Mill Rd intersects Lee Highway, and a third off of Lee Highway between Union Mill and Centreville Road that can only be entered while traveling south on 29. I frequently am forced to circle precisely because all the cars pumping have their gas caps on the same side, and there is no way to enter the station from the other side.

It happens all the time that gas stations have only one entrance that can be entered when traveling a particular direction, particularly when one road that the gas station faces is a one-way street or has a median or when the gas station is on a corner. I have seen several gas stations with a single entrance and a single exit, where one is forced to turn right out of the exit (and turn right to enter), and the curb is designed in such a way that makes it impossible to enter via the exit.

Might the gas cap thing be the result of unintelligent, non-design? Are we really to believe that auto manufacturers considered the fueling up problem faced by consumers? And on the Florida Turnpike they solve this circling problem by making hoses long enough to reach either side of the car.

"Neoclassical economists argue that however people behave, they are expressing their true preferences."

This assumption is the problem, because this leads to the argument that people always behave rationally. Nobel prize winner Rober Aumann gave this definition of rationally in his nobel lecture.

"What do I mean by “rationality†? It is this:
A person’s behavior is rational if it is in his best interests, given his information."

http://nobelprize.org/nobel_prizes/economics/laureates/2005/aumann-lecture.pdf

And behavioral economics shows that people DON'T behave consistently in this way. Read the
Mullainathan article in the link above. So neoclassical models do fall short in explaining
human behavior because we are unable to "optimize" in every situation.

So Thacker, the first three out of five comments actually do make pretty darn good sense!

This piece by Caplan just demonstrates that he is great at taking a little knowledge and making up a great story about it that has little or nothing to do with reality. I challenge anyone to come up with any evidence to support Caplan's "theory" that the auto manufactures try to balance the number of cars with fuel tank caps on opposite sides. I am willing to bet that the location of the fuel cap -- and don't forget there are cars with caps in the middle of the back side -- is strictly driven by engineering and cost considerations and for all practical purposes is random.

Interestingly, I now see rental cars with a tag on the dashboard controls indicating which side the gas cap is on.

This story seems to demonstrate that Caplan had lots of "book learning" but not a bit of common sense.

I agree with John Thacker.

>"On a homework problem in intermediate micro, you can eliminate poverty with lump-sum transfers. In the real world, matters are not so simple."

I am reading the poverty and discrimination book that tyler recommended. The naive view of the elimination of poverty by government payouts is not unsupported. In the 60's the poverty rate was cut in half to a large extent by government payouts. Since 1970 the poverty rate has fluctuated around about 10% and there has been no major increase in per capita transfers to the poor. There hasn't been much political will to increase per capita transfers, largely because concerns over social pathologies.

Liberals now promote policies such as single payer health care, payments through the income tax system and in-kind transfers that are less efficient in reducing poverty but are believed to not to have as much of perverse incentive on the poor.

Someone left the italics tag on.

martin: If you read Caplan's work on irrationality, he shows that it is education/intelligence that makes the difference, not wealth. Perhaps society should attempt to improve the condition of the poor, but that is a different problem than "poor choices". Trying to shove everything into a morality tale can get you wrong answers.

spencer: It is not Caplan who came up with the gas-tank story but Robert Frank in The Economic Naturalist.

Bill Harshaw: I thought the grandmothers that took care of the kids were usually the maternal grandmother, of which there is only one no matter how many kids the mother has. It would certainly seem odd if it were the paternal grandmothers, but on the other hand I haven't read the book which you refer to.

Tyler: What do you mean by "support the status quo"? Not start a revolution? That doesn't seem plausible to me in modern America. Do you mean how they vote? The poor and uneducated are among the least likely to vote, and they are limited to the two main candidates selected in the primaries, which they are even less likely to vote in. Is there any evidence that they vote differently with transfers vs without? I don't know of it, but maybe you could point some out.

Thacker writes: "Now, neoclassical economics may be flawed, and you may believe that people do not behave rationally. But in that case, it is you who wish to decide for people, who wish to "dissolve the people and elect another," not the neoclassical economists."

But doesn't that presuppose that economics boils down to policy fights? In the midst of it, what about the pursuit of knowledge?

For instance: Isn't it possible to 1) acknowledge that people routinely act in all kinds of bizarre ways that neoclassical economics has a hard time accounting for, yet 2) not be a big fan (for a variety of reasons) of hugely-coercive government?

That didn't work!

Perhaps this will?

"It suggests that behavioral economics in fact provides microfoundations for the conservative critique of the welfare state."

This is clearly one of those "that's what you would say" situations. There is a huge assumption regarding the direction of causation in your conclusion. And there is a slightly schizzy tension between "suggests" and "in fact" that you might want to look into.

There is a strong tendency among all political religions to look at some research result or world event and say "and that proves I was right all along," but the poor beleaguered libertarian faith seems to need to say it more than most.

There is a huge literature on the behavioral gap between the results we are looking at and the conclusion you want to draw. Until you can show that better economic outcomes don't lead to more "normal" behavior, you're just engaging in a "you're it - I quit" declaration of victory. You don't know the direction(s) of causation here, and should stop pretending you do.

Go find the "two cookie" study and see if you can't find a rather different set of implications than the one built into your priors.

Tyler,
I am sorry you find this one of the more interesting papers in a while. Before you post you should either check with a few psych people to make sure the article is actually thoughtful before linking it to your favorite theory. And it would take about ten minutes of your time to obtain thoughtful questions. You could make the claims interesting and rewarding to explore, but not this way. Too bad you missed the chance.

There is a strong tendency among all political religions to look at some research result or world event and say "and that proves I was right all along," but the poor beleaguered libertarian faith seems to need to say it more than most.
The revelation of behavioral economics Caplan discusses is that expanding choice sets can make people worse off and (conversely) decreasing choice sets can make people better off. As he notes in this paper, this can justify paternalism, which would be anti-libertarian. What he notes in favor of libertarianism is that government programs have costs, so removing welfare programs would cost less than nothing (since implementing them costs money) while implementing paternalist policies would cost more than nothing. That's something, but still a weak objection in favor of libertarianism rather than paternalism. Note that Caplan used the phrase "conservative critique" rather than "libertarian critique".

There is a huge literature on the behavioral gap between the results we are looking at and the conclusion you want to draw. Until you can show that better economic outcomes don't lead to more "normal" behavior, you're just engaging in a "you're it - I quit" declaration of victory. You don't know the direction(s) of causation here, and should stop pretending you do.
Would you be referring to something like twin adoption studies? They generally show that IQ has a heritability over 0.7 while having a well-off family raise you does not matter all that much. I don't know if that's what you were after. If you want to focus on stereotypically underclass behavior, we have evidence that the crime-rate is higher among middle class suburban blacks than lower class whites. We also have shows like this. With regard to Caplan's theory of irrationality, he shows that wealth does not have nearly as much of an impact on "thinking like an economist" as education does. As he says "The typical cab driver with a Ph.D. in philosophy shares the economic outlook of other Ph.D.'s, not other cab drivers. Given the strong correlation between income and education, though, widespread misconceptions about the "beliefs of the rich" are quite understandable".

Go find the "two cookie" study and see if you can't find a rather different set of implications than the one built into your priors.
I don't know which one you're referring to, could you provide a link?

If there is no pathology (absent quotation marks) in the behavior described, and (neo)classical assumptions are themselves just assumptions and not necessarily virtures
Neoclassical economic theories of rationality are not normative. The standard neoclassical view has long been the theory of "rational expectations". It is behavioral economics that asserts people are not rational.

I would hope that people would be very careful with the claim that potential welfare recipients have characteristics that make them deviate further from what is typically considered economic rationality. At the very least, one should poke around to see if that notion is supported by research. One example:

http://personal.psc.isr.umich.edu/~danben/files/Who_is_Behavioral_Cognitive_ability_and_anomalous_preferences.pdf

This paper is authored by a former classmate of mine, and shows that how "behavioral" someone behaves is correlated with cognitive ability but not specifically with wealth. Therefore saying that potential welfare recipients exhibit particularly non-rational behavior makes a sometimes unfair assumption, namely that low income people are less intelligent.

Dan, feel free to provide more information. I don't doubt that even wealthy people behave stupidly. I do doubt that wealth will not have some negative correlation with "behavioral" behavior. I did not know that Krugman wrote about behavioral/psychological issues, so it would be helpful if you linked to pieces of his that do so.

saying that potential welfare recipients exhibit particularly non-rational behavior makes a sometimes unfair assumption, namely that low income people are less intelligent.
Is it merely an "unfair" or also an inaccurate assumption? According to Wikipedia welfare recipients and those in poverty (among non-Hispanic whites) tend to have lower than average IQs. You might be interested to note that Caplan cites that very paper you reference here.

kharris, I apologize for my assumption. I haven't read as much from Tyler on this, so I don't have as good of an understanding of his stance.

Comments for this post are closed