Bryan Caplan  

The Prudence of the Poor

PRINT
Is Losing Your Job in Your Eco... Eighty years later and the res...
Ari Fleisher in the WSJ:
Given how deep the problem of poverty is, taking even more money from one citizen and handing it to another will only diminish one while doing very little to help the other. A better and more compassionate policy to fight income inequality would be helping the poor realize that the most important decision they can make is to stay in school, get married and have children--in that order.
John Cochrane demurs:
"[H]elping the poor to realize" is pretty hopeless as a policy prescription. They poor are smart, and huge single parenthood rates do not happen because people are just too dumb to realize the consequences, which the see all around them.
But why on earth should we believe that the poor are "smart"?  There is overwhelming evidence that the poor have substantially below-average IQs.  And even without these empirics, it would be very surprising if low cognitive ability failed to sharply reduce income in high-tech societies. 

Cochrane would have been on much firmer ground if he'd said, "While the poor do have below-average IQs, they have more than enough brains to see the consequences of single parenthood."  If he said that, I'd agree.  But this revised position still neglects the possibility that people who foresee bad consequences of their behavior will fail to exercise self-control.  As a result, they predictably make imprudent decisions when their choices have pleasant short-run effects - even if the long-run results are predictably awful.

The empirics on the poor's lack of self-control are not as abundant as the empirics on the poor's low IQ.  But the empirics are out there.  And even if there were no empirics at all, it would be very surprising if low self-control failed to sharply reduce income in a high-tech society.

Does this undermine Cochrane's claim that "[H]elping the poor to realize" is pretty hopeless as a policy prescription"?  At first glance, no.  When someone has low intelligence, it's hard to make him realize stuff; when someone has low self-control, it's hard to make him act on what he realizes.

On further reflection, though, there are multiple ways to make people "realize."  The most popular - and, I suspect, the one Cochrane dismisses - is publicly-funded nagging.  An alternative route to realization, though, is simply cutting government subsidies for imprudent behavior. 

Such cuts have two effects.  First, cutting subsidies for imprudent behavior makes the imprudence even more blatant than it already was.  Second, cutting subsidies for imprudent behavior makes the behavior's unpleasant consequences happen sooner, potentially deterring even the highly impulsive.  As Scott Beaulier and I put it:
What do these behavioral findings have to do with the poor? Take the case of single mothers.  On the road to single motherhood, there are many points where judgmental biases plausibly play a role. At the outset, women may underestimate their probability of pregnancy from unprotected sex. After becoming pregnant, they might underestimate the difficulty of raising a child on one's own, or overestimate the ease of juggling family and career. Policies that make it easier to become a single mother may perversely lead more women to make a choice they are going to regret.

A simple numerical example can illustrate the link between helping the poor and harming them. Suppose that in the absence of government assistance, the true net benefit of having a child out-of-wedlock is -$25,000, but a teenage girl with self-serving bias believes it is only -$5000. Since she still sees the net benefits as negative she chooses to wait. But suppose the government offers $10,000 in assistance to unwed mothers. Then the perceived benefits rise to $5000, the teenage girl opts to have the baby, and ex post experiences a net benefit of -$25,000 + $10,000 = -$15,000.
As far as poverty policy goes, I suspect that Cochrane and I are on the same austerian page.  My fear is that he's discrediting the correct conclusion with implausible justifications.  The Chicago descriptive view that everyone is "smart" has little to do with the Chicago prescriptive view that government is way too big.  Indeed, as Donald Wittman has shown, it's hard to argue that government is way too big unless you're willing to insult the intelligence of a great many people.  I'm happy to bite that bullet.  Cochrane should do the same.



COMMENTS (23 to date)
Pajser writes:

Many unemployed people develop harmful habits, attitudes, even mental disorders. Correlation is easy to prove; causation can be proved as well. (Diette et al., Causality in the relationship between mental health and unemployment, 2012) Full employment approach of Eastern block is, I think, adequate. It introduces new problems, perhaps less malign and solvable. Assistance that demotivates work is harmful, still lesser evil than leaving people without job and money.

Philo writes:

“[C]utting subsidies for imprudent behavior makes the imprudence even more blatant than it already was.” Well, it makes the behavior *even more imprudent*. On the other hand, sufficiently generous subsidies would make the behavior *not imprudent*.

Again: “Policies that make it easier to become a single mother may perversely lead more women to make a choice they are going to regret.” Of course, this is possible. But note that it would also be possible to make the lives of single mothers *so* easy that most single mothers would *not* regret their choice.

The actual policies are bound to have different-valued effects on different potential single mothers. Do you have some evidence about this range of effects? And do you have some evidence that making life harder for single mothers would be better on the whole than making it even easier than it is being made now?

Ross Levatter writes:

Perhaps Bryan does so only for rhetorical effect, but his link at the end relating to him biting the bullet to the effect people's intelligence should be insulted does not go to some clever blogpost where he notes half the people have an intelligence of under 100, but to his book on voter irrationality. As he well knows, his voters are rational to vote irrationally, and it seems difficult to me to use a rational choice as evidence of acting stupidly.

Am I missing his point?

Enial Cattesi writes:

To be honest I don't understand the point of the post. I know a number of "poor" people who aren't stupid, just
1. plain lazy
2. have something better to do with their life than get a job and so live from the government doll.
3. what they are smart at doesn't sell particularly well
4. have personality problems (obnoxious rascals)
etc.

Jonathan writes:
I know a number of "poor" people who aren't stupid

Wow really, then it most be true. I wish we could all know like you.

Am I missing his point?

Maybe he is referencing to the fact that he already have in the past talked about taboo subject. Other than that, if that is all you got from this post then I would say yes you missed the point mate.

John B writes:

School can be a huge waste of time and getting married can be very dangerous for a man these days. Here's my proposal to fix social problems. Create a program with 4 planks. This will cover all social insurance.

1. Legalize all drugs. Anyone familiar with the economics of prohibition will understand why this will make us a fairer, safer, and rich society

2. Get rid of the minimum wage, any disincentives to hire, and any obstacles to firing people. People of any skill level need to be able to work.

3. Anyone who is able to document a full-time work schedule over the previous year receives a lump sum from the government of about $15,000. This subsidizes work and production driven by entrepreneurs rather than wasteful government projects and, more importantly, it doesn't subsidize undesirable things like single motherhood, not working to qualify for benefits, staying unemployed for long durations while collecting UI, etc.

4. The maximum income that qualifies for this lump sum changes every year and gets selected by a random number generator that picks a number somewhere between 75K and 150K. This way people cannot structure their incomes to be certain of qualifying unless they are already relatively comfortable at 75K.

This way real wages rise because there is so much more production. In addition even if they get paid pennies, a couple makes at least $30,000 a year which puts them in the top 75% of earners in many places in the country.

The biggest drawback to this is that it would become prohibitively expensive if any immigrant qualified for it. Perhaps a waiting period for collecting the lump sum would allow for open immigration and get the best of all worlds.

Enial Cattesi writes:

@Jonathan:

OK, I give up. Clearly poor people are stupid. Does this mean that all rich people are super geniuses?

Terra Desolata writes:

(Coming here from a link from FTAlphaville)

I realize this is anecdotal, but in my personal experience, the primary reason young women from disadvantaged backgrounds have children early and out of wedlock is emotional. Growing up in a home with little love or affection, having a child is the one way to ensure a loving relationship, with the child if with no one else. Also, there is often a deep-seated hope that the child will secure the affections of its father - which rarely materializes, but the hope is still there.

Economic incentives and government programs are not going to significantly mitigate that reality. But they can make it worse, by making the decision to have a child easier, by guaranteeing financial support for unwed mothers, as previous iterations of the dole and welfare did. In prior generations, having a child out of wedlock was a rite of passage for a young woman - the first child guaranteed a home in the housing projects and a monthly check from the government, so the earlier a young woman had her first child, the earlier she could be independent.

The fix to those unintended side effects is to continue going the way we have been going: tying government assistance to work. Having a child at a young age will seem less like an easy way out of a bad home life, if it means having to start working to support the child. If there are a shortage of jobs, then the government should provide them.

Ultimately, the poor make a mix of rational and irrational decisions - just like the rest of us. What can be done is to consider the intended and unintended ways that government policies and programs interact with the incentives.

Eric Falkenstein writes:

Many project themselves onto their agents out of the thought that 'the but for the grace of God go I', and note the lack of disincentives in welfare because it offers a pretty base existence. They can't imagine someone being short-sighted or unambitious enough for welfare to do anything but provide momentary relief. They can't imagine someone wanting to have more than 2 children when they don't have any income or spouse.

Lots of people aren't that way, and look at children more like fish than penguins (ie, no investment). Charles Murray's Losing Ground argued that for such people, disincentives are more important than incentives, and these were being softened by welfare.

RPLong writes:

I don't know when IQ reclaimed its academic vogue, but for my money it's one of the most disappointing and untenable trends among academic bloggers, replete with implications that the bloggers themselves would never endorse if they took them to their logical conclusions.

Enial Cattesi does a good job of pointing out that poverty cannot possibly boil down to a stupidity problem. The many wealthy psychopaths and risk-loving investors out there provide ample evidence that lack of self-control is not a strong indicator of income.

Clearly the problem is not stupidity, but lack of a certain kind of knowledge. Clearly the problem is not lack of self-control, but lack of a certain kind of self-control.

The burden of proof is on Caplan and others to demonstrate that IQ actually, successfully measures that which it claims to measure, better and more reliably than all other means of measuring whatever it is that people claim it measures. That must be accomplished first, before we ever start drawing conclusions from demographic IQ data.

For example: Researchers believe that human intelligence peaked thousands of years ago, before our wealth and standard of living increased by such a high degree. Any theory based on IQ must be consistent with all facts of human intelligence, not just the ideologically convenient ones.

Jay writes:

@Enial Cattesi

Nobody said all of anything was anything, but yes IQ is correlated with income on both sides of rich/poor.

sam writes:

Fifty years ago, the illegitimacy rate was under 10%, and the male workforce participation rate was well over 90%.

Now the illegitimacy rate is over 40%, and the male workforce participation rate is under 90%.

I do not believe that the IQ of Americans has changed that drastically over two generations. (For the Sailers out there, this analysis still holds if you exclude Hispanics and Asians and look only at non-Hispanic whites and blacks)

IQ is not the issue. Culture is.

Floccina writes:

A billion people in the world live on less than $2 a day so there is a possibility that these young girls do not worry so much about having children because in the rich countries it is easy to get enough money to live and they love their children. It is difficult to say what would happen if all welfare were eliminated? In the short run people might move in together causing rents in the parts of town with the worst housing to plummet. The poor would learn to eat differently(eat the way they eat in poor countries). They might learn how to conserve electricity and heat etc. Some that do not work. It would be tough but I think norms would slowly arise that would allow most to live and cope.

@Enial Catessi writes: "OK, I give up. Clearly poor people are stupid. Does this mean that all rich people are super geniuses?"

One of the surest signs that people feel emotionally threatened by an argument is that, rather than responding to what was said, they immediately turn into a distorted straw man which they can readily dismiss without any further reflection.

That is what you did here.

Bryan did not say that all poor people are "stupid." He did not say that all rich people are "super geniuses." He did not even say that no rich people are "sociopaths" (to respond to @RPLong).

He said that, on average, within rich advanced societies like the U.S., poor people tend to have lower IQ (and worse self control) than rich people.

Does anybody with any experience in the real world seriously deny that? And, no, "I once knew this poor smart dude/rich idiot" is not a refutation, even if true.

In that case, why not get on and argue about the more interesting question about what this fact about the world implies or explains, or doesn't?

RPLong writes:

Sub Specie AEternitatis, you misunderstood my point. Let me clarify.

My point in bringing up psychopaths is that (1) one of their defining features is compulsive risk-seeking, (2) wealthy CEOs and fund managers score significantly higher on tests that profess to measure "the dark triad," and (3) compulsive risk-seeking is the opposite of self-control.

From this, we can quite effectively call into question any theory that asserts that self-control in general translates into higher incomes. Clearly assuming great risks is the hallmark of both the drug-addicted pauper and the successful entrepreneur, as well as many other kinds of people in the center of that continuum.

So, if we wish to advance a self-control theory, we have to account for this kind of variation. Cattesi's point is that if we wish to advance the IQ argument, we have to account for IQ discrepancies within demographic groups as well as across them. It may be easy to show that IQ is lower among paupers than the rich, but how easy is it to show that, among only the rich, higher IQ translates to higher income? Probably impossible.

Finally, IQ scores are highly correlated to performance in school. To report that the poor are categorically less successful at school isn't exactly man-bites-dog, and doesn't exactly say much about intelligence, either.

Jeff writes:
Clearly the problem is not stupidity, but lack of a certain kind of knowledge. Clearly the problem is not lack of self-control, but lack of a certain kind of self-control.

What kind of knowledge and what kind of self control? If you think the notion of differences in IQ is too simplistic to explain the pathological behavior of the poor vs. middle or upper class people, which may indeed have some merit to it, I don't see how substituting the equally simplistic idea that poverty is caused by lack of knowledge of a kind you don't even bother to specify represents any improvement.

Arthur_500 writes:

This is shown repeatedly but the policy wonks always utilize the idea that people all have the right to make excellent choices. Motherhood, for example is sacrosanct.
In other words, we must help out mothers. Why? Because of the children. So, we give al sorts of advantages to mothers and many individuals choose to play the system.
Our attitude towards stay-at-home moms is similar. Just because mom stays at home does not mean she is cleaning, caring for the children, cooking, shopping, etc. She is all too often simply not working. Kids are in school or daycare and mom is watching Oprah. We portray the stay-at-home mom as if they were the Butler overseeing the household. In reality, this is seldom the case.
Our policies are based on an ideal of motherhood but we have nothing that says, if you take this money we expect this in return.

Richard Besserer writes:

A notorious Canadian blogger who is occasionally invited on cable shout-shows (not at all a libertarian, and not famed for either her empathy or depth of thought---she is actually proud of the fact she never went to university) is fond of saying "Poor people are lazy and stupid---that's why they're poor."

Frankly, this reminds me of the same argument in more intellectual clothes.

1. Most poor people are of low intelligence and suffer from poor impulse control, preventing them from improving their lot.

2. Hence, any assistance in cash or kind to the less fortunate will only encourage further irresponsible behaviour.

This isn't a good argument against the welfare state. It's far too strong. No, government isn't good at picking winners, and clearly welfare states can be accused of not doing a good job of targeting those who are most likely to benefit from the help.

But writing off all poor or simply unlucky people as a lost cause is much like saying that most small businesses fail, so nobody should lend money to a small business. Private charity has more leeway to target aid towards the most likely to benefit, much as banks are better able to target loanable funds towards the ventures most likely to repay with interest than are individual depositors (or governments).

If the argument against helping the poor is that poor people are lazy and stupid and if you bother trying to help them all you'll have to show for it is more poor people with a bigger sense of entitlement, presumably that applies to any aid regardless of source. If anybody on Econlog seriously proposes banning private charity, I missed that blog post. It would be a pretty paternalistic move, to say the least.

MingoV writes:
Given how deep the problem of poverty is...
What is this 'given'? Who says the problem is deep? I already am uninterested in the rest of the editorial. He then jumps to income inequality without showing any linkage to poverty. That's because no linkage exists. I read one and one-half sentences and won't waste time reading more.

@RPLong My apologies. I should not have thrown your argument in with that of @Enial Cattesi. It is more sophisticated and serious.

That said, I'm not convinced by it.

It is my understanding that you are correct that certain very high paying professions (e.g., CEOs and investment bankers) show a positive correlation with certain forms of sociopathy and perhaps even poor impulse control in certain limited settings.

At the same time, it is true that certain poor subgroups (e.g., graduate students) show a positive correlation with intelligence and conscientiousness.

But that does not throw Bryan's overall argument into doubt. The vast majority of the poor (the bottom 25% of the population, let's say) are not graduate students. The vast majority of the rich (the top 25% of the income distribution, let's say) are not CEOs or investment bankers. Both of these subgroups are vanishingly small proportions of these larger groups.

Overall, when speaking of the rich and poor groups overall, Bryan's statistical correlations hold extremely well.

RPLong writes:

Jeff: You can interpret my original comment as a call for greater specifics. You're right to say that I haven't provided them myself; but you're wrong to take Caplan's explanation by default solely by virtue of the fact that I myself have failed to fill in the holes left by Caplan's theory.

Sub Specie AEternitatis: I'm not arguing against the correlations; surely they do hold true. I'm arguing against (some of) Caplan's conclusions, especially the ones that imply that IQ = intelligence or that self-control in the abstract improves outcomes for the poor. Correlations, we must remember, are little more than data artifacts until they are paired with an adequate theory.

Consider the following:

There was an experiment by Rosenthal in which school teachers were casually told at the beginning of the school year that certain students (mentioned by name) were "spurters," that, according to some tests designed to measure "spurting," they would blossom in the coming year. Actually no such test had been given. In fact, no such test exists. The information was actually given about 20% of the students, chosen at random.
These kids not only did well academically (which we might expect, with teachers having some control over that), but actually increased their IQ test scores!

Note that the children's IQ scores improved. Not only does this call into question Bryan Caplan's claim that IQ = intelligence, but it ALSO blows a large hole in Caplan's claim about the signalling model of education.

If, on the other hand, you view IQ tests as measuring little more than the ability to excel in a modern schooling environment, then we have a far simpler and realistic theory than the one Caplan provides: Education levels are a primary determinant of income. Occam's Razor wins.

Note that this alternative theory is both consistent with many of Caplan's other claims, including the signalling model of education, and keeps us from making morally objectionable claims about the general intelligence of the poor.

JohnB writes:

The science behind IQ is extremely flawed. Does anyone really believe that entire nations have lower IQs as the finding suggest. This is complete non-sense. IQ does not measure innate intelligence. The greatest physicist of the second half of the 20th century, Richard Feynman, supposedly tested as having an IQ only slightly above average.

isomorphismes writes:
But why on earth should we believe that the poor are "smart"? There is overwhelming evidence that the poor have substantially below-average IQs.

That may not be the relevant kind of "smart", Bryan. Nor the kind he means.

And even without these empirics, it would be very surprising if low cognitive ability failed to sharply reduce income in high-tech societies.

Are you proposing that high salaries of computer-programmers put a gap between low-IQ and high-IQ? That would be testable, although I'm sure you mean a slightly wider class.

Nevertheless it seems like you're conflating [a] linear relationship between IQ and earnings with [b] threshold effects—clearly −5σ will struggle and +2σ will have advantages in certain respects. Not sure why this needs any kind of "high tech" qualifier as I'd plausit this statement in my imagined environment of evolutionary adaptation as well as my imagined environment of Developed North now.

Comments for this entry have been closed
Return to top