The Persistence of Poverty

February 14, 2011

Traditional thinking about what motivates the poor is precisely backwards.

Karelis’ book contains one big idea, and he spends most of his pages in defense of the idea and the rest on the idea’s implications. The idea, packed densely in economic terms, is that marginal utility is increasing amid scarcity. Let’s unpack it a bit and give it some background. Marginal utility is the amount of good you get from an additional unit of something. For example, the marginal utility of cookies is usually decreasing because each additional cookie you eat makes the next one less desirable.

According to Karelis, traditional economic wisdom holds that marginal utility is always decreasing. On this view, a lot of patterns among the poor are pretty baffling, such as why poor people, who presumably have the most to gain from it, would refrain from working as much as possible, or drop out of school, or break the law. If cookies get less and less good the more of them you have, shouldn’t people with the fewest cookies be really motivated to get more cookies, and not risk losing the ones they’ve got? This is, of course, only if you assume marginal utility to be decreasing (i.e. the usefulness of something goes down the more of it you have). Karelis’ big idea is that while marginal utility is decreasing when you’re talking about having more of a good thing, the reverse is true when you’re talking about having less of a bad thing. Remember the cookie metaphor, but now imagine a special salve that when you put it on bee stings it makes them stop hurting. Let’s say you’ve got 5 bee stings, all hurting to high hell (they’re stings, after all), and I give you one dab of salve. You go from 5 throbbing stings to 4–not a a big difference, and you may not even notice it. But what if, on the other hand, you’d found 4 dabs of salve and you’ve only got one bee sting left. Now, when I give you one dab of salve, you go from 1 painful sting to no pain at all–hi ho! Here the marginal utility of salve is increasing, which is to say that the more dabs of salve you have, the more benefit you derive from each one. Karelis makes a case that poverty is more like having too many bee stings than not enough cookies, and the fact that marginal utility is increasing in the one case and decreasing on the other has far-reaching implications. In fact, if you think that marginal utility is decreasing among the poor (as traditional wisdom holds), then all your assumptions about what motivates poor people will be precisely backwards.

Poor People are Not Irrational

Karelis points to five very general patterns of behavior that tend to perpetuate poverty:

  1. Not working much for pay
  2. Not getting much education
  3. Not saving for a rainy day
  4. Abusing alcohol and other drugs
  5. Taking risks with the law

Obviously, the point isn’t that all poor people are lazy, drug-addled scofflaws (at least not any more than the rest of us). But according to common wisdom it might seem especially baffling that poor people would engage in this behavior at all. After all, dropping out of school or not saving money means less cookies, and people with the fewest cookies should derive the greatest pleasure from those they have, right? But if poverty is like having too many stings (bills, aches, broken windows, etc.) instead of not enough enough cookies, everything changes. Intuitively, when you’re truly poor and can’t make ends meet, it’s like having ten bee stings. Working all day for a dab of salve or two, or staying in school for the sake of an extra dab of salve many years in the future–it makes little difference when you’ve got stings throbbing all over your body. You could save up your dabs of salve, but life is hard and chances are you’ll get even more stings in the meantime. Hell, you might even consider robbing a bank and risking your useless few dabs of salve on a chance at enough to alleviate all your stings and those of your kids.

The point here is that when you’re so poor you can’t possibly make ends meet, you’ve got a lot of painful problems: hungry kids, creditors demanding payment, medical problems you can’t afford to fix, exhaustion, neighborhood crime–the list goes on and on. As a result:

  1. Solving any single problem brings little to no comfort, since all the others remain. As a result, there is little incentive to further exhaust yourself chasing a slightly higher income.
  2. There is a great incentive to take risks, even bad ones, because a small chance of all your problems going away is surer path to happiness than playing it safe and keeping yourself and your kids poor.

In other words, poverty has built into it a disincentive toward solving your problems and an incentive toward risky behavior, so the most significant cause of poverty ends up being poverty itself. The obvious negative here is that poverty is self-perpetuating. If we do nothing, poor people responding rationally to the incentives before them will stay poor. Fortunately, this also means that by giving poor people a little help, we have it exponentially easier for them to help themselves. If you’ve got 5 stings, you don’t have much reason to work for a dab of salve that’ll only reduce your pain 20%. But if someone gives you two dabs of salve free-of-charge, suddenly working (or staying in school, etc.) for an additional dab of salve will reduce your pain by 33%. Put literally, by helping the poor solve some of their problems, is greatly increases their ability to solve the rest of them.

So What Should We Do

Karelis starts with what we shouldn’t do: cut government assistance to the poor. If you understand all the above business about marginal utility, it will be clear that making poor people poorer will drastically reduce their incentive and ability to get themselves out of poverty. In fact, the dead simplest way to alleviate poverty is to give money no-strings-attached to the poor and let increasing marginal utility take care of the rest. However, there are even more cost-efficient means of alleviating poverty than straight handouts. “Make work pay” programs like the Earned Income Tax Credit that incentivize work have a double benefit: they not only make working a more efficient path to happiness for the poor, but the increased income will help solve some of their problems making the remaining ones more manageable.

Jesus told his would-be followers in Matthew 19:21 that “if you wish to be perfect, go sell your possessions, and give the money to the poor, and you will have treasure in heaven.” While this would be a pretty remarkable way to end poverty, we can’t all follow that example. But can we at least stop calling anyone a socialist who says we should help the poor? And maybe stop gutting government assistance to the most needy to subsidize tax cuts for the nation’s wealthiest? (Especially when the poor spend a larger portion of their income keeping the economy going than the rich do.) Or, going just a little further, donate to organizations that help the poor, knowing that your gifts will be multiplied by the law of increasing marginal utility amid scarcity.


Since you’re reading this on the internet, chances are you’re not living below the poverty line. If you think you could handle it, try playing Spent, a clever game by the Urban Ministries of Durham, North Carolina. If you’re not persuaded by the sheer indecency of poverty amidst the richest nation the world has ever seen, at least consider the argument from self-interest: with some bad luck you could end up in poverty, and wouldn’t you want your society to help you get out?