At a conference I attended in May 2012, Charles Karelis, author of “The Persistence of Poverty,” demonstrated what is wrong with much thinking about poverty, using a simple analogy. Suppose you are stung by a bee, and you are offered enough salve to relieve the pain of that sting. Most people would consider that daub of salve to be worth quite a bit.

Now suppose you have seven other bee stings. Will you still value a

daub of salve sufficient to relieve one sting as much as you did when you had only one sting? If you think about it, you will value that one daub less because it will do nothing to relieve the seven other stings that remain.

Now suppose with these eight stings that you are given salve for seven stings. Now you have increased motivation to relieve the one sting that

remains because that additional daub will free you from pain. This simple example is an important exception to a widely accepted principle of economics, the principle of declining marginal utility.

According to this principle, the more you have of something, the less each additional unit is worth at the margin. For example, after you have had one piece of cake, the second is worth less to you than the first. After two, a third is worth even less.

The principle of declining marginal utility applies well to what Karelis calls “pleasers,” such as the dessert example. But it does not apply to what he calls “relievers,” such as the sting salve. In the case of relievers, the more you have of something, the more an additional unit is worth at the margin. The utility of that last daub of salve is worth more to you, not less, than the first daub, because the last daub is the one that gets you out of misery.

Now it turns out that many of the goods that matter to poor people are relievers, not pleasers, or they are hybrids, functioning like relievers when one has less than enough, and like pleasers once one crosses a threshold of sufficiency. Transportation is an example of a hybrid. If you have a 20-mile journey to work, you are not apt to pay bus fare for the first mile of the journey, leaving 19 miles to go on foot. But you might be willing to pay bus fare to relieve you of the last mile after having walked 19. And transportation beyond what you need declines in value.

Poverty means troubles, and like multiple bee stings, these troubles drown each other out. Relief from one problem will not necessarily be pursued by someone if she is left in other troubles. If we keep this in mind, Karelis argued, we can explain much of the behavior of poor people, not as due to some character defect, but rather as what any reasonable person would do in such circumstances.

Consider low-work effort. If money from work were a pleaser, than the first dollar should be the most valuable, and a rational person should be eager for work, no matter how poorly paid. But if money from work is only a

reliever, as it is for someone in poverty, then that first dollar won’t be worth much, like the first daub of salve, since it leaves one in a sea of troubles.

Or consider a failure to save. Saving makes sense as a means of deferring

consumption and as a way of insuring for unexpected shocks to one’s income from layoffs, illness and emergency home repairs. But when one’s current consumption is taken up with basic needs, the value of deferred consumption is much less. And it may be more rational to address the ups and downs of income shocks than to try to smooth out these shocks through saving.

Going back to the bee sting analogy, suppose you are getting two stings a day but have only enough salve for, on average, one sting per day? Are you going to relieve only one sting every day, or relieve two stings, every other day? The latter makes more sense, but it is the opposite of the smoothing-out strategy that saving makes possible. Yet it is more reasonable, given that one is dealing with relievers.

What are the policy implications of this analysis? Karelis argued that a

guaranteed income would be counterproductive for people who are not poor, as it would undermine work motivation. However, a negative income tax, guaranteeing income up to the poverty line, would actually increase the incentives for a poor person to get out of poverty.

It would supply the reliever goods up to the point where the additional unit of income is worth more, and so, in pursuing it, one is stepping out of poverty, not remaining stuck in it. The poor are just like everyone else, except that they have less money. Once policymakers begin to understand this, we may begin to shift from our current counterproductive policies of punishing the poor and blaming them for their condition, to an effective strategy that will get people out of poverty.

Michael W. Howard is Professor of Philosophy at the University of Maine, coordinator of the U.S. Basic Income Guarantee Network, and co-editor, with Karl Widerquist, of two books: Alaska’s Permanent Fund Dividend and Exporting the Alaska Model.