From Charles Karelis, author of The Persistence of Poverty.

Here are some responses to critics who were kind enough to participate in Bryan Caplan’s “book club” on my The Persistence of Poverty. I am grateful to everyone who gave my ideas some attention, starting of course with Bryan, and I would be happy to continue the conversation if Bryan doesn’t mind forwarding other questions or rebuttals.

Caplan. “You say that single parenthood is not a “global and perennial” feature of poverty.  I suggested that we generalize this to, “not refraining from having children you are not ready to support.”  I add: “In First World countries, this usually takes the form of single motherhood; in the less-developed world and in earlier times, this instead simply took the form of having too many children too early in life.”  Would you accept my amendment?”

My response. Yes. Good amendment.

Caplan. “Karelis is highly confident that unconditional transfers increase work effort.  Virtually no empirical researcher shares this view; they argue about the magnitude of the disincentives, not the sign.”

My response. A 2016 study of seven government transfer programs in developing countries by Abhibit Banerjee and three other MIT and Harvard economists concluded that unconditional transfers to the poor had no effect on work effort. ( This big landmark study confirms other studies listed in the BIEN website (

However, the Banerjee research is consistent with the thesis that unconditional transfers don’t just leave work effort the same but increase it. That’s because Banerjee is aggregating individual country studies that exclude educational effort—generally expended for the sake of later income—from the category of paid work. Much research shows that even people who leave the regular workforce after receiving transfers seek further education. Thus the view that unconditional transfers increase work effort is empirically supportable. My hypothesis about the utility function for income that alleviates the miseries of poverty is obviously not proved by the fact that it predicts behavior that we actually see, but it is to some degree supported.

Caplan. “In any case, you don’t need empirical research to see the absurdity of Karelis’ position.  If unconditional transfers increased work effort, then current recipients of these transfers – many of whom don’t work at all – would work even less if the government cut them off completely.  Are we really supposed to believe that able-bodied welfare recipients would voluntarily starve to death – and watch their children starve to death – rather than work at McDonald’s?”

My response. This is a straw man. As my book clearly states, the hypothesis of increasing marginal utility concerns the shape of the utility function for income levels between subsistence and sufficiency. I don’t discuss the utility of behavior to avert actual starvation. My sense is that the concept of utility is not well-suited to the consideration of life-or-death choices.

Caplan. “He’s wrong on relievers; it is child’s play to list important examples with decreasing marginal utility. We normally take our vision for granted, so fixing blind eyes is a clear-cut reliever.  Now suppose you choose between: (a) one good eye for sure, or (b) a 50/50 gamble between normal vision and blindness.  Karelis’ story implies that most of us would choose (b)! If that seems extreme, sleep is another clear case of a reliever.  Now suppose that tonight you choose between (a) sleeping seven hours a night for the next eight days, or (b) sleeping eight hours a night for seven of the next eight days, and sleeping zero hours on the eighth day.  Karelis’ story again implies that most of us would choose (b)!”

My response. It’s not obvious that the marginal utility of marginal improvements in the ability to see is diminishing. Rather the question is a puzzling one. That’s because there are no clear units for measuring equal small improvements in the ability to see. (It’s certainly not done by counting the number of sighted eyes!) Is gaining the ability to see dark and light when you’ve previously been blind the same size improvement as gaining the ability to see vague shapes when you’ve previously been able only to see dark and light? Is gaining the ability to see colors too when you’ve previously been able only to see vague shapes the same size improvement as the first two? You tell me. So how can we say that equal improvements in the ability to see make for ever-smaller improvements in utility? We can’t. Even a good theory, such as I think mine is, can withstand some good counterexamples, but this isn’t one of them.

My response, continued. As for sleep, the question is again puzzling, complicated by the existence of thresholds like the minimum number of hours to reach REM sleep, and by the fact that sleep isn’t a “more is better” good in the first place, but a “sweet spot” good, where you can have too little or too much. That said, I’d say it’s roughly a reliever at low levels and  sometimes a pleaser at higher levels. With some exceptions (Leonardo da Vinci, reportedly), people don’t generally interrupt their waking activities even when they’re tired to take little naps because the relief they get from fatigue by doing that is small (or non-existent: remember those REM thresholds).