Karelis Responds to You
By Bryan Caplan
Here are Charles Karelis’ responses to readers’ questions.
Would I prefer EITC or a guaranteed basic income?
My response. I’m not sure, but here are the key pros and cons, as far as I can see.
EITC is great because it combines two upward vectors pressing on work effort: an income effect (according to my theory of IMU poverty relievers) and a substitution effect (because it makes non-work more expensive). However, EITC is invidious. Not everybody gets it, and it exacerbates class antagonism between people who see themselves as pulling the wagon (taxpayers) and people they see as riding the wagon (recipients). It’s also expensive to administer and a hassle for the recipients themselves. UBI has negatives too. It would be way, way more expensive, pushing US tax rates into the middle of the EU pack. There would be no positive substitution effect, since it’s not work-contingent. The pros are that it would eliminate the class antagonism inherent in EITC, since everybody would get it. What’s more it would be easy to administer and receive. Above all, it would eliminate poverty, which is a miserable condition and an embarrassment to a rich country like ours. And as a cash program, it would get government out of the business of telling people what their spending priorities should be. People could spend their money on health care, or education, or leisure, or whatever they themselves thought was most valuable.
(From the threads.) Isn’t CK’s theory just Kahneman and Tversky’s prospect theory with sufficient income rather than current income as the reference point?
My response. Yes. I hit on my theory independently, but the comparison is correct. I think I am right, and that prospect theory is wrong. If prospect theory were right that people are generally risk averse over gains and risk loving over losses, then there would not be broad differences in risk tolerance according to income level. After all, people at every income level are equally likely to see an increase or a decrease. But there are broad differences in risk tolerance by income level, which is consistent with my theory and not theirs. Why did Kahneman and Tversky make this (I think) error? I think their lab tests and thought experiments were too artificial to mirror real life. Trinkets (or small, out-of-context sums) gained or lost aren’t in any obvious way enhancers of pre-existing happiness or relievers of pre-existing misery being felt by student experimental subjects. They’re unconnected. So the subjects have nothing to measure by but whether they are gains or losses relative to current holdings of e.g. trinkets. Try the same experiments with real sources of unhappiness! If you had eight nasty dents in your new car and were offered a choice between a 50/50 chance of getting them all repaired and the certainty of getting four repaired, which would you choose? I think you’d take the risk. Here, contrary to prospect theory, is a clear case of being risk loving over gains, as is typical when the gains are gains in reliever goods.
(From the threads.) Haven’t I underestimated the importance of weak impulse control in accounting for the poverty-prolonging behaviors of the poor?
My response. Weak impulse control is certainly a factor in many poverty-prolonging behaviors, but my theory claims to account for weak impulse control too. As psychology has stressed in the past few years, exercising will-power is like doing work with one’s muscles: it’s tiring. Like other work, exercising will-power will tend to occur only if the perceived gain is greater than the exhaustion it brings on. But now poverty means trouble, and troubles (like pleasures) drown each other out. So the “will-work” needed to alleviate the first few in a poor person’s stack of troubles, whose repair will hardly be noticed, is less likely to be expended than the (let’s suppose equivalent) will-work involved in alleviating the last few troubles in the stack, where the gained utility per unit of effort will be greater.
As this leads us to expect, it’s less likely that an obese person will muster the effort to lose the first five pounds that he or she needs to lose than it is that they will muster the effort to lose the last five. The same unit of effort beings more and more utility as the subtraction of problems becomes progressively more noticeable.
From the threads: What can an individual do to escape the poverty trap?
My response: First, compare down, not up. By comparing yourself with people who are worse off, some of the circumstances you regard as troubles will lose that negative charge. If you only have one bathroom, think about people who have none, not people who have three. If your bathroom is subtracted from your problem-list, the remaining problems on your list will get more attention, and removing them will bring more relief. So you’re likelier to expend the effort to do so. (Just don’t take it too far, to the point of complacency.) Ever notice how many bootstrap success stories start with something like, “looking back, I now realize my family was poor, but we didn’t think of ourselves that way at the time, because we knew so many people who were even poorer”?
Second, recognize that productive effort in one’s own behalf will change what is going to seem rational to you down the line.
Here’s a simple example. Washing a few dishes in a sink full of dirty dishes, however much that feels like a waste of time, will enlarge the appeal of washing the next few, because when you’re done with the first few, then the subtracting of each dish from the remaining pile of dirties will make a more noticeable difference, and so the relief will be greater. “Well begun is half done” was already a proverb by the time of Aristotle (Politics, 5), and where the job is relieving a negative condition, the proverb makes sense. The effort involved in each step of digging out may not diminish, but the felt relief increases, making the task seem more worthwhile as you work through it. Getting through the early steps feels psychologically like half the battle. In other words, the marginal utility of goods or actions that bring relief is increasing.
So why don’t more poor people follow this second strategy already? (Some threads mentioned choice bundling, which I think is the same issue.) That’s because this second strategy requires someone to look beyond what is rational in his or her immediate situation to what will seem rational down the road. That’s hard for all of us, regardless of income. But in the case of people acquiring pleasers, i.e. non-poor people, it’s unnecessary. Non-poor people don’t have to imagine what will seem rational to them if they make an effort now. They’re already on a steep segment of their utility function, so their effort/reward ratio makes effort rational right now.
From the threads: Isn’t a lot of poverty-prolonging behavior just a matter of individuals’ conforming to what they see going on all around them in their communities?
My response: Sure, but that just pushes the question back a step. Why are the behaviors in question (failing to work, save, avoid risk, etc.) common in poor cultures in the first place? My answer is that they are common because—given the circumstance of poverty—they are actually rational, utility-maximizing behaviors, absent the rare capacity to look ahead at what will seem rational in the future if one bites the bullet in the short run. Wash those first dishes.