On Poverty (or illiteracy), Maybe We’re All Wrong
I generally do not read newspapers, but this one strikes either a chord or a dischord, something I am still trying to figure out.
Steven Pearlstein - On Poverty, Maybe We’re All Wrong - washingtonpost.com
… after doing lots of reading and giving it extensive thought, Karelis concluded that the reason some people are perpetually poor is that they don’t have enough money.
The argument is in a new book, "The Persistence of Poverty," by Charles "Buddy" Karelis, a professor at George Washington University. As Pearlstein puts it, self-sustaining economic behaviors such as schooling, saving, etc. have a goal — to make more money (Uh?). And per standard assumptions, poor folks should value $$ more and therefore should pursue these long-term money-making goals more vigoroursly. But,
In economics, this insight — that the fifth ice cream sundae is less valuable than the first one — is enshrined in the law of diminishing marginal utility.
But what if this iron law of economics is wrong? What if it doesn’t apply at every point along the income scale? If you and everyone around you are desperately poor, maybe it’s perfectly rational to think that an extra dollar or two won’t make much of a difference in reducing your misery. Or that you won’t be able to "study" your way out of the ghetto. Or that if you find a $100 bill on the street, maybe it’s logical to blow it on one great night on the town rather than portion it out a dollar a day for 100 days.
Duh. If this is a new insight, economic departments must have been paying too much salaries, way beyond the sweet spot of rational thinking, penciled by Pearlstein as between 1/2 and 2x median income. And I dare to predict a whole suite of neuroeconomic studies will pop up soon with glowing images of rich and poor brains.
I don’t know how to think about the policy implication of Karelis’ theory, but my neurons are attracted to the analogy between self-sustaining economic behaviors and literacy decisions.
For example, is the 7th Harry Potter book more or less rewarding than the first 6?
I guess the answer is IT DEPENDS. It depends on how much you are into this self-sustaining phenomenon. (And obviously, the life expectency of the HP phenomenon depends on how many self-rewarding neurons Rowling can attract and sustain; but that’s besides the point). If you are in the sweet spot — as Rowling and company have carefully calculated: having read the first 6 volumes and more or less into the plot — then you should weigh the return of the immediate future (reading HP7) more than the return of the past (that you read the HP 1-6; duh) and distance future (e.g., your SAT scores).
(Another digression: without the least background in Econ, I see the law of diminishing return as a discount function — one of many possibilities — in the sense in Temporal Difference calculation in reinforcement learning. )
But if you are not at the sweep spot — you think wizadry is a taboo, or you can’t read — then it appears intuitive to me that you have a different assessment of HP7’s worth, which leads to a different set of value functions that predicts different behaviors, which are completely optimal given your own assessment of reward and punishment. A whole different set of strategies is to be expected.
In English, it struck me as insane to assume that people with different backgrounds will value X (x=$, books, etc.) the same way. I am particuarly surprised — and bothered — that we do not understand how poor people make economic decisions. Neither do we know much about how poor readers make literacy and other decisions. Seeing all the lures of $ or fame and not going after them, are they simply irrational?
It appears to me that educational research has a long tradition of focusing on environmental factors, but makes simple-minded, fatalistic linear models individual choices (if you call them choices ar all) — if you have the following risk factors, then you are unlikely to read — with no regard of the child, how the child makes decisions and how changes can be brought about. Interviews and studies that attempt to understand the "psychology" of (il)literacy stop short of a normaltive theory of how decisions are made. This is an even sadder state of affairs than that in economics. We need a normative, ratonal account of literacy behaviors.
Understanding why someone decide to rely on foodstamps rather than to find jobs, or why some child avoid books while others devour them, requires more than the standard assumption. And if a "rational" model does not describe human behavior, the model is irrational, not the human.