Book review: The Moral Economy: Why Good Incentives Are No Substitute for Good Citizens, by Samuel Bowles.
This book has a strange mixture of realism and idealism.
It focuses on two competing models: the standard economics model in which people act in purely self-interested ways, and a more complex model in which people are influenced by context to act either altruistically or selfishly.
The stereotypical example comes from the semi-famous Haifa daycare experiment, where daycare centers started fining parents for being late to pick up children, and the parents responded by being later.
The first half of the book is a somewhat tedious description of ideas that seem almost obvious enough to be classified as common sense. He points out that the economist’s model is a simplification that is useful for some purposes, yet it’s not too hard to find cases where it makes the wrong prediction about how people will respond to incentives.
That happens because society provides weak pressures that produce cooperation under some conditions, and because financial incentives send messages that influence whether people want to cooperate. I.e. the parents appear to have previously felt obligated to be somewhat punctual, but then inferred from the fines that it was ok to be late as long as they paid the price.[*].
The book advocates more realism on this specific issue. But it’s pretty jarring to compare that to the idealistic view the author takes on similar topics, such as acquiring evidence of how people react, or modeling politicians. He treats the Legislator (capitalized like that) as a very objective, well informed, and altruistic philosopher. That model may sometimes be useful, but I’ll bet that, on average, it produces worse predictions about legislators’ behavior than does the economist’s model of a self-interested legislator.
The book becomes more interesting around chapter V, when it analyzes the somewhat paradoxical conclusion that markets sometimes make people more selfish, yet cultures that have more experience with markets tend to cooperate more.
He isn’t able to fully explain that, but he makes some interesting progress. One factor that’s important to focus on is the difference between complete and incomplete contracts. Complete contracts describe everything a buyer might need to know about a product or service. An example of an incomplete contract would be an agreement to hire a lawyer to defend me – I don’t expect the lawyer to specify how good a defense to expect.
Complete contracts enable people to trade without needing to trust the seller, which can lead to highly selfish attitudes. Incomplete contracts lead to the creation of trust between participants, because having frequent transactions depends on some implicit cooperation.
The book ends by promoting the “new” idea that policy ought to aim for making people be good. But it’s unclear who disagrees with that idea. Economists sometimes sound like they disagree, because they often say that policy shouldn’t impose one group’s preferences on another group. But economists are quite willing to observe that people generally prefer cooperation over conflict, and that most people prefer institutions that facilitate cooperation. That’s what the book mostly urges.
The book occasionally hints at wanting governments to legislate preferences in ways that go beyond facilitating cooperation, but doesn’t have much of an argument for doing so.
[*] – The book implies that the increased lateness was an obviously bad result. This seems like a plausible guess. But I find it easy to imagine conditions where the reported results were good (i.e. the parents might benefit from being late more than it costs the teachers to accommodate them).
However, that scenario depends on the fines being high enough for the teachers to prefer the money over punctuality. They appear not to have been consulted, so success at that would have depended on luck. It’s unclear whether the teachers were getting overtime pay when parents were late, or whether the fines benefited only the daycare owner.
You linked to the daycare study, so I read it before continuing the post…and it answered one of your questions: the money went to the owner, not the teacher. I think that the teachers were not paid overtime, which is a separate issue. The teachers just recorded who was late, as they had been doing before, and the fine was added to the monthly bill. It probably would have been more aversive if the fine had been demanded immediately (though for all I know the teachers had already been scolding the parents).
Before the fines, there were an average of 2 late arrivals per day, so it was pretty rare for teachers to leave on time. So bumping it up to 4 didn’t create much more work for them, unless the severity of lateness also increased, which it probably did, but the paper doesn’t seem to talk about, just dichotomizing at 10 minutes. At a fine of $2.72, they probably could pay for an hour of teacher overtime. Probably they should double the fines. So I think the system passes on a willingness-to-pay metric.
(The shekel to dollar conversion is given in the paper. Perhaps a more relevant comparison is that the fine was 2/3 of the minimum hourly wage.)
The teachers did not know that it was an experiment.