Cooperate or Defect?

A game-theory perspective on morality

Rob Henderson

writes,

Many products are marketed with the promise of making people hotter or smarter. But there are no pills or exercises or brain teasers marketed toward making people more ethical. Shows you where people’s priorities lie.

But from a Darwinian fitness standpoint, morality differs from looks or intelligence. Strictly speaking, it is not in my interest to actually be more ethical. It is in my interest to appear to be more ethical. So you would expect products to be marketed that help people appear to be more ethical. And, if you think about it, there are. Virtue signals are everywhere you look, and some of them are quite expensive (I will argue below that a college degree is an example).

I am looking at this issue in terms of game theory. I am thinking of ethical decision-making as the choice of whether to cooperate or defect.

• Seeing a package left on my neighbor’s doorstep, do I cooperate (leave the package alone) or defect (steal the package)?

• When a group of us are working on something, do I cooperate (contribute my fair share) or defect (shirk, and free ride on others’ labor)?

• When presented with an opportunity to cheat on my spouse, do I cooperate with my spouse (spurn the opportunity) or defect (have an affair)?

• If I own a bank, do I cooperate (only take risks proportional to my skin in the game) or defect (pass my losses along to creditors or taxpayers)?

When we pass moral judgment on these choices, we say that cooperation is ethical and defection is unethical.

A society as a whole is better off if people are making ethical choices. Cooperation leads to positive-sum outcomes. When stealing is common, this acts as a tax on economic activity. When stealing is rare, fewer resources have to be used to try to protect property.

My decision about whether to cooperate or defect is affected by incentives. I will be more inclined to defect if defecting offers a potentially enormous payoff and/or I am unlikely to be caught and punished.

Social institutions can help to foster cooperation by ensuring that defectors are caught and punished. As I wrote in Trust and Accountability, “a high-trust society is one that has an effective process for identifying and punishing cheaters.”

People are more likely to treat me well if I have a reputation as a cooperator. That gives me an incentive to strive to obtain such a reputation.

Actually being a cooperator is one way of obtaining a reputation as a cooperator. Thus, one of the incentives to cooperate is that it enhances my reputation, and that in turn will lead people to treat me well.

But another way to obtain a reputation as a cooperator is to get a “seal of approval” of some sort. Membership in the Better Business Bureau, or a professional society, or a church, or an alumni association, will convince some people that you are likely to be a cooperator.

Bryan Caplan has argued that college graduates are treated well in the job market not because of what they learn but because of what they signal. Many people assume that the college degree signals intelligence, but I am skeptical. Compared with other signals of intelligence that are available, a college degree is noisy and costly. Instead, I would argue, and Caplan would agree, that college graduates have signaled that they are willing and able to cooperate in certain endeavors that simulate a work environment. Employers want to hire cooperators, not defectors.

A society is corrupt when the signals of cooperation become misaligned with the actual propensity to cooperate. Becoming a priest or a coach signals that you want to mentor young people. Sexual abuse scandals indicate corruption. Going into government signals that you want to provide “public service.” Becoming wealthy and dictatorial indicates corruption. Samuel Bankman-Fried signaled that he was a cooperator by donating money to various causes, but it turned out that this was money obtained by illicit means.

There are worrying trends in higher education. Students in college seem to learn to exaggerate grievances and acquire a sense of entitlement. Organizations now have to worry that hiring a college student means getting a defector, not a cooperator. Employers may find it necessary to discard the college diploma in favor of other signals.

We want virtue signals to actually signal virtue. But that is a complicated issue. I may disagree with you that “eating local” is virtuous, and yet I might still regard your pledge to “eat local” as a signal that you care about virtue and will deal honestly with me. (Alternatively, I might regard it as a signal that you are not a reliable partner in a transaction, because you over-estimate your virtuousness.)

Society has had centuries to experiment with various approaches for cost-effectively separating cooperators from defectors. It remains an imperfectly solved problem.

This essay is part of a series on human interdependence.