Of Men and Mice

Being skilful at something, and being able to signal it to the world, can be a great source of pleasure. Think of top tennis or football players, or all kinds of artists. Humans, after all, like achievement: it is good for their self-image. But being morally upright is also good for their self-image. Which imperative is stronger? Could the drive to excel also drive people to cut moral corners?

Two CESifo Fellows, Armin Falk and Nora Szech, set out to shed light on this question in their latest CESifo Working Paper. They recruited 301 graduate students and assigned them randomly to one of four tests: one that was morally problematic, one that was neutral, and two used as controls. All participants were paid equally, regardless of results, so monetary incentives played no role.

The morally problematic test is the interesting one. On the one hand, the subjects were told that this was an IQ test, which immediately piqued their interest in scoring as high as possible. After all, IQ is one of the most important and universal personal characteristics in determining overall success in life.

On the other hand, they were told that each correct answer to the 52 questions would increase the probability, by a clearly specified percentage, that a mouse (the furry, four-footed kind) they were “endowed” with would be killed by gassing it. To drive the point home, they were shown a picture of their healthy young mouse, and a video showing how such mice were killed. These, by the way, were so-called “surplus mice”, bred for animal experiments but found unsuitable or unneeded. Common practice in laboratories conducting such experiments is to gas them, since they are expensive to keep alive. (The surplus undergraduate students were not gassed.)

An identical test was conducted with another group of undergraduates, but presented as a simple questionnaire, with no talk of IQ scores at all.

It turned out that the students answering the questions that would lead to determining their IQ, faced with competing sources of self-image, showed significantly less concern about the probability of their mice being killed: getting a high score reduced their moral qualms noticeably, compared to the “neutral” group. They scored about 22 percent higher.

It was heartening to see, however, that a few scored zero, on purpose, thus eschewing what the researchers labelled “pleasures of skill” to make sure that their mice survived.

But the overall findings were conclusive: the temptation to excel provides a powerful motive to detach oneself from moral consequences. And in this case, bear in mind, the consequences were transparent and morally unambiguous: the mouse’s life was clearly at stake.

So what happens when the moral consequences are more ambiguous and less immediate? And, if on top of pure self-signalling, you add monetary incentives, public recognition or reputation effects? Well, then you get doping in athletes, scientists producing landmines and nuclear bombs, bankers creating toxic asset-backed securities, gifted writers producing a tyrant’s propaganda, and emissions-cheating automotive engineers.

Makes you look at skills displays rather differently, doesn’t it.

But there are still those guys who scored zero, bless them. The Mahatma Ghandis of our times.

Armin Falk and Nora Szech: Pleasures of Skill and Moral Conduct, CESifo Working Paper No. 5732

Other CESifo Working Papers by Armin Falk
Other CESifo Working Papers by Nora Szech
Other CESifo Working Papers dealing with morality