Skill vs morality

There’s a trade-off between skill and moral behaviour, according to new research.

Armin Falk and colleagues got people to sit a form of IQ test, except that some subjects were explicitly told it was an IQ test whereas others were told it was just a questionnaire. Some were then told that for each question they got right, the probability would increase of a mouse being gassed to death. They found that the subjects sitting the IQ test were more likely to answer questions correctly – at the possible expense of the mouse’s life – than those only filling in the questionnaire. They conclude:

Striving for pleasures of skill can have negative moral consequences and causally reduce moral values.

This is not the only paper to establish a link between technical skills and bad behaviour.  Francesca Gino and Dan Ariely have found that creative people tend to be more (pdf) dishonest than others.

There is, I fear, no lack of external validity here. To take a few diverse examples:

 - Scientists sometimes work on projects with potentially dangerous effects, such as enhancing viruses. Falk quotes Sir Mark Oliphant, a member of the Manhattan Project: “if the work’s exciting, they’ll work on anything.”

 - Soldiers can fight well even if a war is unjust.

 - Civil servants can work diligently to implement policies we might find repellent (though it’s questionable how often this is happening at the DWP).

 - Bank traders rarely worry that their work might be socially useless.

 - Accountants and lawyers can be untroubled by – or even take pleasure in – devising tax-dodging wheezes.

 - Even the best footballers, as well as Diego Costa, sometimes dive.

Readers might object that I fall into this category. If I do well in my day job – a matter which I leave others to judge – the best that happens is that already-wealthy people get even richer. I honestly confess to being untroubled by this.

Herein, though, lies a paradox. All this seems to imply that societies with large numbers of skilled workers will have lower ethical standards than others. But this is not the case. There’s little evidence that wealthier nations, which have higher human capital, are in aggregate less moral than poor ones. In fact, I strongly suspect the opposite: richer nations are more tolerant and less crime-ridden than poorer ones. I’d rather walk through the streets of Zurich at night than those of Lagos. This tells us that some other mechanisms offset the tendency for skills to diminish moral conduct. Ben Friedman and Deirdre McCloskey have described these.  

Comments

[cjcjc]

"Some were then told that for each question they got right, the probability would increase of a mouse being gassed to death."

Were they expected really to believe such a thing?

[Luis Enrique]

this kind of stuff strikes me as exposed to the problems currently causing a crisis in social psychology research:

http://andrewgelman.com/2016/03/03/more-on-replication-crisis/

(I haven't read paper for all I know they do everything right)

[Miguel Madeira]

I think there is nothing in the expriment saying that "skilled" people are more amoral; what the expriment says is that people behave in in more amoral ways if their skills are being measured (a very different thing, I think).

Then, is not a conflit between skills and morality, but between pride and morality.

[chris]

@Luis - I don't know if their research is replicable, as it hasn't been tried yet. However, the list of examples I gave suggests that the phenomenon they describe is common in the real world. For me, that informal external validity matters more that p-values. (I suspect we under-rate the power of informal statistical thinking, and over-rate that of formal tests).
@ Miguel. That's a good point. But the real world examples I gave are also all cases where skills are being measured, & where results matter. The phenomenon seems real, even if Falk and I have misdescribed the precise mechanism.
@ cjcjc - yes: mice are often gassed to death after being used in other experiments.

[Christiaan Hofman]

I really don't see how this experiment shows such a connection between skill and (im)moral behavior. That connection seems to be a result from the design of the test rather than the way the test objects behaved. It's the test's design that, artificially, connects "doing well in a test" to "immoral behavior". And finding that skill increases "doing well in a test", well, that's not a big surprise. So the moral part is just a irrelevant side effect from the test. What if the situation was reversed, and the test objects were told the chance of a mouse being gassed was *reduced* instead? I am pretty sure the conclusion would now have been that skill increases *moral* behavior instead, no trade off! See what I did?

[Endrew]

I agree with the others. I can't see the implication you draw from this paper.

[Sukh Hayre]

The reason you feel less like to be a victim of crime in a richer environment probably has more to do with the fact that it's just that the risk isn't worth the reward.

How much of this is a result of the fact that the institutions in place allow the citizens of these countries to benefit from what they are able to 'legally" extract from poorer nations due to their stronger bargaining position and the fact that they benefit from in effect supporting the few corrupt "leaders" in other countries. The developed nation's citizens benefit, the few in a poor country benefit, and the vast majority in the poor country get screwed, so it may feel less safe because these individuals do not have the luxury of having had all their basic safety needs met.

[Matt Moore]

Only the accounting example is valid.

There is no question that people can perform in morally dubious environments. The contention is that personal enjoyment of performance is a benefit that people consider in making moral decisions.

Which, under a utilitarian perspective, is completely correct.