Attribution fallacy, incentives, and income inequality

It is difficult to read anything on the justification of high salaries these days without running into catch phrases such as “the hunt for talent”, “attracting the best people to this job”, or “retaining human capital.” The core idea underlying this kind of discourse is one that has got a lot of traction in political philosophy in recent decades, too: It is justified to pay certain individuals – be they neurosurgeons, lawyers, or CEOs – financial incentives, because the productive contribution they will make in response benefits us all.

The justificatory mileage we can get out of the incentives argument is much less significant than standardly assumed. As a starting point, consider the debate between philosophers John Rawls and GA Cohen on the issue. Rawls’s difference principle famously argues that income inequalities are just provided they benefit the least advantaged in society. Applying this to labour income, paying the neurosurgeon, lawyer or CEO a premium is justified to the extent that it results in economic growth and higher incomes for those at the bottom of the income distribution.

Cohen counters this reasoning with the following analogy. Suppose someone has kidnapped a child and will only release his victim for a ransom. Cohen points out that holding out for the ransom is clearly a choice the kidnapper makes. In the same way, the high-earners of contemporary society choose to (or, at least, threaten to) withhold their prized productive contribution unless they receive an incentive payment. This, Cohen concludes, reveals a rather unsolidaristic disposition on their part and suggests that incentive payments, though often expedient, are unjust.

This seems like a frustrating place to end up for critics of income inequality. We have diagnosed the injustice, but there is not much we can do about it. I believe this is conceding too much. To see why, consider two important blind spots of the incentives debate both in Rawls and Cohen but also more generally.

First is the assumption that what makes highly skilled and paid individuals special is some innate talent they possess. While talent differentials are certainly real, the high productivity of the individuals in question is arguably just as much a function of the role they play in a highly specialized division of labour and the skills they have acquired in that context. To the extent that their high productivity is due to acquired skills rather than innate talent, someone else could have acquired the skills and done the job equally well. In fact, if they are more intrinsically motivated rather than concerned largely about their paycheck, they might even do a better job.

Second, once someone has acquired a highly specialized skill set that makes them highly productive, they have significant bargaining power. This is precisely why Cohen concludes that paying financial incentives is often expedient. Yet, what if we adopt a longer-term perspective on bargaining? Perhaps society should bargain with potential future high-earners before they acquire the skills in question. Before the neurosurgeon or the lawyer has been trained, any threat that they will not do the job without being paid the big bucks will ring hollow because we can just train someone else to do it instead – again to the extent that productivity is due to acquired skills rather than innate talent. Note that similar kinds of ex ante bargains already exist; for instance, people can get their university fees covered provided they commit to a number of years in rural medical practice, the military, or whatever it might be. Here, they would commit to some form of a salary cap.

Awareness of these two blind spots raises two important questions: Whence the confidence that we will find someone else who will be prepared to do the highly qualified job for less money? Let me just point out that the existence of less inegalitarian social norms in different time periods (the 1960s compared to today) and different places (Scandinavian countries compared to the US, for instance) suggests there is room for manoeuvre here.

How much of an individual’s productivity is due to innate talent versus acquired skills respectively? This is a tricky, perhaps even impossible question to answer. However, it is worth noting that our standard way of thinking about the justification of incentives does not usually acknowledge the role of acquired skills for which the individual does not deserve neither praise nor income.

We are faced with a case of what psychologists call the attribution fallacy. Once we realize that the productivity of all members of society, including those in crucial roles with high responsibility, is in part due to the fact that we are part of a sophisticated division of labour, then we see that incentive payments are not necessary to achieve that productivity. The challenge in operationalising this insight is not to wait until the roles in the division of labour have been assigned. Once that is the case, high-skilled individuals have such bargaining power that they can hold the rest of society ransom.

–

This piece is informed by a recently published article. For more details, see Peter Dietsch, “If It’s Not Your Talent, How Come You’re Getting an Incentive?”, Oxford Studies in Political Philosophy volume 9.