The Ethics and Economics of Incentivizing the Uninformed (2/2017)

Sandro Ambuehl
Department of Management UTSC &
Rotman School of Management

Economists often espouse incentives, since they can lead to desirable outcomes simply by enlarging the set of voluntary choices available. Becker and Elias (2007), for instance, argue that allowing incentives for living organ donation would be a Pareto-improvement. Ethicists, by contrast, are typically queasy about incentives, in particular as they apply to transactions like organ donation, medical trial participation, or surrogate motherhood.

My ongoing research agenda (with various collaborators) attempts to bridge this gap between disciplines. Assertions in the ethics literature are sometimes based on untested behavioral hypotheses, and the toolbox of economics is tailor-made to test them. Economics, on the other hand, is often quick to dismiss any ethical concerns with behavior that may be rational. The ultimate aim of this research program is to help design institutions for the exchange of goods and services that are not only efficient but also ethically sound.

The agenda starts with the question of how incentives affect behavior. To economists, there might appear to be a trivial answer – if we pay people more for doing something, more people will do it, and that’s the end of the story. Often, however, incentives change much more.

To see why, notice that transactions for which incentives appear problematic to ethicists are often highly complex, unfamiliar, and often have irreversible consequences. If I offered you $500,000 in exchange for one of your kidneys, for instance, would you accept? Most likely, you would not want to commit to a decision before you have gathered some information about what living with a single kidney entails. But neither would acquire a PhD in nephrology (kidney science) before making the decision; doing so would be too costly. Instead, you would get some, but not fully encompassing information, and then make a decision. This is where incentives come in – they may change what kind of information you want to see before making a decision. This is why American Society for Reproductive Medicine conjectures that “Payments to women providing oocytes … should be … not so substantial that they … will lead donors to discount risks … The higher the payment, the greater the possibility that women will discount risks.” [1]

Such arguments have led to the restriction of incentives for transactions such as living organ donation, surrogate motherhood, egg donation, medical trial participation and more around the world, even though unpaid participation in these activities is often allowed, or even encouraged. These restrictions do not come for free. For example, about five to ten thousand people are estimated to die prematurely each year because they cannot obtain a donor kidney in time. But there is no direct empirical evidence that incentives actually affect behavior in the way that the ethics literature hypothesizes.

This is the first gap I fill, by running laboratory experiment in which real people make actual decisions in exchange for money. Of course, the law prevents me from purchasing human kidneys in the laboratory, but I can look for a transaction to which a similar behavioral mechanism would apply, if it exists. One important feature that characterizes transactions like kidney donation is that they are unfamiliar, offer room for selective information acquisition, and are hard to evaluate in dollar terms. A transaction that has all of these features, but that I can do with undergraduate experiment participants in the laboratory is the following. I pay people actual money – $3 or $30 – if they ingest a specified quantity of insects such as mole crickets, silkworm pupae or darkling beetle larvae. Importantly, I want to give participants the opportunity to selectively inform themselves about the transaction, just as the ASRM predicts would happen. Hence, before deciding whether to eat the bugs, but after learning about the incentive, participants can choose to watch one of two videos about insect eating. One is called “why you may want to eat insects”, the other “why you may not want to eat insects.”

So what do people do? As the ethics literature hypothesizes, once I promise people a larger amount of money for bug-eating, they become significantly more interested in learning about the upsides of eating insects, and less interested in learning about the downsides. Importantly, this biased information gathering changed the decision subjects ultimately made about whether to eat the insects; it did not merely serve to help people follow through with a decision they would have made anyway. To show this, I withheld the opportunity to watch videos from some participants, allowing me to measure what it is “that they would have done anyway.” And it turns out that giving people a higher incentive is a much more effective way of increasing participation if people have the opportunity to convince themselves to eat the bugs by selectively gathering information than if they do not.

Now imagine that this experiment was not just about eating insects, but about selling kidneys. Imagine you observed that as you pay people more, they become less interested about the downsides of living with a single kidney, and begin to want to learn more about the upsides. Suppose as a result, they have different beliefs about what the transaction entails, and make different decisions. This kind of behavior surely cannot be rational, and must thus be cause for concern?

Not so fast! It turns out that once I carefully deploy economic theory to study how a perfectly rational individual would behave in a setting like this (where one must acquire information, at a cost, about what a transaction entails) I find that they would behave almost exactly like my experimental subjects!

This mathematical model has further implications that I am currently exploring together with Axel Ockenfels (U Cologne) and Colin Stewart of UofT’s economics department: Not everyone can equally easily process the kind of complex information required to understand a transaction like kidney donation. If we increase incentives for such a transaction, who will be most likely to react, those who can easily inform themselves, or those who cannot? The mathematical model predicts that it is often those for whom understanding the transaction is the most difficult whom we should expect to respond the most. This is not entirely far-fetched. Suppose you perfectly know what living with a single kidney is like. Then, it is unlikely that I would change your decision by offering you an additional $1000 for your kidney. It is when you do not have a good understanding what the transaction entails when the higher money amount becomes highly salient, and thus likely to influence your decision. The fact that this may be rational is important. It is nonetheless in potential conflict with the important principle of informed consent, which maintains that the decision to participate in a transaction is ethically sound if it is made not only voluntarily but also based on a good understanding of the consequences of the decision.

In another ongoing project (with Axel Ockenfels) we study when and why people decide to prevent voluntary transactions like paid kidney donation, even if they will not be personally affected by this decision. Professional ethicists worry about large incentives, because they fear those may be coercive. By contrast, most of the experiment participants in our pilot experiment worry about small incentives. They attempt to protect others from accepting bad deals, but they are quite happy if someone receives a large amount of money for engaging in aversive visceral activities.

Overall, I hope this research agenda will help bridge the gap between the disciplines of economics and ethics; and the Centre for Ethics at the University of Toronto is the perfect venue for this. Equally hard at work at bridging this gap are UofT’s Nicola Lacetera and Cendri Hutcherson, whose recent work explores the role of sacred values in economic transactions.

Further Literature

Ambuehl, Sandro, Muriel Niederle, and Alvin E. Roth “More Money More Problems? Can Incentives Be Coercive and Repugnant?” American Economic Review Papers & Proceedings, 107(5), 2015

Ambuehl, Sandro, “An Offer You Can’t Refuse? Incentives Change What We Believe,” CESifo working paper, 2017

Ambuehl, Sandro, and Axel Ockenfels, “The Ethics of Incentivizing the Uninformed. A Vignette Study,” American Economic Review Papers & Proceedings, 109(5), 2017

Ambuehl, Sandro, Axel Ockenfels, and Colin Stewart, “For They Know Not What They Do. Incentives with Costly Information Processing,” Work in progress

American Society for Reproductive Medicine, “Financial compensation of oocyte donors,” Fertility and Sterility, 2007, 88 (2), 305–309.

Becker, Gary S and Julio J Elias, “Introducing incentives in the market for live and cadaveric organ donations,” The Journal of Economic Perspectives, 2007, 21 (3), 3–24.

Hutcherson, C.A., Montaser-Kouhsari, L., Woodward, J., & Rangel, A. (2015). “Emotional and utilitarian values of moral dilemmas are encoded in separate areas and integrated in ventromedial prefrontal cortex,” Journal of Neuroscience, 35, 12593-12605.

Lacetera, Nico, “Incentives and Ethics in the Economics of Body Parts,” Osgoode Hall Law Journal, forthcoming.

Lacetera, Nico, Julio Elias and Mario Macis, “Efficiency-Morality Trade-offs in Repugnant  Transactions: A Choice Experiment,” NBER Working Paper 22632.

Lacetera, Nico, Julio Elias and Mario Macis, “Sacred Values? The Effect of Information on Attitudes toward Payments for Human Organs,” American Economic Review Papers and Proceedings, 105, 5, 361-65, 2015.

[1] In ASRM v. Kamakahi, this provision was challenged on antitrust grounds. A settlement was reached, in which ASRM agreed to remove the specific language “sums of $5,000 or more require justification and sums above $10,000 are not appropriate” from its recommendations.