Reverse Prostitution: cognitive biases and conditional cash transfers

Stuart Rennie writes a thoughtful blog on bioethics.net, Can you buy changes in health behaviours? on how the World Bank backs an anti-AIDS experiment paying young people to not contract sexually transmitted infections. The basic idea is not new, conditional cash transfer programs in poor countries have had successes in improving health and education but applying it to sexual health moves it into more controversial territory. Is there anything wrong with “reverse prostitution”?

A FT editorial wonders if "Are the funders saying young Tanzanians cannot be trusted to do what is good for them without a bribe?"

The overwhelming evidence unfortunately seems to be that humans are not great at doing what is good for them. In an unsettling review Christopher K. Hsee and Reid Hastie (Decision and experience: why don’t we choose what makes us happy? Trends in Cognitive Sciences, Vol 10:1 31-37, January 2006) show that in many situations we systematically fail to predict what maximizes our happiness. Worse, even when we have an accurate idea of what makes us happy we still might not do it.

In particular we are bad at predicting how we will feel in a future state we are not in right now - our thinking is quite different when we are dispassionately discussing health, when we are sexually aroused or when discovering we have been infected with something. We rationalize past actions and persist in maladaptive behaviours to retain internal consistency. We are also often impulsive, foregoing later big rewards for smaller present rewards.

However, economic incentives can actually make use of some of these biases. "Lay economism", a tendency to base decisions on financial aspects of a decision rather than happiness aspects, supports conditional cash transfers. If there is a medium such as money in a decision people often maximize the medium rather than expected happiness. Biases that often works against human happiness can be made to support rational behaviour (and indirectly, happiness).

If there is anything economists like to point out, it is that people respond to incentives. Non-economists often regard incentives as a form of coercion: are not these poor people coerced into behaving in a certain way by being offered something valuable? However, that requires stretching the concept of coercion far outside its normal use as compelling people to behave in a certain way using threats of some form of harm. In a coercive situation an outside agent restricts our range of options by making some impossible, or adds strong disincentives to options we would otherwise want to make. Adding value to certain options is not coercive: we can still choose to ignore the difference or not take the added value (while the coercive reduction cannot be foregone if we wish). There might be an element of opportunity cost in the incentive situation that reduces the value of the options because of regret over the options not chosen, but in the case of preventing the spread of a lethal disease the regrets caused by the program appear quite small compared to the gains if it prevents the spread. Yet it is likely that people to some degree will notice their regrets more than their health, and not see the true benefit they have gained.

Another criticism against the program might be that it, just like prostitution, commoditizes people or behaviours. But fighting STDs tends to medicalize sexuality (unless it is done from a moralistic perspective - in which case it often becomes guilt-ridden). In both cases social values may be replaced by economical or rational values, but the good of avoiding the STD is sizeable and may overrule them. A more psychological problem is that the program could replace the internal motivator to keep healthy with an external motivator, making people dependent on outside support for what was previously a natural behaviour. These concerns cannot be judged theoretically, they have to be studied empirically among real people in a real culture. Past experience shows that it is likely that the program needs to be shaped to fit in and that we may discover unexpected side-effects of apparently simple incentives.

There are many ways this project may fail, but the most likely is through unwise outside criticism before the facts are in, making the World Bank discontinue it. The project challenges many moral intuitions (that good people naturally will do what is good for them, that paying for something is bad or makes it bad) and sacred values (that sex and health do not mix with money). But these moral intuitions and values likely biased. Even if they were entirely due to rational thinking rather than emerging from shared cultural outlooks and evolutionary quirks of our brains they would still be built on fallible human thinking. The only way we can check whether we are correct is to test them, and for that the project needs to go on.