Unwanted pregnancy and default options

It is well known that the default option for a choice hugely influences the outcome of peoples’ decisions. Governments exploit this regularly by, for instance, using opt-out rather than opt-in schemes for pension plans. The idea here is to encourage people to choose the ‘right’ thing without actually constraining their decisions making in any way. People tend to be comfortable with an arbitrary choice of default option for a new scheme. For existing schemes, a decision to change the default option from the status quo might meet with considerable opposition. However, using the power of the default option to influence peoples’ choices could potentially have a huge impact on problems that our society currently faces.

Consider the problem of unplanned pregnancies. They are hugely costly in terms of the emotional and monetary impact on both the families and the health system. Steven Levitt has famously written a paper connecting the number of unplanned pregnancies to the crime rate. Unplanned pregnancies result from the default state of fertility of both men and women. Imagine a solution in which a drug was intruduced to the water supply which rendered either men or women infertile. Suppose that there was a freely available pill which fully restored fertility, so the choice whether or not to have a baby was still unrestricted but the default option was changed. The problem of unplanned pregnancies would essentially be eliminated.

For a more ethically difficult question, consider the problem of HIV transmission. HIV transmission occurs through unprotected sexual contact. Ideally we would inoculate people against HIV to prevent the spread of the disease. However, if that were not possible (and it currently isn’t) then we could prevent the sexual contact in the first place. Suppose the drug in the water now removed peoples’ libido and the free pill restored it. Again, the choice set hasn’t been restricted and the transaction costs of the decision are very low, but the default option has changed: now people default to having no sexual contact. This change in behaviour would, no doubt, hugely reduce the rate of HIV transmission.

Note that in both of theses cases there are massive costs, to both individuals and to society, that would be eliminated if the default option were changed. In both cases huge amounts of money are presently spent to prevent pregnancy or HIV transmission. Yet there would likely be huge resistance to the implementation of either scheme on ethical grounds. Do humans have such a huge status quo bias that we avoid implementing obvious solutions to immensely important problems simply because we’re resistant to changing default options that clearly aren’t working for us? Unfortunately, the answer appears to be yes.

  • Good post.

    I agree with you on the unwanted pregnancy front.

    However, on the libido front I would also like to add that the benefit to society would have to exceed any loss that occurs from a the strategic complementarity between sexual partners. After all, if introducing the pill for libido creates a game where consuming the pill and having sex are strategic complements between people, then in free market interaction we are going to have a sub-optimal amount of sex. In the case without the stuff in the water we are effectively enforcing compulsory taking of the pill – which may solve this issue.

  • I think they are likely to be strategic complements as you say. However, the cost to people of havig less sex would have to be weighed against the benefits of reduced HIV transmission. Note also that we’re not changing the choice set, only re-framing the decision, so it can’t be said that the pill prevents people from utility maximising.

  • “so it can’t be said that the pill prevents people from utility maximising”

    Definitely, I 100% agree with that. Which is why the existence of strategic complements is so important – as the utility maximising choice does not necessarily lead to the highest level of utility (given the strategic element).

    If that did not exist then there would be no problem, because as you say it is a framing issue (unless we argued that the reference dependence was a result of significant transaction costs instead of cognitive bias – which I’m not going to do).

  • Hmmm, interesting: is it unethical to re-frame a decision in such a way that peoples’ ‘irrationality’ causes them to choose a pareto inferior option?

    I think any argument about framing being about transaction costs flies in the face of a significant body of empirical research so I can see why you’d avoid that one 🙂