Abortion is a contentious issue. There is a diversity of moral systems that have a variety of opinions on the matter (many arguments are internally consistent – I’ve argued on both sides).
[Kaldor-Hicks] efficiency is a pretty sterile approach to policy making. I don’t necessarily think that it should totally govern anyone’s decision making – but it’s an illustrative framework all the same.
It is necessary to assert that more people isn’t a bad thing. “…The Division of Labour is limited by the Extent of the Market…” More people results in more trade, higher productivity, and higher standards of living. This is pretty broadly accepted. One could even argue that we ought to subsidize births. But for the moment let’s continue with the mere premise that more people is either a good thing or at least not a thing which is a net “bad on average” (I’m thinking externalities here).
Simple, one period KH analysis states that those who value something more should be the ones who get their way (utilitarianism). Pareto optimality would require that, in addition, a transfer be made to anyone who was harmed in the process such that they are no worse off. Implicit in both frameworks is that people who value something more also have a corresponding willingness to pay [ignoring strategic value revelation]. This is the basic version of the analysis.
Enter our pregnant mom who seeks an abortion.
One Period Model
Using the primitive analysis described above, we should ask “who has the higher willingness to pay?” Obviously, fetuses and babies can’t reveal their preference or make bids in reality – but let’s put that aside for the moment. For illustrative purposes, let’s describe the costs and benefits. The mom is facing pain, stretch marks, limited drug consumption and physical activity, and a belly-bulge if she doesn’t abort. The baby gets to live for one period, not particularly in pain or enjoyment – just kinda growing and developing in the womb.
It seems clear that the mom should be the one to get her way because fetuses don’t have any productive opportunities and total utility would be lower if the mom had to bear a cost rather than get her way. The fetus would be compensated for the costs borne with Pareto optimality, and the outcome would be more efficient. In most cases the model has the same outcome as a world in which the mom has the “right” to decide what to do with her body.
If this seems ridiculous, then you are right. Something runs afoul.
There are two things which provide the above result:
- A one period analysis
- Neglecting 3rd party welfare
Multiple Period Model
If we extend the model to include intertemporal analysis, then we can see a very different picture. The mom is willing to pay enough to avoid one period of pain, a few periods of discomfort due to pregnancy, and a lifetime of stretch marks & a depreciated body. The baby on the other hand, is willing to pay enough in order to experience all of the net joys of life from birth until death.
The mom has maybe 40 years of life to enjoy while the baby has about 80 years. Further, the child will likely live longer and have a higher standard of living that the mom. So now, we get an obvious conclusion that’s contrary to the one period model. The baby is giving up FAR MORE than the mom is. Now, the baby would be willing to pay the mom a sum much greater than the costs borne by the mom, and everyone is better off.
But this analysis has some implications. For most people, the result would be identical to a world in which people believe in the “right” of a fetus to live. For children who live shorter, miserable, unproductive lives, it may by that the mom’s willingness to pay is higher. The model implies that children with severe disabilities or sicknesses should not get their way and that the mom, who would forego much more than the child, should. Indeed, the analysis implies that eugenic outcomes are among the set of optimal policies.
3rd Party Welfare
Maybe other people have a stake. Maybe they value one outcome over another for any variety of reasons. So we shouldn’t be comparing the total willingness to pay of just the mom and baby. We should be comparing it for everyone with a stake. How does the entire value of one side compare to the other? I strongly suspect that some eugenicists would pay in order to prevent another person with [insert attribute here] from coming into the world. I also suspect that some religious folks would pay to stop particular (all?) abortions.
The outcomes can still be the same as the intertemporal model. There may be an adequate value for all aborted pregnancies such that a birth never happens. But given the moral fervor of those who appose abortion on principle, I highly suspect that there would be far fewer abortions if such transfers and welfare considerations were possible.
I don’t mean for this post to be a policy prescription. Incentive effects, strategic bidding, and technological hurdles all change the analysis. And a too often neglected public choice question is “What is the welfare implication of allowing policy to be guided by such sterile analysis rather than rights?” This is simply the framework that I find enlightening as a consequential analysis of the abortion issue. In moral matters and real cognitive constraints, maybe we should adopt a rule that best reflects the most efficient outcome. And I don’t know what that rule is.
You know what would be cool? If someone did a blog post discussing the design of a transfer system which would allow people to pay pregnant mothers not to abort. Any takers?