I recently used cost-benefit analysis, and estimates of the dollar value of life, to consider the sensitive issue of covid masks, lockdowns, etc. While it can be emotionally hard to compare money and lives, we must do so if we are to think carefully about such things. Hey, as long as you and I have already paid that cost, why don’t we continue on to deal with another even more sensitive topic: abortion.

In the US about 18% of pregnancies end in abortion, for a total of 862K abortions per year. Many see this as a terrible moral wrong, while others consider it more wrong to limit mothers’ choices. When framed as a moral issue, we get stuck, and so one side or the other just wins if they have more political power. I’d like to frame this issue instead in terms of economic efficiency, and see if we can’t make more progress.

First, is abortion efficient? In the US today, mean GDP per capita is $67K, while mean lifespan is 78.5 years. Given the reasonable estimate that one life year is worth three years of income, the total value of a life at birth becomes $16M. All abortions in a year are then a loss of $13.7T, which is more than the $10.9T estimated loss that the US will suffer from covid from all causes for all time. So abortions are objectively a big deal. (Total US annual GDP is $20.5T.)

Middle income parents apparently spend $234K to raise a kid over the entire childhood, not including labor costs. And a poll I just did suggests that about 2/3 of abortions would be prevented by offering the woman $100K cash. So if the total cost to create a kid is about $0.5M, while the value created is $16M, that’s a 32 to one return on investment! Yes, taking interest rates into account would reduce the rate of return. And if we used the median income of women who abort today (75% of them make <$31K), instead of average income, that might cut life value to $4M, though it may also cut the costs to raise a child by a similar factor.

But still, it sure looks like abortion is quite often inefficient in a cost-benefit sense. And there may be even larger inefficiencies associated with failing to create more children more generally. After all, the calculation seems similar; the value the child gains from their life seems to often be far more than the cost the mother incurs to create that child.

This analysis suggests that we should seek ways to reduce abortions, and encourage fertility. Perhaps even via finding a way make these into win-win deals. You see, there’s a general econ theorem to the effect that it is always possible to find a set of cash transfers to make everyone prefer the package of those transfers added to any economically efficient policy. So if avoiding abortion is efficient, there should be win-win deals to make that happen.

The obvious win-win deal to consider here is to have the child later pay back the the cost of creating and raising them. That is, if not for legal barriers, pregnant women not inclined to raise their children would in effect pay others to raise their kids and endow those kids with debt that they must pay back over their lifetimes. If children usually end up valuing their lives more than the cost of creating and raising them, then we expect people to find the win-win deals possible here. In these deals, the mother prefers to have the kid, someone prefers to raise them, and the kid is grateful to exist, relative to the alternate scenario of having been aborted.

However, in actual practice we don’t allow adoptive parents to pay the original mother more than limited expenses, nor do we allow parents to endow their kids with debt that they must repay and cannot easily evade via bankruptcy or emigration. So it seems we have two choices:
A) expand freedom of contract to allow such deals between moms, child-raisers, and children, or
B) have the government step in and try to produce a similar effect, suffering its usual reduced flexibility and adaptation to context.

A government-based solution could look like this. The government pays each mother per child born, and pays people to raise those children, giving preference to the mother if she’s willing. When children are grown, they pay taxes to compensate for these expenses. Maybe these policies are very uniform, with every mother getting the same payment and every person paying the same taxes. Or maybe payments and policies vary by context, such as by the income of the mother, or the estimated future income of the child. Such adaptation to context is something that private contracting tends to do better, but still the government maybe be able to do some things.

Note that I haven’t talked about whether to allow or ban abortion, which seems a separate issue. I’ve instead talked about how to entice mothers into not having abortions, and perhaps to even plan to have more kids.

This sort of government policy looks a lot like “infrastructure investment”. Except instead of paying to build roads or power plants, it pays to build people. The usual problem with infrastructure investment is trusting the government to choose the types and amounts of different kinds of spending, and trusting them to choose efficient suppliers, instead of taking bribes to approve inefficient suppliers. Relative to roads, etc., investing in babies looks easier to manage, and requires citizens to less trust the details of government choices.

So, win-win babies seem an especially good government infrastructure investment.