Abortion Bet with Bryan Caplan

And thoughts on betting and the future of abortion

I

I have made an abortion bet with Bryan Caplan similar to the one I discussed here. The final bet is with different odds – 8:1 (but adjusted for inflation) rather than the 10:1 I initially proposed. The bet is “I, Michael Crone, send you $100 right away. You, Bryan Caplan, pay me the inflation-adjusted equivalent of $900 (my $100 plus your $800) in 2024 dollars if on June 1, 2034, surgical abortion is illegal in all states of the US except in cases of risk to the life of the mother.”

Understandably, Bryan requires payment up front when betting people he doesn’t know very well, so the bet required up-front payment from me (which I have already done), The effect of requiring pre-payment (which I did not anticipate when I originally discussed the bet) is that the risk of non-payment due to mortality¹ or any other reason is on me. I calculate my bet to be equivalent to a 6:1 bet on the main abortion question²

With the change from the 10:1 bet I proposed to the bet I made, the bet is basically an even-sided bet, in my estimation. I view the expected dollar value of the bet to be approximately $0 to me. If I were to estimate the odds against abortion becoming illegal in the US, 6:1 is roughly close. So why make a bet where I don’t expect to make money?

II

Well, first off, I don’t expect to lose money either. But that doesn’t answer the question of why I chose to bother with it. I chose to do so as a way of responding to the pessimism that I sometimes run into with my fellow pro-lifers, especially once our side started losing referenda. By putting a number on this, my goal is to counter the pessimism (Aren’t 1:6 odds of saving millions of lives already worth the effort? And, remember, the effort would improve the odds.) without it coming across as the optimistic puffery common to grassroots campaigns (not that I’m running a grassroots campaign anyway). I recognize that there are things going wrong and wish I could honestly predict better odds, but I also have reasons to expect eventual success, including:

1. I maintain that abortion is against human nature and its acceptance has to be carefully taught to each generation.

2. Of all the changes from the 1960’s and 70’s generally classified as socially liberal, abortion is unusual, and possibly unique in that it has retained sustained opposition. Something is going on.

3. Contrary to what you may have heard, the referenda have not been winning by margins that preclude the reasonable possibility of a flip in the majority. Abortion won 57% to 43% and 56% to 44% in Ohio and Michigan respectively – both purple states – but a net change of more than 6 or 7% is well within the changes of opinion that happen over the course of less than a decade. Current polling numbers on pro-life vs. pro-choice are similar to mid-1990’s levels, from which pro-life polling numbers rebounded quickly.

4. There remain a number of strategies to be tried. My current favorite, for what it’s worth, is a campaign using Jay Watts-style apologetics. Of course, our opponents will also work on improving their tactics. My point is that this is another reason to believe the current situation is not stable.

III

Finally, some thoughts on prediction betting in general. The process of making the bet involved much more back and forth communication than I expected. For that reason, I have a much greater appreciation of why individual bets are uncommon. Presumably, a lot of this friction is removed when the bets are made on an established prediction market.

Betting on our predictions is supposed to keep us honest turn us into homo economicus. But my idea for this bet was revived because Frank Atwood made a different bet that was very obviously not designed for his homo economicus drive to expected financial benefit and then was honest enough to agree to be interviewed by me for over 30 minutes on the subject of “Why I made all these silly bets that are obviously not designed for my expected financial benefit”. Then I proposed this abortion bet that I did figure to have my own expected financial benefit, only to decide to accept it after terms became such that I think of it as financially neutral. All this provides actual evidence to me that bets are keeping us honest, but it would take actors using only financial benefits for me to not add at least a grain of salt to, say, prediction market estimates.

Also related to prediction markets: I would not accept the other side of this bet even if the odds were changed so that I expected a financial gain on the other side. I can’t see betting to win on a result that I find odious. This is probably a common enough sentiment that I do not trust prediction markets to correct for on issues where there would be a directional bias of the odious-seemingness. (To be clear, I don’t think there is anything immoral or even inherently distasteful about betting on a result one thinks is odious as long as one doesn’t act to bring about that result. I just think a lot of us wouldn’t do it.)

None of this is to say that prediction markets should be ignored. Perhaps they are fine for now, but they would not scale to widespread use by “normies” like me³. I see their predictions referenced in the parts of the internet I frequent, and I will probably take the results at face value most of the time. I even think this experience moves me in the direction of favorability towards prediction betting, but it will also cause me to add a little more of a that grain of salt for directional biases.

1. I am told it is typical that, in the case of mortality, no further money changes hands in these bets. This makes a lot of sense. I would feel very weird contacting heirs for collection in the case of Bryan’s death.

2. Bryan and I are both in the first half of our 50s. I used the most recent mortality statistics by age that I found online, which are admittedly from an usual year, 2020, although a brief look at overall mortality showed not that much of a decline since. I went through a few different ways of estimating mortality risk using that data. I settled on a pessimistic estimate for mortality that used the 2020 data and the somewhat too old age group of 55-64 year old males and, as a counter-balance, I ignored any other reason for payment risk. That gives a mortality risk from the table of 1.3233% for each of us each of the next 10 years, or an overall pessimistic mortality risk of 1-(1-0.013233)^20 = 23.4%. Using 23.4% as the complete payment risk, the 8:1 bet has a zero expected value if the probability of surgical abortion being illegal in 2034 is 14.5% since (1-0.234)*0.145 = 11.1% which is 1/9 or 8:1 odds. 14.5% is equivalent to about 5.9 to 1, which I will call 6:1.

3. Hard to get over the discombobulation of being in the normie group on this one.