[Edit since this is blowing up -- I did not mean to imply that the *only* Democratic policy that decreased abortion rates was increased access to contraception, but the first bullet point kind of reads that way. I regret the implication. Second, under "update" at the bottom I have added an analysis of the Guttmacher abortion rate data, which differs somewhat from the CDC data]
One perspective I regularly hear from Christians in my life is that while Republican policies are "not ideal", we nevertheless have no choice but to vote Republican because Democrats are the party of abortion.
There are very good arguments against this perspective, many of which focus on showing that other issues can outweigh the effects of abortion. And I certainly agree with them! But we've been making these arguments for years, and we're still stuck in a world where more than 80% of White Evangelicals are planning on voting for Donald Trump, many specifically because of the abortion issue.
I think there's an even simpler argument to be made: if you're a single-issue pro-life voter, you should vote for Biden because Democratic policies reduce abortion and Republican policies do not.
To establish this, there are three separate questions to examine:
Do Republican and Democratic presidents affect the national abortion rate?
Do Republican and Democratic presidents affect the international abortion rate?
What about Roe v. Wade?
The answers are perhaps surprising: Democratic presidents tend to reduce the domestic abortion rate more than Republicans, Republican presidents tend to increase the international abortion rate in ways Democratic presidents do not, and it's complicated and unclear what would happen if Roe were overturned.
Do Republican and Democratic presidents affect the national abortion rate?
It's hard to be sure, but let's look at the CDC's statistics on abortion rate per 1000 women between 15 and 44 years of age, color-coded by the president's political party: (It doesn't look like the statistics have been updated post-2016)
There's a clear trend: a steady increase throughout the 70's, after which the abortion rate stagnates during Republican presidencies and significantly drops during Democratic presidencies.
Why might this be? It turns out that one of the most effective ways to reduce abortion is by increasing access to contraception, which in American politics tends to be associated with the left.
One way to test this theory is to look at the birth rate. If the abortion rate is falling because of things like abortion restrictions that cause more live births, we'd expect the birth rate to rise during the periods when abortion rates are falling. On the other hand, if the abortion rate is falling because of things like contraception that lower demand for abortions, we'd expect the birth rate to fall somewhat.
The trends are a bit less clear here, but again we see dips during Clinton's and Obama's presidencies, which support the theory that increased access to contraception drives the falling abortion rates.
We'll come back in a bit to the question of what happens if Roe is overruled, but if that doesn't happen, the data are clear that, at least in recent years, Democratic presidents have saved many more unborn lives than Republican presidents.
Do Republican and Democratic presidents affect the international abortion rate?
Outside of war, the primary way in which a president affects global abortion rates is through foreign aid, most notably by choosing whether or not to enact the so-called "Mexico City Policy".
On its face, the Mexico City Policy appears to be pro-life: when it is in place, the US Government will not fund NGOs that provide any sort of abortion counselling or referrals, advocate for abortion rights, or otherwise come across as "too pro-abortion."
The situation is made more complex, however, by the fact that the vast majority of family planning NGOs (among others) also provide abortion counselling or similar pro-choice services. Since, as we saw above, access to birth control is one of the only policies that actually decreases abortions, the Mexico City Policy risks inadvertently increasing abortion rates by blocking access to contraceptives.
Does this happen in practice? A paper by Bendavid et al studies this question in Sub-Saharan Africa, and finds that people exposed to the policy are two-and-a-half times as likely to have an abortion than they otherwise would have been. This is an insanely large effect size -- even Roe v. Wade didn't double the US abortion rate!
Here's a graph from their paper showing abortion rates in countries less affected by the policy (because they receive less family planning aid) relative to those more affected. Countries with little US aid have little change in abortion rates, while abortion rates in high-exposure countries absolutely skyrocket.
This fits with a common theme in economics: plausible-sounding interventions may or may not actually lead to the outcome they're aimed at, and sometimes they cause the exact opposite.
Once again, we see that while Republicans talk about their opposition to abortion and even (attempt to) act on it, the Democratic policies are the ones with the actual effects.
What about Roe v. Wade?
A significant chunk of the pro-life movement has pinned their hopes on the possibility that the Supreme Court might overrule Roe v. Wade. It's hard to put actual numbers to a policy that hasn't been tried, but it's not at all clear that overturning Roe would decrease abortions much at all:
Overturning Roe wouldn't make abortion illegal on a national level, but rather would leave states to decide. Pro-life laws would probably be put in place in red states, which already have lower abortion rates.
There are now fairly effective abortion pills that work at an early stage, and are easy to send (and hard to detect) in the mail. In a Roe-free world, expect the "illegal abortion pill" market to take off. (It's even plausible that this increase in infrastructure could make abortion more common, since abortion access is currently difficult for many people even where abortion is legal.)
Expect a huge blue-state backlash, to the point of "people paying for other people to travel to blue states to get legal abortions". This will probably have a negative effect on other nonprofits by taking up funding they otherwise would have received.
In countries where abortions have been banned, the birth rate often spikes (due to the lack of abortions and, typically, birth control), then mysteriously falls as people find more illicit (and dangerous) ways to procure abortions. It seems like the long-term effects of abortion bans on abortion are not very large, while the effects on people seeking abortions are horrific and often permanent.
What is a Pro-Life voter to do?
If legal restrictions do a poor job of preventing abortions (but a good job of endangering people seeking abortions), what can a pro-life voter do to achieve their goals?
It's difficult to reduce the supply of abortions, but it's possible to reduce demand! Increased access to birth control, reduced poverty rates, better adoption services, and access to affordable childcare all reduce abortion rates in ways that anybody, pro-life or pro-choice, can get behind.
Stand up for your convictions, but do it effectively. Vote for Biden, then find ways to help your community and beyond. Support babies by supporting their mothers.
(Please vote for Biden.)
UPDATE (much more math ahead, sorry)
A few people have pointed out to me that the Guttmacher institute has an in-depth estimate of abortion rates, which differs somewhat from the CDC estimate. Here's a typical graph:
There's a bit more variation in the data here, which makes it harder to "eyeball" the effects of Democratic and Republican presidents, so we will have to apply some math to see what's actually going on.
I fit a regression to their data post-1980 following what we could see in the CDC model's data -- namely, a linear function whose slope at Republican and Democratic presidents is allowed to be different. Here's what we get:
The coefficient of "year" is the expected yearly change in abortion rates under a Republican president, so on average we see about a .26 point decline in abortion rates per year when a Republican president is in the White House.
The coefficient of time_dem is the expected additional yearly change in abortion rates under a Democratic president, so on average we see a .26 +.38 = .64 point decline in abortion rates per year when a Democratic president is in the White House.
The fit of the model is good (R^2 > .99, meaning roughly that the model explains more than 99% of the variation in domestic abortion rates), and the difference between the effects of a Republican and Democratic president are highly statistically significant (p < .001), so we have strong evidence that abortion rates have declined quite a bit more when Democrats control the presidency than when Republicans do.