Abortion Restrictions Will Not Cause Overpopulation

Some Basic Statistics About A Sensitive Issue

Yesterday afternoon, I got curious about what percent of the people born in 1991, my birth year, I had outlived. I really had no idea just off the top of my head what the survival curve might look like for my birth cohort. So I looked it up in the Social Security Administration “Life Tables.” It turns out, of those born in 1991, about 98% are probably alive today. That is higher than I’d expected, but fair enough!

But then I got more curious. I’ve heard pro-life campaigns give off-hand estimates of how many abortions have occurred, and structure the claim like a cohort-specific mortality event, making claims like, “30% of the people in my generation were murdered before birth.” With that structure of claim in mind, I decided to produce abortion-adjusted mortality survivorship data.

One way to do this is to simply assume all abortions would have become births. But that’s probably unfair. We know that abortions skew towards pregnancies that may never have come to term anyways, we know that in the absence of abortion maternal and infant mortality usually rises somewhat, and we also know that when abortion is restricted, contraceptive usage rises significantly. So if abortion were illegal in 1991, my birth year, it simply is not the case that 100% of those abortions would have actually shown up as births.

It’s much harder to say what percentage would have, or, rather, what the net effect on births would have been. Illegal abortion might make individuals more sexually cautious, perhaps producing fewer accidental pregnancies, which could actually lead to illegal abortion reducing the birth rate! But that’s sort of a fringe view; it’s not likely that the elasticity of fertility with respect to abortion is that large, especially if contraception remains available.

So below, I show the survivorship for the 1991 birth cohort as the Social Security Administration estimates it, with two alternate scenarios. In one, I count 100% of abortions as deaths. In the other, I count only 25% of abortions as deaths. Neither of these should be interpreted as moral statements about abortion; rather simply as ways to conceptualize the demographic impact of abortion. Indeed, I’m going to try to write an entire post here about abortion and avoid making any statements about the moral/political question of abortion. Let’s see if I succeed.

So, depending on whether you want to take a “net fewer lives” perspective (orange) or a “actual abortions carried out” perspective (red) or an “abortion has no survivorship relevance” perspective (green), the survivorship curve varies substantially. If you think abortion is not relevant for thinking about survivorship, then 98% of the 1991 birth cohort is probably alive today. If you think abortion is relevant, and every abortion should be counted as a net survivorship decline, then about 73% of the 1991 birth cohort is probably alive today. And if you think abortion is relevant, but think an abortion ban would be partially offset by behavioral changes and other health factors which are also relevant for estimating a counterfactual survivorship curve, then about 90% of the 1991 birth cohort is probably alive today.

But wait. That’s using CDC data, which has incomplete coverage. Most abortion experts suggest that Guttmacher Institute data, based on surveys of clinics, is more accurate. Below is a chart comparing the two sources.

Here’s the survivorship curves shown above, using Guttmacher Institute data.

To the naked eye, identical. But the numbers turn out to be a bit different. Based on Guttmacher data, our two abortion-inclusive estimates lower survivorship of the 1991 cohort today to 71% or 89%, so marginally lower.

When I presented this on Twitter, I received a response suggesting that my data showed that we should be thankful for abortion, because otherwise the United States would be overpopulated. I would link to the tweet, but it has been deleted. I am not 100% sure if the tweet was serious or parody. But it’s an interesting question, and one I’ve heard many times: without abortion, we would have overpopulation in the United States.

Is this claim true? Well, the first question to answer is, if abortion had been illegal since 1970 in every state, how many births would have occurred?

I won’t bore you with all the stats, but basically, I assume that less than half of those abortions would ever have manifested as births. The exact ratio varies by year. I use a “contraceptive offset” factor that rises over time from 1970, and a “miscarriage offset” factor that falls over time from 1970, and a “cultural change” factor that rises with time; i.e. I consider that with fewer abortions, pressure to adopt contraception and otherwise postpone births will be greater. The result is this line showing the share of abortions I reclassify as births:

But that’s not all. With more births, population on the whole rises. I assume the birth rate among the population is otherwise identical to without abortion, which is a reasonable enough assumption. I do slightly elevate the death rate, on the assumption that some of the kids now born will be less healthy, and parental resources more divided. I slightly lower the immigration rate and slightly raise the emigration rate, essentially positing that higher fertility would partially replace immigration. I will use the Guttmacher Institute data, since it is considered to be more reliable, with some very minor extrapolations in early/late missing years.

Here’s U.S. population since 1970, showing the actual history versus a plausible counterfactual where abortion is banned.

As you can see, under my plausible parameters, an abortion restriction would produce a noticeably different population time series. Population in 2016 would be about 18 million people higher, or about 5.6%.

This estimate is highly sensitive to the specification made about what percent of abortions would have been births were abortion restricted. If we assume a larger percentage of conversions, the effect is larger. The chart below shows different conversion rates for abortions-to-births.

As you can see, at 100% conversion of abortions to births, the effects on population are quite substantial: we would have over 430 million Americans today. But let me be crystal-clear: that yellow line is not a plausible scenario. No demographer with any credibility as an analyst would endorse that line as a plausible population scenario under an abortion-restricting set of policies. The actual plausible forecasts range would range from a bit above the orange line to a bit above the black line. Had abortion been federally restricted in 1970, population would probably be between 330 and 370 million people, versus 323 million currently. I think 350 million is a good number to have as your mental benchmark for “population in 2016 had abortion been made illegal everywhere.”

The question, then, is whether any of these figures credibly represent “overpopulation.” This is a thorny question because we can define overpopulation many different ways. The most rigorous definition would be a plain Malthusian reading: overpopulation is the point at which the caloric needs of the population exceed the trade-adjusted caloric capacity of the land. The US food system, with no particular strain, produces about 4,000 calories per person. It is reasonable to believe we could escalate as high as 6,000 or 7,000 under pressure conditions. Let’s say 6,000. Assuming the average person can survive on 2,000 calories, that means that we could support up to about 1 billion people in the United States. So by a simple caloric definition of overpopulation, none of these abortion scenarios get us anywhere close to the danger zone.

But that’s not the only viable definition. Maybe we could support these people in caloric terms, but maybe they would exhaust other resources. I don’t have good data on how to calculate this; but I should note that there’s fairly little evidence that population changes have direct impacts on society-wide energy use. That is to say, a poor country getting 15% richer has a vastly larger impact on resource use and energy use than a same-size rich country experiencing 15% higher fertility. Population effects are extremely small compared to income-level and intensity-per-dollar effects. Furthermore, such population effects as do exist are extremely delayed, and are likely to occur far beyond any plausible tipping point. Ergo, “overpopulation” as a resource-use concern is probably mistaken. At any plausible population, we are going to exhaust our resources and all perish in fire or flood, unless we find ways to reduce energy intensity per GDP dollar. So I think resource-depletion generally is not a great angle on this.

But there’s another way we could look at this, I believe the most responsible way. Overpopulation can be conceived comparatively: would we be more densely settled than other countries?

Well, current U.S. population density is about 86 people per square mile. At the 100% conversion ratio (extreme upper limit of possible effect of an abortion ban), U.S. density would be 115 people per square mile.

The population density of the European Union is approximately 300 people per square mile. In other words, under the absolute most extreme possible outcome of an abortion ban, the United States would still be just barely over 1/3 as densely populated as the European Union.

Americans should always remember an important fact: we have a very large amount of land. Granted, much of it is in Alaska; population density in the continental US would be about 140 people per square mile, still not even half of current European population density.

While many advocates on both sides of the debate may have many arguments about abortion, overpopulation should not feature as a significant concern with abortion restriction. It is not a serious risk in the United States under any plausible definition of the word. This is not the same as saying global overpopulation is not a risk, it is simply noting that the United States’ population trajectory will be a rounding error of any major population question. What matters is population in the developing world. Fundamentally, there is no reason the United States could not have 600 million people without breaking a sweat. Whether that would be desirable depends on a whole host of factors, but it is not debatable whether it would be economically possible.

One last note. Abortion restriction regimes could radically alter birth rates. Here are crude birth rates under the various different conversion ratios for abortions that I’ve described.

This chart should make two things clear. First, even a small abortions-to-births conversion ratio would create an extremely different fertility experience for women and, ultimately, for all of society. 20% higher fertility at peak is nothing to sneeze at. And if you adopt the high-plausible conversion rate, the effect is even larger. Indeed, we can see that without abortion, there is a very noticeable “echo-boom” generation, whereas historically “echo-boomers” (i.e. Millennials) are not as prominent in the birth rate data. And of course the 100% conversion factor, which, let me reiterate, is not actually plausible, would put crude birth rates in the 1980s higher than during the 1950s baby boom. The point is though, while no plausible abortion scenario can actually create overpopulation risks, different abortion regimes could have extremely different impacts on the lives of women and families.

Second big thing: Even if 100% of abortions were converted into births, crude birth rates would still fall dramatically. The decline in crude birth rates is not driven by abortion. It’s driven by an aging population alongside fertility falling for other, non-abortive reasons. Pharmacological or behavioral pregnancy prevention or avoidance is becoming more widely adopted and effective, ranging from increasingly scientifically supportable “Natural Family Planning” techniques, to better long-term contraceptive devices. Furthermore, sexual behavior is changing, with young people having less sex, resulting in sharply falling youth-fertility. When people of any age do have sex, it is more common nowadays for it to be sex of a variety which cannot produce children, whether sex between potentially reproductive pairs opting for different sexual practices, or else due to the higher incidence of sex between pairs with no reproductive potential. Falling fertility since 1990 can be blamed on many things, but abortion is not one of them.

My hope is that, as I did recently for immigration, I’ve waded into this thorny issue with a reasonable amount of neutrality, and informed you a bit about a hot-button topic in a way that was useful for you, regardless of your prior beliefs. If I’ve failed, my apologies, leave me a comment telling me why. I have no doubt many readers know or can rapidly discern my personal views on this topic, but I do hope that won’t color your ability to trust the information I’ve presented. And for anybody who wants the underlying data, I am, as always, happy to share. Just ask.

Check out my Podcast about the history of American migration.

If you like this post and want to see more research like it, I’d love for you to share it on Twitter or Facebook. Or, just as valuable for me, you can click the recommend button at the bottom of the page. Thanks!

Follow me on Twitter to keep up with what I’m writing and reading. Follow my Medium Collection at In a State of Migration if you want updates when I write new posts. And if you’re writing about migration too, feel free to submit a post to the collection!

I’m a native of Wilmore, Kentucky, a graduate of Transylvania University, and also the George Washington University’s Elliott School. My real job is as an economist at USDA’s Foreign Agricultural Service, where I analyze and forecast cotton market conditions. I’m married to a kickass Kentucky woman named Ruth.

My posts are not endorsed by and do not in any way represent the opinions of the United States government or any branch, department, agency, or division of it. My writing represents exclusively my own opinions. I did not receive any financial support or remuneration from any party for this research.