US Abortion Statistics: An Honest Analysis

the debate over abortion is rightly fought over morality, although I think a better grip of the statistics on abortion would aid the discourse

An Analysis of Abortion Facts And Statistics, 1970-2016

Last month I wrote an article debunking Planned Parenthood‘s claim that “1 in 3 American women have had an abortion”.

You’d think that an article on abortion would’ve been easy to research, since it’s such a hot button issue.  But it wasn’t.

In truth, getting accurate statistics about abortion is a pain in the ass, because most of the sources either (i) misrepresent the data to score political points, or (ii) have never heard of the phrase “user friendly”.

That’s why I wrote this article.

I’ve collected (and organized) some of the most basic statistics on abortion, and also answered some of the most common statistical questions, and corrected some common misconceptions—we can’t have a debate unless we get our facts straight.

First-thing’s-first: the following table records the number of abortions per year in the US, how many of which were “first time” abortions, and also the cumulative number of abortions since 1970.

Also, if you’re interested in exploring the data yourself, you can download the abortion statistics as an excel file—it’s much more detailed than what you see below, which I parsed for brevity and clarity.

US Abortion Statistics: 1970-2016

 YearLegal Abortions1st-Time AbortionsCumulative Abortions
1970*193,491106,420193,491
1971485,816267,198679,307
1972586,760322,7181,266,067
1973*615,831338,7071,881,898
1974763,476417,9112,645,374
1975854,853470,1693,500,227
1976988,267543,5464,488,494
19771,079,430593,6865,567,924
19781,157,776636,7766,725,700
19791,251,921688,5567,977,621

 

19801,297,606713,6839,275,218
19811,300,760715,41810,575,978
19821,573,920865,65612,149,898
19831,575,000866,25013,724,898
19841,333,521733,43615,058,419
19851,328,570730,71316,386,989
19861,328,112730,46117,715,101
19871,353,671744,51919,068,772
19881,371,285754,20620,440,057
19891,396,658768,16121,836,715

 

19901,429,577803,42223,266,292
19911,388,937750,02524,655,229
19921,359,145733,93826,014,374
19931,330,414705,11927,344,788
19941,267,415671,72928,612,203
19951,210,883653,87629,823,086
19961,221,585654,76931,044,671
19971,186,039581,15932,230,710
1998884,273465,12733,114,983
1999861,789448,13033,976,772

 

2000857,475454,46134,834,247
2001853,485460,88135,687,732
2002854,122461,22536,541,854
2003848,163457,15937,390,017
2004839,226450,66438,229,243
2005820,151438,78039,049,394
2006846,181456,93739,895,575
2007827,609462,63340,723,184
2008825,564459,01341,548,748
2009784,507433,83242,333,255

 

2010765,651424,17043,098,906
2011730,322392,18243,829,228
2012699,202389,45544,528,430
2013664,435365,43945,192,865
2014*540,537297,29545,733,402
2015340,255187,14046,073,657
201677,98342,89046,151,640

*Notes On The US Abortion Statistics:

  • This data only reflects legal induced abortions, and makes no effort to account for illegal induced abortions, for which there is no accurate data.
  • The majority of the data came from voluntary reports of legal abortions to the Center For Disease Control And Prevention (CDC).  This data was supplemented (when necessary) by data from multiple sources, compiled by Wm. Robert Johnston.
  • First time abortion statistics between 1970 and 1989 are estimated using rates from 1986, 1987, and 1990 (55% average), due to a lack of data.  Given that this number remained remarkably consistent over the remaining period, this is a reasonable assumption.
  • 2014 marks the last year we have accurate data yet compiled.  The reason the 2015 and 2016 abortion statistics are unrealistically low is because the numbers are still being reported and compiled.
  • Regarding the conditional probability of a woman having an abortion in the USA: over the period for which we have good data (1990-2013), the average likelihood that a woman would have an abortion was 1.75%.
    • The average likelihood that an abortion was a “first time” abortion (performed on a woman who had not had a prior induced termination) is 53.92%.  This means that (i) 46.08% of all abortions were performed on woman who had a previous procedure, and (ii) the average likelihood that a woman would have a first time abortion was 0.945%.
    • The average likelihood that a woman had 1 prior abortion was 25.78%.  This means that a woman who received 1 prior abortion had a 47.8% chance of having a second abortion.
    • The average likelihood that a woman had 2 prior abortions was 11.03%.  This means that a woman who received 2 prior abortions had a 42% chance of having a third abortion.
    • The average likelihood that a woman had 3+ prior abortions was 7.4%.  This means that a woman who received 3+ prior abortions had a 67% chance of having additional abortions.

1. How Many Abortions Are Performed Each Year In The US?

Let’s start with the most basic question: how many abortions are there per year in the US?

Between 1973, when Roe v Wade made abortion a constitutionally guaranteed right, and 2014, the most recent year for which we have full data, an average of 1,058,746 abortions were performed each year in the US.

But that doesn’t give us the full story.  We also need to look at how the abortion rate’s changed over time.

As it turns out, the number of abortions per year have declined every year since 1990, when 1,429,577 were performed.

In fact, abortion is currently at historic lows: only 540,537 abortions were performed in 2014, which is the smallest number since 1972.

To put it in perspective, that’s only one-third as many abortions as were performed in 1982, the peak year for abortion in America (1,573,920 fetuses were aborted).

2. Why Is The US Abortion Rate Declining?

Both pro-choice and pro-life supporters argue that their methods have been the most successful at reducing America’s abortion rate.

But as is often the case, the truth lies somewhere in between.

Pro-Choice Argument

The pro-choice camp says that the abortion rate has declined primarily because of increased awareness, and use, of contraceptives.

In other words, more women use birth control, and more men wear condoms than they used to.

In turn, this lowered the number of unintended pregnancies, and therefore reduced demand for abortions—they didn’t get them because they didn’t need them.  Simple.

Although this is probably a contributing factor, it’s not nearly as big as they’d have you believe.

chart showing the historical use of contraceptives in the US: contraceptive use is only loosely correlated to the U S abortion statistics

According to the Center for Disease Control and Prevention, in 1982, 95% of sexually active women aged 15-44 had, at some point, used contraception.

By 2008 this increased to 99%.

Of course, given the prevalence of contraception in 1982 (when the abortion rate was at its peak), this 4% increase in the intervening years can’t possibly account for the difference.

But perhaps it’s not the prevalence, but the type of contraception used.  For example, in 1982 only 52% of sexually active women’s partners used condoms, whereas in 2008 93% did—could an increased use of condoms explain the declining abortion rate?

No, not totally.

The abortion rate was still near its peaks in the mid-1990s, even though condom use was at 90% during this period—that extra 3% is pretty irrelevant.

All in all, the prevalence of contraceptives, and the type used, doesn’t fully explain America’s declining abortion rate (mostly because they haven’t changed all that much in the last 30 years).

Pro-Life Argument

Pro-lifers often attribute the declining abortion rate to the passage of anti-abortion laws at the state level, which limit access to abortion by placing incomprehensible restrictions on abortion clinics, or denying funding etc.

By and large, these laws have successfully reduced legal induced abortion rates—even the pro-choice Guttmacher Institute admits this.  After all, the underpinning logic makes sense: if there are no clinics, then there’s nowhere to get the procedure.  This applies to pretty much anything—no watering hole, no water.

However, as  the Institute points out, said laws only affect legal abortions, not illegal abortions, nor do they prevent someone from getting an abortion in an adjacent state (no clinic in New Jersey, no problem, just drive to New York).

For these reasons, they’re not as effective at reducing the abortion rate as their proponents claim, although they are a contributing factor.

The pro-life side also cites the growth of the pro-life movement as a cultural force: they say they are “winning the culture war“.

While I think this is true in general (millennials are the most conservative generation yet), it doesn’t appear to have much of an impact on abortion rates.

Why?

First, the rates of premarital sex have remained remarkably consistent since the 1950s, regardless of whether the abortion rate was high or low.  The moral of this story is that people are getting laid, regardless of the prevailing cultural winds.

Second, the historical ebbing and flowing of the pro-life and pro-choice movements don’t correlate strongly with abortion rates, according to polling from Gallup—people do what they’re going to do.

statistics about abortion showing the how likely Americans were to support pro-life or pro-choice movements throughout history

Speculations On The Declining Abortion Rate

I’ll be honest with you, I can’t offer you one solid, unassailable answer as to why the rate is dropping; I think it’s a combination of factors.

I think that people today are generally more diligent about avoiding unplanned pregnancies than they were in the past.

Both sides have contributed to this, albeit in different ways.

On the one hand, pro-choice people have pushed for better sexual education in school, and although the rate of contraceptive use hasn’t changed much, people today use contraception more diligently (they’re less likely to skip the condom, or forget the pill), and are more likely to use it correctly (which prevents unintended pregnancies) than they used to.

We’ve all seen the “condom on the banana” demonstration, and that’s probably a good thing (even if it was weird).

On the other hand, pro-lifers have brought much-needed attention to the darker side of abortion (including the existence of an underground fetal organ market), and have ensured that people have greater respect for the unborn.

For example, although half of Americans are pro-choice, two-thirds of Americans believe a fetus is a human life—all Americans, even many pro-choice people, now take abortion much more seriously than they used to.  People are more reticent to abort today than they used to be, and are often more willing to explore other options (such as adoption).

Fundamentally, the abortion rate is declining because American culture is changing: people are better at using contraception, and have a greater respect for fetal life than in the 1990s—this is good for everyone.

3. How Many Abortions Have There Been Since Roe v Wade (1973)?

Of all the statistics about abortion, this one’s the most disquieting (no matter where you stand on the issue).

There have been 44,0112,775 legal abortions since 1973, when Roe v Wade gave abortion constitutional protections.  This, of course, does not include abortions that were performed in private.

Interestingly, this is greater than the total number of legal immigrants in the USA (38 million), and the total number of illegal immigrants (11-33 million) as well.

4. How Many American Woman Have Had Abortions?

Planned Parenthood claims that “1 in 3 woman have had an abortion” in the US and UK.  That’s a lie.

They created this fake “fact” by looking at the total number of abortions (46 million) relative to America’s female population (159 million)—rather than the number of individuals who’ve had abortions.

Basically, they counted women with multiple abortions multiple times, which makes abortion seem more prevalent than it is—Planned Parenthood deliberately inflates the abortion statistics.

For example, Irene Vilar had 15 abortions (and brags about it), and therefore was counted 15 times by Planned Parenthood—people who use abortion as contraception distort the data.

People like Irene Vilar inflate the statistics on abortion by having multiple abortions
Irene Vilar has aborted 15 pregnancies—she must hate condoms (and babies) with a burning passion.

If we look at the actual data, we find that 46.08% of all abortions were performed on women who had previous abortions.

In actual fact, only 25,012,076 million American woman have had abortions between 1970 and 2016 (not even close to 46 million).

That equates to 15.7% of American women, or just over 1 in 7.

This is the same trick that people use to make the divorce rate look so high: we’re told that that 50% of all divorces fail, which makes marriage seem like a bad bargain.

However, what they forget to mention is that only 25% of first-time marriages fail—the difference is made up because people like Ross from Friends just can’t keep it together, and end up getting divorced 3 times.

5. What Is The Likelihood That An American Woman Will Have An Abortion In Her Lifetime?

There are a few ways to look at this question.

First, we could say that since 15.7% of American women have had abortions, it’s reasonable to conclude that there’s a roughly 1 in 7 chance that a woman will artificially terminate a pregnancy sometime in her life.

But that’s rather crude, since it doesn’t account for the fact that the abortion rate has been declining for decades—we have to account for trends if we’re going to make our claim remotely useful (although if I’m being honest, I think predicting stuff like this is a fool’s errand).

For example, in 1980 there were 25 abortions for every 1,000 woman per year; in 2013, that number dropped to 12.5.  And remember, this includes woman with multiple terminations.

Let’s get a more accurate number.

Assume that a woman has 29 child-bearing years, between the ages of 15-44 (this is what the data we’re working with assumes, so we’ll stick with it).

Also assume that the abortion rate continues to decline at a steady rate, and reaches 6.25/1,000 at the end of our 29 year period.  This means the average rate will be 9.38/1,000.  We’ll run with that.

Finally, remember that only half of those abortions are performed on different individuals, while the other half are repeat procedures.  This gives us an average annual abortion rate of 4.69/1,000.

Given all this information, we can calculate that the likelihood that a girl aged 15 (in 2013) only has a 12.7% chance of having an abortion in her lifetime, or just over 1 in 8.

Again, Planned Parenthood isn’t even in the right ballpark.

6. What Is The Probability That A Woman Will Have Multiple Abortions?

This question regards something called conditional probability for dependent events.  In other words, once a woman’s had one abortion, how likely is she to have a second?  Or a third?

Luckily we don’t have to do any calculations, we have the data to answer this question.

Between 1990 and 2013 some 46.08% of all abortions were performed on women with previous terminations.

After a woman had 1 abortion, there was a 47.8% chance that she’d have a 2nd at some point (this is many, many times higher than the base rate of less than 1%).

Likewise, if she already had 2 abortions, there was a 42% chance that she’d have a 3rd.

And of course, if she had 3, there was a whopping 67% chance she’d have another, and another, and another, until she gets too old.

Statistically, women that have abortions often have more than one, and when they have a few, they’ll probably have a few more.

When it rains, it pours.

This is why abortion statistics conform to a fat-tailed distribution, as opposed to a tidy bell curve.

In a bell curve (normal distribution), the further away from the mean you get, the more unlikely an event becomes—at an increasing rate.

However, because the more abortions a women gets, the more abortions she’s likely to get, the decline isn’t as steep as you would expect.

The below graph is exaggerated, but it illustrates my point.  The blue line is what most people think abortion statistics look like, the red line is what they actually look like (again, exaggerated so you can see the difference).

My point is that there are many more women who’ve had 10 abortions walking around than you’d expect, and fewer women with 1.

the statistics about abortion follow a fat-tailed distribution, rather than a normal, thin-tailed distribution
The statistics on abortion follow a fat-tailed distribution, rather than a thin-tailed distribution, because once a woman gets one abortion, she’s more likely to have another.

7. Why Do Pro-Choice People Overestimate The Abortion Rate, While Pro-Lifers Underestimate It?

If you ask your average pro-choice feminist how many women have had abortions, she’ll probably give you a number that sounds unreasonably high (like 1 in 3); on the contrary, if you ask an ordinary pro-lifer, she might say that she’s never even met a women who’s had one, so they must be rare.

It’s not that they’re lying, and they’re probably not being intentionally deceptive, it’s that they’re falling prey to the anchoring and adjustment and availability heuristics (or biases).

Heuristics are mental shortcuts we use to make quick decisions.

For example, if you’re eating at a new restaurant, but don’t have time to read the menu, you might just order a burger because you like burgers, and assume you’ll like this burger too (even though you haven’t tried it).

On the plus side, heuristics usually work pretty well (odds are you’ll like the burger), but they do have some blind spots.

As it happens, estimating probabilities is one of the human mind’s most notorious blind spots, as the behavioral psychologist (and economist) Daniel Kahneman notes in his book Thinking Fast and Slow

Returning to the problem at hand: most people over, or underestimate the abortion statistics because most people live in areas where abortion is either relatively common, or relatively uncommon—there are few states in America that actually reflect the national average.

For example, in 2010 there were 81,282 abortions performed in New York City, out of a population of 8.2 million.

In contrast, in 2010 there were only 77,422 abortions induced in Texas, out of a population of 25 million.

Basically, abortions were over 3 times as common in New York City as they were in Texas.  Also, there were relatively few woman in Texas who’d had more than 3 abortions (only 6.1%); in New York City, this number was 16.5%.

How does this impact how people see abortion?

Simple, in New York City, a person is much more likely to know someone who’s had an abortion, and that person is much more likely to have had multiple terminations.  In Texas, you may have never met someone who’s had one, and if you did, she probably only had the one.

Therefore, someone in New York City would be inclined to think that abortions are relatively common because: (i) memories of people who’ve had abortions are more salient, and this observation is implicitly extrapolated by the mind (availability heuristic), and (ii) people may anchor their understanding of the abortion rate in reference to New York levels, and then walk it down, which will make the number higher than if they started at a lower base rate (anchoring and adjustment heuristic).

Heuristics are the biggest reason why people tend to get the numbers wrong.

The US Abortion Statistics Paint A Different Picture Than You Might Think

People have strong opinions about abortion.  That’s good.  They should.

The questions regarding the sanctity of life, and of freedom, are at the center of what it means to be human.

However, I think the debate’s passion often poisons the facts: people and organizations, like Planned Parenthood, distort the data in order to make their side look more compelling.  This doesn’t help anyone.

In fact, it hurts the quality of discourse as a whole, and can only lead to bad policy.

In the spirit of intellectual integrity, and honesty, I hope you’ve found this article, and information, useful.  I did my best to handle the data objectively, and to analyze the statistics on abortion fairly.

Although I suppose I should disclose my personal biases: I am pro-life, although I think there is room to carve out some exceptions.  I’m not a hardliner by any stretch of the imagination, I’m pragmatic.

If you have any questions, or want me to look at anything else in the article, let me know in the comments below.  I’m open to expanding this article if there’s demand for it.

Share Me

About Spencer P Morrison 62 Articles
JD candidate, writer, and independent intellectual with a focus on applied philosophy, empirical history, and practical economics.Author of "America Betrayed" and Editor-In-Chief of the National Economics Editorial.

Leave a Reply

4 Comments on "US Abortion Statistics: An Honest Analysis"

Notify of
avatar
Sort by:   newest | oldest | most voted
HC
Guest

Enjoyed the article (was a long, but good read). You should add in a section on the racial component of abortion, and just break down the actual numbers. Also, has it changed over time? Thanks again.

JW
Guest

“All in all, the prevalence of contraceptives, and the type used, doesn’t fully explain America’s declining abortion rate (mostly because they haven’t changed all that much in the last 30 years).”

This seems wrong. The types of contraception used have drastically changed in effectiveness over the last 30 years. IUDs, Implants, and Shots are far more widespread and enormously more effective over the long term than condoms and even pills. Condoms have an 85%+ failure rate in normal usage over 10 years. Pills are at ~45% over 10 years. People constantly mess up those methods. Meanwhile IUDs and Implants and outright sterilization, which have become much more common in the last 30, are at 2% over 10 years. Dramatically more successful over time.

The widespread use of much more effective contraception methods has got to have had a large impact on the unwanted pregnancy and abortion rate. Only considering condoms isn’t a reasonable way to judge the effect of contraception.

wpDiscuz