Introducing the WeCount report.
You may have recently seen headlines like these:
These articles are based on the October 2023 WeCount report. WeCount describes themselves as a “national abortion reporting effort.” Their report focuses on the number of abortions in the United States in the 12 month period after Dobbs (July 2022 through June 2023), calculated by comparing monthly averages to the average number of abortions two months pre-Dobbs (April and May 2022).
Based on these calculations, WeCount estimates that there were 2,200 more abortions in the 12 months after Dobbs than in the 12 months before.
Does this result mean abortion bans and restrictions don’t decrease abortions? Does it mean overturning Roe v. Wade didn’t make any difference in the total number of abortions performed? Is it a waste of time for those of us who oppose abortion to focus on legal and political changes?
In a word: No. The WeCount report leads to none of these conclusions.
Let’s assume the WeCount numbers are accurate.
They almost certainly aren’t, but we will discuss that in a minute.
WeCount reported a 0.2% increase in abortions, far lower than the 8% increase seen in recent years.
WeCount calculated 987,580 total abortions in the country from July 2022 through June 2023–and estimated that number was 2,200 higher than the 12 month period preceding Dobbs. This would mean that in the first 12 months after Dobbs, abortions increased in the U.S. by 0.2%.
Assuming this number has any statistical significance (and WeCount doesn’t claim that it does), it would make the headlines above technically true: abortions increased in the US by 0.2% post-Dobbs. But the headlines don’t raise the question: compared to what?
The Guttmacher Institute reported that from 2017 to 2020 (the most recent year for which Guttmacher has reported a national total) total abortions in the U.S. had increased 8%, 40 times more than WeCount’s reported increase. If abortions were increasing 8% before Dobbs, and in the first 12 months after Dobbs they increased only 0.2%, we already see suggestions of a massive impact from Dobbs.
The national total increased infinitesimally despite massive efforts by abortion advocates.
WeCount explains that the 2,200 year-over-year difference in national abortion numbers masks huge changes in abortion totals state by state. WeCount divides states into three legal paradigms:
- states that banned abortion (“banned”),
- states that restricted abortion to six-weeks of gestation (“gestational limit”), and
- states that permitted abortion (“permitted”)
They calculated that, in the 12 months after Dobbs, banned states had 94,930 fewer abortions; gestational limit states had 19,660 fewer abortions; and permitted states had 116,790 more abortions. In other words, there were approximately 117,000 more abortions in some states compared to approximately 115,000 fewer abortions in others. So the small total change (2,200) represents very large state-by-state changes offsetting each other.
Why were there 115,000 fewer abortions in some states?
WeCount grouped states based on their abortion laws, and found that the states with the largest decreases had the most restrictive laws (and vice versa). This is not to suggest it is only laws that decrease abortion rates. There are many factors at play. But laws are a major one.
WeCount acknowledges this reality all throughout their report. Abortion advocacy groups often acknowledge this fact, albeit with very different language than anti-abortion groups use. Anti-abortion groups will describe the lives saved by abortion laws, meaning the embryos and fetuses who were not killed and instead grew, were born, and (in the vast majority of cases) were raised and loved by their mothers.
Abortion advocacy groups will describe people forced to continue pregnancies and/or forced to give birth. WeCount is no exception (emphasis added).
People in states with abortion bans were forced to delay their abortion, to travel to another state, to self-manage their abortion, or to continue a pregnancy they did not want.#WeCount Report, released October 24, 2023
The very different descriptions allude to the same fact: laws against abortion mean some people who would have gotten abortions instead carry their pregnancies to term.
Why were there 117,000 more abortions in other states?
State level abortion restrictions and bans passed since Dobbs have created major road blocks to abortion. Abortion advocacy groups have expended enormous energy and resources trying to undo these effects. WeCount describes some such efforts (emphasis added):
Abortion-providing facilities have increased efforts to see patients quickly, particularly in states with early gestational limits. Many facilities in states where abortion is permitted have expanded their capacity and new clinics have opened in surge states, including in Illinois, Kansas, and New Mexico. Many facilities have added capacity for telehealth services, and new virtual clinics, which offer only telehealth care at substantially lower price points have opened in these states. Additionally, several states have passed laws to expand abortion access and protect providers; such laws have likely led to increases among both residents and people traveling from states with bans. Such laws include funding to cover abortion care for out-of-state residents, and policies that mandate insurance coverage of abortion.#WeCount Report, released October 24, 2023
There are many factors that may be contributing to increases in abortion in some states; among those factors are more permissive abortion laws. But, of course, if it’s true that more permissive abortion laws lead to more people getting abortions, the reverse will also be true: more restrictive abortion laws lead to fewer people getting abortions.
In any case, if the WeCount numbers are accurate, the results suggest that these massive efforts by abortion advocates have so far fallen far short of sustaining the abortion increases we were seeing in recent years. National totals went from 8% increases to 0.2% increases even with concerted efforts to expand access in pro-choice states, and even before some pro-life states had further restricted abortion. These are the profound effects of abortion legislation.
The WeCount numbers are unlikely to be accurate.
This is not a criticism of WeCount specifically. It’s difficult to measure total abortion numbers, and WeCount is upfront about their data limitations.
First, they received reports representing 82% of all abortions provided by clinicians, and had to impute (estimate) numbers for the other 18%. In some states, such as Alabama and Arkansas, WeCount didn’t have to do any imputation. In others, imputation ranged from less than 1% (e.g. Delaware and Georgia) to between 30-44% (e.g. Washington DC and New Jersey). In Florida, WeCount imputed over 45% of the final abortion number.
Receiving reports representing 82% of abortions is not at all a bad response rate. But when the final results amount to a difference of 183 abortions per month, nationally, compared to about 82,000 abortions happening each month, it wouldn’t take much uncertainty to render the results inconclusive. Notably, WeCount doesn’t report confidence intervals or attempt to verify statistical significance.
Aside from the imputation rate, there are many variables missing from WeCounts calculations that, if known, could paint a different picture–in either direction.
Factors that could mean abortion rose even more despite legal restrictions.
WeCount is only counting clinician-supervised abortions. They don’t account for abortions that occur “outside the formal healthcare system,” meaning abortions from illegally obtained abortion pills or from other medications, drugs, or herbs. This is a huge unknown so far, which I’ll examine in more detail below.
Additionally, WeCount calculates pre-Dobbs levels based only on April and May of 2022. But in May of 2022 the Supreme Court draft decision for Dobbs was leaked. WeCount speculates (plausibly, in my opinion) that in the wake of the leak there was a surge in people seeking abortion, which may have artificially inflated the pre-Dobbs numbers and thus artificially diminished the increase in abortions post-Dobbs.
Factors that could mean abortion fell with legal restrictions.
Abortion restrictions in place before Dobbs.
The WeCount pre-Dobbs numbers don’t account for the fact that some states–most notably Texas–already had major decreases in abortions due to restrictions or bans in place before Dobbs. In fact, WeCount estimates that if Texas hadn’t already had the September 2021 heartbeat bill in place, the state would have seen 25,230 fewer abortions for the year, more than eliminating the national 2,200 increase WeCount reports.
There are smaller but similar effects in states like Oklahoma, which had closed several abortion clinics and then passed an abortion ban by May of 2022, or states like Missouri, which, despite technically having legal abortion pre-Dobbs, had already had minimal total abortion numbers due to other abortion laws (e.g. parental consent, limits on insurance coverage, banning telemedicine abortion).
Abortion restrictions that had already decreased abortion numbers pre-Dobbs would thus artificially amplify the apparent increase in abortions post-Dobbs.
Abortion restrictions implemented after the WeCount study period.
WeCount refers to states as “surge states” if they saw the largest cumulative increases in abortion. “Surge states” were not per se states with the fewest abortion restrictions. Instead they were states that bordered restrictive states. WeCount’s report in particular emphasizes the “important access points” of North Carolina and Florida.
However WeCount’s report concludes in June of 2023. In July of 2023, North Carolina passed a 12-week limit. In the first month after this restriction, per Guttmacher, North Carolina saw a 31% decrease in abortions.
Meanwhile Florida is waiting for its state supreme court to rule on a 15-week ban, and in the interim Governor DeSantis has also signed a 6-week ban. On the other hand, pro-choice groups in Florida are working to get an amendment on the 2024 ballot to enshrine the right to abortion into the Florida constitution.
The uncertainty in Florida is a microcosm of legal battles across the nation. It’s not at all clear how many states in the next several years will expand versus restrict abortion. But the fact that pro-choice and pro-life groups alike are fighting so fiercely to implement state policy underscores the same axiom: abortion laws affect the number of abortions.
It’s difficult to estimate abortions from illegal pills, but they are unlikely to have negated Dobbs.
The WeCount report focuses on clinician-supervised abortions “within the formal healthcare system.” It does not (and in fairness, cannot) account for abortions obtained illegally.
Abortion rights groups have worked overtime to promote “self-managed” abortion, meaning women (typically illegally) obtain abortions without medical oversight. In fact, some abortion advocates have urged the pro-choice side to move away from the talking point that illegal abortions are dangerous; instead they want people to view abortion pills as extremely safe.
And requests for illegal abortion pills have certainly increased. WeCount cites research estimating Aid Access received 6,500 requests per month following the Dobbs decision, but Aid Access is not the only supplier. An April 2023 New York Times article claimed overseas abortion pills suppliers are on track to have delivered pills to 100,000 people in the year after Dobbs.
How can we calculate how off-the-record abortions via illegal pills affect the pre- and post-Dobbs numbers? Realistically we can’t, not with any certainty. There are too many unknowns:
- Have suppliers actually shipped abortion pills to 100,000 people in the year since Dobbs?
- How many of those people received the pills?
- How many used the pills?
- How many who used the pills had completed abortions?
- How do these numbers compare to the same numbers pre-Dobbs?
Still, we can at least do some uncertain, back-of-the-envelope calculations in order to consider the possibilities.
First, of course people were requesting pills from suppliers before Dobbs as well. One study found that Aid Access received on average 83 daily requests for pills from September 2021 to May 2022 compared to 214 daily requests from June 24, 2022 through August 2022. This would suggest 39% of the post-Dobbs requests were a continuation of pre-Dobbs behavior, and the other 61% were new.
If we assume for the sake of argument that suppliers did ship pills to 100,000 people in the year after Dobbs, and we apply the above percentages, approximately 61,000 of those shipments were new in the post-Dobbs era.
If we further assume for the sake of argument that (1) all 61,000 shipments reached the correct destination, (2) all 61,000 people took the pills within the 12 months after Dobbs, and (3) these resulted in 61,000 completed abortions, then the updated numbers for WeCount would be 2,200 + 61,000 = 63,200 more abortions after Dobbs. WeCount estimated 987,580 abortions in the 12 months after Dobbs and further estimated this was 2,200 more than the previous period, meaning the previous 12 months would have seen 985,380.
From that number, an increase of 63,200 would represent a 6.4% increase in abortions in the year after Dobbs—still a lower increase than the 8% increase seen in recent years.
Now, all of this is very uncertain. The 8% increase reported by Guttmacher was through 2020. We don’t yet know what abortion numbers they will have calculated for 2020-2022. We don’t have a good pre-Dobbs comparison in changing abortion rates.
We also don’t know how many pills were actually shipped from overseas in the 12 months since Dobbs, or how many were already being shipped before Dobbs. And it’s highly improbable that all of the pills requested were shipped, received, taken, and resulted in complete abortions. But these are the kinds of unfounded, generous assumptions you have to make to get numbers that come close to representing the same rate of increase of abortions seen at least in 2017-2020.
All this to say: the WeCount data is dubious, and ideas about illegal abortions even more so. This doesn’t mean, however, that we can’t find any indicators of what’s going on. At minimum, WeCount seems to reliably indicate that clinician-supervised abortions have plummeted in states with bans or restrictions and surged in states without them. Furthermore, we have other data points to give insights.
When considering effects of abortion restrictions, it’s important to examine live birth data.
As seen above, it is very difficult to directly analyze abortion rates accounting for illegal abortions. But it’s not as difficult to analyze birth rates. People can have secret abortions, but secret births are a much higher hurdle.
Fertility rate research over the decades
Previous research has suggested that, at least in the near-term, after abortion laws get more permissive, births decrease, and after laws get more restrictive, births increase. For example:
States legalizing abortion experienced a 4% decline in fertility relative to states where the legal status of abortion was unchanged.“Roe v Wade and American fertility,” Levine, Staiger, Kane, & Zimmerman, American Journal of Public Health, 1999 February; 89(2): 199-203
Approximately one-fourth of women who would have Medicaid-funded abortions instead give birth when this funding is unavailable … Studies have found little evidence that lack of Medicaid funding has resulted in illegal abortions.“Restrictions on Medicaid Funding for Abortions: A Literature Review,” Henshaw, Joyce, Dennis, Finer, & Blanchard, Guttmacher Institute, June 2009
I estimate an increase in the birthrate of 4% to 12% when abortion is restricted. In the absence of anti-abortion laws, fertility would have been 5% to 12% lower in the early twentieth century.“The Effect of Anti-Abortion Legislation on Nineteenth Century Fertility,” Laney, Demography, 2014 Jun; 51(3): 939-948
We find that a 100-mile increase in distance to the nearest clinic is associated with 30.7 percent fewer abortions and 3.2 percent more births.“Undue Burden Beyond Texas: An Analysis of Abortion Clinic Closures, Births, And Abortions in Wisconsin,” Venator & Fletcher, Journal of Policy Analysis and Management, Volume 40 Issue 3, 03 November 2020
Birth rate research after the Texas Heartbeat Law
We’ve seen this phenomenon with data arising after the Texas heartbeat law. You may recall headlines like this from a few months ago:
Johns Hopkins Bloomberg School of Public Health explains that “researchers expect the number of abortion procedures to drop and live births to increase when abortion restrictions go into effect within a state.” Their research used statistical modeling accounting for live birth data from all 50 states as well as other confounding factors, and ultimately calculated that there were 9,799 more births than would be expected in Texas absent the heartbeat law. The researchers, abortion rights advocates, describe these babies’ births as evidence that “a considerable number of pregnant individuals in Texas were unable to overcome barriers to abortion access.”
Birth rate research after Dobbs
We’ve seen similar birth data even more recently, with a November 2023 report on the birth rates of pro-life states compared to controls since Dobbs.
The report suggested the number of births rose 2.3% on average in states where abortion was banned compared to states where it stayed legal.William Skipworth, Forbes
Whether you view this outcome primarily in terms of women’s rights violated or babies’ lives protected, there is at least one point of agreement: laws against abortion stop some abortions.
The effects of abortion legislation are profound.
Keep in mind Roe v. Wade was the legal status quo for 50 years. We’re only a little over a year into the post-Roe era, and we’re already seeing immense impact.
Pro-lifers have enormous work ahead of us: cultural changes to humanize children in utero, better support pregnancy and parenthood, improve efforts to prevent pregnancy, and countless related endeavors. But make no mistake: laws against abortion are unequivocally a crucial part of this puzzle.
We fight on.