Does restricting abortion lead to more unwanted births?

Who will adopt all the extra children?

“Pro-lifers are hypocrites and pro-birthers”

“Unwanted children lead to educational and career problems“

“Unwanted children lead to poverty”

Unwanted children and child abuse

Does abortion reduce crime rates?

Does restricting abortion lead to more unwanted births?

This question is important for three main reasons: it is alleged that unwanted births harm those children (by being unwanted and having lives not worth living), that they harm the mothers (physically, economically, educationally, economically, and so on), and that they harm society (by increasing the welfare cost of raising children, and so on).

I will respond to each of those specific allegations elsewhere, but since all of these claims depend in part on the seemingly obvious claim that restricting abortion leads to more unwanted births, I thought this was worth a question in itself.

The reason is that the answer is not at all obvious. It is often assumed that every abortion – 200,000 a year in the UK, many more in other countries – would instead result in a child. But in fact that is not what the empirical evidence shows, as summarised best by Philip Levine’s Princeton University Press volume on the economics of sex. While the evidence that pro-life laws reduce abortions is widely known within the relevant circles of academia (admittedly relatively small circles), the evidence concerning the impact of these laws on sexual behaviour and birth rates is far less widely known.

It turns out that there is a lot of evidence showing that laws protecting women and children lead to changes in sexual behaviour so as to make people more careful – having sex with fewer partners, using contraception more often, or better contraception, and so on. For example, in Nepal the expansion of abortion services led to a reduction in the use of modern contraception, while in Russia a significant minority of women explicitly cited the availability of abortion as a reason for not using contraception. Klick and Stratmann have repeatedly shown across multiple samples and countries that abortion legalisation leads to an increase in risky sex.

It can therefore be expected that fewer unwanted pregnancies will occur in the first place when pro-life laws are in place. This is a good thing for everyone, especially women who disproportionately suffer the burdens of unwanted pregnancies. Since many unwanted pregnancies end in births, there is a reason to think that fewer unwanted pregnancies means fewer births, other things being equal.

Whether pro-life laws increase unwanted births is, therefore, a question that can only be settled empirically. In fact, the empirical evidence clearly shows that moderate restrictions on abortion, while reducing the abortion rate, do not increase the birth rate at all. Complete bans on abortion do appear to increase the birth rate, but only by a relatively small amount, and the effect is often temporary – see, for example, Romania’s 1967 ban on abortion, which, even with a joint ban on contraception at the same time, led to only a brief spike in birth rates for a couple of years before birth rates gradually returned back to baseline rates. Neither Chile nor Poland, the countries to have banned abortion completely in recent times, saw significant increases in birth rates following their bans on abortion.

Arguments suggesting that restricting abortion lead to many unwanted children therefore rest on a pillar which is generally empirically false – and have other problems in addition (see the questions on those arguments). In fact, legalisation of abortion certainly leads to more unwanted pregnancies, which burden women disproportionately, and may even lead to more unwanted children (by increasing unwanted pregnancies and decreasing the ‘acceptability’ of pregnancy).

Who will adopt all the extra children?

It is often claimed that if women and children are protected by pro-life laws, there will be a huge number of children available for adoption, since there are already so many children in foster care. This is sometimes linked with the ‘hypocrisy’ question: why haven’t you adopted all those children?

I’ll say a lot more about the hypocrisy under ‘Pro-lifers hypocrites and pro-birthers’. But it’s worth saying one thing in response to this specific allegation: namely, that opposing killing a certain group of human beings doesn’t entail a duty to provide for all of them (though of course society has a duty to do what they reasonably can to provide for the vulnerable). During the migrant crisis in Europe a few years ago, a huge number of people protested the conditions to which they were subject, and advocated (rightly) for their wellbeing. Almost none of the people who did so actually took a migrant in to their own homes. This was not hypocritical. It would be even less hypocritical to oppose the killing of those migrants while at the same time not providing for them. This is not at all to say the affluent and powerful (which includes the average person in the West) have no obligations to ensure the basic needs of vulnerable human beings are met – as a welfarist and a Christian I think both the government and private individuals have a duty to do so.

I certainly encourage adoption and think that some people have a duty to adopt. There are certainly some children very much in need of parents. It is also a profoundly important alternative to abortion for those women who are not in a position to raise a child. But that is clearly a distinct (though related) issue from whether it should be legal to end their lives.

And there are a number of other reasons why this question misses the mark:

First, as shown in ‘Does restricting abortion lead to more unwanted children?’, abortion restrictions do not typically result in higher birth rates – only when abortion is fully restricted, and even then only by a relatively small amount and for a temporary period. Instead, people adapt to the laws and are more careful about whom they have sex with and use of contraception. This is better for everyone: it prevents many unwanted pregnancies, which overwhelmingly burden women. So there is no empirical evidence there would be large numbers of unwanted children being born.

Second, the large majority of women denied abortions choose to raise their children themselves. In the Turnaway study, of the women denied abortions who gave birth, 91% raised the child themselves (and of those, 98% were glad they were refused abortion by the child’s 5^th birthday). Interestingly, choosing adoption was not associated with age, race or poverty status, and women choosing adoption were actually more likely to have completed high school than women raising the baby themselves, though less likely to be in employment. For these first two reasons, the number of children available for adoption would be very, very small.

Third, by contrast, the number of willing parents vastly exceeds the number of newborn babies available for adoption. Ann Furedi, former CEO of BPAS, the UK’s leading abortion provider, notes that while there are over 200,000 abortions in England and Wales each year, there are only 200 newborn babies placed for adoption. By contrast, although it is difficult to find estimates, there are thought to be around 1-2 million parents in America waiting to adopt, and there are already 140,000 adoptions in the US each year (adjusting for population, that would make about 200,000-400,000 parents waiting to adopt in the UK, far exceeding the available 200 a year). In the US, the waiting list to adopt a child with Down Syndrome is typically around 1-2 years, but in some cases up to 5. While children with Down Syndrome are deeply treasured and equally valuable to other children, they are not always seen that way by prospective parents, hence there being a 90% abortion rate for children with Down Syndrome in many European countries. Disabled and non-disabled children alike are in short supply. Hence it is pretty clear that regardless of the exact numbers, the number of parents wanting to adopt far exceeds the number of extra children who would be placed for adoption if abortion were to be prohibited.

So why are there so many children in care? The reason is that most children in care are from wanted pregnancies and were taken from their parents after an unforeseen or unwanted family breakdown – abortion would have made no difference in these cases. Indeed, most children in care are hoped to be reunited with their birth families after some time and are never even placed for adoption. In the UK, out of 80,000 children in care, only 6,000 are available for adoption. The reasons not all of those are adopted are far more complex than a lack of prospective parents. Even if there were a lack of parents for older children in care, however, there has never been a lack of parents willing to adopt a newborn baby, which is the relevant consideration in the case of abortion. Hence, the short answer to the original question is: the many thousands (or millions) of parents who are already seeking to adopt a baby and unable to do so because there are so few available.

“Pro-lifers are hypocrites and pro-birthers”

As I write this, I am (to my shame) in a Twitter conversation with someone claiming that pro-lifers are inconsistent for opposing abortion while not opposing war, the death penalty, and guns, and while opposing parental leave, welfare, universal healthcare, and so on. Let me say clearly that it is only an inhumane society which leaves women to carry the burdens of a crisis pregnancy alone. I believe that governments should step in to ensure as much as possible that no woman ever feels she needs to have an abortion. Fathers have even more of a duty to ensure this, and it is a disgrace that so many men have been permitted throughout history to walk away from these responsibilities, leaving the woman to deal with them alone.

But there is a bit of an irony in the allegation that pro-lifers are myopic, since it implicitly assumes that all (or even most) pro-lifers are American Republicans who agree with that party platform. As I’ll show, the association between those issues is largely an historical accident unique to the American context and has little to do with the global pro-life movement generally. Nevertheless, as an English pro-lifer who is pro-welfare, anti-capital punishment, pro-universal healthcare, pacifist, and so on, it is surprising how much I am criticised for being short-sighted and ignorant by pro-choicers who assume that I must be an American Republican who opposes all these views. The world is far bigger than America, and in most countries there is little to no association between these views. In the UK, for example, Labour voters are just as likely to be pro-life as Conservative voters.

Of course, the same (mutatis mutandis) was true in the US too until fairly recently in history, as chronicled by Daniel Williams in his Oxford University Press book on the history of the pro-life movement in the US. Until the 1970s, the pro-life movement was largely based among pro-life Catholic Democrats who supported Roosevelt’s New Deal massively expanding welfare. In the 1960s and 1970s, both parties had very mixed records on abortion, and it wasn’t until the later 1970s that Evangelical Republicans began to be more heavily involved in the pro-life movement. It wasn’t until the Democrats began to shift firmly towards the pro-choice side that many pro-lifers felt they had no choice but to reluctantly move to the Republican Party, despite deep disagreements on various issues. This has only increased in recent times as the Democrats have purged the last few remaining pro-lifers from the party. For my own part, as someone who knows a huge number of American pro-lifers, there is an incredible desperation for a third party – or even just a place for pro-lifers in the Democrat Party, to give a real choice. The marriage of pro-life issues and the rest of the Republican platform was never intentional, and to this day is hardly a consensus. There is barely even a hint of it anywhere outside of the US.

And in fairness to the Republicans, they are not always as bad as their reputation. In recent weeks, a pro-choice Democrat State Representative from Oklahoma came out with the perfect ‘gotcha!’ to Republican pro-lifers: ‘This week I filed HB3129, which codifies that a father’s financial responsibility to his baby & their mom begins at conception. If Oklahoma is going to restrict a woman’s right to choose, we sure better make sure the man involved can’t just walk away from his responsibility.’

The problem? Pro-lifers unanimously supported it. In fact, the only people who rejected the bill were the Representative’s own team, who castigated him and forced him to backtrack, saying that he ‘obviously’ isn’t ‘moving forward with this bill as written’ and would ‘go back to the drawing board’. More embarrassing still, the Republican Party had literally already proposed essentially this bill at a federal level two years earlier. At that time too, it was met with unanimous applause from Republican pro-lifers, and not a whiff of support from pro-choice Democrats.

Clearly there are a few other issues, however. Are these positions hypocritical? And are pro-lifers hypocritical for not doing more practically for women experiencing crisis pregnancies?

Let’s take the first question. Of course, not all pro-lifers even in America support these positions – they often just take the package deal as the least bad of two realistic options. That’s not hypocritical – it’s just politics. What about those pro-lifers who do support those positions? Again, although I personally do not, it is hard to believe that they are hypocritical. Being pro-life in the context of abortion means that you oppose the intentional killing of innocent human beings. That is perfectly consistent with supporting, for example, the killing of guilty people in capital punishment, however wrong the latter position might be. Likewise, the morality of killing is completely distinct from (though related to) the morality of saving people through healthcare, and so on. In other issues, full-blooded Republican pro-lifers could be mistaken without being hypocritical. They might think that universal healthcare makes healthcare worse and lead to more deaths, for example. However wrong that position might be, it is not hypocritical. Thinking that welfare is best given privately than by the state might be wrong, but it is not hypocritical. Likewise, given the evidence that welfare can increase the abortion rate in pro-choice states (see ‘Does welfare reduce abortion?’), it is easy to see how opposing welfare but supporting pro-life laws could be consistent, however wrong.

What about the work that individual pro-lifers do? In fact, they do a great deal more than they are given credit for. Laura Hussey has documented in comprehensive detail the work that pro-lifers do in practice to support women in crisis (and non-crisis) pregnancies – and afterwards. She notes that there are nearly three thousand crisis pregnancy centres across the US. Each of these has an average of 40 volunteers contributing an average of 5 hours every week. This is on top of an average of 5.7 paid staff, only a tiny proportion of whose funding comes from the state. The overwhelming majority of centres either provide or refer women for a wide range of goods and services, ranging from baby care products to furniture, food, housing, childcare, medical care, and even cash. This work is not just about discouraging women from abortion: in fact, the majority of women helped by these centres were never considering abortion to begin with, while 79% of centres continue offering help after the baby’s first birthday, and 20% continue support even after the child’s 5^th birthday. Indeed, some pro-life pregnancy centres have even been criticised for not doing enough to discourage abortion! As Hussey shows, the empirical research available shows very high client satisfaction from these centres. And this work doesn’t include the many other kinds of pro-life pregnancy help, such as the myriad adoption centres or maternity homes across the US.

The scope of this charitable work is enormous. And it far surpasses the political pro-life work which goes on in the US – in direct opposition to common perception. As Munson puts it, ‘more individuals are involved in volunteering more time in the individual outreach stream than in any other [pro-life] movement activity. The number of organizations involved in individual outreach is greater than the number involved in all the other streams combined’.

Despite all this, such centres are constantly maligned in the media, especially by abortion advocates. Indeed, there continue to be a variety of attempts to have them defunded or shut down, either directly or by forcing them to refer to abortion, which could clearly contradict their fundamental values. Likewise, many Western countries are trying to ban people from offering support outside abortion clinics, which is known to be needed by some women, and many Christian adoption agencies are likewise under threat. Fortunately, such laws have so far largely been struck down, but attempts continue.

None of this takes into account the far greater work still done by pro-lifers outside of the pro-life label. Perhaps part of the reason pro-lifers are not perceived as addressing other issues is because they often compartmentalise their work on other issues. But we know, for example, almost the entire volunteer base for crisis pregnancy centres is religious, and almost entirely Christian specifically. And Christians (even the nasty evangelicals) do a huge amount of work looking after people after birth, even if in America they often have a scepticism of doing this through government. This is not just church work: even on secular metrics, Christians give more than non-religious people – in addition to what they give to their churches and faith-based organisations.

Christians – who compose the overwhelming majority of the pro-life movement in the US – give far more money even to secular causes and do far more volunteer work than non-Christians, and adopt at over twice the rate of the average citizen. In the US, 91% of religious conservatives give to charity, compared to 67% of others, and they volunteer at a rate 10 percentage points higher than the general population. Religious believers are far more likely to donate blood, to give food or money to homeless people, and so on. In Europe, Christians are 30 percentage points more likely than a non-religious person to volunteer, and 15 percentage points more likely to volunteer for non-religious charities. John Dickson’s excellent book Bullies and Saints, chronicling the good and bad of Christian history since the very first years, has a number of other cited examples.

Hence, it is clear that, in fact, under the pro-life label and aside from it, those involved in the pro-life movement – most often religious conservatives – are in fact quite a bit kinder than they are typically portrayed, when the empirical evidence is on the table. I say this not to blow the trumpet of Christians – but to show that certain allegations of hypocrisy do not always hit the mark, despite the Church’s many failings throughout history.

This is particularly clear when contrasted with the pregnancy support work offered by abortion providers themselves. Planned Parenthood’s own annual report, for example, cites over 350,000 abortions performed, but only 8,626 instances of ‘prenatal care’, an odd distribution of services for an organisation dedicated to women’s ‘choice’ (since the large majority of pregnancies are not aborted). This is despite receiving vastly more funding from the government – over $600 million in 2019-2020 even under a Republican government, for example. Evidence of pressure and coercion to abort is discussed under ‘How common is coercion?’

Unwanted children lead to educational and career problems

Of course, if the child is a person with equal rights, then educational and career reasons could never justify ending its life. While this seems like an easy and accurate response to the question, it’s not enough in itself – not because it’s true, but because given the ways women are often treated, especially the ways they are discriminated against for becoming pregnant, we all have a duty to think seriously and carefully about the educational and career challenges women face, and how best to resolve them. Hence, it is important to explore this question further.

There is no doubting that unwanted children can present challenges for the mother’s education and career. Wanted children do as well! But I do want to suggest that these issues have been overstated – and where they cause significant problems, there are solutions other than abortion. Hopefully this will be a helpful starting point for developing better policies and practices to protect women and girls in schools and workplaces who may face formidable challenges because of their situation.

As I say in many of these questions, an important factor to bear in mind is that abortion restriction does not necessarily lead to more unwanted children. If so, most of what I say here may well be irrelevant – when abortion is restricted, many women do not get pregnant in the first place, which is better for everyone. For more details, see ‘Does restricting abortion lead to more unwanted pregnancies?’.

Perhaps surprisingly, there is relatively little empirical evidence showing that childbirth hinders education, on average. We start with Upadhyay et al.’s 2015 paper from the Turnaway study, showing that women receiving abortions were more likely to achieve their one year aspirations than women giving birth. It will be helpful to consider this alongside McCarthy et al.’s 2020 paper from the same study looking at five year aspirations. In this study, the authors asked women one week after the abortion, “how do you think your life will be different a year/five years from now?” The researchers then coded these into positive, neutral, and negative expectations, and assessed their achievement insofar as they were able using data they collected separately (i.e. without asking the individuals whether they considered their expectation achieved). They coded positive expectations as ‘aspirational plans’.

The study in fact found that among women setting plans, there was no difference between the groups in achievement. But since women receiving abortions had more positive expectations, more of them overall were fulfilled. Further, there are a number of problems with the inference that women denied abortions are less able to achieve their aspirational plans:

1. The measurement of aspirational plans or hopes is extremely suspect. Participants were not asked for their hopes; they were simply asked what they thought would happen. Women giving birth probably felt less able to predict their lives, but were not asked what they hoped for from their lives. Life with a child is often less predictable; but this does not mean the parent has fewer hopes, ambitions, or sources of happiness.
2. Participants were not asked to appraise their expectations as positive; this was left to the researchers. But some of these decisions are extremely vulnerable to the researchers’ biases: for example, they coded the expectation: “my future is dedicated to my kids and their education” as a neutral expectation and “[the child] will probably be with me for the rest of my life” as a negative expectation, but it is perfectly possible that the participant thought of these in positive ways.
3. No attempt was made to weight expectations according to ambition/realism or importance. For example, “hopefully by then I’ll have a baby” and “hopefully I’ll be married by then” were effectively considered just as important as “I hope to get a dog”. This will bias the results towards favouring many more trivial positive expectations over fewer more profound positive expectations (such as having a baby).
4. The researchers included a bias towards positive changes by asking how the woman thought her life would be ‘different’ in a year or five years. But women who were already happy with elements of their lives had no way to answer this, and no way to show that their happy lives remained the same at follow up. Perhaps women giving birth were already somewhat happier with more elements of their lives, for example, feeling less need to begin a new relationship, perhaps already being in one, and so on (hence the rarer relationship expectations among women giving birth); indeed, perhaps these considerations are what caused them to give birth rather than seek an abortion at another facility. Importantly in connection with this is the fact that they found ‘stability’ was a common aspiration, and yet the precise question of the study selected against measuring any stability at all (moreover, there is some evidence that many women see having a child as a crucially important means of stability).
5. The setting of expectations is relatively unimportant; what is more relevant is actual attainment. The setting of expectations is likely to be significantly affected for women giving birth by the fact that Western society generally sends the message that having an unintended pregnancy is a very bad thing and will ruin one’s life; but as these studies and others have shown, what a woman expects during an unintended pregnancy, and especially one week after being denied a sought abortion, is not an accurate indicator of what women are actually able to achieve. Modern society also makes parenting needlessly labour-intensive, so the negative expectations of parenting need not be met.
6. Within one year, there was no difference in attainment of the plans actually set – the number of plans set may be irrelevant, for example, if women giving birth were already content with their lives.
7. The researchers didn’t assess women’s original goals prior to pregnancy, so were unable to say how denial of abortion affected prior plans.
8. There are other possible selection biases for which the researchers could not control; for example, it is widely thought that when women are deciding to continue an unplanned pregnancy, the opportunity cost is a salient factor. Perhaps the women who decided to keep the baby rather than seek an abortion elsewhere were less ambitious to begin with, and this is precisely why they decided to continue the pregnancy after being initially denied.
9. On the five year scale, while in multivariate analysis women giving birth had fewer aspirational plans, they were far more successful at achieving them, such that they actually had an overall higher probability of both setting and achieving an aspirational five year plan than women receiving abortions, though this difference was not statistically significant. This does presumably mean that women giving birth were less likely to suffer the disappointment of failed 5 year aspirations as well.
10. The study did not include positive outcomes the women did not expect but nevertheless achieved. These are particularly important for women giving birth, given a culture where women having unintended pregnancies are routinely told that their lives will be effectively over.
11. Finally, what is surely of importance is that life satisfaction overall was no worse long term for women giving birth (this point is also relevant for the other outcomes – income, education, mental health, and so on). While one week after the abortion, women giving birth had marginally lower life satisfaction, throughout the rest of the five years, they had higher life satisfaction than those receiving abortions. Likewise relevant to each of these outcomes is the fact that by 5 years, 96% of women giving birth were glad they were denied an abortion, and a further undocumented percentage were unsure.

A related Turnaway study paper by Ralph et al. in 2019 studied women’s educational attainment after having or being denied an abortion. They note that there is disagreement in the literature about whether women giving birth at young ages have lower education for pre-existing reasons or as a result of giving birth, and draw attention to some evidence showing that even the possible negative impact of childbearing on teen education does not extend to women in their 20s. They note too that much of the existing literature is from before 1990, when educational options for pregnant or parenting teens were virtually non-existent. They then find in their own study that being denied an abortion had no discernible impact whatsoever in the educational attainment of women seeking abortion, consistent with various other studies. They conclude that the Turnaway study suggests adverse educational outcomes are largely due to pre-existing factors.

Beyond this, there are a number of points to make very briefly:

1. A child can often be the stimulus for people to ‘get their life together’, particularly people in poorer socioeconomic positions. In their magisterial work on motherhood among less affluent women, Edin and Kefalas summarise: “Middle-class observers often believe that the lives of poor youth could be salvaged if not for the birth of a child—but this is seldom the case… The poor women we came to know often describe their lives prior to motherhood as spinning out of control. Over and over again, mothers tell us their children tamed or calmed their wild behavior, got them off of the street, and helped put their lives back together. Children can banish depression, calm a violent temper, or serve as do-it-yourself rehab from alcohol and drugs. Children—and the minute-by-minute demands they make on their mothers’ time, energy, and emotions—bring order out of chaos… there were startlingly few “if only” tales of how “coming up pregnant” wrecked dreams of education, career, marriage, or material success. Instead, mothers repeatedly offered refrains like these: “I’d be dead or in jail,” “I’d be in the streets,” “I wouldn’t care about anything,” “My child saved me,” and “It’s only because of my children that I’m where I am today.” For all but a few, becoming a mother was a profound turning point that “saved” or “rescued” them from a life either leading nowhere or going very wrong.”
2. Much of the ‘motherhood penalty’ comes not from motherhood itself, but from the rampant discrimination against mothers in workplaces – including, as mentioned elsewhere, at Planned Parenthood. This is a reason to fight that discrimination, not encourage women who would otherwise keep their child to abort because of it.
3. While it is uncontroversial that having small children usually involves a productivity decline, the evidence at later stages is far more equivocal, suggesting that the benefits of parenthood at later stages can often compensate.
4. Plausible mechanisms have been identified for this compensation in productivity: for example, Graves and Ruderman show how parenthood and work can facilitate each other, and family commitments strengthen leadership, other task skills and overall-wellbeing, as well as positively affecting mood and providing sources of support. Scientific American has a helpful summary.
5. Part of the reason for work and family conflict is because of modern gratuitously hyper-intensive parenting. Astonishingly, a working mother today spends as much time on childcare as stay-at-home mothers did in 1975. In light of this, it is no wonder balancing work and childcare are difficult.
6. Finally, recall the impact that family breakdown has had on families’ economic situation, particularly among the poor. The marriage bonuses to one’s career and one’s finances are well known, as are the economic difficulties of single parenting. These costs were particularly high for ethnic minority groups, lower socioeconomic classes and women. Hence the description of Akerlof et al. that abortion and contraception jointly, by dramatically changing family structures, had led to the ‘feminisation of poverty’.

There is a huge amount more that could be said – unfortunately I cannot do so yet as much of this work is part of a working paper. Please do let me know if more information is needed. I will try to make the paper available soon.

Unwanted children lead to poverty

Clearly, all of the last answer is relevant to this specific question: poverty depends in large part on education and career milestones. I recommend anyone interested in this question read that answer first, as the information there is key. There is often a tendency to project the preferences and expectations of affluent men and women onto less affluent men and women: as Edin and Kefalas, among others, show, the reality is quite different. In fact, polling in the UK shows that those in the lowest socioeconomic group were twice as likely to support a 12 week abortion limit (or less) than those in other socioeconomic groups. But what else can be said here?

There is another Turnaway study paper suggesting that women denied abortion were more likely to be in poverty, less likely to be employed, and more likely to receive public assistance, both 6 months and 4 years afterwards, than women who were permitted an abortion. Of course, this study suffers the same issues as the Turnaway study generally (described under ‘What about the Turnaway study?’). But there are other important limitations:

It is true that women who were denied abortions were more likely to be unemployed at 6 months and 4 years, but women denied abortions had far lower full time employment, and much higher unemployment, to begin with. Examination of the trends shows, in fact, that while both groups increased full time employment over time, women denied abortions who gave birth actually gained full-time employment at a faster rate – so much so that there was no significant difference between the groups by 4.5 years. Likewise, while it is true that women giving birth had higher unemployment rates at 6 months than women receiving abortions (as one might expect, not least because of the number of women choosing to look after their child full time at least for some months), they had far higher unemployment rates at baseline. Again, both groups decreased unemployment over time; but women giving birth’s unemployment decreased so much that by five years, they had marginally lower unemployment rates than women receiving abortions. If anything, the evidence suggests that being denied an abortion improves one’s chances of becoming employed over a few year period.

A second main outcome of the study showed that women giving birth had higher receipt of welfare. This is exactly as it should be and is no cause for concern; if anything, we should be concerned if the women, many of whom had a low income, did not have significant public assistance after having a child. The only measure of final outcome as opposed to benefits received in this category is having health insurance; in fact, women giving birth were more likely to have health insurance at 6 months than women receiving abortions, though this was no longer significant at 1 year.

Personal income was lower at 6 months for women giving birth. But again, this was true at baseline. And again, by 5 years, women giving birth had increased their personal income at a higher rate, so that the two groups had the same average personal income. Household income was the same in both groups throughout the study, but because of the changing household sizes, women giving birth were more likely to live below the federal poverty line. This was the only difference which was not already present at baseline. However, this gap narrowed over time so that the difference was non-significant by 5 years. Subjective poverty (reporting having enough money to meet basic living needs) decreased in both groups over time at roughly similar rates; there remained a difference at 4 years but primarily because there was a difference at baseline. So while there was some evidence that women giving birth were more likely to live below the federal poverty line, this does not appear to have caused a noticeable difference (given the differing baselines) in ability to meet basic needs.

So the Turnaway study’s results on employment and poverty are far more nuanced than has been publicised. If anything, women who gave birth appear to have gained employment at significantly higher rates.

Of course, all this neglects the many ways in which legalised abortion has made women and children worse off financially. This is perhaps best seen through the impact of abortion on family breakdown, which has led to the ‘feminisation of poverty’, disproportionately affecting women. To quote Akerlof’s conclusion again: “Just at the time, about 1970, that the permanent cure to poverty seemed to be on the horizon and just at the time that women had obtained the tools to control the number and the timing of their children, single motherhood and the feminization of poverty began their long and steady rise. As a result, United States poverty rates have been stubbornly constant for the last quarter century.” Again, family breakdown has been estimated to cost the US at least $112 billion each year.

Likewise, this neglects the other costs of abortion, individually or societally: increased costs from suicide, IVF, abortion and its complications, preterm births, STDs, mental health consequences (and effect on career). Those having an abortion may also miss out on the financial benefits of motherhood: motivation for one’s career, the ‘motherhood advantage’ (described in the previous question), increased efficiency, and increased emotional support available within the family. None of these are trivial, and many have considerable non-financial costs as well. For all these reasons, it is highly doubtful whether women are better off financially with legal abortion; in fact, there are many reasons (and empirical evidence) to think they are financially worse off.

In any case, the answer for women in poverty as a result of having a child is surely not to end the life of their child. The answer, as with women with born children, is surely to give them the financial and other support they need so that they are no longer in poverty.

Unwanted children and child abuse

A traditional argument for abortion is that it is compassionate for the children themselves: no child deserves to be unwanted. This is true – though we would never draw these implications for born children. We wouldn’t assume that because they were unwanted, they would be better off dead and therefore we should kill them. Rather, we should do as much as we can to find them a loving home where they are wanted, and to create a society where they are valued equally.

As I’ve remarked on a number of questions, the evidence that more unwanted children would be born is, in fact, very slim. If abortion legalisation increases risky sex (as it clearly does), and if it makes pregnancy/children less ‘acceptable’, then abortion legalisation could even increase the number of unwanted children (see ‘Does restricting abortion lead to more unwanted births?’).

In any case, this is an extremely unrealistic picture of the empirical evidence. In the overwhelming majority of cases, unwanted pregnancies result in wanted babies. Recall the words of Diana Greene Foster, lead author of the Turnaway study:

‘How do women feel about having been denied an abortion? Initially, bad. But over time, most of the women who ended up carrying the unwanted pregnancy to term reconciled themselves to their new reality, especially after their babies were born… Women don’t often say they want an abortion for fear of what an unwanted pregnancy would do to their mental health. And mental health rarely seems to suffer, even when abortion is denied… Most of the women turned away, over time, reported that they were happy they had the baby.’

Foster reports further that only 9% of women who were denied abortions had poor maternal bonding, and even then, the overwhelming majority of these were still happy they had the baby.

On the whole, children from unwanted pregnancies end up no worse off than the average child on a wide variety of outcomes, and marginally worse off on a few. Let’s look at the recent literature. Jessica Houston Su found that children from intended births scored an average of 4.04 on a depressive symptoms score, compared to 4.77 for children from unintended births. The worst possible score was 60, and best was 0. Given the range, this is a negligible difference – just 0.73. To give an indication of how much worse this would make someone’s life on average, 1 point could be ‘scored’ by feeling hopeful only 4 days in the last week rather than 5-7, or having one night’s restless sleep in the last night. The difference between children from intended and unintended births was even less than that. And the most important point? Even then, the author concluded that results from the more complex models showed that ‘there is little to no causal relationship’.

The Turnaway study also compared children born after being denied an abortion with children born to a mother who had previously obtained an abortion. They also found no significant difference at all in perinatal or child health outcomes. Children born from unintended births had marginally worse gross motor outcomes, and had a significantly higher likelihood of poor maternal bonding – but as mentioned, the rates were still very low (9%) and even most of these women were glad they had the baby by the child’s 5^th birthday. Hence this study offers no support for the idea that ‘unwanted children’ suffer miserable lives and are better off not living.

Now, recall that abortion was supposed to be a solution to the problems of child abuse, and out-of-wedlock births. Larry Lader, the founder of NARAL, the leading abortion lobbying group in the 1970s, promised in 1974 that “The impact of the abortion revolution may be too vast to assess immediately. It should usher in an era when every child will be wanted, loved, and properly cared for.” NARAL then went on to vow in 1978 that “A policy that makes contraception and abortion freely available will greatly reduce the number of unwanted children, and thereby curb the tragic rise of child abuse in our country.”

Then recall what I said about abortion decreasing the ‘acceptability’ of pregnancy. It is not unreasonable to think that abortion has contributed to the general coarsening of attitudes towards children over the last half-century. And it is uncontroversial that it has led to a huge increase in risky sex and decline in shotgun marriages, and hence a rise in unconventional family structures, which are associated with significantly higher rates of abuse (as a child of divorced parents, I make no judgments, and am well aware that most unconventional families do not involve abuse – nevertheless, the statistics clearly show higher rates). Likewise, abortion’s legalisation led to higher rates of male delinquency. All of these have multigenerational effects.

Abortion was never going to do much to solve child abuse, since the large majority of children subject to abuse originate in the context of a wanted pregnancy. Studies on the relationship between abortion and child abuse have shown mixed results – while some show that children born from unwanted pregnancies are more likely to be abused (though most are still not), others suggest an association in the other direction. Most studies on this topic are old, and many look only at child abuse likelihood within a given family – without looking at how abortion legalisation can affect child abuse on a societal level.

What we do know is that infant deaths from child abuse have steadily been rising since the early 1970s in the US, despite far greater efforts to prevent them. Finkelhor and Jones note that the data show big increases in reports of child physical and sexual abuse in the 1970s and 1980s in the US, and not all of these indicators can be explained by increased reporting: “trend data for child well-being indicators mostly show a deterioration in the 1970s and 1980s”.

Hence abortion has uncontroversially made the problem of out-of-wedlock births worse. The evidence about its impact on child abuse in individual families is mixed, but this generally ignores the long-term effects of abortion on family structure, male delinquency, and so on. There is some provisional evidence that abortion has made the problem of child abuse worse on a societal level.

Does abortion reduce crime rates?

A famous paper from 2001 argued that legalised abortion led to reduced crime rates. I hope to comment on this in more detail in time, but for now I highly recommend Michael New’s brief discussion of this paper, and for a far more detailed analysis of some of the relevant evidence, Jonathan Klick’s paper summarising the evidence on this topic. It is safe to say that the original paper has received a substantial amount of criticism from a wide variety of sources.

In fact, as New points out, some have argued that abortion has increased crime rates. Akerlof’s papers, which I discuss elsewhere, are highly relevant here. Abortion legalisation is known to have contributed substantially to family breakdown, a key determinant of criminal activity. It has likewise caused an ‘extended adolescence’ of men, leading them into higher delinquency rates for more of their adult lives than before. Hence there are good reasons to support that abortion legalisation has had the opposite effect, substantially increasing crime. Recall that Akerlof is a Nobel prize-winning economist, and his wife (who co-authored one of these papers), Janet Yellen, is President Biden’s Secretary of the Treasury – presumably no pro-life activist!