Someone who sounds like Megan McArdle

If we cannot discount the interests of the fetus simply because it is not yet with us as a person, then how can one morally justify legal abortion as a coherent national policy?…I find it hard to construct a really compelling argument in favour of abortion which does not rest in some way on discounting the utility of the fetus-as-future-person.

There is much more at the link.  This is a real ouch, her barbs are directed at left-liberals but they do not stop there.  In my view we should subsidize births, keeping in mind that the long-run is the relevant time horizon.  I also believe a free and wealthy society will, at some point, have many more people than the alternatives, and on an ongoing basis.  As for what kind of restrictions on abortion are a good way to subsidize births, that is a very tricky question, especially keeping in mind I am not a pure utilitarian but rather a pluralist…I am not in Chicago to debate it with all the other economists as we are celebrating Yana’s 17th birthday in Miami…

Comments

I don't see the inconsistency here at all. Discounting is only relevant if you think that the utility of something already counts in the welfare calculus. There's a perfectly reasonable strand of thought on this that says we should treat non-existent individuals (which I’ll assume, arguendo, to include foetuses) as having no interest in existing. So, if we’re comparing a state of the world x in which a foetus is aborted, and y, in which it is not, the foetus’ utility is irrelevant, regardless of one’s attitude to discounting the utility of future lives. (This is a little simplified, but I think the point is robust.)

On the other had, if we are comparing two states of the world y and z where the foetus is not aborted, then one’s attitude to inherent discounting can affect which we prefer. The upshot of this is that there’s no necessary link between one’s attitude to discounting on the one hand, and one’s attitude to abortion on the other (at least unless one believes, as many do, that the foetus is a person.)

All of the serious environmental ills that we now confront are exacerbated by the increasing population of our species. Our increasing numbers will, if left unchecked, result in the mass extinction of other species with which we share the planet, and whose complicated web of interactions provide the services we need to live comfortably. With the population at close to six and half billion
and on its way to nine or ten, to assert that births should be subsidized is hard to understand.

(I initially posted this under the wrong topic.)

Subsidizing births seems like another potential consequence of a "basic income". If you don't want to subsidize births, you could, of course, directly penalize them with deductions from parental basic income.
In general, subsidizing births as an independent policy seems like the sort of policy that should count as "too utopian". One way of quantifying this is that the precedent value in terms of unlimited government exceeds the direct utility value of the policy.

As Tyler suggests, this is actually an argument against all forms of contraception or reproduction-prevention, equally, not against abortion in particular. The argument has two main parts: first, it asserts that we have a duty to increase the population, and then it considers the ways in which the government could legally enforce that duty. She only considers an abortion ban, not bans on contraceptives, various government provided incentives to encourage reproduction, bizarro-China quotas that mandate at least X offspring per person, government ad campaigns with messages like "Be a good person: make babies" or "Sex is fun!", or any other policy that might increase the birth rate.

Although some of these other policies may be unattractive, abortion bans are probably not a very effective population enhancement technique. One problem, if I'm remembering my Levitt correctly, is that it takes something like five legal abortions a year to lower the birthrate by one birth per year. This is largely because people get abortions in order to delay childbirth by a few years, a decision which is morally neutral according to a simple zero-discount calculation (giving equal value to each person). Increasing the population by forcing particular individuals to bear children also seems costly and inefficient, like other forms of conscription. The quality of life for the child, and for the rest of the family, is likely to be lower when the mother is an unwilling parent, and, Levitt's abortion research suggests that children born in these government-mandated childbirths tend to have a more negative (or less positive) impact on the lives of the rest of society (e.g. due to more criminal activity). So it's not clear whether trading off five of these births to women who couldn't get abortions for four births to women who got abortions earlier on is a good deal for society (or whatever the ratio is).

It is also an open question whether we want the population to be increasing more quickly right now, or how important that is. On a zero discount view, one extra person next year is equivalent to one extra person 1000 years from now (again, ignoring differences in their lives and their impact on others). Given that we're currently using up resources that we don't know how to renew or replace at a rapid rate, it's not clear what the long-term effect on the population and quality of life of future generations is from each added birth right now. Having fewer babies now might allow us to support a larger population a few hundred years in the future, once our technology has advanced so that we can use resources much more efficiently.

What's so magic about the moment the sperm actually meets the egg? Doesn't this argument apply equally well to birth control?

If we cannot discount the interests of the egg and sperm simply because it is not yet with us as a person, then how can one morally justify legal birth control as a coherent national policy?...I find it hard to construct a really compelling argument in favour of birth control which does not rest in some way on discounting the utility of the egg-and-sperm-as-future-person.

In fact, going further, doesn't this argument apply equally well to every baby you could have had, but did not because you abstained?

If we cannot discount the interests of the possible-egg-and-sperm-had-we-had-sex simply because it is not yet with us as a person, then how can one morally justify legal abstaining as a coherent national policy?...I find it hard to construct a really compelling argument in favour of abstaining which does not rest in some way on discounting the utility of the possible-egg-and-sperm-had-we-had-sex-as-future-person.

Think of it as a "state's rights" kind of argument. Each and everyone one of us is the sovereign rulers of our own bodies.

While me may morally oppose, for example, the death penalty, we may also think it unwise to
impose that morallity on all other countries.

Likewise, you can consistently believe for yourself that abortion is wrong while still wanting others to have the right to do so. (For reasons of sovereignty or utilitarian its-better-than-the-alternatives grounds.)

Studies in the U.S. show that women who get legal abortions tend to be better organized and more future-oriented than women of similar backgrounds who don't. That's why Levitt and Donohue only cited European studies to support their theory that abortion-cuts-crime by improving the qualities of mothering -- because in America the opposite is true.

Abortion deals with the welfare of potential individuals. Stern deals with the welfare of the entire species.

I don't care which of the nearly infinite permutations of the human genome are actually realized at any point in the population. If embryo
X is aborted in the first trimester because the mother is 14 years old and embryo Y is never conceived because the parents used birth control, it doesn't matter. The species will continue and there will be lots of people who are brought to term who will enjoy the experience of living - as long as we don't screw up their world. (I don't want to sound flippant; I'd much rather people used birth control methods other than abortion).

I do care very much that conditions on earth remain such as to support a high and sustainable quality of life for those who are born now and in the future which requires that pay attention to Stern-like warnings and also limit our numbers.

Megan, everybody, I need some help in understanding this. I may not be an economist, but I really cannot see the contradiction between not discounting the interests of future persons who will actually be brought to life (as the first commenter in free exchange explains very well) and considering we should have the right to decide whether we bring to life another person or not. Could you help me understand how these two are connected to each other?

I, like, many others here, don't see the contradiction. The opposition to a ban on abortion rests on a simple assertion: until a fetus can survive outside the womb, it is not a person. Therefore our ethical obligations to a fetus are different from those we owe to a person.

In the case of the Stern Report, we are talking about what we owe to people who will be walking around on earth, surviving outside the womb, in the future. Our obligations to them do not start until they are no longer fetuses, but once that condition is met, we should treat their utility as roughly equivalent to our own.

Now, it is true that there is a very slight -- in my view, vanishingly small -- chance that humanity will become extinct, in which case it will have made no sense to make changes in order to protect the interests of those future people. So we can increase the discount rate, as Stern does, to take that into account. But that doesn't affect the ethical question that you think raises such a contradiction.

I, like, many others here, don't see the contradiction. The opposition to a ban on abortion rests on a simple assertion: until a fetus can survive outside the womb, it is not a person. Therefore our ethical obligations to a fetus are different from those we owe to a person.
In the case of the Stern Report, we are talking about what we owe to people who will be walking around on earth, surviving outside the womb, in the future. Our obligations to them do not start until they are no longer fetuses, but once that condition is met, we should treat their utility as roughly equivalent to our own.

That's the rub. Our obligations to them do not start until they are no longer fetuses, under the liberal framework. People who will exist, if at all, only in the year 2200 cannot in any sense be described as "no longer fetuses." As of today, they do not exist. Therefore, by your logic, we do not have any obligations to them whatsoever. Far from treating these hypothetical future individuals as having interests equal to our own, we should treat them as having zero interests in anything.

Jane,

I'm dreadfully sorry, but I'm afraid I don't understand either your claim of inconsistency, or the relevance of Stern's radical discounting to said claim. I'll try to respond as best I can, but please excuse me if I miss the mark. Perhaps you could clarify your objection?

My formulation is consistent with protecting the welfare of future generations, on the assumption that they're going to exist regardless of what we choose to do about global warming. What it claims is that the potential utility of people whose existence we *can* affect is not relevant to determining whether we should prefer a state where they exist from one where they don't. (So in that sense, we would have no obligation to take into account the lost utility of people who might not be born because of global warming. But we would be obliged to take their utility into account in weighing up two alternative states in which they are born.)

The point of the radical discounting Stern advocates is indeed that we shouldn't be willing to sacrifice anything today for the sake of people who don't and will never exist in the future. Extending that probabilistically, if we're only 90% sure that a given population is going to exist in 100 years, their utility should count for 90% of what it would have otherwise. Hence the discounting.

Just as an additional thought, it is worth noting that most Americans take very different views of a woman who gets an abortion and a woman who drinks or uses illegal drugs during pregnancy but allows the pregnancy to go to term.

The reason, again, is that damaging a fetus but allowing it to be born is cruelly undermining the future utility of a human being who will actually exist. Abortion, on the other hand, is cutting the lifespan of a fetus short, before it is capable of feeling happiness or pain.

I'm not a utilitarian so I agree that there are problems with maximizing average utility. I do think, however, that the problems run even deeper for maximizing total utility.

It suggests that if we could double the U.S. population and maintain everyone at a standard of living slightly better than half of what it is currently, we should jump at the chance. Moreover, it seems to suggest that people (or to be more specific, women, since they are the ones who impose a binding constraint on population size) who do not reproduce to capacity are committing a grave crime almost as serious as murder by denying life and happiness to a potential person.

Also, it's important to distinguish between persons who might have been (e.g., aborted fetuses) and persons who actually will be (e.g., the aforementioned future Bangladeshis). Might-have-beens don't count because they never actually will be.

You seem to think I'm trying to "prove" that abortion is wrong. I'm not. I'm exploring the implications of a zero rate of pure time preference, one of which is that I think abortion becomes impermissible. So might birth control, although that wasn't the issue at hand; decisions to have sex/not have sex are so much more complicated that I think they probably would escape under the "sins of omission" clause.

Brandon, my personal attitudes towards abortion are rather fraught, but abortion seems to me to be wronger than causing birth defects. All of the disabled people I know prefer living disabled to not living. Your (and Conchis's) position seem to me equivalent to the burglar who gets away with his crime because he murders the witness. More interestingly, you're not just getting away with it; you're claiming it's legal because there's no one walking around who was hurt by the burglary. Also, you haven't explained why under your system we shouldn't remove the probability of an extinction event from the sample, since there will then be no Bangladeshis or whatever walking around to be hurt by global warming, and no shifting ground to the species, which is just an agglomeration of individuals.

Under any system I devised, an extinction event would be a major negative. Also, your system is certainly not endorsed by the report's writers, since they too regard the deaths of nations/species as a major drawback.

Can't we just say we value future generation because surely *somebody* will exist in the future, but value aborted fetuses not at all because surely it won't be them.

This is freshman dorm room level logical stuff, and remarkably silly.

And misery, not death is the opposite of utility.

Why stop at the fetus? Sperm and egg have rights too! Let's outlaw menstruation!

I find it kind of ironic that people who claim to be libertarian are so interested in poking their noses into the lives of the others, and passing laws that prevent those people from doing what they wish with their bodies.

May you all never be in a position where laws are imposed on you that turn your life into a living hell.

Jane:
I understand that you think that abortion is more wrong than causing birth defects. But can you understand--not agree, just understand the logic--why I it is that I regard causing birth defects as wrong, but am perfectly okay with abortion? This distinction is equivalent (but less abstract, I think) to the one you're objecting to in the Stern Report.

"Your (and Conchis's) position seem to me equivalent to the burglar who gets away with his crime because he murders the witness."

No. And this is the crux of the disagreement, I think. There's a world of difference between killing someone who already exists, and preventing someone who does not yet exist from ever being born. Again, if you accept the premise that a fetus is not a person, then there's no moral distinction between abortion and contraception. All either one does is prevent someone who does not yet exist from ever being born.

"Also, you haven't explained why under your system we shouldn't remove the probability of an extinction event from the sample..."

Do you mean an extinction event caused by global warming, or an unrelated extinction event that renders global warming obsolete? Obviously an extinction event caused by global warming is a huge negative. But if there's a non-negligible probability of an unrelated extinction event that renders global warming irrelevant, then we should discount accordingly. I'm not sure I understand why this is relevant, though.

There are infinitely many people that could possibly exist. Any utility function that assigns an utility at all to those people is useless. Jane only has a point if you distinguish aborted feti(?) as having rights.

John Doe,

An extinction event will also kill billions of people who exist at the time it occurs. that's a pretty massive reason to think it's bad, even if you don't care about the billions more potential people who won't be born (which I don't for precisely the reason josh points out).

If you think foetuses are people, then clearly abortion is extremely problematic. But that wasn't Jane's argument, at least as I understood it. What she seemed to be claiming was that, even if you don't think foetuses are people, believing in a radical discount rate of 0% forces you to take their interests into account. This claim makes no more sense to me than saying that a 0% discount rate obliges us to weight the interests of non-human animals on par with humans. There may be good reasons for thinking that foetuses or non-human animals should count for something, but *discounting has nothing to do with them*. Discounting only matters if you already care about their interests.

And now I will endeavour to stop repeating myself...

I, like many of you, claim not to be a utilitarian. But it seems only Jane and John are appealing to a procedural theory of justice, while the rest of you are maximizing some social outcome, be it population size, or average utility, etc. Individual rights do not extend to a right to abortion, not completely anyway, for the same old reason claimed by the wacky pro-life crowd: it impinges on the rights of another individual, the fetus. Even assuming, as Michael does, that a fetus has no preferences, the effect of the fetus on the mother's preferences is important. Isn't the mother the best judge of these factors? I'm not sure. To me, that is equivalent to assuming the murderer is the best judge. Murder is worse than abortion, I suppose, but only because norms against murder are well established.

What she seemed to be claiming was that, even if you don't think foetuses are people, believing in a radical discount rate of 0% forces you to take their interests into account.

Yes. Exactly. Even if you don't think fetuses are "people," there can be no question that they WILL BE PEOPLE in a very short time, MUCH SOONER than anyone who lives in the year 2300 will be a "person." Thus, if you are going to base current environmental policy on the hypothetical future interests of hypothetical future people, then of course you have to base abortion policy on the future interests of not-at-all hypothetical fetuses. If you don't do so, you're being facile.

By the way, the whole debate over whether a fetus is a "person" seems jerry-rigged to me. There's absolutely no scientific question that a fetus is a developing human life. The only reason that anyone brings up whether it's a "person" is because the concept of "human non-personhood" is very useful if you want to kill something.

Anyway, more errors:

John, as Conchis keeps saying, everything you're writing depends on your assumption that a fetus is a person who's just more undeveloped than the rest of us.

No, while I think that fetuses are human beings with interests in not being killed, I'm not assuming anything whatsoever about personhood, for purposes of this debate.

But this is precisely the assumption that I reject. Until the fetus is born, it's not a person, and therefore it has lesser claims upon us than real people do.

Until the hypothetical year-2200 person is born, it's not a person, and therefore has no claim on us. Right?

The fact that people in the future will, at some point, have been fetuses is irrelevant to the question of what we owe them.

That's amazingly missing the point. The point is not that future people will pass through the fetal stage. The point is that if completely imaginary people have interests, hundreds of years before they even rise to the level of a fetus, then of course fetuses have interests. Entities that are fetuses right now are more directly sure to be "future persons" than any entity in the year 2300.

But the fact that there will be people in the future -- assuming extinction doesn't happen -- means that we have obligations to them.

But the fact that a fetus will indisputably be a "person" in a matter of a few months -- assuming that abortion doesn't happen -- means that we have obligations to it.

John Doe: You tell it:)

I don't like the attained personhood support for a pro-choice position it doesn't depend on any relevant characteristic of the fetus but rather on a rather arbitrary time table and event that does not correlate very precisely with any relevent change of the fetus. The 9 month old fetus and the 9 month old and 1 day newborn are not very different from one another in the ways that ought determine our consideration.

However it is obvious that during the human gestation period and into childhood the factors that ought determine consideration/rights emerge on a somewhat predictable time table. Before a certain point (exactly when is murky) there is no consciousness. There is no intellect, there are no preferences.

Now unlike aborted fetus, there will in the future (very probably) be actual people with real preferences and intellects. Because these intellects will be they deserve consideration. The notion that there be intellects is not a moral question. We have no duty to ensure that there will be future thinking human beings, we do have a duty to the interests of thinking future human beings that we believe will be.

Someone just reminded me there are worse things to worry about:

http://www.ebaumsworld.com/2007/01/why-idiots-prevail.html

But Conchis, you've now reduced your article to a tautology: those who have an interest in not being killed, have an interest in not being killed. Okay, now that we've wrapped that up, we need to determine who or what belongs in that category, and how we are compelled to treat their interest. IF you apply a zero rate of pure time preference, I do not see how fetuses can be excluded from the category of those who have an interest in not being killed, since in less than nine months it is overwhelmingly likely that the entity currently called a fetus will enjoy all the attributes that we both agree entitle him/her to personhood.

By the terms of your formulation, I, as an agent burning carbon in America, have some control over whether future Bangladeshis exist or not. Your formulation gives me the privilege of deciding that they shan't exist, because allowing them to exist would require great inconvenience on my part. If that is not all right, why not? If it is all right, how many future unpersons may I eliminate? What if I prevent all but two people from being born, who then form a small but stable breeding population that doesn't know enough about other states to be unhappy?

It is not that it is illogical to be pro-choice; I am. But it seems that the moral intuition that intergenerational time preference is wrong mandates banning abortion, and possibly birth control. I do not think that your formulation can consistently be applied in the case of global warming, which will after all undoubtedly involve a substantial falloff in birth rates in affected areas.

Actually, and maybe somebody has already mentioned this, I do think you can be consistently pro-choice and believe in a low discount rate, so long as you believe there will be future generations. You can value those future generations who will be born and who will walk the earth.

However, to the extent there's some decision that involves a phenomenon that could wipe out humanity entirely in the future, you could only count the generation that got wiped out in your utility calculations, and not count any generations beyond that. After all, following the pro-choice logic, you can't value the utility of people who won't exist. In fact, if there's some phenomenon that wipes out a lot of people and their descendants but leaves the remaining people very well off, a consistent pro-choicer should even favor such a phenomenon. After all, a lot of people got wiped out, but you can't count the utility of their kids since now the kids won't exist. But you can count the utility of the well-off survivors' kids.

So, if global warming ends up wiping out Bangledesh but also creates more comfortable Chicago winters, the logically consistent pro-choicer might favor global warming.

After all, the wiped-out Bangladeshi's children won't exist, so they can't count for future utility comparisons, but Chicagoans kids will, so they can count.

"provided you think that the increased utility of the future generations of Chicagoans outweighs the deaths of the Bangladeshi's who are alive when Bangladesh is wiped out"

Under zero discounting, the utility of ongoing generations of Chicagoans will definitely eventually outweigh the one-time loss in utility to Bangladeshis.

"Provided you're talking about a consistent pro-choice *consequentialist*"

Once we're discussing the choice of discount rates, we're into consequentialism. It's silly for a non-consequentialist to even care about the appropriate discount rate, because what is right or wrong is right or wrong independent of any discount rates.

At most, you can say you like the zero social discount rate for this particular issue because it yields the answer you believe is morally correct on this issue. But then on another issue, you may choose a different social discount rate based on another moral intuition.

That's fine, but I think we could get straight to the point if you just state your moral intution on a given issue, so then we can derive your social discount rate du jour.

Hey Keith,

I pretty much agree with what you've said.

"Under zero discounting, the utility of ongoing generations of Chicagoans will definitely eventually outweigh the one-time loss in utility to Bangladeshis."

Sort of. We need to be careful to distinguish discounting on the basis of pure (intergenerational) time preference, and discounting on the the basis that there's an (exogenous) chance that the Chicagoans might not be around in the future either. As as long as you allow the latter, I think my condition makes sense.

"Once we're discussing the choice of discount rates, we're into consequentialism. It's silly for a non-consequentialist to even care about the appropriate discount rate, because what is right or wrong is right or wrong independent of any discount rates."

Well, there are hybrid theories that follow essentially consequentialist type maximisation, but subject it to deontological constraints that pure consequentialists would buy. It might make sense for such people to think about discount rates in general, but be willing to outlaw the sacrifice-the-Bangladeshis scenario on deontological grounds. I'm not sure there's anything inconsistent in that. Indeed, I think it's pretty close to how lots of people think about stuff.

But that wasn't really my point, which wasn't to disagree with you at all. It was to head of a potential argument against the specific form of consequentialism I'm advocating (call it "Conditional Caring", on the basis that it cares about individual interests only conditional on their existence). My fear was that someone else would try to use something like the following argument to disprove Conditional Caring:

(1) Conditional Caring & (Pure) Consequentialism => Sacrifice-the-Bangladeshis
(2) Sacrifice-the-Bangladeshis is false.
Therefore:
(3) Conditional Caring is false.

Not only is this clearly invalid, but it can't even be rephrased as an abductive argument, because the best argument against Sacrifice-the-Bangladeshis is that (Pure) Consequentialism is false. The point of which is simply that believing (2) gives one no way to resuscitate the claim that a 0% Rate of Intergeneration Time Preference => Pro-Life, which requires Conditional Caring to be false.

Perhaps that was overkill.

"(1) Conditional Caring & (Pure) Consequentialism => Sacrifice-the-Bangladeshis
(2) Sacrifice-the-Bangladeshis is false.
Therefore:
(3) Conditional Caring is false. "

Yes, that was my argument, and it works. A->B does not imply B->A, but it does imply ~B->~A.

Now, you seem to say that the argument augurs more against pure consequentialism, so you have your conditions and constraints. You said your condition was existence, but you clearly also have some ethical constraints that you're adding in there.

You seem to be claiming that the Bangladeshis who will never exist if their parents die in a flood deserve more consideration than fetuses. That's fine, but essentially you're down to choosing your constraints based on your moral intution, which seems to be "possible future Bangladeshis count more than unwanted fetuses. They just do."

Comments for this post are closed