Pro “Slavery” OpEd

Imagine my surprise to read a pro-“slavery” Post oped:

Imagine a kid from rural Montana who, after scoring high on the SAT, has investors clamoring to finance his college education. … 17 percent … That’s the astronomical potential return on investment of educational intervention on young children, according to the Nobel laureate economist James Heckman. …

Consider … A mother works two jobs, dropping her toddler off at a friend’s house early each morning and picking her up late at night. The mother can’t afford high-quality child care and, because of that, statistics show, years from now that child will be more likely to repeat grades, become pregnant while a teenager, commit crime, visit the emergency room and depend on welfare. …

If [the toddler] could sell a percentage of her future income in exchange for a coupon to receive child care and if the government offered tax credits to investors to compensate them for the decreased social cost that they finance, investors might compete to pay for her education. …

Of course, daunting logistical and moral questions would have to be answered. But before we rule out this proposal on logistical or moral grounds, let us also consider the millions of Americans who will suffer inequity and injustice as a result of our inaction.

If we allowed this, observers would happily laud the initial influx of money to these kids and their families.  But a few decades later, many would complain loudly about the “exploitative” and even “racist” extra “tax” these kids must pay as adults.  Media coverage would focus on those who sold the largest fraction of their future income and made the worst educational investments, becoming a new “slave underclass” of “company town” parents who feel they can’t afford to raise kids without selling off most of those kids’ future.

Even so, I’d accept this outcome as an improvement over the status quo.  I question the assumption that the net education externality is positive rather than negative, but types of “slavery” can indeed be efficient, and good, even a Pareto improvement.  But I’d be reluctant to be an long-term investor, fearing that later political pressure would be irresistible to “free” these “slaves” from their payment obligations.  In the meantime, however, liquid markets in income shares could be a great basis for many interesting prediction markets.

Added: Parents now have great discretion to choose the city and nation where their kids reside, who schools their kids, the language, religion, and culture kids will learn, and kids habits of housekeeping, etc.   These choices greatly constrain and influence the future options available to those kids.  I don’t yet see a good clear principle (besides “the status quo is best”) that endorses these discretions, but forbids parents to choose kid debt levels or equity shares.

Added 7June: I am especially interested in this issue because of its close relation to the whole brain emulation I have discussed before (here and here).  Note that the em case seems easier since em copies would be adults immediately, and should be very good at estimating each others’ preferences.

GD Star Rating
loading...
Tagged as: ,
Trackback URL:
  • Peter Twieg

    I don’t understand why equity would be preferred to debt here, except for the old-fashioned bias against carrying debt (which apparently doesn’t extend to selling self-equity.) If the debt is defaulted on, maybe then it can be swapped for equity at some agreed-upon rate.. but perhaps Epstein just wanted to discuss an “exotic” concept without regard for its relative merit.

  • http://sophia.smith.edu/~jdmiller/resume.pdf James D. Miller

    You could take this a step further and apply it to abortion decisions. Imagine a woman is considering having an abortion. But then an investor offers to pay the woman to carry the baby to term. In return, the investor will get a percentage of the baby’s future income. The baby will now be born into slavery, but he will at least be born.

    The investor would demand a sample of the baby’s DNA before deciding whether to finance the baby’s birth. Only baby’s with the genes for high expected earning potential would get financed out of abortions.

    • http://hanson.gmu.edu Robin Hanson

      Yes this is an interesting variation, to which I find it hard to object.

      • http://www.philosophyetc.net Richard

        Assuming that fetuses aren’t yet moral persons, the proposal is equivalent to paying a couple to conceive a child and hand it over to be the investor’s slave. The resulting child isn’t exactly benefited by this, since there’s no basis for comparison (they never would have existed otherwise). So we need to ask whether the world as a whole is better if it contains extra slave-children (of slight positive welfare; the negative case is a no-brainer). I’m not a total utilitarian, so I find it easy enough to object to this as a bad outcome. YMMV.

  • Nathan Cook

    These kids would still be citizens, right? So they would get passports and leave, permanently. Not all of them, sure, but certainly the ones who made something of themselves, i.e. the ones who would otherwise provide most of the return on investment.

  • http://www.weidai.com Wei Dai

    I don’t think parents are legally allowed to incur liabilities on their children’s behalf. There must be reasons for this, which I don’t see addressed in the op-ed, other than in the phrase “daunting logistical and moral questions”. Surely, when this law or legal principle was created, it was anticipated that underinvestment in human capital might occur as a result. So what has changed that we should reconsider it now?

    • Jayson Virissimo

      “I don’t think parents are legally allowed to incur liabilities on their children’s behalf.” -Wei Dai

      Ever heard of Social Security?

      • ad

        Social Security allows you to incur liabilities which will be paid off almost entirely by other peoples children.

        If your own children had to pay off the liability, it would be a lot less popular.

      • http://macroethics.blogspot.com nazgulnarsil

        ad is clearly a poor person or expects to have poor children. of course low earners don’t pay the full amount for their parents/ But on the flip side high earners are paying for their parents plus other people’s parents.

  • kevin

    Do student loan companies offer loans to students based on earning prospects? For example, do they use IQ testing (or SAT/GRE as an IQ proxy) to predict their future income prospects? Do they use personality testing (or some proxy of personality testing) to determine if the potential lendee is likely to repay loans?

    If not, I have a good risk mitigation strategy to offer them. 😀

    • http://macroethics.blogspot.com nazgulnarsil

      illegal due to the white black IQ gap. Such a scheme would be shot down as racist.

      • tim

        It’s not the “white black IQ gap” that would torpedo this scheme. Simply the fact that, say, men earn more than women on average would be sufficient, as men by default would be “worth” more investment. One significant consequence: if a particular segment of the population tended to earn less, and if this was due to lower education, then providing less in student loans would mean that segment’s earning power could constantly degrade.

  • http://rhollerith.com/blog Richard Hollerith

    In the U.S. almost all student loans are guaranteed by the U.S. government: if the borrower defaults, the government makes to bank whole.

  • Psychohistorian

    This also assumes these returns would remain constant were this investing system implemented. Part of the reason a college education has that 17% return to investment is because not everyone has it. If you made it easier to get the capital, the result may (and should, theoretically) be the reduction of those returns. When you consider that investors would be buying up parts of large blocks of investments in students, and when you consider how asymmetrical information is and how difficult some enforcement may be,
    “daunting logistical problems” seems a rather severe understatement. And that’s not even getting to moral issues.

    I also agree with the debt v. equity. The primary advantage of equity appears to be that it sounds more controversial because you get to say “slavery.”

    • http://hanson.gmu.edu Robin Hanson

      Yes the 17% would only be realized rarely; most returns would be lower.

  • http://www.athousandnations.com Mike Gibson

    Somewhat related, there’s a new sci-fi novel out called “The Unincorporated Man”, in which the authors envision a future in which this policy is the norm. They take a section from Friedman’s Capitalism and Freedom on this kind of financing for education and then use it as a springboard to set up a dystopian future.

  • Jess Riedel

    Doesn’t everyone here agree that it would be wrong, as Wei Dai points out, for parents to incur liabilities on their children’s behalf? Children aren’t fully capable agents, therefore they can’t enter into most binding contracts, take on debt, etc. What’s the debate?

    Much more interesting is the question of whether 18-year-olds (or perhaps a year or two younger) should be able to sell a percentage of their future income for college loans and the like. And, in fact, I believe this has been done in the past, although I can’t seem to find any evidence for it online.

    The Income Based Repayment program seems to be nearly equivalent to this. However, it is retroactive, so it basically amounts to government funded debt forgiveness. But it’s a step in the same direction.

    • CannibalSmith

      But parents already are incurring liabilities on their children’s behalf through inadequate parenting etc.

  • http://hanson.gmu.edu Robin Hanson

    I just added to the post.

    • Jess Riedel

      I don’t yet see a good clear principle (besides “the status quo is best”) that endorses these discretions, but forbids parents to choose kid debt levels or equity shares.

      Can we not draw a distinction between (a) choosing where geographically we raise a child, what habits we give the, etc., and (b) saddling a person with long-term contractual obligations? One is a restriction upon the child when they are a child, the other is a restriction when they are an adult.

      Just because a parent gets to choose where a child goes to school doesn’t mean they get free rein over their entire lives.

      • Peter Twieg

        I think the point is that a parent’s control over a child’s education could have greater long-term impacts than saddling the child with moderate long-term debts. If we view the educational process as one of imbuing human capital (which many do), and we consider it morally repugnant for adults to be negligent in imbuing their children with human capital (which many do), then it seems odd to conclude that adults shouldn’t have access to options to finance this process that are legitimately expected to make children better off.

        I understand that there does seem to be an essential difference between neglecting a child’s education and giving a child a good education but also saddling her with debt – but if we look at the welfare of the child, it’s not clear that one option is better than the other for her, just that the debt option seems to have repugnantly constrained her choice set… which ignores the effects of education on the same choice set.

  • Robert Koslover

    Hi folks. I’m not a social scientist, but I agree with the comments by both Wei Dai and Jess Reidel above. Also, isn’t it true that one of the more objective ways in which humanity is often considered to have advanced, in social not just technological terms when compared to our uncivilized past, is that human slavery is now almost universally regarded as morally unacceptable?

  • Raphfrk

    The income based payment seems like a reasonable policy. Presumably, the theory is that if you aren’t earning a certain level of income then you aren’t getting the benefit of the college education.

    It would also provide lots of info about what courses are actually worth doing. If a $20k loan cost 1% of your future earnings for 1 course and 5% for another, it is clear than the 2nd one is a much less marketable skill. It would also allow different courses in different colleges to be compared.

    Btw, this isn’t technically slavery, it is closer to indentured service, as the person is agreeing to be bound. I think as long as there was a cap (say 15%) for the total amount of your income that you can commit, then it isn’t likely to result in problems.

    I think as long as companies aren’t allowed sue you if you take a lower paying job, then it doesn’t count as slavery. You would pay them x% but still be free to choose what jobs you take.

    Also, there is the issue of what happens if someone leaves the country. This is less of an issue for the US, but for a small country, the effect could be to push graduates to leave.

  • http://mark.reid.name Mark Reid

    The Australian government already acts as an investor in students’ university education through a scheme called HELP – the Higher Education Loan Programme. The loan is only available to students having already secured a spot at a university through their grades.

    The loan is paid pack as an extra tax on the student’s income once he or she is earning above a certain threshold. The interest is set to the CPI. Early repayment is possible and encouraged via discounts.

    Even though the scheme has been around for at least 20 years and many people have paid loans back, the general attitude towards them seems to be positive and I am not aware of any observers discussing them in terms of “slavery” or “exploitation”.

    Given my parent’s financial situation at the time I started university, these loans made it possible for me to get a degree. I entered into the loan when I was 17 and have had no regrets.

    • http://meteuphoric.blogspot.com Katja Grace

      I’ve seen more negative attitudes to the system. At ANU over the last few years there seemed many complaints and small activist efforts in favor of free education ‘because it’s a human right’. The National Union of Students seems to call for lower debts, and the Greens and I think socialist parties pressure for free education. The one time I spoke in favor of the Australian system at home my parents were shocked.

      Apparently one of the most popular ideas coming out of the Australian 2020 summit last year was to allow students to reduce their debt by doing community service during university (reducing debt at a rate far below the minimum wage as far as I read). That anyone wants to reduce an interest free debt years early by working a far worse paying job than they could get later (while many complain they are living in poverty, and while they will only have to pay it back later if they are not) suggests they have especially ill feelings about the debt.

  • gwern

    Jess: parents are allowed, without moral censure, to do plenty of things that affect their kids’ futures negatively.

    For example, just about any health-related failing will negatively impact fetuses, to say nothing of what the parents name their kids, or where they raise them, or whether they homeschool them, or how old the parents are when they conceive the child in question.

    We don’t feel there’s any moral issue with a 45 year old having a kid, even though we have studies saying that older parents mean worse health and lower IQs; we don’t feel there’s any moral dimension to a black parent picking their kid’s name out of the latest _Ebony_ even if ‘black’ names will cost them in the future, and so on.

    In what way does a pregnant mom failing to take her vitamins or exercise sufficiently not be equivalent with a more diligent mom indenturing her child? In both cases the child struggles under a burden not of its choice; that one burden is physical/mental and the other financial doesn’t seem like a very important difference to me.

    • Jess Riedel

      This is completely besides the point. Robin asked for a clear principle which distinguished between normal child rearing choices and contractually obligating them to service a debt.

      Yes, if you assume some version of consequentialism, then you can draw conclusions like the ones you are suggesting (although these become empirical questions which are far from obvious). But most people are not consequentialist. Just as they wouldn’t accept slavery even if it led to the maximization of utility, most people wouldn’t equate these two very different ways parents could effect their children.

      • gwern

        Jess, Robin asked for a clear principle. You still haven’t given one; your closest attempt seems to be

        > Doesn’t everyone here agree that it would be wrong, as Wei Dai points out, for parents to incur liabilities on their children’s behalf? Children aren’t fully capable agents, therefore they can’t enter into most binding contracts, take on debt, etc. What’s the debate?

        Well, as most people here are consequentialists (or the deontologists and the rest are hiding out), no, it’s not a universal matter of agreement that we Just Shouldn’t Do That.

        As I tried to show above, it’s perfectly plausible that a suboptimal upbringing could be equal in goodness to an optimal upbringing with liabilities. Given that society will not intervene in most cases of suboptimal upbringing, why should it intervene in the latter case?

        Parents already have powers over their children that are best described as tyrannical, and those powers have always been fiercely guarded. Incurring debts binding on them wouldn’t seem to be too great. (eg. parents can imprison their children until age 18 by homeschooling them and just not letting them leave etc. And in another comment I point out that parents legally have tremendous scope for using violence against their children.)

        Better than acting outraged and appealing to people’s intuitions that this is a Bad Thing, would be an attempt to justify the status quo by, I dunno, arguing that modifying the legal system to remove the abrogation-of-all-contracts-at-age-18 would have bad consequences.

  • Eric Johnson

    I don’t yet see a good clear principle (besides “the status quo is best”) that endorses these discretions, but forbids parents to choose kid

    The only argument I can think of is that we may have evolved to accept that our parents determine our residence, mother tongue, religion etc. If we are less well adapted towards being entered into the proposed sorts of contracts by our parents, then the contracts may cause more umbrage than our parents’ determining our mother tongue.

  • http://www.weidai.com Wei Dai

    Society doesn’t prevent parents from making choices for their children (except in cases of clear harm), but it also doesn’t allow parents to use violence to enforce their choices. Violence or threat of violence would be necessary to enforce kid debt levels or equity shares.

    While most parents are altruistic towards their children, some aren’t and would exploit them if they could. Presumably, not allowing violence prevents the worst abuses, and that’s worth the lost opportunities where some parents might achieve more optimal outcomes for their children by using violence.

    • gwern

      > Society doesn’t prevent parents from making choices for their children (except in cases of clear harm), but it also doesn’t allow parents to use violence to enforce their choices. Violence or threat of violence would be necessary to enforce kid debt levels or equity shares.

      Were you never spanked as a kid? As it happens, I was once told that my state’s child abuse laws only covered violence that left semi-permanent traces (eg. bruises). Quickly googling, this seems correct:

      “Injury by other than accidental means causing death, disfigurement, impairment of physical or emotional health; deliberate indifference causing such injury; creating substantial risk of such injury; sexual abuse, permitting sexual criminal behavior”
      http://law.findlaw.com/state-laws/child-abuse/new-york/

      • http://www.weidai.com Wei Dai

        Yes, I was spanked a couple of times. If I had known to claim “impairment of emotional health,” maybe I could have saved myself. 🙂

        Ok seriously, Robin asked for a “good clear principle” that explains the line between what parents are and aren’t allowed to do, and I offered one. Upon reflection, it doesn’t really make sense that there would be a good clear principle that explains the line, since it’s actually the result of compromise between political forces backing a variety of moral and legal principles.

  • Josh

    An even better solution has already been proposed.

    We don’t feel there’s any moral issue with a 45 year old having a kid, even though we have studies saying that older parents mean worse health and lower IQs . . .

    We don’t?

    In what way does a pregnant mom failing to take her vitamins or exercise sufficiently not be equivalent with a more diligent mom indenturing her child?

    First of all, a pregnant mother ought to eat well and exercise sufficiently. None of us are perfect, and we can understand when people have moral failings; but pregnant mothers and their partners do have a special obligation to care for themselves and their future children.

    More importantly, the question here isn’t whether it is moral for parents to sell their children into indentured servitude. The question is whether it is moral for a society to bind an individual to a debt they never incurred. I’m not well-read enough to know exactly what I am, but I’m a type of consequentialist if not a utilitarian, and I am at a loss trying to understand how this maximizes happiness.

    But okay, I’ll play. If we’re willing to allow parents the incentive to loot by selling their children’s freedom, which never belonged to them in the first place, why not have the moral courage to allow them to loot us all by publicly funding childcare and a decent primary and secondary education? That would have the added benefit of avoiding the potential for slaveowners, err, shareholders, to manipulate financially illiterate parents into making poor decisions on their children’s behalf. (And I’m sure teaching kids to read and write properly has positive externalities. yeesh.)

    • gwern

      > We don’t?

      I’ve asked a few people questions along the line of ‘if you knew that the older the parents the worse off the kids in terms of IQ/health, and some parents A elected to have their kids at 45 instead of 40, would you consider this choice to have any moral dimension?’ (in various simplified or nuanced ways). Most of them didn’t see it as an ethical issue, or if it was one, it was a very tiny one.

      Unscientific? Sure, but in accord with intuition. Most people don’t seem to take correlations or causations seriously unless the links are very strong. (A comparison here might be: Big Tobacco = murderers, environmental polluters = not murders, but just polluters.)

      > First of all, a pregnant mother ought to eat well and exercise sufficiently. None of us are perfect, and we can understand when people have moral failings; but pregnant mothers and their partners do have a special obligation to care for themselves and their future children.

      Exercise sufficiently, I said. What if she should exercise 60 minutes a day and only does 20 or 30? Who’s really going to see that as a moral issue? 20 or 30 is Good Enough.

      > The question is whether it is moral for a society to bind an individual to a debt they never incurred. I’m not well-read enough to know exactly what I am, but I’m a type of consequentialist if not a utilitarian, and I am at a loss trying to understand how this maximizes happiness.

      If we judge based on what people do, apparently it’s perfectly moral to bind people to debts they didn’t contract (and Constitutions, and…). At least, I’m hard-pressed to think of a functioning civilized country with no national debt.

      > If we’re willing to allow parents the incentive to loot by selling their children’s freedom, which never belonged to them in the first place, why not have the moral courage to allow them to loot us all by publicly funding childcare and a decent primary and secondary education? That would have the added benefit of avoiding the potential for slaveowners, err, shareholders, to manipulate financially illiterate parents into making poor decisions on their children’s behalf. (And I’m sure teaching kids to read and write properly has positive externalities. yeesh.)

      In the US, apparently half the electorate considers that it’s fine to loot their children, as long as it’s for something like tax cuts or invading Iraq. Between looting for Iraq and bailing out financial elites, and splurging on education, I like your suggestion better. So we agree there.

      • Josh

        Most of them didn’t see it as an ethical issue, or if it was one, it was a very tiny one.

        But originally you claimed the following:

        We don’t feel there’s any moral issue. . .

        The latter could be used to imply a contradiction if someone says parents shouldn’t have the moral discretion (as opposed to the legal discretion) to harm their children needlessly, but the former cannot be. I happen to think it’s an ethical issue, and depending on how bad the health outcomes are, it could be small or large.

        Exercise sufficiently, I said. What if she should exercise 60 minutes a day and only does 20 or 30? Who’s really going to see that as a moral issue? 20 or 30 is Good Enough.

        Well, immoral does not mean the same thing as requiring moral censure. No one is perfect. I still see it as a moral issue, though.

        . . . Apparently it’s perfectly moral to bind people to debts they didn’t contract (and Constitutions, and…). At least, I’m hard-pressed to think of a functioning civilized country with no national debt.

        Social contracts are a different beast; it’s hard to say exactly what voluntary and involuntary mean in this case. I don’t think social contracts need the consent of every individual they bind, but I can’t think of many good reasons to bind people to individual contracts they didn’t consent to.

    • Ben

      If the parents are given vouchers for education, rather than cash, their ability to loot is significantly reduced. (not to zero, as they could substitute vouchers for cash they would otherwise invest in their child’s human capital. But if they’re investing anyway, they’re not likely to loot.)

      Additionally, investors wouldn’t want to ‘buy in’ to children likely to be looted.

  • Robert Koslover

    I suggest that the line-in-the-sand is a two-part rule: Any decision that results in “indenturing” a person should satisfy the following: (1) the decision must be made by the actual person to be indentured; and (2) the person must be “of age,” such as 17, 18, or some similar number of years, that the society in question recognizes as marking when a person is deemed able to make informed judgments about his/her future. My reasoning here is, quite deliberately, consistent with long-standing rules in the USA for signing up for ROTC programs, which pay for a student’s college education in return for a multi-year commitment to serve in the military reserves. I understand that similar contracts have been negotiated by municipalities, e.g., contracts to the effect of “we’ll pay for your medical school, but then you have to offer medical care in this small town for X number of years.” Note that none of these contracts are signed by, or on behalf of, anyone who is below a particular age. Hence the two-part rule.

    • InLimine

      The problem with your two-part rule is that by the time the proposed indenturee becomes “of age” no benefits from Epstein’s proposal would be left. The point is that this type of plan needs to be instituted early in life when it will make a true impact on the child.

      • Joe Teicher

        The most obvious solution that I can think of is that the parents retain the liability. They agree to pay a fraction of their child’s future earnings in exchange for cash up front. If they loot it then their child is unlikely to assume the liability as an adult, but if they make a genuine investment the child would presumably help out their parents. It seems like this would increase the default risk somewhat but I think it would still leave profitable investment opportunities and it would greatly reduce at least my moral concerns.

  • michael vassar

    Agreed with raphfrk that this is definitely not slavery and with Wei Dai about parents not using violence to enforce many of their choices.

    The key to this whole discussion is that we have nothing even remotely close to a generally accepted ethical norm about creating new people and Robin is trying to smuggle that lack of consensus about creation of new people out of its tidy box and claim that it implies a total lack of consensus about existing people.

  • http://www.cawtech.freeserve.co.uk Alan Crowe

    If the question is what principles should be used to analyse this kind of question, I have two answers.

    First separate the politics-level language from the object-level language. For example, at the political level we say “Lets us prohibit alcohol; this will completely solve the problem of alcoholism.” and the object-level reads “Let us attempt to prohibit alcohol by police work obstructed by collusion between buyer and sellar and bribery of the police, resulting in an expensive increase in the prison popular and little reduction in alcoholism.” The proposal doesn’t separate out how it is supposed to be and how it will really work. That leaves it unclear as to what is actually proposed. Are we asked to consider a pure hypothetical (would it be a good idea if it worked like it is supposed to?) or are we being asked to consider it as a practical proposal with unintended consequences? Little good can come of a muddled conversation.

    Second answer is the principle of triple distrust. Work through how this proposal goes wrong when people are venal, weird, and incompetent.

    One imagines that Ying-co provides the money to pay Yang-co for the child’s education. Ying-co has an incentive to get value for money, least the uneducated child fail to repay the debt. In real life Ying-co and Yang-co are both owned by Whole-co. This destroys the intended incentive scheme. Yang-co give Ying-co money for old rope and Whole-co acquires indentured workers for free.

    One imagines that the Wahhabi Education Fund insists on a particular kind of Islamic education. The overall system relies on the inbuilt incentive system: if the education provided leads to poverty it also leads to defaults on the loans. But the Saudi backers of the Wahhabi Education Fund do not in fact see it as a commercial activity but rather as an opportunity to promote their values.

    One imagines that Ultimate-Stupid-co with its mission statement of leveraging synergies to achieve ultimate educational performance is actually responsive to the intended incentives, but what if they are hopelessly incompetent? Eventually they go bust, and the debts, the assets of Ultimate-Stupid-co are sold to repay the investors. The investors get pennies on the pound. The children remain burdened by huge debts from a useless but expensive education.

    Perhaps that is more of a checklist than a principle and anyway I have only done a small part of the job by suggesting a triple-distrust analysis of the proposal but not doing it for issues such as parental location. Nevertheless one must use some such lens for scrutinising the issue if the discussion is to make contact with human experience.

  • Max

    We already have a similar thing – student loans. You don’t have to be a slave, at least not for life, and the interest rate is much lower than 17%.

    Why would anyone agree to the slavery proposition that charges usurious rates? The debt is better, since you can pay it off, after which no additional commitments are required.

  • http://www.weidai.com Wei Dai

    BTW, I do think it would be useful to create markets that allow adults to sell a fraction of their future income stream, not primarily to correct for underinvestment in human capital, but to allow more diversification of risk. I’d justify making this change to the status quo by pointing to the lower costs of securitization that we have today as a result of advancing information technology. This also satisfies Robin’s wish for a liquid market in income shares to serve as a basis for prediction markets, without introducing any thorny moral issues.

    I think I proposed this idea (or maybe just read about it?) years ago on some mailing list, but can’t find the original post now.

  • http://macroethics.blogspot.com nazgulnarsil

    I think this hits on the deeper issue that we’re uncomfortable formalizing certain relationships in terms of cold hard cost benefit analysis because we prefer not to signal that we’re opportunists looking for the best deal.

  • http://hanson.gmu.edu Robin Hanson

    I just added again to the post.

  • josh

    What is the principle difference between the slaves relationship with his mater and my relationship with the US Government that makes slavery always immoral. I agree with Robin that it should be a question of efficiency.

  • Josh

    Jess, Robin asked for a clear principle.
    . . . Well, as most people here are consequentialists (or the deontologists and the rest are hiding out), no, it’s not a universal matter of agreement that we Just Shouldn’t Do That.

    If we are not willing to agree to any universal ethical principle, how can someone present us with a universal ethical principle to determine what discretion parents should have? And why is such a principle needed?

    As I tried to show above, it’s perfectly plausible that a suboptimal upbringing could be equal in goodness to an optimal upbringing with liabilities. Given that society will not intervene in most cases of suboptimal upbringing, why should it intervene in the latter case?

    It could be that we would like to reduce some of the discretion that parents already have, but it is politically less feasible than denying them this new discretion. There may not be a clear way to distinguish.

    But I really do think this is different. The best principle I can give, which is far from clear or universal, is that parents are more likely in this case to choose incorrectly for their children. Without publicly funded childcare and a solid primary and secondary education everywhere, parents have an obligation to work hard to provide these to their children. By enforcing contracts for indentured servitude, society would be giving parents an incentive to sell part of their children’s future when they didn’t need to. Investors would have an incentive to manipulate parents into doing so. Yes, I’m sure similar incentives exist in other cases, but in this case they seem especially likely to lead to bad outcomes and unlikely to do much good. And it’s a case where society has to intervene dramatically in order to give parents this discretion, whereas most of the other cases mentioned would require dramatic intervention to take it away.

  • JWB

    This isn’t all that different from student loans except for some radical rights infringement here and there but before we change our laws to allow parents to bind children to service note that any return would be greatly reduced if we did this across a society (because the advantage of a high quality education is reduced when more people have it). Also investments that require a 20 year wait before anyone can see any sort of return is a hard sell especially when we are dealing with an investment fraught with all sorts of risks (we already discussed that the kid may default or perhaps fudge on obligations, the laws could change etc). There’s a reason we don’t make loans to someone with a high interest rate that they don’t have to start paying for 20 years from present.

  • Jeff

    Turns out there’s something like this already:

    My Rich Uncle