To which Sailer responds:

Sure, but the bigger question is: What’s in society’s long-term interest? I mean, where do the next generation of poor children come from? Does the stork bring them?

Obviously, with the exception of immigration, poor children are mostly the product of poor parents.

Society is better off if the kind of young women whose lives wouldn’t be ruined by having a child out of wedlock reproduce less rapidily.

Sailer’s argument, basically, is a eugenic one. Whether you want to attribute it to biologically-heritable characteristics or cultural factors or whatever, poverty runs in families. Not by any means exclusively, but it does. So, yes, one answer to “how can we reduce poverty?” is: “reduce the birthrate of poor people.”

Except . . . how? Assuming we’re not going to entertain profoundly illiberal ideas, what that means is providing some kind of incentives to defer childbirth. Which brings us back precisely to where Yglesias was. His answer is: economic opportunity, and the perception thereof, is the key incentive to deferring childbirth. It’s the flip side of Sailer’s own “affordable family formation” thesis. Women who already aspire to a middle-class lifestyle will defer childbirth until they can afford it – so reducing the cost of such a lifestyle (of housing particularly) will encourage family-formation. Women who don’t aspire to a middle-class lifestyle because they see it as unattainable will have children for psychological reasons (having someone to love, for example) without regard to their economic circumstances – because they aren’t planning to change those circumstances. So increasing economic opportunity, providing a ladder out of poverty for the long-term impoverished, will encourage poor young women to defer childbirth, and start behaving like people who aspire to a middle-class lifestyle.

Makes sense to me. But saying “provide a ladder out of poverty” isn’t much of an answer because we don’t have a huge number of examples of policies that have successfully broken the back of multigenerational impoverishment. Similarly, saying “provide a better education” isn’t much of an answer because we don’t have a huge number of examples of policies that have successfully educated the worst-educated segments of society.

We know a little bit, but we don’t know that much. Yglesias mentions in passing that “stingier welfare benefits” cut birthrates modestly, but if I remember the literature (and it’s been a while), what makes a dent isn’t stingier benefits – less money – but work requirements. There’s a handful of “reform conservative” types – Ross Douthat and Reihan Salam are the ones I know best – who’ve advocated a kind of traditional/paternalist reform of the welfare state, focused not on spending less but on structuring programs to better-incentivize good old bourgeois virtues – hard work, deferred gratification, etc. – which may, in the end, require more money. Yglesias himself comes from a much more “liberaltarian” direction – aware of the danger of service-provider capture of social programs (a criticism usually lobbed from the right) and respectful of the autonomy of the poor, he’s argued in the past in favor of simply “writing checks” as the solution to poverty. Charles Murray has, on occasion, argued for the same, and for similar reasons (though with less conviction that poverty would actually go down that much).

This debate – between more libertarian-minded and more paternalistic schemes of to tackle the problems of long-term poverty – is what’s interesting to me. I’m already sold on the idea that economic aspiration is the key to keeping poor girls from getting pregnant. But how the government can successfully promote that kind of aspiration is not something we agree on the answer to. We need more data – from more experiments.