Some Links to Consider, 8/7
Scott Alexander on population decline; Emily Oster on studying rare events; Antonio Garcia Martinez on Balaji's book
As birth rates rise, you have many hard-working young people supporting a small number of retirees. As they fall, you have fewer young people and more older people who need support. This either burdens the young, or requires cuts in support for the elderly.
And yeah, to some degree this will happen. I think it will look less like an apocalypse and more like increasing effective retirement ages, but that will suck.
I agree with him that population decline is not worth worrying about. Change is happening slowly, and there are many ways to adapt. Plenty of time to substitute robots for labor, for example.
The biggest demographic shock that has a chance of occurring over the next 50 years is a cure for aging. That would stop population decline. The world in 2100 would be full of vibrant, healthy people born late in the 20th century and early in the 21st.
Imagine I want to study a condition that appears in 1 of 10,000 people. In the NHANES, I’d expect one person (on average) to have the condition in each year. That isn’t enough people to say anything useful about even what is correlated with it.
In studying rare outcomes like this, it is therefore common to use a “case-control” approach. This works by beginning with identifying a set of cases — people who experience the rare outcome. The researchers then identify a set of controls — people who are otherwise similar but didn’t experience this rare condition. You then collect information about behaviors and circumstances for both groups, and try to identify behaviors that are more common in the case group.
…Studies of SIDS, stillbirth, and childhood cancer all commonly rely on this approach. This is by necessity, and these studies can be done well. But they are also subject to significant concerns.
…When you recruit people for a study from the general population, you typically get a fairly selected sample. People who are in the case group — with a sick child — usually have high participation because they hope to get some answers. People who are unaffected but volunteer to participate tend to be very different — in some ways that you can see and some that are harder to see. When you compare behaviors between cases and controls, you now worry about differences in the sampling that could drive your results.
She criticizes a study that claims that pregnant mothers who eat hot dogs increase the risk of childhood cancer. I think that a similar criticism could be applied to the SIDS studies that purport to support back-sleeping, although she would not agree.
Antonio Garcia Martinez writes,
The most radical (and underrated) change wrought by technology has been the decoupling of information from physical movement, the flight of bits liberated from the slow lurch of atoms. This dislodges human life from a geographic setting, making what you see, think, and experience independent of the colored shape on the map labeled “San Francisco, California, USA” (or whatever).
This point requires an amendment. Some of human life has been dislodged from a geographic setting. Not all of it.
The main point that Martinez makes is that any state, network or physical, requires some strong tribal psychology to maintain it. Srinivasan shares that view.
In the abstract there are a lot of possibilities for "population decline", and in some of those scenarios there are potential upsides like capital accumulation per capita and relieving scarcity, congestion, pollution, etc.
But if we focus on the specifics of the actual kind of population decline we are observing to occur across the entire developed world, and even most of the developing world, there are legitimate sources of worry.
My impression is that the whole debate is warped by a background consideration that, IIRC, Yglesias made explicit at least once. That is, "Well, what are you gonna do about it?" The implication being that since the politically acceptable tactics like baby subsidies don't seem to do much, all the other government interventions that stand any chance of working are basically "Handmaid's Tale" unthinkably dystopian and so beyond the pale that it makes these complaints like "unactioinable intelligence". And the brain adapts to problems it can't solve by SlatePitching itself into redefining "legitimate problem" to move this issue from the legitimate-problem class to the illegitimate-complaint category, which justifies ostracizing anyone who complains about it. Accepting it as a problem and taking it seriously would mean measures too radical to contemplate.
As for worrying scenarios, for one thing - and as Scott discussed when mentioning Hive Mine - if you believe in smart fraction theory - that the number of people in their prime productive years and above some cognitive threshold both in absolute terms (innovation) and relative terms (leadership roles) has an outsized influence on a population's welfare and prospects - then one needs to pay close attention to any multiplier effects that happen at the right tail of the distribution when that distribution changes.
The issue is that if smart fraction is true, then when they are combined, even relatively small and slow changes in (1) fertility rates, (2) disparities and skews in who has the kids, and (3) in many countries the brain drain effect, can still make things go really bad fast. Scott says "0.3 IQ points per decade" which is just 2% of the SD and way too low an estimate in my judgment. But even a 20% drop in the SD (100 to 97) reduces the number of +3SD people by 6.7%.
If you add in declining births, the net flow of population through "prime productive years" is negative, and makes this decline worse. In South Korea, there are about 12 million people between 35 and 50. In 35 years, that will be cut in half to 6 million. With a little high human capital emigration, by 2060 their smart fraction could be down 2/3rds, and if Hive Mind is right, that's huge and justifiably worrisome. That is, if you don't define away the problem.
Another source of worry is that many people have really bad models and theories for why birth rates are declining so much and so fast in so many places despite all the other big differences in policies, resources, culture, etc. Even if you find the current situation acceptable you can still have a point at which, if current trends continue, the situation would become unacceptable.
If you don't know why the trend went the way it did, you don't have any good reason to believe "here but no further" and why it won't continue to keep deteriorating, then focusing analysis on the derivative is an error when all the action's in the second derivative and what's driving it, indeed, what's been driving it for a long, long time. And something like that doesn't just stop on its own unless we get serious about stopping it.
Something that can't go on forever won't, but the question then becomes, "When it goes, how much else will it take down with it?" The bathwater can't keep draining from the tub forever, but it can take the baby with it down the drain.
Being alive is awesome, not just cause it affects society. I think that more people is probably good and declining birth rates is bad.