Dear Ben,

       We’ll never agree on whether the human population as a whole is about to go into a tailspin or whether (with luck) it will approximately stabilize at a level consistent with a satisfactory standard of living for all. We simply don’t have any evidence about what will happen to fertility in 50 years. We do know, however, that the world population will be at least one-third larger in 50 years than it is today. Given that, it does seem premature to worry about a population implosion.
       You say that developed-world fertility has been trending downward for a couple of centuries (more like one century for most of the developed world, actually). I say that developed-world mortality has been trending downward as well, with the result that numbers of surviving children have increased through most of this period. Over the very long haul, population-growth rates have been very close to zero: Fertility and mortality have been in very close balance. Over the last couple of centuries, mortality fell sharply, at first much more sharply than fertility, with the result that population-growth rates rose to unprecedented levels. Fertility has now declined in the developed world (and is declining in the developing world) to bring them back into close balance again. It is the last century that has been abnormal from the perspective of population growth. The present represents a return to normalcy.
       Your view that it is hard to have children today confuses values with costs. You argue that young people say it is hard to have children because they cost so much, because women have to work, because they have college loans to pay off, etc. Anybody would think that times are tough. But they aren’t, Ben! Per-capita income in real terms is more than twice as high today as it was in 1950. The tax deduction for a dependent is dickering at the margin: The big change is in the overall standard of living.
       Differences in fertility among social groups underline the same point. If money were the major problem, the rich would have lots more children than the poor. They don’t, so money is not the major problem.
       Why, then, do couples have fewer children today than in 1950? And why are individuals less likely to get married? The answer is that values and preferences have changed. People are making different choices. It’s not that it is hard to meet people or to have children but rather that people are choosing not to settle into long-term relationships and are choosing not to have as many children. Is it the business of the state to say they are wrong, or to institute incentives to encourage them to have more children than they would otherwise choose to have? It might be, if the imperatives to the state were of overwhelming importance. French governments have followed pro-natalist policies for decades (with very limited success, by the way) to try to maintain an adequate population for national defense.
       Let’s focus on the United States and its population, since this appears to be your main concern. (On the basis of your remarks to the effect that it is the American experience that has brought into play the best aspects of organized humanity, I should have accused you not of being ethnocentric but of being nationalistic.) So are the imperatives for a pro-natalist policy in the United States overwhelming?
       A recent National Academy of Sciences report on immigration made population projections on the basis of various assumptions about future immigration. If immigration continues at current numbers (which would imply falling rates), the population of the United States will increase to 387 million by 2050–from 270 million today. That is an extra 117 million people in a little over 50 years. In 2050, the annual growth rate of the population will still be half a percent a year. So we really don’t have to worry about the population going into a tailspin. The dependency ratio (defined in this case as the ratio of people under 20 or 65 and over, to those age 20 to 64) will be higher then than now: about 60 dependents to every 100 workers, compared to about 52 today. But the ratio will be substantially lower than it was in the 1960s–when it was close to 70–and will not be rising; indeed, it will have fallen slightly from 2030’s.
       So there are going to be plenty of people around in 2050. The U.S. population as a proportion of world population remains constant at 4 percent. And the age structure does not become greatly more dependent than it is now. Where is the imperative for the state that would justify incentives to childbearing? The pleasures of children and grandchildren are essentially private goods. If parents want children, they can have them–they can certainly afford to. If would-be grandparents want grandchildren, then they can create their own subsidies. There is no necessary role here for the state. The state does have an interest in making sure that children get educated, and we all have an interest in establishing a living-standard safety net for children. Policies in these areas do reduce the cost of childbearing, and as such are pro-natalist, but they are not tied explicitly to births but rather to making sure that all children born have a chance to thrive.
       The last paragraph, then. We both agree that population growth has to stop sometime. You want to postpone the stopping time; I am happy to see it happen more quickly. I prefer the view of a future world in which there are still some empty spaces, open areas, wild places. Where not every inch of land is used to support human activity. A smaller ultimate population rather than a larger one. To some extent, though, these are questions of taste, and your tastes differ from mine.
       I live just outside D.C., so come over for dinner one evening–if you can stand the traffic jams–and we can continue the thread. You may want to do it sooner rather than later, though, since the traffic jams will get worse.

Best wishes,