More On Replacement Fertility

As must be obvious, I've spent a large part of today thinking through some points on the demographic side of things. One of the conclusions I've certainly come to is that there is a big problem: demographers and economists aren't really talking to each other. The economists who understand growth theory (eg Robert Lucas) don't understand that the definition of the demographic transition they're working from is, well wrong. Demographers, who don't understand too much economics think the problem can be addressed by massive capital deepening, without considering the kind of 'diminishing returns' limits to which this will be subjected. Or they think about human capital, without thinking of the narrowing gap we are now experiencing with eg China and India, or they think in terms of capital exports, without thinking about the currency re-alignments which may occur on the way. One of reasons for my time abuse today is Mats who kindly asked me the gollowing question:

Commentators over at Brad DeLong's site briefly touvhed on the question of demographics. I've read your links on it. In one of them they point at the very regular tendency for public statisticians/demographers to assume birthrates reverting to replacement levels. And this reversion often goes with little or no explanation. What's your opinion?

Well my opinion is that there is a big disconnect between those deomgraphers and statisticians who work formulating policy (and economists of virtually every description), and the theoretical demographers. They seem to have much more of a consensus: replacement levels aren't coming any time soon, and certainly not in any way that will meaningfully affect the 'pensions/SS/growth panorama over the next 25 years, which seems to me what we should be focused on. If the numbers come back up over the next century, then that would be fine, although I don't see how, but this is hardly relevant to the real policy options we have today. In the extract below, demographer John Bongaarts tries to explain the observed drop in the Total Fertility Rate. One of the problems he clearly identifies is that of overshoot. That is to say, that in the early and middle phases of the transition people tended to have more children than they desired, now we have less. This situation may well adapt to a level of natality nearer to the desired level. At the same time part of the dramatic decline in TRF below 2.1 in the South European and Asian societies is undoubtedly due to a continual rise in the age of having the first child. When this stops increasing at this rapid rate, numbers in these countries may creep back up. But this process is not done, and we may be many years away from any significant change, so it would seem more sensible to decide what we are going to do on the basis of the data we already have, rather than waiting for that 'lucky eventuality' which, of course, always might turn up just around the next corner.

The future course of fertility in countries where it is already at or below replacement is one of the most controversial issues in contemporary demography. One group of analysts points to the indisputable fact that fertility has dropped below replacement in virtually all countries that have reached the end of the transition. This is the case in Europe and North America, where fertility has been below replacement since the mid-1970s, as well as in the most-developed countries in the South, such as Hong Kong, Korea, Singapore, Taiwan, and Thailand. In a few instances fertility has leveled off above replacement (e.g. Argentina and Chile), but these are exceptions. According to this school of thought, replacement fertility is a theoretical threshold that has little or no meaning for individual couples building their families, and below-replacement fertility is expected to be the norm in post-transitional societies (Demeny 1997).A contrary view is held by analysts who believe that the current low levels of post-transitional fertility are a temporary phenomenon and that concerns about imminent population declines caused by low fertility are misplaced in some countries (Le Bras 1991; Knodel et al. 1996). This perspective is supported by data onDFS (Desired Family Size), which has remained near or above two children in all societies for which measures are available. In this view, the observed below-replacement fertility is largely attributable to ongoing shifts in the timing of childbearing. Once this rise ends—as it eventually must—the corresponding fertility-depressing effect stops, thus bringing fertility back up, presumably to near replacement. These competing views are both partly valid, but incomplete. The actual situation is more complex and a full assessment requires a separate examination of trends in DFS as well as in each of the six factors linking fertility to DFS.
MORE