China’s Demographic Outlook to 2040 and Its Implications AN OVERVIEW Nicholas Eberstadt JANUARY 2019 AMERICAN ENTERPRISE INSTITUTE Executive Summary urprising as this might sound, we already have a growth for the 65-plus group, even as its working-age S fairly reliable picture of China’s population profile population (conventionally defined as the age 15–64 for the year 2040 because the overwhelming majority group) progressively shrinks. While the Chinese gov- of people who will be living in China in 2040 are already ernment has a number of policy options for mitigat- alive and living there today. Consequently, despite the ing the impact of pronounced population aging and uncertainties in population projections and the addi- a shrinking manpower pool, these trends can only tional data problems caused by Beijing’s longstand- make for serious economic headwinds, presaging the ing One-Child Policy, we can actually describe China’s end of China’s era of “heroic economic growth.” impending demographic trends and evaluate their Additionally, a number of demographic changes implications with a fair degree of confidence today. underway now and all but impossible to forestall China’s population prospects over the decades over the generation ahead constitute “wild cards” for ahead are largely shaped by its prolonged episode of China’s future: These factors will have unpredict- sub-replacement childbearing, underway for a genera- able and potentially powerful disruptive effects. They tion at this writing and likely to have been in effect for include (1) the impending “marriage squeeze” due a half century by 2040. Thanks to continuing low lev- to abnormal sex ratios at birth from the One-Child- els of fertility, China’s population is on track to peak Policy era, which may severely prejudice the odds of in the coming decade and to decline at an accelerating family formation for the rising cohort of would-be tempo thereafter. Apart from radically redrawing the grooms in China; (2) the “Soweto with Chinese char- country’s boundaries, Beijing can do nothing to fore- acteristics” problem of mass urbanization under a stall this decline. hukou system that consigns hundreds of millions of China’s impending depopulation will be accom- migrants in urban areas to an officially inferior sta- panied by a momentous transformation of national tus; and (3) the revolutionary changes in the Chinese demographic structure. Between 2015 and 2040, family structure, which portend a dramatic departure China’s population age 50 and older is on course to from previous arrangements on which Chinese soci- increase by roughly one-quarter of a billion people; ety and economy depended. While it is admittedly dif- the under-50 population is set to decline by a roughly ficult to predict how such changes may affect social comparable magnitude. This means China is set to cohesion, political stability, and national security, it experience an extraordinarily rapid surge of popu- would be highly unwise to presume they will have no lation aging, with especially explosive population impact on these matters. 1 China’s Demographic Outlook to 2040 and Its Implications AN OVERVIEW Nicholas Eberstadt or any serious attempt to assess China’s future China’s Current and Future Population: Foutlook, an examination of the country’s popu- What We Know and How We Know It lation prospects is not only advisable but absolutely indispensable. There are two reasons for this. Before presenting the demographic projections First, of all areas of inquiry about China’s future underpinning this report, we are obliged to address that might be of interest in academic, business, two basic questions about China’s demographic out- and policy circles, China’s demographic future is look: What do we know and how do we know it? perhaps the least uncertain over the coming gen- Answering these questions requires us to discuss eration. The reason, quite simply, is that the over- data limitations today and the intrinsic limitations of whelming majority of the people who will be living demographic projections for tomorrow. in China in (say) the year 2040 are already alive, liv- Consider first the limits of current Chinese pop- ing there today. Population projections are far from ulation data. Vastly more population information is error-free, but if we are trying to peer ahead a cou- available for China today than was for most of the ple of decades, they are most assuredly more reliable Maoist era, when a virtual statistical blackout pre- (and empirically grounded) than corresponding pro- vailed. China today also has trained and groomed a jections of economic change, much less political or large cadre of top-rate demographers and population technological change.1 economists who work in the nation’s universities, Second, demographics and demographic change state-sponsored think tanks, and government. On the actually matter—to economic performance and other hand, China has not yet achieved complete or social development and in some measure arguably near-complete vital registration, meaning that ana- to such things as military potential, political stabil- lysts must rely mainly on reconstructions of trends ity, and international security. This is not to invoke from censuses and “mini-censuses”2—and these the “demography is destiny” claim, often attributed counts are far from error-free. to the 19th-century French polymath Auguste Comte. Many errors in China’s population data are essen- A less florid, more immediately defensible reformu- tially politically induced; the data are deformed by lation of that aphorism would be that “demographics mass misreporting due to ordinary people’s attempts slowly but unforgivingly alter the realm of the possi- to avoid the harsh consequences of Beijing’s various ble.” In the following pages, I try to show just how population control policies (using that term broadly). the realm of the possible is being reshaped in China With regard to Chinese household registration data by impending demographic changes over the decades (which are derived from a separate demographic sys- immediately ahead. tem run by the Ministry of Public Security), the 2010 3 CHINA’S DEMOGRAPHIC OUTLOOK TO 2040 AND ITS IMPLICATIONS NICHOLAS EBERSTADT census indicated that at least 13 million Chinese cit- assumptions are crucial. China clearly undercounts izens lacked legal identity papers because they were births—with a reported total fertility rate (TFR) of born “out of hukou”—that is, outside the locality that 1.18 births per woman per lifetime in the 2010 cen- the state mandates to be their residence3 (more on the sus—but no one knows by exactly how much. The hukou system shortly). But that guesstimate is based consensus, for better or worse, is that the actual rate on official assumptions about China’s true population in recent years has been around 1.6, or about 30 per- totals, and China’s vital statistics, census returns, and cent below the level required for long-term popu- sample population surveys have undercounted the lation stability in the absence of in-migration, but nation’s actual numbers for decades due to Beijing’s consensuses are not always correct. heinous One-Child Policy and the familial incentives it established for birth concealment. The United Nations Population Division (UNPD) China 2040: The Population Projections currently suggests that the 2010 China census esti- mate missed the mark by about 30 million, even after Like sausages and law, the making of demographic its own internal undercount adjustments, and that it projections may not look so pretty when seen up close. may have failed to enumerate well over a quarter of Nevertheless, population projections for China are all female children under 15 years of age.4 From the likely to be less problematic than for many other coun- 1982 China census onward, population totals and sex tries or regions of the world. China is a low-migration, ratios for given birth years from one census to the low-mortality, low-fertility society. This means there next have proved unstable for babies, children, and is, so to speak, relatively little “turnover” in the popu- youth. These errors due to politicization of demo- lation from one year to the next. According to UNPD graphic rhythms of life may lessen now that Beijing or US Census Bureau projections, almost four-fifths appears to be scrapping its anti-natal campaign, but of China’s projected 2040 population will be 22 or they are embedded in the data we use for projections older then—meaning they have already been born at to 2040. this writing. This brute fact far outweighs many of the As for demographic projections themselves, these smaller uncertainties highlighted above and in a sense are no more reliable than the baseline data they use rescues us from them. and the assumptions they input about future trends Nonetheless, we need to know the assumptions in fertility, mortality, and migration. International built into the China population projections. Consider migration is negligible for China in relation to its UNPD’s projections, which I will mainly use in this enormous population, and the assumption is this report. (The UNPD’s assumptions, by the way, are will continue to be true—lucky that, since demogra- fairly close to those of the US Census Bureau and for phers have no really defensible method for projecting that matter also the National Bureau of Statistics of international migration trends into the future. Demo- China.) For the 2015–40 period, the UNPD assumes graphic techniques for projecting survival trends for negligible net outmigration from China of 0.2 percent the currently living are fairly good—thanks to actu- per year—a rounding error, essentially. With respect arial mathematics, after all, the life insurance indus- to mortality, the UNPD estimates overall male plus try has not gone out of business—but catastrophes of female life expectancy at birth was a bit over 76 years biblical proportion do take place from time to time, in 2010/15 and that it would rise to 80 by 2040/45 and Providence has already visited a number of them (with detailed “life tables” offering survival proba- on postliberation China.