[ The end of world population growth Nature 412, 543 - 545 2aug01 below ]
The world's population will peak at 9 billion over the next 70 years before beginning a decline into the 22nd century, researchers predict in a new study.
The world population currently stands at 6.1 billion, and the study projects that most of the new growth will continue to occur in developing countries.
It also predicts some demographic changes. For example, the authors say, the number of people aged 60 or older will more than quadruple by 2100.
The study, published in the latest issue of the journal Nature, was greeted with skepticism in many quarters. Some researchers argue the predictions are misleading because of unforeseeable changes in everything from air quality to food supply.
However, the study's authors at the International Institute for Applied Systems in Laxenburg, Austria, say they have developed a statistical computer model that considers uncertainties in migration, mortality and birth rates.
The result: There is an 85 percent chance the world's population will stop growing by the next century.
"Everybody thinks quite correctly, 'You can't predict the future,"' said Warren Sanderson, co-author of the study and a professor at the State University of New York at Stony Brook. "We have much more confidence in predicting a range."
The computer model predicted that in 2070, the world population would peak at 9 billion people. By 2100, the population would dip to 8.4 billion. Both represent the midpoint of the projected population range.
The probabilities were determined from historical demographic data. The researchers also asked outside experts to predict key outcomes, such as the number of children born to each woman.
The researchers also expect the world's population to become older. The study predicts the portion of the population age 60 or older to increase from the current 10 percent to about 22 percent by 2050. It's expected to increase to about 34 percent in the next 100 years.
Nico Keilman of the University of Oslo, Norway, praised the study's forecast. But he warned that using information from historical predictions may be inaccurate and that overconfident experts could give very narrow prediction ranges.
United Nations officials stuck by their 1998 assessment, in which they predicted an increase of 3 billion over the next 50 years to 9 billion. It would be the second largest population jump in history. From 1950 to 2000, world population increased by 3.5 billion.
"The first most rapid growth is over, but we're still growing," said Joseph Chamie, director of the United Nations' Population Division. "It's like a huge freighter. You can't say, 'Stop the boat!' and expect it to stop immediately. It has to come to a slow stop."
But Sanderson believes the United Nations' estimates are off because its calculations only allow fertility rates to fall to 2.1 children per woman. Sanderson said fertility rates are already below that level in many countries. Since those rates typically do not recover, population growth should decline, he said.
At least one population researcher says the study falls short.
David Pimentel, of Cornell University, said there are so many young people in the world today that the population will increase for the next 70 years, even if world fertility drops to 2 children per woman.
Pimentel predicts the world population will hit 12 billion before declining. He attributes the slowing to high mortality rates from shortages of food, fuel and other resources.
WOLFGANG LUTZ*, WARREN SANDERSON*† & SERGEI SCHERBOV‡
* International Institute for Applied Systems Analysis, Schlossplatz 1,
A-2361 Laxenburg, Austria
† Departments of Economics and History, State University of New York at Stony Brook, New York 11794-4384, USA
‡ University of Groningen, PO Box 800, NL-9700 AV Groningen, The Netherlands
Correspondence and requests for materials should be addressed to W.L. (e-mail: email@example.com ).
There has been enormous concern about the consequences of human population growth for the environment and for social and economic development. But this growth is likely to come to an end in the foreseeable future. Improving on earlier methods of probabilistic forecasting1, here we show that there is around an 85 per cent chance that the world's population will stop growing before the end of the century. There is a 60 per cent probability that the world's population will not exceed 10 billion people before 2100, and around a 15 per cent probability that the world's population at the end of the century will be lower than it is today. For different regions, the date and size of the peak population will vary considerably.
Figure 1 shows the probability that the world population size would reach a peak at or before any given year. It indicates that there is around a 20 per cent chance that the peak population would be reached by 2050, around a 55 per cent chance that it would be reached by 2075, and around an 85 per cent chance that it would be reached by the end of the century.
|Figure 1 Forecasted probability that population will start to decline at or before the indicated date.|
There is around a 75 per cent chance that the peak population of the European portion of the former USSR has already been reached in 2000, an 88 per cent probability that it will be reached by 2025, and over a 95 per cent chance by the end of the century. For the China region, the probability of reaching a peak within the next two decades is still low owing to its relatively young age structure. By 2040 the probability becomes greater than half. In sub-Saharan Africa, despite the prevalence of HIV, there is a low probability of peaking before the middle of the century. The probability reaches 25 per cent by 2070, 50 per cent by 2085, and almost 75 per cent by 2100, owing to assumed reductions in fertility.
Figure 2 shows the distribution of simulated world population sizes over time. The median value of our projections reaches a peak around 2070 at 9.0 billion people and then slowly decreases. In 2100, the median value of our projections is 8.4 billion people with the 80 per cent prediction interval bounded by 5.6 and 12.1 billion. The medium scenario of the most recent United Nations long-range projection2 is inserted in Fig. 2 as a white line. It is almost identical to our median until the middle of the century, but is higher thereafter owing to the United Nations assumption of universal replacement-level fertility, that is two surviving children per woman.
|Figure 2 Forecasted distributions of world population sizes (fractiles). For comparison, the United Nations medium scenario (white line), and 95 per cent interval as given by the NRC11 on the basis of an ex post error analysis (vertical line in 2050) are also given.|
Table 1 shows the median population sizes and associated 80 per cent prediction intervals for the world and its 13 regions, indicating major regional differences in the paths of population growth. While over the next two decades the medians are already declining in eastern Europe and the European portion of the former Soviet Union, the populations of north Africa and sub-Saharan Africa are likely to double, even when we take into account the uncertainty about future HIV trends.
The China region and the South Asia region, which have approximately the same population size in 2000, are likely to follow very different trends. Owing to an earlier fertility decline, the China region is likely to have around 700 million fewer people than the South Asia region by the middle of the century. This absolute difference in population size is likely to be maintained over the entire second half of the century and illustrates the strong impact of the timing of fertility decline on eventual population size3.
Our findings concerning the timing of the end of world population growth are robust to plausible changes in parameter assumptions. A detailed sensitivity analysis is provided as Supplementary Information. The forecasts of the World Bank, the US Census Bureau, and the medium variant of the United Nations2, 4, 5 are based on independent assumptions; the median trajectory of our world forecasts is almost identical to these up until 2045. Of these three forecasts, only the UN long-range projections provide scenarios of the world's population to the end of the century. If we define the end of population growth slightly less literally, and take it to correspond with annual population growth of one-tenth of one per cent or less, the United Nations medium projection also shows the end of population growth during the second half of the century. Their medium scenario predicts that world population growth will first fall below one-tenth of one per cent at around 2075.
A stabilized or shrinking population will be a much older population. At the global level the proportion above age 60 is likely to increase from its current level of 10 per cent to around 22 per cent in 2050. This is higher than it is in western Europe today. By the end of the century it will increase to around 34 per cent, and extensive population ageing will occur in all world regions. The most extreme levels will be reached in the Pacific OECD (mostly Japan), where half of the population is likely to be age 60 and above by the end of the century, with the 80 per cent uncertainty interval reaching from 35 to 61 per cent. Even sub-Saharan Africa in 100 years is likely to be more aged than Europe today. The trend of our median proportion over age 60 is almost identical to that of the UN long-range projections2 up to 2050, but shows significantly stronger ageing thereafter. This confirms recent criticism that conventional projections tend to underestimate ageing6, 7. The extent of and regional differences in the speed of population ageing—the inevitable consequence of population stabilization and decline—will pose major social and economic challenges.
However, population numbers are only one aspect of human impact, and in some of the world's most vulnerable regions, significant population growth is still to be expected. Nevertheless, the prospect of an end to world population growth is welcome news for efforts towards sustainable development.
The method of probabilistic population projection that was applied here (see
The substantive assumptions about future trends in the three components of fertility, mortality and migration, and their associated uncertainty ranges are based on revisions and updates of our earlier work12 and the extensive analyses summarized in the recent US National Research Council (NRC) report11.
Fertility The key determinant of the timing of the peak in population size is the assumed speed of fertility decline in the parts of the world that still have higher fertility. On this issue there is a broad consensus that fertility transitions are likely to be completed in the next few decades11. For the eventual size of the population and the question of whether or not world population will begin a decline by the end of this century the key variable is the assumed level of post-transitional fertility. The thorough review of the literature on that subject by the NRC states that "fertility in countries that have not completed transition should eventually reach levels similar to those now observed in low fertility countries" (page 106 in ref. 11). Our fertility assumptions are consistent with this view.
The trends in the means of the regional fertility levels have been defined for the periods 2025–29 and 2080–84 with interpolations in between. The total fertility rates assumed for 2025–29 are similar to those chosen by the United Nations2, but for 2080–84 they are assumed to range between 1.5 and 2.0, with lower levels for regions with higher population density in 2030. The variances in the total fertility rates are assumed to depend on the level of fertility. If the total fertility rate is above 3.0 there is an 80 per cent chance that fertility would be within one child of the mean. When it is below 2.0, the same probability is attached to a range within one half a child of the mean. Between the two total fertility rate levels, the variance is interpolated.
Life expectancy We assume that life expectancy at birth will rise in all regions, except in sub-Saharan Africa, where HIV/AIDS will lower life expectancies during the early part of the century. In general, we assume that life expectancy increases by two years per decade with an 80 per cent probability that the increase is between zero and four years; but there are a number of exceptions to this rule based on specific regional conditions. These assumptions reflect the very large uncertainty that exists regarding future mortality conditions. On one hand, significant biomedical breakthroughs are likely to be made; on the other, AIDS could still become a major issue outside Africa, and new and unexpected threats to human life can emerge.
Autocorrelations Migration is treated as a random vector on the basis of recent interregional migration patterns. The autocorrelation chosen for all components is based on a 31-year moving-average process that seemed the most plausible after we had experimented with 21-, 31-, and 41-year moving averages (see Supplementary Information), and is close to existing national level figures10. We assumed an interregional correlation of 0.7 for fertility and 0.9 for mortality deviations with no correlation between fertility and mortality deviations from the assumed trend, and perfect correlation between male and female life expectancy. These choices followed extensive sensitivity analyses as documented in the Supplementary Information. The main rationale behind our choice is that under post-transition conditions, correlations between deviations from assumed fertility and mortality trends are unlikely to be large, while globalization of communication is likely to bring correlated fluctuations of rates among world regions. Mortality correlations will be higher than fertility correlations, owing to the faster communication of medical technology and the faster spread of new health hazards. The sensitivity analysis documented in the Supplementary Information shows that our main conclusion, that there is around an 85 per cent chance that a peak in world population size will occur in this century, is quite robust to plausible changes in those correlations.
Supplementary information is available on Nature's World-Wide Web site (http://www.nature.com) or as paper copy from the London editorial office of Nature.
Received 30 April 2001; accepted 22 May 2001
1. Lutz, W., Sanderson, W. & Scherbov, S. Doubling of world population unlikely. Nature 387, 803-805 (1997). | Article | PubMed | ISI |
2. United Nations. Long-Range World Population Projections: Based on the 1998 Revision (United Nations, New York, ESA/P/WP.153, 1999).
3. O'Neill, B. C., Scherbov, S. & Lutz, W. The long-term effect of the timing of fertility decline on population size. Pop. Dev. Rev. 25, 749-756 (1999). | ISI |
4. US Census Bureau. International Database 10 May 200 edition http://www.census.gov/ipc/www/worldpop.html (US Census Bureau, 2000).
5. World Bank. World Development Indicators (CD-ROM) (World Bank, Washington DC, 2000).
6. Tuljapurkar, S., Li, N. & Boe, C. A universal pattern of mortality decline in the G7 countries. Nature 405, 789-792 (2000). | Article | PubMed | ISI |
7. Vaupel, J. W. & Lundström, H. in The Future Population of the World. What Can We Assume Today? (ed. Lutz, W.) rev. edn, 278-295 (Earthscan, London, 1996).
8. Lutz, W., Sanderson, W. & Scherbov, S. in The Future Population of the World: What Can We Assume Today? (ed. Lutz, W.) rev. edn, 397-385 (Earthscan, London, 1996).
9. Lutz, W., Sanderson, W. C. & Scherbov, S. in Frontiers of Population Forecasting (eds Lutz, W., Vaupel, J. W. & Ahlburg, D. A.; Pop. Dev. Rev. Suppl. 24, 1998) 139-155 (Population Council, New York, 1999).
10. Lee, R. D. in Frontiers of Population Forecasting (eds Lutz, W., Vaupel, J. W. & Ahlburg, D. A.; Pop. Dev. Rev. Suppl. 24, 1998) 156-190 (Population Council, New York, 1999).
11. Bongaarts, J. & Bulatao, R A. (eds) Beyond Six Billion. Forecasting the World's Population (National Academy Press, Washington DC, 2000).
12. Lutz, W. (ed.) The Future Population of the World. What Can We Assume Today? rev. edn (Earthscan, London,, 1996).
13. Lee, R. & Tuljapurkar, S. Stochaistic population projections for the United States: Beyond high, medium and low. J. Am. Stat. Assoc. 89, 1175-1189 (1994). | ISI |
14. Lee, R. D. & Carter, L. Modeling and forecasting the time series of U.S. mortality. J. Am. Stat. Assoc. 87, 659-671 (1992). | ISI |
15. Alho, J. M. Stochastic methods in population forecasting. Int. J. Forecast. 6, 521-530 (1990). | ISI |
16. Lutz, W. & Scherbov, S. An expert-based framework for probabilistic national population projections: The example of Austria. Eur. J. Pop. 14, 1-17 (1998). | ISI |
17. Alho, J. M. Scenarios, uncertainty and conditional forecasts of the world population. J. R. Stat. Soc. Ser. A 160, 71-85 (1997). | ISI |
18. Keilman, N. in Frontiers of Population Forecasting (eds Lutz, W., Vaupel, J. W. & Ahlburg, D. A.; Pop. Dev. Rev. Suppl. 24, 1998) 15-41 (Population Council, New York, 1999).
19. Stoto, M. The accuracy of population projections. J. Am. Stat. Assoc. 78, 13-20 (1983). | ISI |
If you have come to this page from an outside location click here to get back to mindfully.org