NIH-funded study forecasts growth of 2 billion over previous estimates
Listen to this podcast (MP3 - 6 MB).
Barrett Whitener: World population is likely to keep growing throughout the 21st century, reaching 11 billion by 2100, according to results of a new study. That's about 2 billion higher than previous estimates. The revised forecast uses statistical methods to combine government data and expert forecasts for trends such as mortality rates, fertility rates, and international migration.
From the National Institutes of Health, I'm Barrett Whitener. This is Research Developments, a podcast from the NIH's Eunice Kennedy Shriver National Institute of Child Health and Human Development, the NICHD. With me today is study author Adrian Raftery, a professor of statistics and of sociology at the University of Washington in Seattle.
For about the past 20 years, experts had estimated that world population, which is currently around 7 billion people, would go up to 9 billion and then level off or decline. A rising population could intensify world problems, such as climate change, infectious disease, and poverty, so a reliable estimate of population growth can be important for planning experts in many fields.
The study was led by the University of Washington and the United Nations, and was funded in part by the NICHD. Thank you for joining us today, Dr. Raftery.
Adrian Raftery: My pleasure.
Mr. Whitener: Let's start with the methods you used to reach the higher estimate for a world population growth. What sets them apart from the methods used in previous studies?
Dr. Raftery: Well, the general traditional approach to population projection and forecasting relies on looking at the main components of population change, which are fertility, mortality, and migration, and making expert assumptions about how these will evolve in the future. What we did instead was to look at statistical methods and statistical models for two of these components, fertility rates and life expectancy, and take advantage of accumulated data on all these quantities for all the countries of the world, build a model, and use that to project these quantities into the future. And that gave results that were somewhat different from what previous assumption driven methods had yielded. And as you say, the results are somewhat higher than previously had been found.
Mr. Whitener: One example that you give in the paper is that earlier reports had used scenarios in which women would have 0.5 children more or less than the experts' forecast. And can you explain why that wasn't necessarily the way to get the best findings and how you varied from that approach?
Dr. Raftery: Yes. So the previous approaches, because they weren't statistical, nevertheless wanted to get some idea of uncertainty, and a standard approach to doing this was what they call scenarios, so you imagine variations from the assumptions that are made. And the way the United Nations has been doing that, as you say, is to add or subtract half a child from the fertility rate for all countries in the future, to get what they call high and low scenarios.
And that makes some sense. If you take the United States, for example, we have a total fertility rate of close to 2, a little bit below 2 at the moment. And one could think of adding half a child to that, approximately 2.5, and it's feasible, it's plausible to think that at some point in the future, total fertility rate might indeed go up to 2.5. But what the high scenario, the high—the traditional UN high scenario assumes is this fertility rate will be half a child higher at all time points and in all countries, at the same time. And you can see that's quite unlikely that it would be all time periods into the future. And similarly for the low scenario, which subtracts half a child.
So what the UN scenarios have given are very widely diverging scenarios for total world population, which generally are thought of as not being very plausible and not giving a very good idea of uncertainties. So they haven't been—they haven't featured much in discussions about world population. On the other hand, our statistical model takes account of the possibility that, for example, fertility might be higher in one country and lower in another at the same time, and it gives much narrower bounds.
So the world population going to the end of century, the UN high and low scenarios go from about 7 billion to almost 17 billion, so plus or minus 5 billion, which is just a huge range and nobody takes it very seriously. Whereas our statistical method says the population will probably be between 9 and—well, 9.6 and 12.3 billion, which is a much narrower range. There's quite a lot of uncertainty there still, but it's also narrow enough to be useful in policy and other discussions.
Mr. Whitener: I mentioned at the beginning a few of your high-level findings, and you just mentioned the number between 9.6 and 12.3 billion. Could you share with us a little more of the details about what you found?
Dr. Raftery: So our medium projection, or median projection, which is kind of our middle projection, is that world population at the end of the century would reach almost 11 billion, actually 10.9 billion. But there's uncertainty about that, and as you say, it ranges between 9.6 and 12.3, so just between 9 and a half and a bit over 12 billion. And that means that there's an 80 percent probability, 8 chances in 10, it would be between these two bounds by the end of the century.
An important aspect is what this distribution—how this is distributed across different areas of the world. And the big difference from the previous projection, which is 2 billion less generally, is what's projected to happen in Africa. So the population of Africa is now about 1 billion, and our projection is it would reach about 4 billion, so it would quadruple over the—to the end of the century. And there's uncertainty on that. Our range is about from 3 and a half to 5 billion, but what's almost certain is that there will be a very substantial increase in the African population.
The other continents are, in the case of Asia, currently about 4 and a half billion, projected to go up a bit to the middle of the century and then to decline again, and not much change there. And the Americas and Europe, not major changes predicted. So the main change over the remainder of the century will be in Africa, and that's also why the projection is different from previous ones.
Mr. Whitener: And this has implications I imagine for the ratio of people of working age to older people as well?
Dr. Raftery: Yes. So one of our findings, one of the things we looked at, and that the probabilistic population projections we're doing allow us to study, is the aging of the population, particularly the ratio of older people to people of working age. And a very general trend in that, which is a worldwide trend, is that that's going to go down. So for example, the United States currently has about 4 and a half people of working age for every senior. We're looking at—the age groups are age 20 to 64 and over 65, which they're used as rough markers of workers and retirees.
So the United States currently has about 4 and a half workers per retiree, but that's projected to go down to about 2 by the end of the century. And currently the oldest country in the world is Japan, which has about 2 and a half workers per retiree. And similar kind of patterns in Europe, although Europe probably—some European countries like Germany, the trends will be even more drastic where the number of workers per retiree is lower than the United States, it's about three. That's well known and is the object of a lot of policy discussion, both in North America and in Europe, in fact across the developed world.
What is not so well known is that countries that currently have young populations, but have seen recent declines in their fertility rates, are also likely to see a very similar pattern. So an example is Brazil, for example, where currently there are about eight workers per retiree. So it's a very young population, and people think about the issues of young populations, like, for example, youth unemployment. However, Brazil also is likely to see a massive decline in its number of workers per retiree, and that's likely to go down to below two by the end of the century. So Brazil, for example, will probably face the same problems of the aging population that the developed countries do, and within not too many decades. The same is true, for example, of India and of Egypt. The main area of the world where these problems are not likely to be faced this century is sub-Saharan Africa.
Mr. Whitener: You mention in the paper that population has kind of fallen off of the world's agenda. And could you go into a little bit about that and why you think it remains a very important issue?
Dr. Raftery: Well, up to recently, and probably over the last 10 or 20 years, it has been felt that population growth was largely a solved problem, or a problem that would shortly be solved. And that's probably because population projections were that world population growth would stop pretty soon, and also because fertility had come down massively in many countries of Asia and Latin America.
What we're seeing doesn't contradict that, but it seems that in Africa, particularly sub-Saharan Africa, fertility is high. It averages around five children per woman. And although it's coming down, as had been projected, it's coming down more slowly than had been projected. And in fact in some countries, the fertility decline has stalled completely. And when you project forward these trends into the future, you see a lot more people and a rapid population growth.
And it seems plausible to think that this rapid population growth could generate problems, could certainly exacerbate problems, for the countries that are likely to experience such as these can be health problems, environmental problems, and social problems. And while there's no agreement on exactly what these would be, there seems to be a fairly widespread agreement that countries likely to experience rapid population growth would do better if that population growth was moderated, and particularly if the fertility rate came down. And doing that is an important priority because of these potential problems. That's something which is I think worthy of renewed and sustained attention.
Mr. Whitener: Stepping outside the paper for a moment, what steps do you think could be taken to reduce the rate of world population growth and lessen some of the impacts you just described?
Dr. Raftery: Well, probably the last 50 or 60 years of demographic research has given us answers to those questions, and fairly unequivocal answers. There seem to be two main policies that are likely to reduce fertility in high fertility countries. One is making contraception more generally available, so developing family planning programs. And the other is increasing girls' education. And in many of the high fertility countries, particularly in sub-Saharan Africa, while the rate of access to education for girls or for everybody has been increasing, there's still a high proportion of girls who, both in the primary school level and also the secondary school level, who are not in education. So it seems that these would be the two kinds of policies to focus on. And we know pretty well that these policies do work in helping to accelerate a fertility decline.
Mr. Whitener: Are there any final thoughts on the paper you'd like to share?
Dr. Raftery: Well, just to reiterate that I think the population issues did fall off the world's agenda over the past 15 or 20 years because people felt it was a solved problem. And it now seems that that was somewhat premature. For many areas of the world it's true, but particularly for Africa, it remains an issue of importance, an issue where—and it's an issue for the world, where the world should give attention to it and support families and governments in high fertility countries in helping to moderate population growth. I think we know how to do it, and it's a matter of reorienting attention to these issues.
Mr. Whitener: I've been speaking with Dr. Adrian Raftery, coauthor of the study, "World Population Stabilization Unlikely This Century," which was published in the journal Science. Thanks very much for taking the time to talk with me today, Doctor.
Dr. Raftery: Thank you for having me.
About the Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD): The NICHD sponsors research on development, before and after birth; maternal, child, and family health; reproductive biology and population issues; and medical rehabilitation. For more information, visit the Institute's website at http://www.nichd.nih.gov/.