Should We Worry about Overpopulation?

This posting is stimulated by comments made about my passing reference to overpopulation in Subsaharan Africa in the recent blog on DDT and by an article in the Wall Street Journal last week called "The Coming Crunch" that notes with concern a prediction that the population of the United States will reach 400 million in 35 years.

Concerns about overpopulation are ridiculed by conservatives because of the mistaken predictions by Paul Ehrlich (not to mention Thomas Malthus!) in his book* The Population Bomb* and by other anticapitalists since the first Earth Day (1970), and have spread to liberals because the only way to slow or stop the growth of the U.S. population is by curtailing immigration (e.g., the "fence"). Although I have been strongly critical of the shoddy arguments of Ehrlich and other doomsters (in my book Public Intellectuals), I believe that overpopulation is a serious issue and deserves dispassionate analysis. Just because the problem of overpopulation has been exaggerated in the past doesn't mean it is not a problem today. The future may not resemble the past. The belief that the mistakes of Malthus, Ehrlich, and other past prophets of doom show that current concerns with overpopulation are unfounded is on a par with the belief that we shouldn't worry about terrorism because many fewer Americans have been killed by terrorists than in automobile accidents. Such arguments confuse frequencies (the past) with probabilities (the future).

Economists stress the "demographic transition," that is, the tendency of the birth rate to decline steeply as a nation becomes wealthier. But apart from the fact that not all nations experience significant economic growth, such growth tends, other than in Europe and Japan, not to make the rate of population growth zero or negative. Most demographers forecast that world population, currently somewhat more than 6 billion, will rise to between 9 and 14 billion by mid-century.

I shall address the following questions: what are the costs of population increase (1) to the country in which the increase occurs, (2) if that country is the United States; and (3) to other countries; (4) what are the benefits to the country in which the increase occurs and (5) to other countries; and (6), when the costs exceed the benefits, what if anything should be done to slow or arrest population growth?

1. If the arable and otherwise inhabitable parts of a poor country are densely populated, increased population will result in significantly higher costs of food and other agricultural products by requiring more intensive cultivation, or cultivation of poor soil. It will also increase the cost of water, and time spent in commuting and other transportation. This seems to be the situation in India and much of Africa. And notice that China, though it is en route to becoming a wealthy country, has not abandoned its "one child" policy. That policy is an inefficient method of limiting population growth, but is evidence that China does have a problem of overpopulation. Surely India does as well, though like China its economic output is growing rapidly.

2. The United States is not densely populated, but that is only when density is computed on a nationwide basis, i.e., if the total area of the country is divided by the population. Particular areas, mainly coastal (including the Great Lakes coasts), are densely populated, and further population increases in those areas would increase commuting times, which have lengthened in recent years, and in some of these areas (such as California and Arizona) would place strains on the water supply. In principle, however, these problems can be solved by pricing, including greater use of toll roads. Increased commutes impose environmental costs, but tolls could be based on those costs.

3. The greatest costs of further population increases are likely to be costs external to individual countries and therefore extremely difficult to control by taxation or other methods of pricing "bads," because most of the benefits of these measures would be reaped by other countries. These are environmental costs, mainly global warming and loss of biodiversity, about which I have written at length in my book Catastrophe: Risk and Response(Oxford University Press, 2004). Of course, population growth per se does not increase global warming, but the burning of forests and, most important, of fossil fuels does, and these activities are positively correlated with population. Not only is it now the scientific consensus that global warming is a serious problem, but its adverse effects are appearing sooner than expected; it is by no means certain that a technological fix will be devised and implemented before the effects of global warming become catastrophic.

4. Population growth in productive societies increases the society's total output and hence its geopolitical power. It also has a positive effect on innovation by increasing the size of markets. Innovation involves a high ratio of fixed to variable costs (it costs hundreds of millions of dollars to develop a new drug, yet once it is developed, the drug may be very cheap to produce), so the larger the market for the innovative product or process the likelier are the fixed costs of invention to be recouped in sales revenues.

Some people also believe that the larger the population, the more innovators there will be, assuming that a fixed percentage of the population consists of innovators, whatever the size of the population. This is a questionable argument for population growth, as it ignores the fact that a fixed percentage of the population presumably also consists of potential Hitlers and Stalins and Pol Pots, and thus the absolute number of these monsters grows with population growth. Moreover, a population increase that is due to a higher birth rate (as distinct from immigration) increases the number of young people in a society, who are impressionable and therefore more likely than older people to be drawn to extremist politics, including terrorism. In addition, greater competition among innovators may reduce the potential returns to each innovator by increasing the number of simultaneous innovations, and may thus reduce the incentives to innovate.

The relationship between aggregate population and creativity seems in any event very loose. The citizen population of Athens in the fifth and fourth centuries B.C. was roughly 25,000, but produced intellectual and artistic works that dwarf those of entire continents. Furthermore, technological growth currently favors destructive over beneficial technologies. The increasing lethality and availability of weapons of mass destruction--the proliferation problem--has a greater short-term downside than benign inventions have an upside, especially since much innovative activity is focused on increasing longevity, and thus population. Policies that accelerate the rate of technological advance are dangerous unless the advance can somehow be channeled into productive forms. It cannot be.

A dubious benefit of population growth is that it lowers the average age of the population and therefore the burden of the elderly. That is a Ponzi scheme rationale for encouraging growth of population, since as soon as the growth ceases, the average age will shoot up--especially if it is correct that population growth increases the rate of medical innovation and thus the life span!

5. An increase in one nation's power reduces the power of other nations; so there is again a negative externality. The increase in the world's Muslim population is a negative externality for non-Muslim nations, especially the European nations, with their shrinking or about-to-start-shrinking populations. But by the same token an increase in the non-Muslim population of Europe would probably be a boon for the European nations. And an increase in the rate of innovation in one nation will benefit other nations unless intellectual-property laws are extremely strict (which would have its own negative economic effects).

6. If, apart from poor countries, the major costs of population growth are external to the particular nations in which population is growing, there is very little that can be done, given the weakness of international institutions, which is due in turn to the number and diversity of nations that have to be coordinated for effective action against global problems. Moreover, limits on immigration do not reduce global population growth and thus do not respond to the global-warming problem. Rich countries, however, can aid poor countries to reduce their rate of population increase by encouraging family planning and, in particular, female education, since educated women have higher opportunity costs of fertility, and hence fewer children, than uneducated ones. Where, as in the United States, the costs of population increase are concentrated in particular areas (whether in geographical areas or along highways), the costs can be neutralized by increasing prices proportionally tied to density by taxation or other methods of pricing negative externalities.

I want to reply to some of the comments on both my last posting, which was on the NAS report on women in science, and also the previous one, on DDT.

Women in Science. I notice that the comments in defense of the NAS report tend to be--defensive; and also emotional. One comment suggests that if a committee 17/18 female is likely to be biased, any male who comments on the report is likely to be biased too. But I did not suggest that the committee should have been composed primarily of men, only that it should have been more balanced, and that the fact that the only man on the committee could not, because of his position, dissent from the report, made his inclusion, as the lone man on the committee, entirely unprofessional. Another commenter vigorously denies that there is any difference between men and women, then states that he prefers female doctors because they are more caring!

A number of comments point to the range of differences between men and women, encompassing behaviors (crime, sports), preferences, test results, psychology, and much else besides, including the tendency of women in science to prefer the less mathematical fields (I gave the example of primatology). These differences could I suppose all be the product of discrimination, but that seems highly unlikely.

One comment states that the underrepresentation of women in science may be a result of path dependency (where you start may determine where you end up)--the fewness of women in science in past times. This is not persuasive, because there were virtually no women in academic law when I was a law student in the 1950s, but now about half of all law professors are women.

One last point: a good test for whether there is discrimination against or in favor of a group is its average performance in the profession alleged to be a site of discrimination relative to that of the majority. If women were discriminated against in science, one would expect the average woman in science to outperform the average man in publications, awards, etc., simply because only women who were better than men could overleap the discrimination hurdle. But if there is discrimination in favor of women in science, then the average man should outperform the average woman, because then it is the men who have to overcome the discrimination barrier. (If there is no difference in average performance of men and women in a given field, the inference is that there is no sex discrimination in that field--employers and other performance evaluators regard sex as irrelevant.) Since men outperform women in science rather than vice versa, the inference is that there is discrimination in favor of women.

DDT and Overpopulation I repeat my abject apology for calling DDT a herbicide rather than a pesticide. Some comments suggest that the mistake reveals my complete incompetence to discuss environmental issues. That seems a bit harsh. The reason for the mistake was simply that herbicides play a particularly important role in diminution of genetic diversity--thanks in part to the ban on DDT--so I was thinking about herbicides when I was considering the effects of DDT.

Some comments point out correctly that interior spraying won't eliminate mosquitoes and therefore malaria; and that is true. But complete eradication may not be cost justified. Costs and benefits must be compared at the margin. If 99 percent of deaths from malaria can be eliminated by interior spraying, it may not be worthwhile to spend billions of dollars developing and producing a vaccine. That is why I find the Gates Foundation's campaign to eradicate malaria puzzling. (Actually, I don't think it's very puzzling. There is often a strong political and public-relations dimension to foundation giving, even foundation giving for activities thought nonpolitical, such as saving lives. Somehow giving money to spray the interior of houses with DDT lacks pizzazz and could even be thought politically incorrect.)

Most of the comments fasten on the following paragraph in my posting: "Not that eliminating childhood deaths from malaria (I have seen an estimate that 80 percent of malaria deaths are of children) would be a completely unalloyed boon for Africa, which suffers from overpopulation. But on balance the case for eradicating malaria in Africa, as for eradicating AIDS (an even bigger killer) in Africa, is compelling. Malaria is a chronic, debilitating disease afflicting many more people than die of it, and the consequence is a significant reduction in economic productivity." Many commenters regard "unalloyed boon" as a particularly callous chardacterization. I think some of the commenters don't understand the meaning of the word "unalloyed." I did not say it was a good thing that children die of malaria; I just said that it was not just a good thing, if the deaths reduce population. Now, they may not, as one comment explains, because a family that loses a child to malaria may decide to have another child in its place, and indeed if the family is risk averse it may end up having more children because of the high risk of losing one or more of them to malaria than if there were no such risk. That is an interesting empirical question. I suspect that on balance there will be fewer children surviving to adulthood, simply because of the cost of additional children.

I continue to insist that overpopulation, including in subsaharan Africa, is a real problem. It is true but absurdly irrelevant that New York City has a greater population density than Africa. Overpopulation is not a simple matter of dividing people by square miles. In an agricultural society, population density tends to be negatively correlated with wealth, simply because the land must be worked harder to obtain food. Good land is not the only resource that is in limited supply--so is fresh water, forest products, game, and mineral resources. Scarcities in these resources can be overcome, but only at a cost. It is true as several comments point out that as a society grows wealthier, the birthrate tends to drop (the "demographic transition"), but Africa seems to be trapped by extreme poverty exacerbated by overpopulation.

Is it foolish for China to try to limit its population? If not, the case for limiting the African population is much stronger, because Africa has a far less productive population.

And so far I have been speaking only of the effects of population on the populous country. There are external effects as well. The effects of population on the destruction of forests and on the demand for electricity and cars are major contributors to global warming.