Human Population and its Limits

I was recently (1998 January) asked, "WHY WOULD WE WANT ANOTHER 5 BILLION PEOPLE?". Sorry, the email was all caps.

The question seems to assume that "we" can decide how large the population should be and then figure out how to implement the decision. At present there is no mechanism for making a decision about world population or for implementing any such decision. We will return to the question of decision and implementation, but first we consider population trends. (The main Sustainability article presents evidence that 15 billion can be accomodated at an American standard of living.)

Present Population Trends

Here are the 1999 UN Population projections. The 1999 medium fertility scenario predicts 9.5 billion in 2050, peaking at 11 billion in 2200. You can still get scary figures by projecting 1995 fertility rates, ignoring the declines in fertility.

If the UN projections are based on average birth rates for each country they may underestimate future population growth by not taking into account the fact that a country's population growth will come to be dominated by the growth rates of its fastest growing subpopulations. For example, the US population might come to be dominated by Mormons, Mennonites, and ultra-orthodox Jews all of which have high birth rates. A simple extrapolation won't work, because those groups also have high rates of defection.

There have been large declines in fertility in many Third World countries. In June 1999 it was reported that the average Mexican woman had 7 children in the 1970 but has 3 children today. Bangladesh and India have also had large declines in fertility. The world's most populous country is China. Its rate of increase has declined recently. You can see a graph of China's population and GNP.

The earth and accessible and usable planets place a limit on the possible human population. Population may eventually approach this limit, thus necessitating actions by sovereign countries or other entities to limit their population. We are very far from such limits now as my Sustainability essays show, probably hundreds of years. What form these limits may take we don't know, but they will surely not occur before humanity has a much higher level of science and technology than today. Therefore, I think except in a few countries, nothing should be done to restrict people's desire to have children. Indeed the below replacement birth rates of important countries suggest the reverse.

Some people draw inferences about increase in population from the increased number of people visiting national parks and other tourist attractions. From 1955 to 1995, U.S. population increased by 60 percent where as visitors to Grand Canyon increased 600 percent. Visitors to Yosemite Valley also increased by a factor of about four. Thus the crowdedness is mainly caused by increased equality of income and tastes in the population.

Anyway the problem of access to attractions is a problem of equality of opportunity and not a problem of population.

Ultimate Limits to Human Population

Going to the other extreme, if we imagine humans living on earth and refusing any new technology, we get much smaller limits, and maybe these limits decrease with time as things get used up. If you reject some important existing technologies or insist on allocating most of the planet to non-human life you get still smaller limits. We won't discuss these possibilities either.

Neither of the above scenarios has much to do with present decisions - either personal or governmental. We assume continued use and improvement of technology based on present science. This is conservative, since science will surely advance and offer more technological opportunities.

Here are some relevant contentions.

Population Problems are Local to Countries and Regions.

The basic point is that China probably has a population problem at present, and the U.S. certainly does not. China has a population of 1.2 billion and the U.S. about 270 million. The two countries have about the same area and are at about the same latitude.

If the U.S. were to merge politically with China with freedom of migration throughout the merged country, then one could speak of the possible population problem of the merged country. However, the U.S. hasn't permitted free immigration since about 1920 and is unlikely to do it again. Other countries won't either. Therefore, we must consider possible population problems in the U.S. and China separately.

The advanced country with the densest population is the Netherlands, and the densest U.S. state is New Jersey. I occasionally have occasion to ask people from these places why they moved to California. In no case have they volunteered that they found their place of origin too crowded.

2003 Nov 6: Today's New York Times has a long story on India's relatively new efforts to control its population. Some of the laws are rather coercive, though not as coercive as China's. None of these laws would have a chance in the US, because they are quite unnecessary. They stem from Indians deciding how to deal with India's problem.

While U.S. population is stabilizing, the same is not true of some regions within the U.S., especially California, Texas and Florida. These states remain extremely attractive, and I expect their populations to continue increasing, at the cost of decreases in other states. Within California, Silicon Valley (Santa Clara Valley) remains attractive in spite of the enormous increase in housing prices, which are exacerbated by zoning rules that often prevent going to higher buildings. We can expect continued complaints about increasing traffic, etc. in the more popular regions of the country. [My own reason for having moved back to California from Massachusetts in 1962 was substantially because I prefer mild winters, and this comparative advantage is likely to survive substantially higher population density.]

Overall population trends in a country do not take into account differential fertility of subgroups. White protestants are declining, and maybe white catholics also. Jews are declining except for the ultra-orthodox who are increasing. Mormons are increasing rapidly. Blacks and hispanics have higher fertility than whites. Along a different dimension the eugenics movement claimed that low IQ people increase while high IQ people decline. There are recent statistics on this, but the subject is substantially tabooed among social scientists.

If we project these differential fertility trends over the next century, then we should imagine the Mormons expanding East from Utah to meet the Hasidic Jews expanding West from Brooklyn somewhere in Indiana at the end of the present century. Theologians should work hard in the meantime to reconcile their respective religious beliefs.

There are several more issues.

1. Population control enthusiasts often say that the U.S. should set a good example. However, the U.S. will not bring itself to do what might be required by countries with severe immediate population problems. For example, the US is unlikely to adopt the present Chinese policy of penalizing more than one child per family. Therefore, the U.S. could at most set an example of half-hearted measures. Nor would the US go for the laws recently passed in India.

2. It is argued that an American puts more of a strain on scarce resources than a person in an undeveloped country. Except for oil, there is no vital resource likely to become scarce in the forseeable future. Substitutes for oil will eventually be required, and they exist, e.g. hydrogen produced using nuclear energy. See the main Sustainability FAQ for more on these points.

I have argued that Earth can support 15 billion, but it seems unlikely that this number will be reached, because the countries whose populations are growing rapidly will have to stop soon. Namely, they won't be allowed large scale emigration. The U.S. could support a billion or two, but our birth rate is below replacement levels, so we are unlikely to reach a billion. U.S. agriculture could probably expand to support more people via exports, but this depends on people in crowded countries making goods or performing services for which we will trade food. Teleservice may provide a way for greatly extended trade in services.

Birth control technology has made the number of a woman's children a matter of choice. This unprecedented situation in the human world has already had significant consequences. The birth rates of Europe and Japan are well below replacement. The US birthrate also dropped below replacement, but has picked up again to the level of 2.1 children per woman which is just the replacement level. This average is compounded out of different birth rates in different subpopulations of the US.

It is, or at least was, a commonplace to say that natural selection among humans ended long ago. On the contrary, if there are any genes that lead to wanting large families, they are being vigorously selected, and those which lead to wanting none or one children are being vigorously selected against. Whether there are such genes or whether it's all social is unknown. Perhaps the negative population growth of Europe and Japan may change when this selection has had its effects. Outside of science fiction I know of no discussion of the genetics of philoprogenitiveness.

Until recently a drive for sex sufficed to increase the population. Now that sex can be had without pregnancy, a desire for children will be naturally selected if there are relevant genes. Moreover, the advance of technology (not moral argument) has led to women no longer being considered the property of their families. Besides birth control, the increasing fraction of jobs not dominated by physical strength, and the mobility provided by cars and even bicycles, have given women more freedom. [Quotation: Susan B. Anthony said "the bicycle has done more for the emancipation of women than anything else in the world."] Therefore, genetic effects on women's desire for children (or not) will become more important. The replacement of clubs and swords by firearms has also been a force toward equality of the sexes, albeit a small one.

If and when the population of a country, i.e. a region with the sovereignty needed to pass laws, becomes dominated by people who want children, laws regulating the number of children people can have may be needed.

Here are references on population at IIASA (International Institute of Applied Systems Analysis.

Reversing a moralistic argument

People who have no children or only one are depending on my children (I have three) to support them in their old age, whereas I am contributing my share of people who will be working when I am older.

The population complainers can say that their not having children economizes on natural resources. I say their not having children puts an excessive labor burden on my children.

How can we compare these two complaints?

Economics provides an answer. Natural resources apart from energy, amount to 4 percent of the GDP of the U.S., and energy amounts to 8 percent, but these include labor and capital as well as the direct cost of the resources. In contrast to this, labor amounts to more than 50 percent of the GDP, and probably to 75 percent. Therefore, the burden the population complainers put on the next generation by not having children is much larger than the burden imposed by my children's use of resources.

The complainers say that the present low prices of natural resources are illusory and do not reflect real costs. Elsewhere in these pages (start at the beginning with The Sustainability of Progress), I show that overall shortages of resources are very unlikely, and the present low prices are not illusory, although there may be a moderate increase in energy costs during a transition from cheap oil to full use of nuclear energy.

At least for the next ten generations, Americans who have several children contribute to society, whereas those who don't have children harm society.

The population complainers often produce fierce remedies for what they presume the problem to be. Here's my fierce remedy.

Anyone without children to contribute to the labor pool should agree to be terminated ten years after he quits working. I have to tell some people not to take this literally as a serious proposal.

My actual opinion is that rich nations will solve the problem of taking care of their elderly by admitting many more immigrants than are admitted today. The immigrants will pay the American social security taxes in sufficient numbers to keep the system working.

See also birth dearth for more on that menace.

It was pointed out to me that the above does not take into account the possibility of robots and other forms of automation at the human level of intelligence. I think such robots will be developed, and then all of humanity will have the opportunity to join the idle rich. Not all people will take the opportunity to be idle any more than present rich or the past rich took it.

I don't know how long it will be before the required level of AI is achieved. My guess is that when the transition to having robots that can do all our work occurs, it will be rather sudden. We are not in development range of this level of AI yet, i.e. no-one can convincingly say, "Give me ten billion dollars, and no-one will ever have to work again."

Local Crowding

It occurs to me that psychologically akin to overpopulation is experiencing too many people in one's desired occupation. Many more people would like to be astronomers or movie actors or writers than the world is willing to support in those occupations. For some occupations, biographies tell us that it used to be a lot worse than it is today. Maybe people feeling crowded in their desired occupations tend to believe in overall overpopulation.

Here are some population references.

UN population estimates for countries, UN projections for regions and the world, UN information abou these trends, Population Reference Bureau, POPNET.

Send comments to John McCarthy, jmc@cs.stanford.edu.