Twenty years ago, there were two polarized positions with respect to the implications of population trends: one pessimistic and the other optimistic. The pessimists asserted that rapid population growth constituted a trap for the poorer countries: their best efforts to develop could serve only to maintain an ever larger population under unimproved or even deteriorating conditions. Those holding the optimistic counterposition denied any ominous implications of population growth, asserting that poverty was caused by remediable institutional defects, whether these be a highly unequal division of property, the capitalist system, or the unwarranted interference of the government in a free market. The pessimists offered the classical Malthusian argument that the supply of basic requirements such as food could not expand as rapidly as the population tended to grow; unless the birth rate was reduced immediately, efforts to achieve social and economic development were bound to be ineffectual. The optimists replied by contending that the world was technically capable of providing a much larger population with a much improved diet, and that the concern for excessive population growth was misdirected.

In a book we wrote then,1 Edgar M. Hoover and I accepted neither a priori position: we were not prepared to assume either that population growth, if continued, was bound to lead to an early disaster, nor that population growth made little difference to development. Our strategy was to examine as objectively as we could the prospect for economic development in specific countries in an effort to ascertain how much difference it would make if the future population within the country were to develop according to one or the other of two strongly contrasting possible paths. Specifically, we calculated the future of the population if the rate at which women bear children (up until then essentially unchanging in most less-developed countries) were on the one hand to continue for another generation or more without any change, or, alternatively, were to be reduced by 50 percent in a period of 25 years. The purpose of our two alternative projections was to show how the population would grow and how its age composition would change if the rate of childbearing were to remain constant, or if it were to decline, smoothly but strongly, over a generation.

Our next step was to calculate, on the basis of conventional assumptions about factors that influence the growth of the national product, how different would be the evolution of the economy, given the two different projected populations. We were not interested primarily in predicting precisely what would happen in either instance, but rather in showing what difference it would make to the economy and the society if for any reason fertility were to fall, as opposed to continuing to remain unchanged.

Stated in irreducible minimum terms, our economic argument was as follows: the alternative population projections would necessarily yield the same number of persons above age five at the end of five years, above age ten at the end of ten years, and so on. On the provisional assumption that mortality rates would be the same in the two populations, the difference between the population with unchanging rates of childbearing and the population with falling fertility would be confined to a difference in the number of persons born after the projections are initiated. There would be a widening gap in the number of births in the two populations; after 25 years the difference would be a little more than 50 percent of the larger number of births.

To the extent that effective participation in productive activity does not begin much before age 15, especially in a modernizing economy, there would be little difference in the size of the effective labor force in the two projections until after 25 or 30 years. After 20 years, for instance, the population aged 15-64 would be only four percent greater in the high-fertility projection, and only nine percent greater after 30 years. If allowance was then made for the greater availability of women for economic activity outside the home when the rate of childbearing is 40 or 50 percent lower, it seemed clear that there would be no large advantage for 25 or 30 years in the availability of productive labor in the population that had sustained high fertility, as compared to the population with reduced fertility.

At the same time, there was no reason to suppose that the natural resources available in a given country would be different over a generation according to whether fertility remained constant or was reduced. That is to say that the mineral deposits, forests, and agricultural land that are potentially exploitable are not a function of the size of the dependent population, especially if the size of population at ages of labor-force participation is essentially the same.

Two of the principal factors contributing to the national product (labor and resources) are, then, prospectively about the same for 25 or 30 years, whether fertility remains the same or is reduced. The third classical determinant of national product is capital. We argued that additions to capital (net investments) are at least potentially larger with reduced fertility. The reason is that additions to capital stock are taken from that part of national income that is not currently consumed, and as lower fertility reduces the number of children in the population, it would be easier (without reducing consumption per head for the adults or the other children) to divert part of current output from consumption to net additions to the capital stock.

Since two of the principal factors of production would be essentially the same in the two projected populations, and since the accumulation of capital would be larger in the population with reduced fertility, during 25 or 30 years the total national product would increase more in the population that followed the low-fertility alternative. The low-fertility population, in addition to the evident advantage of dividing the national product among a smaller number of consumers, would have the surprising advantage of a larger product to divide.

This scheme of analysis was applied first to India. In our study of India, the bare outline just stated was fleshed out with nearly 300 pages of detailed data and descriptive information about the Indian population, economy and society and by a formal model of economic growth. Our conclusion was that whatever increase in income per person could be achieved by a projected population with no decline in fertility in a 30-year period, that increase could be surpassed by 40 to 50 percent in a population that reduced its fertility by one-half in 25 years.2

After our study of India was finished, we wanted to examine the same set of relations in a developing country in another part of the world. We chose Mexico as the second example because it was less densely populated than India, had substantially higher per capita income, was experiencing much more rapid growth in real national product, and in general was less impoverished and further along the road to an advanced state of development. Our next step, then, was to repeat the same scheme of analysis, using projections of the Mexican population and data and coefficients from the Mexican economy. The surprising result was that the relative advantage foreseen for a population with reduced fertility relative to one that maintained its fertility unchanged was about the same in Mexico as in India.

This conclusion was a surprise when it first emerged from our analysis, but the reasons for similar conclusions became clear once we thought further about them. We were not making absolute judgments about the relative progress in two different economies; we were not saying that India would enjoy as rapid progress as Mexico, only that both countries would, over a generation, gain about the same relative advantage from a reduction in fertility. Moreover, within the intermediate time perspective of 25 or 30 years, when the size of the effective labor force is not much different in the two projections, we had no reason to emphasize shortages of land or other resources that might make difficult the effective use of a growing labor force. Limits of this sort are an inevitable eventual concern if population continues to grow indefinitely. India is probably closer to a problematic relation between the aggregate size of its population and the area and resources of the Indian subcontinent than is Mexico. But in a single generation, the relation of population to resources is not a relevant consideration in the comparison of two alternative projected populations.


During the more than 20 years that have passed since we analyzed the alternative prospects in India and Mexico, how has the public discussion about the relation of population to development changed? The debate about population and development persists in about the same terms as 20 years ago. The argument, common at the time that Hoover and I were working, between optimists and pessimists on the role of population in development has, if anything, sharpened. The pessimists are found among environmentalists, biological scientists, and supporters of the Planned Parenthood movement, groups that have been more vocal in the past 20 years than they were at the time we began our work. The pessimistic view has been expressed in books with alarmist titles such as The Population Bomb, and Famine 1975!.3 The latter book, in addition to forecasting the inevitability of famine by 1975, proposed that the less-developed countries in the world be put in categories sometimes used by medical personnel at battle stations when casualties are very severe. This doctrine (triage) divides the casualties into those so seriously wounded that the overtaxed medical facilities have little chance of saving them, those lightly enough wounded that their lives will not be endangered if they are not given immediate attention, and a third category of those at neither extreme on whom the limited immediate medical care is concentrated. The authors suggest by analogy that there are countries in which the population situation is so desperate that efforts to help are useless. I am sure that this very gloomy position is not only inhumane, but illogical as well.

The metaphor of triage is simply inapplicable to national populations. Battle casualties so far gone that medical care is useless are neglected because under the stated conditions death is inevitable. But populations do not die out because of their growth; growth is not like a severe wound. If any country falls in the too-late-to-save category of triage, it is Bangladesh. Yet with or without technical assistance, there will certainly be more people in Bangladesh at the end of the century than now. At the worst, poverty will have become more acute, and population growth will have slowed down because of increased mortality rather than reduced fertility. Unlike neglected battle casualties, neglected poor countries do not disappear. To advocate neglect is to contend that the poorer countries are condemned, not to death, but to eternal poverty - an extreme view indeed of man's lack of adaptability and ingenuity.

At the World Population Conference in Bucharest in 1974, the pessimistic and optimistic views were very conspicuous in a debate (among government representatives rather than professional demographers and statisticians) that was highly political. As an advisor to the U.S. delegation, I was very conscious of the influence on the U.S. position of the pessimistic (or immediate crisis) conception of population trends. In my opinion the extreme position taken by the American delegation helped prevent the adoption of a stronger World Plan of Action. The optimistic tradition was upheld at Bucharest by delegations from many countries in Africa and Latin America, with support from the Soviet bloc. These delegations took the position that it is wrong to ascribe any of the serious problems in the world to population trends. What is needed, they argued, is not population policy, but a new economic order, land reform, a more equitable distribution of income, and recognition of the rights of women. Two popular slogans were: "Take care of the people and the population will take care of itself," and "Development is the best contraceptive." I can see no objection to defining and advocating more effective economic and social policies, but cannot understand why it is improper, at the same time, to discuss explicit policies on population, especially at a world population conference.

Neither of the opposing positions expressed at Bucharest left room for the modest position that Hoover and I had offered to the effect that a low-income country with high fertility would achieve better progress if it could reduce its fertility than if it did not do so. If the country began with a very poor, unhealthy, and uneducated population and a stagnant economy, economic progress would be slow, but less so if fertility were reduced. If a low-income country already had a fairly well-educated population, and the organizational ability, the social order, and the natural resources to develop rapidly, even with sustained fertility, it would do still better (about 40 percent better per consumer over a 30-year period) if fertility were reduced.

Some economists have questioned the validity of our assumption that there would be more investment with lower fertility, and I must acknowledge that I am far from certain that our economic model was wholly realistic on this point. Even without substantially greater investment, however, the balance of forces would yield a slightly higher national income and substantially higher income per equivalent adult with reduced fertility.

The purpose of this article is not to review these old arguments themselves, but to see what has actually happened in one of the two countries we studied (Mexico): to reexamine the relationship between population growth and development after 20 years of both.


I will begin this reexamination by comparing the actual changes in the Mexican population since 1955 with our alternative population projections. The comparison is easy: the population of Mexico has followed the higher projection very closely.

The population projected with unchanged fertility was calculated to increase by a factor of 1.93 in 20 years. The actual increase was by a factor of 2.03. The population doubled in just under 20 years! Table 1 below shows the projected and recorded average birth and death rates during the four quinquennial intervals from 1955 to 1975, and a comparison of the number of persons under 15, 15 to 65, and over 65, in 1975, as projected 20 years earlier, and as currently estimated. The projection with unchanging fertility slightly underestimates the actual growth of the Mexican population. Our projected age distribution closely resembles the actual, but the actual has somewhat higher numbers at younger ages, apparently because the assumption of unchanged fertility was an underestimate of its actual course.4 Our projections of the death rate appear to have been basically correct: slightly too low in the 1960s but right in line with official data in the early 1970s.

I have spelled out in some detail a comparison of the projection with unchanged fertility and the actual population in part because among the large number of projections with which I have been involved, this is one of the few to be approximately verified by the subsequent actual population. However, the verification of a projection that was intended to be merely illustrative is not a source of much pleasure. Of the two alternative future populations that we worked out 21 years ago, it was the alternative with the less favorable consequences. If our reasoning was correct, the population in Mexico would in many respects be better off if the other projection (with declining fertility) had been correct. Before describing a number of the concrete ways in which reduced fertility might have improved the current social and economic situation in Mexico, I should like to spend a few minutes on the paradox implied by the continuation of high fertility itself. I have just noted that the actual trend of births in Mexico was higher than in a population projection (otherwise quite close to reality) in which it was assumed that fertility did not change from the level of the mid-1950s.

In Table 2 are shown selected changes in Mexico from the 1950s


A. Birth rates and death rates in Mexico 1955-1975. Projected and recorded.

1955-1959 1960-1964 1965-1969 1970-1974

Birth rate 43.3 42.1 41.6 41.3 Projected

(per thousand) 44.9 44.4 43.2 44.1 Recorded

Death rate 12.2 9.7 8.2 7.1 Projected

(per thousand) 12.2 11.0 9.5 7.2 Recorded

B. Mexican population, 1975, according to projections from 1955 and contemporary estimates (thousands).

Age Projected Estimated

Level (Jan. 1, 1975) (June 30,1974)

-14 26,779 27,876

15-64 30,663 30,198

65+ 2,012 2,071

TOTAL 59,454 60,145

SOURCE: Projections from Coale and Hoover, op. cit.; birth and death rates from Population Index, Office of Population Research, Princeton University; estimated population from Demographic Yearbook 1975, United Nations.




1970s 1950s figure/ Dates Used

figure 20

years earlier

Total population 29.7 million 60.2 million 2.03 1955-1975

Population 6-14 6.0 million 12.4 million 2.07* 1950-1970

Attending school 2.25 million 8.01 million 3.56 1950-1970

Percent attending 37.5 64.4 1.72 1950-1970

Literate, ages 6+ 11.8 million 27.5 million 2.34 1950-1970

Percent literate 56.8 71.7 1.26 1950-1970

Urban population 11.0 million 28.4 million 2.58 1950-1970

Percent urban 42.6 58.8 1.38 1950-1970

Expectation of life 48.1 years 61.4 years 1.27 1951-1970

at birth

Growth in income

at constant


TOTAL 3.69 1955-1975

PER CAPITA 1.89 1955-1975

SOURCES: Population data including school attendance, urban population, and literacy from Mexican censuses of 1950 and 1970; expectation of life at birth in 1951 and 1970 from Demographic Yearbook 1966 and 1973, United Nations; growth in income calculated from Yearbook of National Accounts Statistics 1963, 1972, and 1975, Department of Economic and Social Affairs, Statistical Office, United Nations.

* Estimates, 1955-1975, 2.10.

to the 1970s. Together these data show impressive progress: a 72 percent increase in the proportion of the population of primary school age attending school, a 26 percent increase in literacy, an 89 percent increase in per capita income at constant prices, a 38 percent increase in the urban proportion, and a 27 percent rise in the average duration of life. These large and sustained positive changes in the Mexican economy and society certainly are contrary to the pessimistic view that rapid population growth in a low-income country is inconsistent with progress. They reveal more strongly the error of assertions that to double the population of a country no better off than Mexico was in the 1950s in only 20 years inevitably leads to impoverishment, malnutrition, and social collapse. It has not. The Mexican experience is a very useful counter to the over-simplified attribution to excess fertility of every impediment to social and economic development.

The Mexican experience must also be viewed as paradoxical for those who hold the optimistic position that countries with rapid population growth need have no concern about it, because social and economic progress automatically bring down birth rates. Suppose one were to tell the policymakers in India and Bangladesh that the proper policy is to concentrate on social and economic progress: since development is the best contraceptive, the birth rate will come down without any special interventionist measures. Could we say, to illustrate this point, "For example, look at Mexico?" The Mexican population in 1975 was more than 72 percent literate and more than 60 percent urban, the per capita income had nearly doubled in 20 years, the expectation of life at birth had risen to about 65 years and was still increasing; nevertheless, fertility until about 1975 was, if anything, somewhat higher than it was 20 years before.5 The levels of income, literacy, and mortality are less favorable in many parts of South Asia today than they were in Mexico 20 years ago. Current progress in South Asia is much more gradual, and is unlikely with any measures that might realistically be imagined to duplicate the rapid progress in Mexico. In short, Mexico presents a puzzling picture for those who expound a simple version of the "demographic transition," the name that has come to be given to the theory that modernization brings predictable changes in mortality and fertility.

In the study that Hoover and I made, we summarized the transition theory of fertility as follows:

The theory of the demographic transition asserts that the high birth rates, as well as the death rates, characteristic of an agrarian low-income society are affected by economic development. The changing structure of production, with a declining importance of the family as a production unit, with the growth of impersonal systems for the allocation of jobs, and with the development of economic roles for women outside of the home, tends to increase the possibility of economic mobility that can better be achieved with small families, and tends to decrease the economic advantages of a large family. One of the features of economic development is typically increasing urbanization, and children are usually more of a burden and less of an asset in an urban setting than in a rural. The whole process of economic change, moreover, weakens the force of traditional customs and beliefs. In most countries that have undergone the economic transition from an agrarian to an industrialized, market-oriented economy, the custom of the small family has started in the urban groups at the higher end of the socio-economic scale and has spread to smaller cities, lower-income groups, and eventually to rural areas.

Mexico is one of several examples that show that the relationship between social change and the decline of the birth rate is a complex one, and that each stage of progress is not necessarily accompanied by lower fertility. In an intensive study at the Office of Population Research (in Princeton) of the decline of fertility in Europe, we have found a number of instances in which the decline occurred in populations that were rural, not very well educated, and still subject to high infant mortality (sections of France in the late eighteenth and early nineteenth centuries, provinces in Hungary in the nineteenth century, whole regions of Eastern Europe, such as Bulgaria and Romania, in the 1920s). We have also encountered the puzzling anomaly that the nationwide decline in fertility began in France nearly a century before it began in England, although it was England that was the leader in the Industrial Revolution.

In the contemporary world, Mexico is perhaps the most conspicuous national example of a country in which the theory of demographic transition would certainly seem to indicate that fertility should have shown a major reduction and yet none occurred (at least none before 1974 or 1975). There are other populations within large countries that are as paradoxical in this respect as the population of Mexico. I refer specifically to the populations of the Central Asian republics within the Soviet Union. These populations were traditionally Muslim in religion and culture, and were almost entirely agricultural or pastoral at the time of the Russian Revolution (the republics in question are the Kirgiz, Turkmen, Tadzhik, and Uzbek Republics, with a total population of 22 million). In 1926, fewer than one percent of the women aged 20-29 had been reported in the census as literate. By 1970, at least 85 percent of the women aged 20-29 had completed a primary school education. The expectation of life at birth was well above 60 years. (After all, 1970 was some 52 years after the Russian Revolution.) Nevertheless, the fertility of married women in the rural population of these republics was nearly 50 percent higher in 1970 than in 1926. Marital fertility also increased (after due allowance for underregistration of births) between 1959 and 1970. Fertility in these areas is just as high as in Mexico, and rather than falling during the 1960s, actually increased.6

We find that the Mexican experience in the last 20 years, in short, supports neither the simplistic view that population growth is the principal source of all difficulties in any low-income country, nor the notion that progress in education, increases in per capita income, and concentration of the population in urban areas rather than in a rural setting have an inevitable, automatic effect of causing a reduction in the birth rate.


What does the Mexican experience indicate retrospectively about the analysis that Hoover and I attempted 21 years ago? The extensive social and economic progress accompanying very rapid population growth with no decline in fertility is in no sense a contradiction of our argument. The question is whether there is evidence that in spite of impressive progress, the continuation of high fertility has been a detriment to the quality of Mexican life. To put the point more directly, is there evidence that important additional progress would have occurred if fertility had fallen rather than remaining constant or even rising slightly?

If fertility had fallen beginning in 1955, the striking difference between that hypothetical population and the actual population is in the number of persons under 15 - there would be about 25 percent fewer had fertility fallen according to our lower projection. The smaller numbers would yield many advantages for the whole population, including the possibility of some 15 percent higher income per equivalent adult consumer. But the most evident advantages would be for the children themselves. With four or five children in the completed family rather than six or eight, there would be less crowded living conditions, better food, and better parental care. Recent research has shown that school performance is related to the average educational level within the family in the years of infancy and early childhood. Exposure to parents and to older siblings who have had some schooling apparently helps in the intellectual development of the child; constant contact with children only a little older is less helpful. Hence small families and wider spacing of births produce an advantage in the acculturation of the children.7

Let us look further at the extension of education, as an instance in which progress has surely been inhibited by high fertility, which has led to a very rapid increase in the child population. Because reductions in mortality have been especially large in infancy and childhood, the effect of falling mortality has been not only to accelerate the growth of the population, but to accelerate it especially at the younger ages. Therefore, the increase in the population of primary school age has been even greater than the increase of the population as a whole. While the total population increased by a factor of 2.03 from 1955 to 1975, the school age population increased by at least 2.10 (Table 1). Looking at the data on school attendance in the years of the censuses, we saw that from 1950 to 1970 the number attending primary school had risen from 2.25 million to over 8 million, being multiplied by a factor of 3.56, an extraordinary 6.3 percent average annual increase. Despite the sustained allocation of large resources to education (something on the order of 15 percent of total governmental expenditures), the number of children not enrolled in school increased from 3.75 million in 1950 to 4.42 million in 1970. Very rapid increases in enrollment could not reduce nonattendance.

Had there been a linear decline in fertility beginning 20 years ago, of the sort that we predicted in the lower fertility projection, the number of children of primary school age in 1975 would have been 21 percent less. If the same resources had been devoted to expanding education, the current school attendance ratio would be approaching 90 percent of the population of primary school age instead of attaining merely 64 percent, and the number not in school would have been sharply reduced, rather than increasing.

Population growth has contributed to a similarly regrettable trend in the number of illiterates. While the number of literates rose from 11.8 million to 27.5 million between 1950 and 1970, the number of illiterates rose from 8.9 million to 10.9 million. Again, with a more moderate rate of growth occasioned by reduced fertility, the number of children who have grown up during this period without achieving literacy would have been greatly reduced, and the record of rising literacy would be substantially better.

But the most acute disadvantages that we can now see resulting from the failure of fertility to fall in Mexico during the last 20 years are the built-in difficulties that the past continuation of high fertility makes inevitable in the development of the Mexican population during the next 20 years. Since the persons who will be over 15 fifteen years from now are already born, and since any reduction in fertility is likely to occur gradually, we can foresee the size of the population over 15 in 1995 with some confidence. Mortality rates are already quite low and further declines in mortality will have only the modest effect of slightly accelerating the increase in the population above age 15. We can foresee then, with little margin of uncertainty, that there will be more than twice as many persons over 15 twenty years from now as today.

One of the more intractable problems in designing development programs for rapidly growing low-income countries is the provision of adequate employment opportunities for the increasing population at the ages at which participation in economic activity is at its peak. Modern technology, which offers the potential of much greater productivity for the labor force, often requires disappointingly modest additional inputs of manpower in countries where unemployment and underemployment are already at distressing levels. Rapid increases in output, both in agriculture and in industry, can often be attained with only slight increases in the number of workers involved. In recent years, as the worsening international economic situation has reduced the demand for goods in international trade, the precarious nature of job opportunities in many less-developed countries, including Mexico, has become evident. Estimates of unemployment in Mexico run as high as ten percent; and there is much underemployment (involuntary short hours, work at excessively low pay, work for which the employee is overqualified).

President López Portillo, in an address to the International Union for the Scientific Study of Population in August 1977, emphasized the problems, both for Mexico and the United States, of illegal migrants and noted their origin in the inadequate economic opportunities in Mexico. The number of persons aged 15-65 today would be almost the same if fertility had fallen in the past ten years, but the rate of increase would be less. It would surely be an easier prospect for those responsible for the development of the Mexican economy and society if the increase in the labor force in the next 20 years were expected to be only 60 percent, as would be the case if fertility had begun a substantial decline 20 years ago, rather than the actual prospect of an increase of 105-110 percent!

One of the principles that technical demography has to teach us is the principle of the momentum of population growth. This principle is illustrated by what I have just said about the inevitability of the growth of the adult population in Mexico in spite of any declines in fertility that may begin in the near future. The fact that fertility has remained high means that the recent cohorts of those born each year are twice as large as those born 20 years earlier. The passage of these enlarged cohorts through the successive years of life makes substantial future population growth inevitable, even if from now on, miraculously, Mexican parents were to bear a greatly reduced number of children. If fertility were to be cut in half in 25 years beginning today, and were then maintained at no more than 50 percent of current levels, the population of Mexico would still be multiplied by 11 in 150 years. Multiplication by 11 would mean about 660 million, which exceeds the present population of India. On the other hand, to imagine the unimaginable of 150 years more in which fertility was not reduced is to think about a population multiplied by nearly 250 (to reach some 15 billion) or nearly four times the present total population of the world.

These longer range calculations illustrate the undeniable point that the birth rate must eventually come down and come a long way down, if the low mortality that modern science and medicine can bring is to continue. What Hoover and I tried to show, and what I feel after this review is essentially correct, is that an immediate reduction in the birth rate also brings tangible advantages. Some potential advantages have already been lost because no reduction was achieved in the past two decades. Further postponement of the reduction that must occur sooner or later means more opportunities passed up, and worsening of some difficult long-range problems. No one need worry about too small a future population. Even with much lower fertility, the population of Mexico will reach very substantial size in the not very distant future.


What is the difference between the situation today and 20 years ago? Mexico is richer, has much larger industrial and agricultural output, per capita as well as total; it is much more urbanized, and literacy is more common. But the population is twice as large, unemployment is more acute, and the labor force is growing more rapidly. It requires much more investment to maintain adequate employment and productivity for a rapidly growing labor force than for one growing more slowly. It may be even harder now than during the past two decades to attain a level of investment sufficient to supply productive employment.

One important difference between the mid-1970s and the mid-1950s is in the attitude toward population among Mexico's leaders. Hoover and I were very cordially received in 1956, but our ideas found a less hospitable reception. Now the advantages of a lower birth rate are widely perceived. In 1974 a constitutional provision was adopted asserting that, "Every person has the right to decide in a free, responsible, and informed manner the number and spacing of his or her children." There is now a National Council of Population in the Ministry of the Interior, and governmental as well as private programs for the promotion of family planning.

Finally, it appears that the birth rate has begun to go down. The number of registered births indicate that the rate has fallen from 44 per thousand to perhaps 40 per thousand in the last three or four years. Because births in Mexico are registered after a delay, some of the apparent decline may be spurious. However, to some extent the drop is very likely genuine. Sober estimates indicate that in 1975 nearly 10 percent of women aged 15-49 were clients of one of the public or private family-planning programs.

It is not always easy to make available the information and facilities that can accelerate a decline in fertility. But the magnitude of the resources that might be devoted to such a program are not great. President Johnson was persuaded to insert in one of his speeches a statement that one dollar spent on the reduction in fertility in a high-fertility population was better than 20 dollars devoted to general development. Professor Etienne van de Walle, of the University of Pennsylvania, says this statement is wrong. What is true is that 20 dollars spent on development are more effective if one of them is devoted to a population program. In the long run no country with very high fertility can afford to ignore it.


1 Population Growth and Economic Development in Low-Income Countries, Princeton: Princeton University Press, 1958.

2 Actually we calculated income per equivalent adult consumer, a computation in which children were counted as only half as heavy consumers as adults. This allowance prevents an overstatement of the effect on a per capita basis.

3 Paul Ehrlich, The Population Bomb, New York: Ballantine Books, 1968; William and Paul Paddock, Famine 1975!, Boston and Toronto: Little Brown, 1967.

4 Despite an appearance to the contrary, the slightly declining crude birth rates in the projection (shown in Table 1) are not inconsistent with an assumption of constant fertility. The assumption in the projection was that the rate of childbearing by women at each age would remain fixed. The decline in mortality from 1955 to 1975 has, as demographers know, the somewhat surprising effect of increasing the proportion of the population at the younger ages and diminishing the fraction elsewhere, including the fraction of women at the ages of childbearing. The reduction in the proportion of the population consisting of women at the ages of maternity is the source of the gradually declining crude birth rate in our projection. In the actual population the birth rate did not decline consequentially through the early 1970s. This failure to decline implies either that the fertility of women of childbearing age has risen slightly or that registration of births has improved (implying that the birth rate was understated in the earlier years).

5 Data taken from Tables 1 and 2, with the following exceptions: the expectation of life at birth in 1975 is taken from the U.N. Demographic Yearbook 1975; the urban proportion (58.8 per cent in 1970) was surely above 60 percent by 1975.

6 Data from A. J. Coale, B. A. Anderson, and Erma Härm, Human Fertility in Russia since the 19th Century: the Demographic Transition in a Unique Historical Setting, in preparation.

You are reading a free article.

Subscribe to Foreign Affairs to get unlimited access.

  • Paywall-free reading of new articles and a century of archives
  • Unlock access to iOS/Android apps to save editions for offline reading
  • Six issues a year in print, online, and audio editions
Subscribe Now
  • Ansley J. Coale is Professor of Economics and William Church Osborn Professor of Public Affairs at Princeton University, where he is also Associate Director of the Office of Population Research. He is the author of The Growth and Structure of Human Populations: A Mathematical Investigation and the co-author of The Demography of Tropical Africa and Regional Model Life Tables and Stable Populations, and other works. This article is adapted from a lecture presented at the Instituto Mexicano-Norteamericano de Relaciones Culturales in August 1977.
  • More By Ansley J. Coale