Demography

About 9 pages (2,653 words)

Demography Summary

Questions on this topic? Just ask!

The Social Science Encyclopedia, Second Edition

demography

Demography is the analysis of population variables. It includes both methods and substantive results, in the fields of mortality, fertility, migration and resulting population numbers. Demographers collect data on population and its components of change, and construct models of population dynamics. They contribute to the wider field of population studies that relate population changes to non-demographic—social, economic, political or other—factors. In so far as it reaches into population studies, demography is interdisciplinary: it includes elements of sociology, economics, biology, history, psychology and other fields. Its methods include parts of statistics and numerical analysis. Public health officials and actuaries have had their part in its development. Most demographers have professional knowledge of one or more of these disciplines.

Population variables are of two kinds—stock and flow. The important source of information on stock variables is national censuses, whose modern form goes back to the seventeenth century in Canada, Virginia, Sweden, and a few other places, and which are now carried out periodically in nearly all countries of the world. Among the cross-sectional information collected in censuses are age and sex distribution, labour force status and occupation, and birthplace.

The flow variables, the components of population change, include birth and death registrations, initiated before the nineteenth century in Sweden and in Britain, and now routine in all industrial countries. Efforts to attain completeness are slowly making their way elsewhere. Migration statistics, collected at national frontiers, are less available and less reliable than birth and death registrations. Much additional information, including statistics of birth expectations, is collected by sample surveys.

These four sources (censuses, vital registration, migration records, and sample surveys) differ in the ease with which they may be instituted in a new country. Censuses and surveys are the easiest to initiate. With care the completeness of a census can reach 97 per cent or more. It is true that a large number of enumerators have to be mobilized (over 100,000 in the USA in 1980, over 5 million for China’s 1982 census), but that is easier to arrange than the education of the entire population to the need for birth registration. The USA first attained 90 per cent complete birth records in the first quarter of the twentieth century; contemporary poor countries are unlikely to reach this level of completeness until their residents come to have need for birth certificates. Migration statistics will not be complete as long as many of those crossing international borders can conceal their movement from the immigration authorities. Apart from illegal crossings there is the difficulty that migrants are a small fraction of those passing national boundaries, the majority being tourists, persons travelling on business, commuters, and other non-immigrants. American sentiment that people ought to be able to leave their country of residence without hindrance is so strong that out-going residents are not even stopped at the border to be asked whether they intend to return.

The special characteristics of demography are the quantitative and empirical methods that it uses. Once data in the form of censuses and registrations are available, demographic techniques are needed for valid comparisons among these. Mexico has a death rate of 6 per thousand, against France’s 10; this does not signify that Mexico is healthier, but only that it has a younger age distribution as a result of high fertility; standardized comparison consists in finding what Mexico’s death rate would be if it had France’s age distribution but retained its own age-specific rates.

Partly for purposes of comparing mortality, but originally more for the conduct of pension and insurance business, life tables were developed in The Netherlands and in Britain during the course of the eighteenth century. The first technical problem that actuaries and demographers solved was how to go from statistics of deaths and of populations exposed to probabilities of dying. With data in finite age intervals the probabilities are not uniquely ascertainable, and a variety of methods for making life tables are currently in use.

The concerns of public health have led to the improvement of mortality statistics along many lines, including drawing up the International List of Causes of Death, now in its ninth revision. Unfortunately uniformity in applying the classification among physicians in all countries is still a distant goal. One object of the International List is the making of cause-related tables. The expectation of life in the USA is 75 years. If all deaths from cancer were eliminated this would be increased by about 3 years; elimination of all heart disease would increase the expectation by over 15 years.

Increasing populations have lower proportions of deaths than stationary populations. In effect the age distribution pivots on the middle ages as population growth slows. A sharp drop in the birth rate does not show its full effect immediately; births remain high as the large cohorts of children already born themselves come into childbearing; population growth thus has a kind of momentum. Replacement is the condition where each child is replaced in the next generation by just one child, so that ultimately the population is stationary. After birth rates fall to bare replacement a population can still increase by 60 per cent or more.

Births are not as sensitive to the pivoting of age distribution as are deaths, since the fertile ages, intermediate between childhood and old age, are a relatively constant fraction of a population. Fast-growing countries have more children below reproductive age but fewer old people. But births are greatly affected by a bulge of individuals in the reproductive ages; births in the USA have risen from about 3.1 million in the early 1970s to about 3.6 million currently, almost entirely due to change in age distribution as the large cohorts of the 1950s reach reproduction.

The pioneer in demographic methods and models was Alfred J.Lotka, who in a series of papers extending from 1907 to 1948 showed how to answer a number of questions that are still being asked. A central one was, ‘How fast is a given population really growing, as determined by its age-specific birth and death rates in abstraction from its age distribution?’ Any population that grows at a fixed rate for a long period develops a stable or fixed age distribution which Lotka showed how to calculate, and its increase when it reaches this stage is its intrinsic rate.

After a long period of neglect, Lotka’s work came to be applied and further developed during the 1960s. It turned out that his approach could help the estimation of birth and death rates for countries of known age distribution but lacking adequate registration data.

The techniques of birth and death analysis have been carried over to migration, especially in the form of Markov chains that describe movement or transition between countries, and other areas, just as they describe transition between life and death. Such Markov chains are capable also of representing transitions between the married and single condition, among working, being unemployed, and leaving the labour force, and many other sets of states. A literature has now been built up in which changes of state, including migration, are represented by matrices, particularly easy to handle on a computer. The first extensive calculation of this kind was due to P.H.Leslie in the 1940s.

Communities living under ‘primitive’ conditions grow slowly; their high birth rates are offset by high deaths. The movement of a community from this condition to one of low birth and death rates as it modernizes is known as the demographic transition. Since the fall in the birth rate lags behind the fall in the death rate, very large increases can be recorded during the transition. Britain’s population multiplied fourfold between the censuses of 1801 and 1901. Contemporary less developed countries are increasing even more rapidly.

This effect of rising income is contrary to what has often been thought: that people want children and will have as many as they can afford—a view commonly attributed to Malthus, although Malthus’s writings, after the first edition of his Essay, are much more subtle than this. Apparently at a certain point the causal mechanism flips over: for very poor people a rise of income results in a faster rate of increase; once people are better off a further rise slows their increase.

The modernization that brings down the birth rate affects subgroups of national populations at different times. In consequence the demographic transition shows itself as differential fertility: the rich, the urban and the educated have for a time lower birth rates than the poor, the rural and the illiterate in cross-sections taken during the transition. Such differentials close up as incomes generally rise and income distributions narrow.

Some of the most puzzling questions concern the causal mechanisms that lie behind contemporary demographic changes. In what degree the fall of fertility is due to education, in what degree to income, cannot yet be answered in a way that applies to all countries. Less developed countries have far greater rates of increase now than did the countries of Europe when these were at a comparable stage of development. To what extent is the difference due to higher birth rates among presently poor countries than among the poor countries of the eighteenth century, and to what extent to lower death rates? Population models can provide answers to such questions; they show that birth differences are much more influential than death differences.

More difficult are questions on the direction of causation between two variables clearly related to each other. In most advanced countries more women are working outside the home now than in the 1950s, at a time when their partners are for the most part earning higher real wages; at the same time women’s fertility has diminished, when their income and that of their partners would enable them to have more children if they wanted. Is the fall in fertility the result of women seeking jobs, and so finding it inconvenient to have children, the wish to work being primary, or do they no longer wish to have children and so take jobs to fill their time? A wealth of data exists, but the techniques for answering such questions are elusive. Again, are the present low birth rates a cohort or a period phenomenon? Do they result from present generations intending to finish up with fewer births, or are they a conjunctural phenomenon due, say, to the world recession since the mid-1970s.

A task that demographers are often called on to perform is population forecasting. Professional demographers describe their statements on future population as projections, the working out of the consequences of a set of assumptions. Users believe that the assumptions are chosen by the demographers because they are realistic, and they accept the results as forecasts. Forecasting has gone through many styles, starting with extrapolation of population numbers by exponential, logistic or other curves. More acceptable is extrapolating the components of population—birth, death, and migration—and assembling the population from the extrapolated values of these. Sampling to ascertain childbearing intentions of women has been extensively tried. Demographers have no illusions about the predictability of the long-term future, but estimates made by those who have studied the past are more worthy of attention than the simple-minded extrapolations that are the alternative. Some numbers on future population are indispensable for virtually any kind of economic planning, whether by a corporation or a government.

One aspect of the future that has become of interest to several disciplines since the mid-1980s is the relation of a deteriorating planetary environment to population growth. How far are the loss of forests, emissions of carbon dioxide and other greenhouse gases with consequent danger of global warming, thinning of the ozone layer, declining catches in the fishing areas off western Europe, due to population increase, to what extent to expanding economies, to what extent to inadequate environmental regulation? The facts are the same for everyone, but scholars of different disciplines read them differently, and come to very different conclusions.

Biology works above all with direct observation: the facts are plain to the naked eye; no sophisticated theory is needed to see that forests are invaded and destroyed by the need of increasing populations for wood and food, that crowded countries of Asia with growing economies will suffer more pollution of water, land and air if their populations continue to increase at present rates, that Africa may have been thinly settled in the past but its present population increase with stagnant economies will bring increasing misery and destruction. For many biologists population control is humankind’s first priority; the US National Academy of Sciences and the Royal Society of London introduce their joint statement with the words, ‘If current predictions of population growth prove accurate and patterns of human activity on the planet remain unchanged, science and technology may not be able to prevent either irreversible degradation of the environment or continued poverty for much of the world’. No vote was taken of the membership representing all the sciences, but this statement would surely have commanded the assent of a large majority

Economics finds the matter more complicated. Population growth has advantages that could well compensate for its apparent disadvantages. With more people there will be more inventive geniuses; with larger markets there will be more opportunity for division of labour and so more economic progress; once people are rich enough they will be able to use any necessary small fraction of their wealth to counter environmental damage. So it is increasing wealth rather than population control that ought to have priority.

When biology and economics find such opposite relations of population to environment, only the blandest of statements can be made without contradicting one or another scientific authority. Before such contradiction one ought to examine the contending disciplines more closely, and try to find what of their features and methods lead them to such opposite conclusions. Study of the intimate detail of the several disciplines is the only hope for breaking the deadlock, and the only way of making full use of the positive knowledge of those disciplines.

The richness of demography is in part due to the commitment of scholars from many disciplines. Actuaries developed much of the early theory, and statisticians and biostatisticians add to their work the techniques of numerical analysis and determination of error. Sociologists see population change as both the cause and the result of major changes in social structures and attitudes; they study the increase of labour force participation by women, of divorce, of single-person households, the apparently lessening importance of marriage, and the decline in fertility rates. Economists see fertility rising and falling as people try to maximize utility. Biologists employ an ecological framework relating human populations to the plant and animal populations among which they live and on which they depend. Psychologists have brought their survey and other tools to the study of preferences of parents for number and sex of children. Historians, in a particularly happy synthesis with demography, are putting to use the enormous amount of valuable data in parish and other records to gain new insights on what happened to birth and death rates during the past few centuries.

Nathan

Harvard University and the International Institute for Applied Systems Analysis

Reference

(1992) Population Growth, Resource Consumption and a Sustainable World. A joint statement by the officers of the Royal Society of London and the US National Academy of Sciences. Sir Michael Atiyah, President of the Royal Society of London, and Frank Press, President of the US National Academy of Sciences.

Demography

demography

Reference

Further reading