DEMOGRAPHIC TRANSITION

The term demographic transition refers to the decline in mortality and fertility from the high rates characteristic of premodern and low-income societies to the low rates characteristic of modern and high-income societies. Demographic transition is a central concept in demography, and there is a large literature examining the nature and the causes of the phenomenon. On the face of it, demographic transition is simply a description of a pattern of historical trends in vital rates. The influential discussions of demographic transition, however, interweave description with explanation of mortality and fertility declines, and this has made it difficult to separate the descriptive concept from the far more controversial "theory" of demographic transition.

History of the Concept

Although the term demographic transition originated with Frank W. Notestein in the mid-twentieth century, the first systematic effort to describe distinctive demographic regimes that represented historical stages linked to broader societal changes is credited to the work of the French demographer Adolphe Landry dating back to the first decade of the twentieth century. In Landry's formulation, elaborated in greater detail in a book published in 1934, demographic regimes are a function of the material aspirations of individuals and the productive potential of the economic system. In the "primitive" regime characteristic of subsistence economies, mortality but not fertility is constrained by economic factors, and population size tends to the maximum that economic resources can support. In the "intermediate" regime, in an effort to preserve family wealth, fertility is depressed by late marriage and celibacy, and population size falls below the maximum that the economy can support. The "modern" regime emerges when economic productivity reaches high levels and individuals have well-formulated aspirations for a high standard of living. To facilitate the achievement of those material aspirations, fertility becomes an object of conscious limitation, chiefly through various techniques of birth control but also through late marriage and celibacy. Population size is far smaller than the economy could support were individuals willing to accept lower standards of living–indeed negative population growth rates are a distinct possibility.

An alternative three-stage formulation of demographic transition was offered by the American demographer Warren Thompson in 1929. Thompson classified the countries of the world into three groups: (1) countries with high birth rates and high but declining death rates, facing the prospect of rapid population growth; (2) countries with declining birth and death rates in certain socioeconomic strata, with the rate of decline in death rates outstripping the rate of decline in birthrates; and (3) countries with rapidly declining birth and death rates, with fertility declining more rapidly than mortality, resulting in a declining population growth rate. Thompson assumed that these three groups were representative of historical stages. But by limiting his purview to contemporary demographic regimes, Thompson offered a truncated evolutionary scheme–he described neither a full-fledged pre-transition regime nor a post-transition regime. In addition, Thompson had less to say about the causes of demographic change than his predecessor Landry and his successors Notestein and Kingsley Davis.

Notestein's formulation has probably been the most influential, appearing just at the onset of a five-decade period of widespread concern about the development-retarding effects of rapid population growth in Africa, Asia, and Latin America. Notestein held that the lessons he had distilled from the European historical experience were applicable to other regions and could inform public policies. Like Thompson, Notestein focused on the societal variation he observed at the time and therefore devoted limited attention to pretransitional regimes. He was aware that mortality decline was well underway in Africa, Asia, and Latin America yet fertility was essentially unchanged; these societies with high-population-growth potential constituted his first type of demographic regime. A second were those countries where fertility decline was well established but incomplete (Japan, the Soviet Union, and the southern cone of South America), and the third type were the low mortality and fertility populations of Europe, North America, and Australia. What gave Notestein's piece special power was his succinct yet compelling explanation for the declines in mortality and fertility (discussed below). One crucial element in Notestein's argument was that mortality is likely to respond more quickly than fertility to the forces of change, and therefore it is all but inevitable that societies experience a transitional period during which birth rates exceed death rates by a substantial margin, generating rapid population growth.

The Demography of Demographic Transition

Since the 1950s the standard formulation of demographic transition comprises three stages: pretransition regimes, characterized by high (and fluctuating) mortality and high fertility; transitional regimes, characterized by declining mortality and declining fertility, with mortality decline typically running ahead of fertility decline, resulting in population growth; and posttransitional regimes, with low mortality and low (and possibly fluctuating) fertility. The pretransition and posttransition regimes are assumed to be essentially in long-term equilibrium, with transitional regimes acting as a bridge between the two. In pretransition regimes, life expectancy at birth is less than 40 years and women bear on average between five and eight births over their reproductive lifespan, whereas in posttransition regimes, life expectancy at birth exceeds 65 years and women bear on average 2.5 or fewer births.

As empirical studies have accumulated, it has become apparent that pretransition and posttransition regimes are far from uniform in their vital rates. In general, pretransition mortality was lower in Europe than in Africa and Asia–life expectancy closer to 40 years in the former and 30 years in the latter. Even within Europe there was great variability in mortality rates, with the percentage of children dying in infancy ranging from over 30 percent in parts of Bavaria to 10 percent in southern England at the onset of demographic transition. Mortality was also characterized by substantial variation over time, reflecting nutritional adversity and epidemics of infectious disease. Nonmarriage and late marriage significantly reduced fertility rates in pretransition Europe, whereas marriage of women was close to universal in most African and Asian societies and generally occurred soon after menarche (the first menstrual period). As a result, in African and Asian societies fertility levels were higher, even though postpartum sexual abstinence and extended breast-feeding had a moderating effect on fertility rates. There is evidence, still subject to some dispute, that deliberate and conscious regulation of childbearing–the spacing of births–and perhaps of family size as well was common in pretransition African and Asian societies. Fertility within marriage appears to have been subject to far less control in pretransition Europe, although withdrawal was a widely known method of contraception that later was extensively practiced to control fertility in many parts of Europe.

Posttransition populations also show considerable variability in their demographic rates. Continuing declines in mortality at older ages have led to life expectancies at birth approaching 80 years in some European, North American, and East Asian countries, whereas life expectancy has slid below 70 years in eastern Europe because of deteriorating health conditions. The AIDS pandemic, affecting transitional societies especially in eastern and southern Africa, is further demonstration that improvements in health are not necessarily permanent, indeed that reductions in life expectancy on the order of 15 to 20 years can occur over a period as short as two decades. Such nonuniform trends in mortality in transitional and posttransition populations were not fore-seen in the original formulations of the demographic transition. Furthermore, fertility in posttransition countries has in general failed to settle on the replacement level of an average of just over two births per woman over the reproductive lifespan. For decades, births per woman remained substantially above that level, ranging between 2.5 and 3 in the southern cone of South America (Argentina, Chile, and Uruguay) in what seemed a relatively stable posttransition regime. In contrast, in the decades since 1970, fertility has fallen below replacement in most European countries, and even below 1.5 births per woman in some countries of southern and eastern Europe.

The combinations of death rates and birth rates observed in pretransition and posttransition populations allow for modest demographic growth and decline, although over long stretches of time growth rates in pretransition societies were close to zero (typically less than 0.5 percent per year). The rate of population growth in pretransition and posttransition societies is dwarfed by the rate of growth in transitional societies–a result of the time lag between the mortality and fertility declines during the process of transition and, additionally but not universally, a temporary fertility increase early in the transitional stage. Such temporary fertility increases are in all likelihood a physiological response to improved maternal and child health and changes in postpartum practices. The "transition multiplier"–the ratio of the posttransition population size to the pretransition population size–is determined by the extent to which birth rates exceed death rates and the length of time during which that condition prevails. Transition multipliers are high when fertility decline begins from a high initial level and occurs substantially later than mortality decline and proceeds slowly.

An important aspect of the dynamics of transition is that population growth does not immediately subside once fertility falls to replacement level. The high fertility and low childhood mortality of the transitional demographic regime further accentuates the young age-structure that characterizes pretransition populations. This means that for several decades relatively large cohorts pass through the childbearing years. The additional population growth that occurs while the age-structure shifts to its post-transition shape is called population momentum. Population momentum is a substantial component of population growth over the course of demographic transition, typically contributing 30 to 40 percent of the total growth. Formal demographic analysis and simulation exercises demonstrate that population momentum is inversely related to the level of posttransition fertility and to the pace of fertility decline.

The demographic transitions in European populations differed substantially from the transitions in non-European populations in the magnitude of the rate of transitional population growth. In Europe, where the decline in fertility followed close on the heels of the decline in mortality, both starting from relatively low pretransition levels, the rate of natural increase (birth rates minus death rates) during the transitional period from 1800 to 1950 ranged between 0.5 and 1 percent per year, and the transition multiplier was roughly four (a ratio moderated somewhat by overseas emigration). In most non-European populations, mortality declines began during the first decades of the twentieth century and became steep in the decades after World War II, whereas fertility declines (from relatively high initial levels) began in earnest only after 1960 or later. As a result, many non-European countries experienced population growth rates of 2 to 3.5 percent per year for four decades or longer, and the transition multipliers (calculated using projected population numbers) range from 8 to 20. The highest multipliers are found in those countries with slow fertility declines, for example the Philippines, where the pretransition population size was about 8 million, the 2002 population was 79 million, and the posttransition population size is projected to be as high as 150 million, according to the United Nations, and Guatemala (pretransition population of 1.4 million, 2002 population of 12 million, and posttransition population projected as high as 30 million). In no European country did demographic transition produce population growth on this proportionate scale. Population multipliers of this magnitude, often combined with a pretransition population size that was large in absolute terms, are bound to have many and varied repercussions for social, economic, political, and cultural systems–some positive but no doubt also some deleterious.

Explanations for Demographic Transition

The many efforts, from Landry to the present, seeking to identify the forces generating demographic transition fall into two major sets. One regards fertility decline as an inevitable response to the population growth induced by mortality decline, which is therefore all that requires explanation. The second views fertility decline as a response to a richer and more diverse set of social, economic, political, and cultural forces.

While mortality decline has presented less of an explanatory challenge than fertility decline, there has been ample debate about its causes. Economic transformations that improved standards of living–food, clothing, sanitation, housing–appears to account for much of the decline of mortality in Europe. Samuel Preston argued in 1975, however, that economic change, as captured by growth in income per capita, accounts for only a small fraction of mortality decline in non-European populations in the twentieth century. Political stability and the emergence of effective nation-states complement the effects of economic change by leading to more reliable access to food and improved public sanitation. New medical technologies made a minor contribution to the decline of mortality in Europe in the eighteenth and nineteenth centuries but were a major factor in the sharp reduction in mortality from infectious diseases in the developing countries in the twentieth century. A final factor is improved personal hygiene (hand washing, preparation of food, and so forth), with new habits adopted in response to formal school instruction, public-health education campaigns, and word-of-mouth information.

Some scholars have argued that mortality decline is a sufficient cause of fertility decline and hence accounts for the demographic transitions of the past two centuries. Strictly speaking, the explanatory factor is not mortality decline but population growth. In 1963 Davis described household-level strain created by significantly larger younger generations vying for valued economic and social resources. Successively larger cohorts (in particular, the increase in the ratio of sons to fathers) disrupt the equilibrium of the traditional family. Other scholars have noted that mortality decline, normally accompanied by improved health of the population, should increase economic productivity and through that channel exercise a positive indirect effect on fertility. Finally, mortality decline encourages a change in personal psychologies away from fatalism toward a greater sense of self-control over one's destiny, and this facilitates the exercise of deliberate fertility regulation.

Fertility declines have occurred under widely varying social and economic circumstances but virtually never in the absence of mortality decline, and this can be taken as strong evidence that mortality decline is the primary cause of fertility decline. Theories of demographic homeostasis posit that human societies gravitate toward demographic regimes with growth rates near zero; multiple and diverse societal institutions act as governors on population growth and enforce the tendency to oscillate near zero growth. Marked departures meet with the appropriate demographic response–increases in fertility to make up for mortality crises, decreases in fertility in response to mortality decline, or migration that offsets increases or decreases in rates of natural increase (a key element in Davis's theory of "multiphasic response"). While appealing as a general theory of population dynamics, homeostatic theory is not very informative about the demographic transitions that occurred during the nineteenth and twentieth centuries. The end results of these transitions, as noted earlier, were multifold increases in population size. It is not clear how homeostatic theory accommodates this failure of fertility or migration to compensate for the impact of mortality declines. Moreover, the diversity of the pretransition equilibrium levels of fertility and mortality and of the lags between mortality and fertility declines, as reflected in the large variation in transition multipliers, is a major empirical fact that demands explanation. Surely the explanation lies in the conditioning influence of social, economic, and cultural forces.

In the second set of explanations for fertility decline, mortality decline is not the sole causal agent. Indeed, Notestein, in his seminal 1945 work, hardly mentioned mortality decline as a motivation for fertility decline. Instead he argued that both mortality and fertility decline in response to urbanization and changes in the economy (which changed the costs and benefits of children and led to rising standards of living and increased material aspirations) and to growth in individualism and secularism. Notestein's argument has been elaborated in a large subsequent literature on the causes of fertility decline that has featured economic forces, cultural changes, and changes in birth control costs.

Economic theories of fertility decline focus on the causal impact of changes in the costs and benefits of children and childrearing. The fundamental cause of fertility decline is the (perceived) decreasing affordability of large numbers of children. Demographers have resisted giving pride of place to microeconomic changes in models of fertility decline, perhaps because of disciplinary biases but more importantly because of weak empirical associations between macroeconomic changes and fertility decline. The Princeton European Fertility Project, for example, uncovered no systematic relationship at the provincial level between the onset of fertility decline and socioeconomic variables such as levels of urbanization and nonagricultural employment. But other empirical research that has had access to a larger number of economic variables that provide a more complete portrait of the economic system, as well as studies conducted at lower levels of aggregation (the local community or the household), attribute much greater causal impact to economic change. This includes studies on fertility declines in England, Italy, Bavaria, and Prussia. Moreover, it seems likely that cognitive dimensions–in particular, economic aspirations and expectations–mediate the relationship between economic change and fertility. The causal force may not be economic circumstances per se but rather the relationship between economic aspirations and expectations (that is, what individuals want as opposed to what they expect). This can explain why fertility declines have occurred in the presence of both improving and deteriorating economic conditions.

Mortality decline and economic change are the core elements of a model for fertility decline. High fertility is compatible neither with low mortality nor with high-income, modern economies. Both mortality regimes and economic systems have been transformed during the past two centuries, to an extent and at a rate that are extraordinary by any measure. If one wishes to go back further in the causal chain and ask why this has occurred, inevitably one is led to the scientific and technological revolutions of the past four centuries. Ultimately it is these revolutions that lengthened life expectancy and made bearing large numbers of children inconsistent with modernity.

Another stream in the literature on the causes of fertility decline emphasizes the determining role of attitudes about and values related to family life. Ron Lesthaeghe has proposed that the decline of fertility in Europe was caused by the synergistic effects of economic changes and changes in the moral and ethical domain. Lesthaeghe stresses the emergence of secularism, materialism, individualism, and self-fulfillment as dominant values that in combination undermine the satisfactions derived from having children. John C. Caldwell argued in 1982 that a shift in the morality governing family life–in particular, a higher valuation of the conjugal relationship and of investments in children–leads to a dismantling of high-fertility reproductive regimes. Fertility decline is triggered by emotional nucleation of the family, itself a response to broader economic and cultural changes. For both scholars, the critical cultural change has less to do with the value of children narrowly defined and more to do with the nature of intergenerational relations and the perceived contribution of childbearing to the achievement of a desired standard and style of living. But whether changing mentalities and moralities about family life are themselves a sufficient cause of sustained and substantial fertility decline is doubtful, absent the precondition of mortality decline. Certain cultural changes, of course, might provoke both mortality and fertility declines, for example an increase in the value placed on investments in children, per child.

A final cluster of determinants of the timing and pace of fertility decline can be gathered under the heading "costs of birth control." The argument is that various economic, social, psychic, and health factors can make birth control practices prohibitively costly, and hence the reduction or elimination of such costs is a prerequisite for fertility decline. Ansley Coale and Richard Easterlin both highlighted the potentially important causal role of the costs of birth control, and the empirical record now contains numerous studies that demonstrate that reduction in birth control costs can accelerate fertility decline. In the period since 1960, the most prominent strategy for reducing birth control costs has been the provision of contraceptives free of charge or at nominal price through public and private family planning programs. But limited access to contraception is by no means the only obstacle to use, and some scholars have argued that personal knowledge and social legitimacy of contraception are perhaps more critical than the mere provision of contraceptive technology.

Conclusion

The debate that began in the 1950s and is still continuing about the aims of population policy and the nature and scope of interventions can be viewed as a debate about how to weight the various determinants of demographic transition. If one follows Notestein's reading of the European historical experience, then the decisive factors are social and economic change, and the availability of contraceptive technology is of little importance. Davis as well leaves one less than sanguine about the likely contribution of programs that make family planning and reproductive health services more accessible and less expensive. Coale and Easterlin provide a stronger rationale for investment in such programs. Rapid population growth–a function of the gap between mortality and fertility declines–has been a primary public policy concern. But the demography of pre-transition and post-transition populations differ in many other respects–posttransition, the age-structure of the population is older, individuals' lives are far lengthier, and childrearing occupies a much smaller portion of those lives. These outcomes of demographic transition increasingly are the focus of public policy debates about population dynamics.

DEMOGRAPHIC TRANSITION

History of the Concept

The Demography of Demographic Transition

Explanations for Demographic Transition

Conclusion

BIBLIOGRAPHY