Population Cycles And Demographic Behavior Research Paper

By contrast, theoretical developments have proliferated rapidly during recent decades. Entirely new models have emerged to elucidate the interaction between the population and the economy. This has given rise to population economics as a new and vigorously expanding field of scholarly endeavor. Its reasoning is usually strongly rooted in traditional economic theory and often bears the imprint of Malthusian thinking. These developments have meant that the shaping of theory has taken quite another turn and has also become more diversified, with much greater breadth and depth than before. The major characteristics of modern population economics are discussed below.

First, interest has shifted from one demographic variable to another. The strong link which Malthus believed to exist between changes in the economic environment and mortality has been called into question. Mortality variations are influenced by a number of other factors, the means of subsistence and standard of living forming only one of these, and not always the most important one. Fertility—along with the factors which determine it, of which particular mention may be made of varying age at marriage, the use of birth control, abortion, and nutrition—has been regarded instead as being determined by economic factors in a way quite different from that found in Malthus. This does not apply only to modern society. Several studies have shown that even in preindustrial society, the link between fertility and economic factors was often stronger than the relation between mortality and the economy, in both the short and the long term.

Second, population variables have been linked more closely to economic theory, especially neoclassical micro theory, and have then usually been analyzed in a more stringent manner than is the case with the Malthusian theory. Here we may point to the ‘microeconomic theory of fertility’ or, as it is sometimes termed, the ‘new household economics’ or ‘new home economics.’ In this theoretical construction, fertility is viewed as the result of economically rational decisions (the demand for children). Established economic methods and conceptual mechanisms are employed in order to explain variations in fertility. For example, models are devised in which fertility is regarded as a function of income and the ‘price’ of children in relation to the price of other goods and services. In this reasoning children are regarded as ordinary goods, exchangeable in the choice situation for other goods and services. Fluctuations in fertility are explained using only prices and incomes. The pioneering theoretical work in this area was done by Gary Becker (1960) and was then elaborated, for example by Becker and Lewis (1973); it has often been cited as one of the basic explanations of fertility variations in the Western world since World War II. Models have also been formulated along highly neoclassical lines and have been based on a traditional investment approach: cost–benefit analysis. This method regards fertility as being determined by how costs and revenues associated with children change over time. From this standpoint, accordingly, the strong decline of Western fertility which set in at the end of the nineteenth century should be regarded as a consequence of the falling revenues and rising costs which having another child would entail for the parents.

Third, Malthus’ view that there is a one-way causality between the population and the economy has been abandoned. What is now emphasized is that the interaction between the population and the economy is considerably more complex than Malthus believed. In many instances, demographic events can be one of the most important factors underlying the process of economic change. In other words, the population factor may also, in many cases, be the independent variable, which in its turn influences the economy and the means of subsistence. According to this approach, numerous other elements of social change must be viewed in the context of changing population size and composition (e.g., Boserup 1965, Simon 1977, Easterlin 1980).

Fourth, the various models which have been developed to represent the interaction between the population and the economy have become considerably more dynamic in character. One important element marking this trend has been the wide-ranging research conducted into the incidence and causes of various population cycles and into their effects on social change. Intensive discussions have been conducted concerning the extent to which population cycles with varying durations and strengths can be identified; what demographic or other factors can be cited as explanations of their genesis; and what effects the various population cycles have had on demographic behavior and on the national economy as a whole.

2. The Mechanisms Of Population Cycles

The study of population cycles goes a long way back in time. Its origins can be traced back to the northern European countries in the nineteenth century. As early as in 1855 Ejlert Sundt, a Norwegian embryonic sociologist and statistician, observed that a strong rise and fall in a country’s birthrate is repeated during the next generation, though in a less marked fashion. On reaching marriageable age an extraordinarily large birth cohort will bring into the world more children than an age group of normal size. The same law applies in reverse: after one generation an undersized cohort will produce a new undersized group. As a result, regularly recurring waves or ‘echo effects’ are generated in the size and age composition of the population.

Ten years later, in 1865, the Swedish demographer Fredrik Berg paid tribute to Sundt for the idea of cohort analysis, and emphasized the importance of keeping different birth cohorts apart and studying them separately. He also argued that it ought to be possible to use variations in birth cohort sizes to compute the future number of births. Berg foreshadowed modern population projection techniques.

The Swedish demographer Gustav Sundbarg was a pioneer in expressing the view that a substantial change in birthrates may have far-reaching repercussions, not only on the economy but also on institutional and social change for a long time ahead. He may also have been the first to point out that the size of a cohort can have an impact not only on individuals’ demographic behavior, but also on their behavior in numerous other spheres. Sundbarg developed these ideas in his analysis of the strong rise in the number of births in Sweden between 1815 and 1830, and its aftereffects. Little by little the rise in births transmitted itself to higher age levels and grew in strength compared with earlier cohorts because mortality was also falling dramatically. Sundbarg believed that this rise had a far-reaching influence in many areas of social and economic development in Sweden all through the nineteenth century. He wrote in 1903:

As they reached school age these cohorts filled up all the school rooms so that their education and upbringing became a burning issue. As they reached working age they shattered the guilds (in the 1840s) and compelled the introduction of a more liberal trade legislation, without which these masses of people would have been unable to survive. As they reached marriageable age their large numbers enforced delayed marriage, and in the end (though not until the second generation) the great emigration movement. The immense class of ‘all others’ who were excluded from democratic representation at the time destroyed the constitution based on the four estates (in the 1850s and 1860s). And in our own day, having reached old age they fill our poorhouses and make the problem of disability and old age a matter of topical concern.

In 1937, the German economist August Losch noticed that sudden and dramatic changes in the population structure, caused by war for example, generated long waves in population growth, which in turn brought about far-reaching changes in the economy. Great wars are characterized by a deficit of births during the war and a surplus of births in the immediate postwar period. In a now-classical article in the Quarterly Journal of Economics, ‘Population Cycles as a Cause of Business Cycles,’ Losch described how these features caused a population cycle which in turn gave rise to strong fluctuations in German economic growth and in the business cycle.

This discussion of the origins of population cycles and their influence on demographic and economic factors transmitted itself to the American scholarly arena after World War II. American researchers had been able to identify long cycles in US economic activity since the middle of the nineteenth century. How these cycles were generated and propagated became the object of extensive empirical and theoretical studies by American scholars during the 1950s and 1960s. They found that rather than being a passive response to changes in the economy, population cycles played an active role in determining the long cycles that had been observed in US economic activity. These long cycles could be found not only in aggregate production growth but also in most of its subcomponents. This made it possible to identify long cycles in the growth of the labor force, in capital formation, in productivity, and in utilization of capacity. In particular, there was an extensive discussion on the existence of, and mechanisms underlying, Kuznets cycles—named after Simon Kuznets—or ‘long swings,’ as they are often called. Models were devised to show how changes in the age structure and population growth over about 20 years generated long-term fluctuations in investment activity, mainly in building construction, and corresponding fluctuations in economic growth. In explanations of the underlying mechanisms the primary stress was laid on the interaction between demographic and economic factors. In this way long cycles in marriage frequency, birthrate, and immigration were seen to give rise to long cycles in both the growth and the utilization of resources through their effect on such factors as capital formation, housing construction, and labor supply. The causal relation in the reverse direction was also important. Through the link with unemployment and effective demand, changes in the utilization of resources intensified immigration and gave rise to long-term cycles in the pattern of immigration. In the same way, the variations in unemployment influenced job opportunities for the labor reserve (mainly married women, ethnic minorities, and older workers). In other words, there was a long-term cycle in the size of the labor supply. A side result of this wide-ranging discussion of the interaction between demographic and economic factors was that economists increasingly incorporated demographic variables into their reasoning.

After World War II the long-term cycles assumed a different character both in duration and strength. A number of the suppositions applied to earlier periods of history were changed. Some of the mechanisms which formerly generated Kuznets cycles changed character and others weakened or vanished. In particular, the great influence that immigration to the USA had exerted in the generation of Kuznets cycles after 1850 was modified. When immigration was restricted during the interwar period, its cycle-generating function vanished. Instead, this function was replaced by the marked rise in the number of births during the 1940s and 1950s and by an equally marked decline during the next two decades (e.g., Kuznets 1958, Abramovitz 1961, 1968, Easterlin 1968).

2.1 The Easterlin Hypothesis

One of the most prominent figures in this earlier research into the genesis of population cycles and their influence on demographic and economic phenomena was the American economic historian Richard Easterlin. It was also Easterlin who drew attention to the changed demographic conditions that applied to the Kuznets cycle after World War II. It therefore seems quite appropriate that it would be Easterlin who made the most trail-blazing analysis of the causes of the dramatic changes in births in the USA after World War II and their effects on the demographic and economic behavior of individuals. A number of other features of modern social change must also be viewed, he argued, in the context of demographic events.

A group of people born during a given period are marked by the same code of values and attitudes in many respects. Easterlin believes it possible to identify an important underlying reason why values and attitudes may vary from one cohort to another. Accordingly he asserts that an individual’s behavior in a number of areas will be influenced by the size of the cohort to which he or she belongs, since it determines economic situation and future prospects to such a large extent. At this point Easterlin ties in with earlier thinking by Gustav Sundbarg and others, but he conducts the analysis with considerably greater precision and stringency (e.g., Easterlin 1968, 1978, 1981, Artzrouni and Easterlin 1982.)

For young individuals who belong to a large cohort, prospects for the future appear somber. Their incomes will be lower in general and the likelihood of becoming unemployed is higher than for individuals belonging to a small cohort. As a consequence they marry later and also divorce more frequently. Besides they have a greater propensity to commit suicide and to manifest higher frequency of criminality. Married women in a large cohort will participate in the labor force more often because of the family’s difficult economic situation. Unmarried women have high nonmarital fertility, also resulting from social stress factors. Alcoholism is also regarded as more common in a large cohort. Values, too (political opinions, attitude to work, savings, etc.), depend on cohort size.

One key element of Easterlin’s reasoning is the ‘relative income hypothesis.’ This holds that the individual’s behavior is determined by the relation between, on the one hand, the material standard and position in society which he or she has attained, and on the other hand, what he or she had expected. The expectations are in high degree determined by family background, i.e., by the parents’ economic situation, while the possibilities of translating them into reality are dependent on the prevailing labor market situation and relative wages. In other words, it is not the absolute level of living standard that determines behavior, but the relative level, in which the crucial factor is the relation between expectations and the possibilities of realizing them (material aspirations). During a period when high aims are set but the possibilities of realizing them are low, as is the case with a large cohort, the social stress factors are magnified. The stresses and strains associated with large cohort size are likely to continue throughout the whole course of a cohort’s life.

In a small cohort the reverse applies. Firms have to compete for the relatively few new entrants to the labor market, which means that starting wages become high and opportunities of advancement good. Social stress is reduced considerably, and divorce, suicide and criminal frequencies become low. Moreover, in a small cohort expectations are low, because they are determined by the parents’ (the preceding large cohort’s) economic situation. On the other hand the possibilities of realizing expectations—and even of surpassing them—are high. Here, too, it happens that these favorable circumstances follow the small cohort all through its life.

Easterlin makes a particular point of analyzing how cohort size influences childbearing. He observes that a large cohort will bear relatively few children per woman while a small cohort will give birth to many. The linkage between birthrates and adverse economic and social effects causes the emergence of ‘crowding mechanisms’ operating within three major social institutions: family, school, and labor market. The result of the ‘crowding mechanisms’ is that the appearance of social stress factors are quite different in large and small cohorts. A large cohort adapts in order to avoid deterioration of its living standard compared with that of its parents. One such adaptation is that family formation takes place later in life. Another is that a woman in such a situation finds herself compelled to combine labor force participation with childbearing, which reduces her natality. By means of such demographic adaptation the large cohort strives to maintain its economic status compared with that of its parents. For a small cohort the situation is again reversed. Its members have no need to make any demographic adaptation in order to preserve an economic status that is favorable from the outset.

These demographic mechanisms produce what is called the ‘Easterlin effect,’ and they have been used as an alternative to the explanation suggested by the ‘microeconomic theory of fertility’ for the large fertility variations observed in the USA and Western Europe during the postwar period. The most essential difference between the two approaches is that, according to the Easterlin model, demographic behavior is determined by incomes and consumption relative to expectations, whereas in the neoclassical theory demographic behavior is determined by absolute incomes and actual levels of consumption. Where the neoclassical economists regard the cost of children as a response to exogenous changes in female wages, Easterlin regards tastes as response to exogenous changes in relative income status. Instead of assuming a constant preference structure as in Becker’s theory, Easterlin introduces a shifting one; the shift being determined by incomes and prices. The Easterlin model is regarded by many as an attempt to introduce a sociological mode of thought into economic theory.

The observation that large cohorts would give birth to relatively few children and small cohorts to many children also creates the potential for continuing fluctuations in the relative size of birth cohorts. As a result there emerges a self-generating and to a certain extent self-correcting fertility cycle of around four to five decades in length and with associated changes in a wide range of social and economic phenomena induced by fluctuations in relative cohort size. Pre-World War II Kuznets cycles of 15–25 years caused by aggregate demand fluctuations would be replaced, according to Easterlin’s argument, by post-World War II swings of double the length originating in fertility changes operating with a lagged effect on the age structure of labor supply. Some scholars have included such population cycles in mathematical models (e.g., Lee 1974, Samuelson 1976, Chu and Lu 1995). With the relative income hypothesis as a starting point, Easterlin describes inter alia important elements of social change in the USA during the postwar era, his main objective being to explain the baby boom of the 1940s and 1950s and the baby bust of the 1960s and 1970s. As a result of low birthrates during the interwar period—and of low immigration—relatively few young people entered the labor market during the 1940s and 1950s. Labor market conditions were favorable, since firms were competing fiercely for the limited numbers on the labor market by such means as increased starting wages. The future beckoned brightly too, for the prospects of advancement appeared good in the prevailing conditions of labor shortage. As a result of the fall in the proportion of young people in the total labor force during the 1940s and 1950s, the relative incomes of the young cohorts increased during this period at the same time as fertility was rising.

Conditions for the next generation were quite the opposite. Because of the high relative incomes of their parents and the subsequent rise in fertility, the new generation was comparatively large and at the same time entertained expectations of a high material standard. Thus, when the new generation emerged on the labor market during the 1960s and 1970s its expectations were high. Because of the cohort’s size, starting wages were low, the likelihood of becoming unemployed was high, and opportunities for advancement seemed rather limited. The chances of fulfilling their high expectations therefore appeared small. As a consequence, families were formed later, and childbearing was reduced. In other respects, too, as regards marriage and divorce patterns for example, female labor force participation rates, nonmarital fertility, suicide and criminal frequencies, and political alienation (students riots in the late 1960s, etc.), the 1960s and 1970s exhibited characteristics quite different from those of the previous two decades. Easterlin also predicted that a repetition of the baby boom would take place during the 1980s and early 1990s. In reality, however, fertility remained largely unchanged in these small cohorts.

2.2 Criticism Of The Easterlin Hypothesis

Though much admired, the Easterlin hypothesis has not been exempt from criticism. Some scholars contend that the analysis is conducted at far too high a level of aggregation. They assert that the information available on the individual level concerning numbers of children and relative wages at a given time does not substantiate the Easterlin hypothesis, though there also are micro studies that report findings consistent with it. A number of scholars maintain that the theoretical ground for believing that changed relative incomes should influence fertility, labor force participation rate, and so on, are not particularly well developed either. Easterlin is also said to fail to take account of women’s contribution to family income. If men’s relative incomes are low at the same time as the labor force participation rate of women is high, it is fully conceivable that household incomes—which are the relevant factor at this point—are even higher than if the man had a relatively high income and the woman participated little in the labor force.

Other scholars argue that Easterlin’s view of the modus operandi of the labor market is too simple and not rooted in any accepted labor market theory. In particular, one has criticized the idea that substitutability between older and younger workers is as limited as the Easterlin hypothesis holds. Easterlin is even accused of being careless with empirical data. He is said to pay too little attention to alternative explanations and to let his hypothesis rest on fragile empirical foundations, underpinning it only by one observed cyclical covariance between age structure relative incomes, and fertility.

Furthermore, the efforts made by Easterlin to link varying economic conditions to individual behavior in fields other than that of childbearing have not had any profound impact on current research. The studies which have been carried out give incomplete or conflicting results.

Nevertheless, the view that generation size significantly affects the attitudes and actions of individuals in a variety of areas, especially fertility behavior, has formed the basis of a number of international studies. As noted above, the relationships postulated by Easterlin seem to fit rather well for the USA. Some other countries (Australia, New Zealand, Norway, Canada) show similar patterns. In other countries there is a connection, but it is considerably weaker (Finland, France, The Netherlands, England and Wales, Sweden). For certain countries (Germany, Italy, Denmark) no such connection has been found at all. It should be observed, however, that usually it is only the reduced relation between age structure and fertility that has been studied, which is a ‘second best’ solution in this connection. Empirical studies have had to be based on age-specific incomes of different age groups to test the hypothesis properly. (For an overview of the 76 empirical analyses of the ‘Easterlin hypothesis’ that have been undertaken so far, and also of the criticism against the hypothesis, see Macunovich 1998.)

The first link of Easterlin’s key relation between age structure relative incomes and fertility, has shown itself to be the more tenable and has inspired some scholars to develop the basic lines of thought both empirically and theoretically. In Easterlin’s original analysis the relation is constructed on the basis of a relatively simple model of how the labor market functions. Thus, Easterlin does not pay much attention to possibilities for substitution between younger and older workers, nor does he differentiate between educational categories. An increased supply of younger labor will, therefore, result automatically in reduced relative incomes and increased unemployment for that younger element, all other things being equal. Later studies have shown with greater precision and detail how changes in the age structure have influenced relative incomes, and have established inter alia that cohort effects on incomes have been stronger for higher-educated categories than for lower. It has further been shown that the cohort effect on incomes has a tendency to wane over time, and that it is essentially absent for female workers. This suggests that the possibilities of substitution between younger and older women are large.

Easterlin’s thesis that economic conditions can vary very widely between different demographic groups, and that a group’s size influences its relative economic situation, has been clearly substantiated. But studies concerning the period after 1980 suggest that the effect applies only under special conditions. According to Easterlin, the relative income of the young ought to increase when the ‘baby bust’ cohorts entered the labor market in the late 1980s. Yet what one could observe was a continued decline in the relative wages of the young. It seems that changes in the supply of male workers may be able to explain the pattern of wages prior to 1980, but that a change in the structure of labor demand—in particular an increase in the demand for more educated labor—offers a better explanation of wage patterns after 1980.

Easterlin himself has noted that the fortunes of a cohort depend on a number of other factors besides cohort size alone. One would therefore expect cohort size effects to dominate a cohort’s experience only when other factors are relatively constant. This was the case in the period after World War II, when governmental monetary and fiscal policies stabilized growth of aggregate demand. In addition, the relative impact of immigration on the labor force was greatly reduced as a result of restrictive legislation adopted in the 1920s. In Easterlin’s view the general slowdown of economic growth after 1973 and the widespread retreat from full employment as an objective of macroeconomic policy are more important than changes in cohort size as factors explaining observed changes in fertility during the most recent decades. Furthermore, high immigration and the rise in female labor force participation also tend to limit the importance of birth cohort size, and so will government welfare programs that offer policies of full employment. According to Easterlin this is the reason why stronger cohort effects can be observed for the USA than for European nations, since the latter have more general welfare systems (see also Pampel and Peters 1995).

Just the same, it has proved possible to single out one factor—cohort size—that underlies variations in relative income and in fertility during certain periods and in some countries after World War II. This finding is moderate in scope, but solid and robust in its moderation. Only in exceptional cases can it be adduced as a dominant explanation of the variations observed. Nevertheless, the ‘Easterlin hypothesis’ and the many studies it has inspired, has given us new insights into the kinds of different connections and responses of economic, sociological, demographic, and psychological character which generate population cycles.

Bibliography:

1. Abramovitz M 1961 The nature and significance of Kutznets cycles. Economic Development and Cultural Change 9: 225–48
2. Abramovitz M 1968 The passing of the Kutznets cycle. Economica 35
3. Artzrouni M A, Easterlin R A 1982 Birth history, age structure, and post World War II fertility in ten developed countries: An exploratory empirical analysis. Genus 38(3–4): 81–99
4. Becker G S 1960 An economic analysis of fertility. In: Universities-National Bureau, Committee for Economic Research. Demographic and Economic Change in De eloped Countries. Universities-National Bureau Conference Series No. 11. Princeton University Press, Princeton, NJ
5. Becker G S, Lewis H G 1973 On the interaction between the quantity and quality of children. Journal of Political Economy 81(2): S279–88
6. Boserup E 1965 The Conditions of Agricultural Growth: The Economics of Agrarian Change Under Population Pressure. Aldine, Chicago
7. Chu C T C, Lu H C 1995 Toward a general analysis of endogenous Easterlin cycles. Journal of Populations Economics 8: 35–57
8. Easterlin R A 1966 Economic–demographic interactions and long swings in economic growth. American Economic Review 16: 1063–104
9. Easterlin R A 1968 Population, Labor Force and Long Swings in Economic Growth: The American Experience. Columbia University Press, New York
10. Easterlin R A 1978 What will 1984 be like? Socioeconomic implications of recent twists in age structure. Demography 15(4): 397–432
11. Easterlin R A 1980 Birth and Fortune. Basic Books, New York
12. Easterlin R A 1987 Easterlin hypothesis. In: Eatwell J, Milgate M, Newman P (eds.) The New Palgrave: A Dictionary of Economics 2. The Stockton Press, New York, pp. 1–4
13. Easterlin R A, Schaeffer C M, Macunovich D J 1993 Will baby boomers be less well off than their parents? Income, wealth and family circumstances over the life cycle in the United States. Population and Development Review 19(3): 497–522
14. Kuznets S 1958 Long secular swings in the growth of population and in related economic variables. Proceedings of the American Philosophical Society CII: 25
15. Lee R 1974 The formal dynamics of controlled populations and the echo, the boom and the bust. Demography 11: 563–85
16. Macunovich D J 1998 Fertility and the Easterlin hypothesis: an assessment of the literature. Journal of Population Economics 11: 53–111
17. Pampel F C, Peters H E 1995 The Easterlin eff Annual Review of Sociology 21: 163–94
18. Samuelson P A 1976 An economist’s non-linear model of self-generated fertility waves. Population Studies 30: 243–7
19. Simon J S 1977 The Economics of Population Growth. Princeton University Press, Princeton, NJ