DEMOGRAPHY AND DEMOGRAPHIC TRENDS

DEMOGRAPHY AND DEMOGRAPHIC TRENDS. Demography is the study of the growth, structure, and movement of human populations. It focuses on enumerations (censuses), which take stock of a population at a moment in time, and also flows of vital events—births, deaths, marriages, and migratory movements. These two sources furnish the basis for the statistical underpinnings of this article.

The Census

The demographic history of the United States can readily be divided into two segments based on the availability of "modern" demographic data, mainly before and after 1790, the date of the first federal census. The United States was the first nation to have mandated regular census enumerations. The Constitution specified a census every ten years for apportionment of the federal House of Representatives. From a very modest enumeration of heads of households with only a few questions in 1790, the census grew into a large-scale operation. In 1850, the census became nominal; that is, every person was enumerated separately by name, instead of a summary of information by household (as had been done for the censuses of 1790 to 1840). In 1820, a preliminary effort was made at an economic census of manufactures. From the 1840 census onward, censuses of manufactures and agriculture have been taken regularly. Mining and minerals were usually counted along with manufactures. Beginning in the late 1920s, wholesale, retail, and service establishments were also enumerated. Later, censuses of governments were undertaken. The population census is still

taken decennially in years ending in zero, but the economic and government censuses are taken more frequently.

Vital Processes

The study of vital processes has been more difficult. Un-like census enumeration, registration of births, deaths, and marriages was left to state and local governments. Consequently, it was done unevenly. Some cities (New York, Boston, Philadelphia, Baltimore, and New Orleans) were already registering deaths by the early nineteenth century. Massachusetts was the first to institute statewide vital registration in 1842. A federal Death Registration Area was created in 1900 with ten states (the six New England States, New York, New Jersey, Indiana, and Michigan) and the District of Columbia, all of which were deemed to have had at least 90 percent completeness. A parallel Birth Registration Area was only set up in 1915 with ten states (the six New England states, New York, Pennsylvania, Michigan, and Minnesota) and the District of Columbia, again with the criterion of at least 90 percent registration completeness. Both areas were only completed in 1933 with the admission of Texas. By the 2000s, the nation still lacked comprehensive registration of marriages and divorces. For international migration, statistics have been collected at major ports of entry to the United States since 1819. Arrivals across land borders were not regularly counted until 1906, while alien departures were only reported from 1907 to 1957. No direct registration of internal migration has been undertaken. From 1850, the federal census has asked a question on place of birth (state within the United States, country if foreign-born) as well as current residence, which allows a view of "life-time" migration to that point. A question on residence five years prior to the census was instituted in 1940. Since 1947, the Census Bureau has collected monthly data on a variety of detailed demographic, social, and economic topics with an interview system know as the Current Population Survey.

To determine demographic history before the "statistical era" (1790 to the present), other sources must be used. There were several dozen colonial and early national state censuses. In addition, family reconstitutions have been undertaken using parish records, genealogies, tax lists, muster rolls, and other such sources. We know more about the New England and Middle Atlantic colonies than those south of the Chesapeake. The population of the British North American colonies increased from several hundred Europeans in the early seventeenth century to about 2.5 million by 1780 (2 million whites and about half a million blacks). In contrast, the Amerindian population of the region experienced a serious decline, from about 1.1 million in 1650 to about 700,000 in 1780. This decline continued into the early twentieth century as a consequence of new diseases, warfare, loss of economic territory, and changes in way of life. (See Table 1.)

Since 1790, the American population grew from about 4 million (about 3.2 million whites and about

Table 1

760,000 blacks, of whom about 700,000 were slaves) to about 281 million in 2000. This represents an average annual growth rate of 2 percent per year. Over the long run, natural increase (births minus deaths) has accounted for approximately three-quarters of this growth, while net in-migration contributed about one-quarter. Net in-migration is the difference between total gross inmigration and total gross out-migration.

Race and Ethnicity of the Population

The shares of the population by race and ethnicity have varied over time. Prior to 1860, only whites and blacks were enumerated. For the first several decades of the republic, whites comprised 81 to 82 percent of the population. This share began to rise as white immigrants from Europe contributed substantially to population growth, while the black population grew almost entirely through natural increase, since the slave trade was officially ended in 1808. In the decades of the 1850s and the 1900s, the contribution to population growth from immigration rose as high as 33 percent. The share of the white population peaked at about 90 percent in the period 1930 to 1950, after which it began to decline. This was related to shifts in immigration from Europe to Asia, Latin America, and Africa. The 2000 census indicates that whites constitute 77 percent of the population, blacks 13 percent, and other races 10 percent, of which Amerindians are 1.5 percent and Asians 4.5 percent. Also, Hispanics (of all races) are about 13 percent of the total, up from about 1 percent in 1910. (See Table 2.)

A related issue is the foreign-born population. The 1850 census revealed that, at that moment, about 10 percent of the American population had been born abroad. This peaked at 15 percent in 1910, from which point the share diminished to 4.8 percent in 1970. Thereafter, changes in immigration legislation have resulted in large net inflows of migrants, such that the share of the foreign born as of 2002 was about 9.5 percent.

Population Distribution

An important feature of American history is the redistribution across space. In 1790, only 5 percent of the population resided in urban areas, defined as incorporated places of 2,500 persons and over. By the time of the Civil War (1861–1865), this had risen to 20 percent. The census of 1920 was the first to show that more than half of the American population was urban, a fact that led to such a protracted congressional conflict over reapportionment that the 1920s was the only decade in which the House of Representatives was not redistricted. The official urban proportion stood at 75 percent in 1990, but this tends to understate the true urban population because of as-yet unincorporated areas growing up on the fringes of metropolitan areas and other urban places. (See Table 2.)

There have been shifts in regional shares of population. The original thirteen states were spread along the Atlantic Coast, mostly east of the Appalachian Mountains. Population was about evenly divided between the Northeast (New England and Middle Atlantic States) and the South Atlantic and East South Central regions. The well-known westward movement to the frontier (mostly along east-west latitudes), as well as immigration from Europe, which went mostly to the North, began to shift the balance. By 1860 the North (the Northeast, Midwest, and West regions) had about 65 percent of the population, while the South had only about 35 percent. The westward movement ended in about 1890 and was replaced by an increasing rural-to-urban migration. In the late twentieth century this turned to a movement from central cities to suburbs and a regional migration to the "Sun Belt." About half of the American population lived in suburban places and another quarter in central cities. In 2000, 19 percent of the American population resided in the Northeast, 23 percent in the Midwest, 22 percent in the West, and 36 percent in the South. The state of California alone held 12 percent of the nation's people.

Immigration and Population

The United States has long been an immigrant destination. Overall, between 1821 and 1997, approximately 64 million immigrants officially entered the United States. Between 1820 and 1920, a period of virtually unrestricted

Table 2

immigration, approximately 33.7 million immigrants were counted. Serious immigration restriction began with the Literacy Test Act of 1917 and continued with the Emergency Immigration Act of 1921 and the "National Origins" Act of 1924. Immigration quotas were established and continued, in modified form, until 1965. The depression decade of the 1930s was the only period in U.S. history to show net out-migration. After World War II (1939–1945), however, immigration once more picked up. With modification in immigration laws, notably the 1965 Amendments to the Immigration and Nationality Act, and with increasing numbers of undocumented aliens (especially from Mexico and other areas of Latin America), an estimated 25.5 million immigrants entered between 1945 and 1997.

There have been two major shifts in the composition of immigrants. The first took place in the 1880s when the focus of departures from Europe shifted from northern and western European nations (Britain, Ireland, Scandinavia, and Germany) to southern and eastern European nations and areas (Italy, Austria-Hungary, eastern Germany, Russia, the Balkans, Spain, Portugal). For the period from 1821 to 1890, northern and western Europe furnished 82 percent of immigrants, but for the following three decades (1891–1920), southern and eastern Europe contributed 64 percent. The second great shift occurred in the 1960s and 1970s, when the focus of departures shifted from Europe (89 percent of all arrivals in 1821–1920) to Latin America, Asia, and Africa (83 percent of all immigrants in 1971–1997).

The Demographic Transition

Every developed nation has undergone a process known as the demographic transition from high to low levels of fertility and mortality. The United States began its transition at least in the early nineteenth century, but America was also distinctive. First, whereas its fertility transition began in the late eighteenth or early nineteenth century at the latest, other Western nations began their sustained fertility declines in the late nineteenth or early twentieth century. The sole exception was France, whose decline also began about the same time as the United States'. Second, the fertility rate in the United States commenced its sustained decline long before that of mortality. This may be seen in the fertility measures in Table 3, including the estimated crude birthrate and the wholly census-based child-woman ratios. This contrasts with the stylized demographic transition in which mortality decline precedes or occurs simultaneously with fertility decline. American mortality did not experience a sustained and irreversible decline until about the 1870s. Third, both processes were influenced by America's very high level of net inmigration and also by the significant population redistribution

Table 3

to frontier areas and later to cities, towns, and suburbs.

In the late eighteenth century, birthrates for the white population were quite high by European standards, with the crude birthrate (births per 1,000 population per year) in the range of 40 to 55. This attracted comment by Thomas Malthus, Benjamin Franklin, and other observers. Mortality was moderate, with crude death rates (deaths per 1,000 population per year) likely in the range of 20 to 40. Table 3 shows the sustained decline in white birthrates from at least 1800 and of black fertility from at least 1850. Family sizes (as indicated by the total fertility rate) were large early in the nineteenth century, being approximately seven children per woman at the beginning of the century and between seven and eight for the mainly rural slave population around 1850. The decline continued uninterrupted until the late 1940s, when the United States experienced the rather unexpected postwar baby boom. Birthrates rose and peaked in the late 1950s (with total fertility rates in the range of 3.5 to 4.5 births per woman). Thereafter, the fertility decline began again until it reached relatively stable levels in the 1990s (with total fertility rates of about 2 births per woman).

Table 3 reveals that mortality did not begin to decline until about the 1870s or so. Previously, death rates fluctuated, being affected by periodic epidemics and changes in the disease environment. There is now evidence that death rates rose in the early nineteenth century, likely in response to rapid urbanization, the nationalization of the disease environment, and possibly worsening distribution of income. A measure of the biological standard of living, adult heights, exhibited declines for the age cohorts born in the 1830s and 1840s. Since a high incidence of infection early in life is related

to shorter final heights, this supports the view of a worsening disease environment.

Table 3 also shows that American blacks had differentially higher fertility and higher mortality relative to the white population, although both groups experienced the fertility and mortality transitions. Both participated in the baby boom as well as the subsequent resumption of birth-rate declines in the 1960s.

Structural explanations for the fertility transition have involved the rising cost of raising children because of urbanization, the growth of incomes and nonagricultural employment, the increased value of education, rising female employment, child labor laws and compulsory education, and declining infant and child mortality. Changing attitudes toward large families and contraception, as well as better contraceptive techniques, have also been cited. Late twentieth century literature suggested that women were largely responsible for much of the birthrate decline in the nineteenth century—part of a movement for greater control over their lives. The structural explanations fit the American experience since the late nineteenth century, but they are less appropriate for the fertility decline in rural areas prior to about 1870. The increased scarcity and higher cost of good agricultural land has been proposed as a prime factor, although this is controversial. In addition, the increased secularization of social values has been hypothesized as playing an important part in convincing families that they could have control over important processes such as fertility. The standard explanations also do not adequately explain the post–World War II baby boom and subsequent baby bust. More complex theories, including the interaction of the size of generations with their income prospects, tastes for children versus material goods, and expectations about family size, have been proposed.

The mortality decline since about the 1870s seems to have been the result particularly of improvements in public health and sanitation, especially better water supplies and sewage disposal. The improving diet, clothing, and shelter of the American population over the period since about 1870 also contributed to improving health behaviors. Specific medical interventions beyond more general environmental public health measures were not statistically important until well into the twentieth century. It is difficult to disentangle the separate effects of these factors. But it is clear that much of the decline was due to rapid reductions in specific infectious and parasitic diseases, including tuberculosis, pneumonia, bronchitis, and gastrointestinal infections, as well as such well-known lethal diseases as cholera, smallpox, diphtheria, and typhoid fever. Nineteenth-century cities were especiallyun-healthy places, particularly the largest ones. This began

to change by about the 1890s, when the largest cities instituted new public works sanitation projects (such as piped water, sewer systems, filtration and chlorination of water) and public health administrations. They then experienced rapid improvements in death rates. As for the present, rural-urban mortality differentials have converged and largely disappeared. This, unfortunately, is not true of the differentials between whites and blacks.

Aging and Marriage

One of the consequences of the decline in fertility has been the aging of the population. Population age structure depends mostly on fertility rates: high-fertility populations are younger and low-fertility populations are older, provided that migration does not intervene in a major way. Table 2 shows that the median age of the population rose from about 16 years in 1800 to almost 33 years in 1990. Since male mortality is usually higher at all ages than female mortality, an older population tends also to have a preponderance of females. The final column of Table 2 indicates that the sex ratio (males per 100 females) declined during the twentieth century from well over 100 to about 95. The ratio was kept high through 1910 because of sex selectivity of immigration—more males than females entered as migrants up through World War I.

Although no data on marriage are presented in the tables, the United States has experienced some major changes in nuptiality since 1800. Marriage in colonial North America was notable for being early (for women) and marked by low percentages never marrying. This was different from the distinctive northwest European pattern of late marriage and high proportions never married, although the underlying family formation pattern ("neo-local," in which newly married couples form new households) was the same in both colonial North America and the areas of origin of this population. Thus, Malthus was correct. Abundant resources rather than basic behavioral differences made early and extensive marriage possible in the colonies.

Between 1800 and the early twenty-first century there were long cycles in nuptiality. Since about 1800, female age at first marriage rose from relatively low levels to a peak around 1900. The female age at first marriage was less than 20 in 1800, but rose gradually to about almost 24 years (and about 27 years for males) in 1900. Thereafter, a gradual decline began, with a trough being reached about 1960 at the height of the baby boom, when the female age at first marriage had fallen to 20.3 years and the male age to 23.4 years. There then began another, and rapid, upswing in both male and female marriage ages. Proportions never married at ages 45 to 54 replicated these cycles with a lag of about 20 to 30 years. Male nuptiality patterns generally paralleled female patterns. Male marriage ages were higher than those of females

with proportions never marrying also usually higher. Black ages at marriage moved from being lower to being higher than those for whites. So, for example, in 1990 white females married for the first time at an average age of 24.8 years, while black females married at an average age of 28.7 years. Marriage behavior of the Hispanic population looked rather more like that of the white population. Much of the change was due to delayed childbearing, more single-parent households, and more couples living in nonmarital unions and for longer periods of time.

Much of the work of the discipline of demography involves creation, evaluation, and analysis of a variety of data sources. The creation of standard demographic measures, such as those found in Table 3, is a frequent goal. Historical demography, however, faces a number of different challenges. Historical demographers work in areas and on time periods for which regular censuses, surveys, and/or vital registration do not exist or are quite deficient in coverage, accuracy, or completeness. In the case of the United States, fortunately there have been records of decennial federal censuses from 1790, as well as some state and local censuses. With the exception of the 1890 census, the original manuscript schedules of the federal censuses have survived virtually intact in the National Archives. The Integrated Public Use Microsample (IPUMS) project at the University of Minnesota has created or improved microsamples for the censuses of 1850–1880, 1900–1920, and 1940–1990. As of 2002, samples of the 1930 census were under way, and the Census Bureau was producing samples for the 2000 census. These are enormously valuable resources for historical population re-search. But prior to 1850, the federal censuses were not nominal and hence have not been sampled. Before 1880, there were no questions on the schedules regarding marital status or relationship to head of household. As already mentioned, registration of births and deaths did not cover the entire nation until 1933. For the colonial period, while there were some censuses, they had quite limited information. Civil vital registration was not in place.

Consequently, a variety of sources and methods have been used. For data, parochial or town registers of births, deaths, and marriages; genealogies; wills and probates; military records; tax lists; and college and school records are among the sources used. The technique of family re-constitution, using parochial or local records and genealogies, was developed by Louis Henry and his colleagues in France in the 1950s and 1960s. Although it is enormously time-consuming and labor-intensive, it has been used for a variety of community studies. From these we know a good deal about fertility, mortality, marriage, and family structure in the New England colonies, rather less about the middle colonies and the Chesapeake, and least about the South.

Techniques of indirect estimation of fertility and mortality have also been created, originally for use in developing nations, which have missing or deficient data. These include "own children" methods, which have been applied to the microsamples of the federal censuses of 1900 and 1910 to estimate fertility and mortality for the whole nation. Age-specific marital rates and overall birthrates are estimated from the microsamples of the 1850–1880 censuses. Questions on deaths in the household in the year prior to the census in the censuses of 1850–1900 have been used to create life tables for the United States for that time period. Some of these indirect estimates appear in Table 3.

Work is continuing on such topics as fertility and marriage, migration, social mobility, rural and urban environments and their influence on demographic behavior, and differentials between blacks and whites, between the native born and the foreign born, and among specific ethnic groups in various settings. The availability of small area data (counties, cities, and towns) allows matching of macrodata to local conditions. These same small area data are being placed in a National Historical Geographic Information System (GIS) for aggregate analysis. Advances in computing technology and data storage and retrieval have made this work much easier and more accessible to a wider variety of researchers. And there can be expected to be further cross-fertilization from the disciplines of demography, sociology, history, economics, geography, anthropology, and others that should inform the work in historical demography methodologically and theoretically and with more and better data.

BIBLIOGRAPHY

Anderson, Margo J. The American Census: A Social History. New Haven, Conn.: Yale University Press, 1988.

Daugherty, Helen Ginn, and Kenneth C. W. Kammeyer. An Introduction to Population. 2d ed. New York: The Guilford Press, 1995.

Haines, Michael R. "Economic History and Historical Demography: Past, Present, and Future." In The Future of Economic History. Edited by Alexander J. Field. Boston: Kluwer-Nijoff Pub., 1987.

———. "Long-Term Marriage Patterns in the United States from Colonial Times to the Present." The History Of the Family: An International Quarterly 1, no. 1 (1996): 15–39.

———. "The American Population, 1790–1920." In The Cambridge Economic History Of the United States. Edited by Stanley Engerman and Robert Gallman. Vol. 2. New York: Cambridge University Press, 2000.

Haines, Michael R., and Richard H. Steckel. A Population History of North America. New York: Cambridge University Press, 2000.

Preston, Samuel H., Patrick Heuveline, and Michel Guillot. Demography: Measuring and Modeling Population Processes. Malden, Mass.: Blackwell Publishers, 2001.

Shryock, Henry S., Jacob S. Siegel, and Associates. The Methods and Materials of Demography. Washington, D.C.: Bureau of the Census, U.S. Government Printing Office, 1971.

U.S. Bureau of the Census. Historical Statistics of the United States from Colonial Times to 1970. Bicentennial Edition. Washington, D.C.: Department of Commerce, Bureau of the Census, U.S. Government Printing Office, 1975.

Thompson, Warren S., and P. K. Whelpton. Population Trends in the United States. New York: McGraw-Hill, 1933.

United Nations. Indirect Techniques for Demographic Estimation. Manual X. New York: United Nations, 1983.

Vinovskis, Maris A., ed. Studies in American Historical Demography. New York: Academic Press, 1979.

Wells, Robert V. Revolutions in Americans' Lives: A Demographic Perspective on the History Of Americans, Their Families, and Their Society. Westport, Conn.: Greenwood Press, 1982.

———. "The Population of England's Colonies in America: Old Englishor New Americans?" Population Studies 46, no. 1 (March 1992): 85–102.

Willigan, J. Dennis, and Katherine A. Lynch. Sources and Methods of Historical Demography. New York: Academic Press, 1982.

Michael R.Haines

See alsoCensus, U.S. Bureau of the ; Immigration ; Migrations, Internal .