POPULATION AGING: IT’S NOT JUST THE BABY BOOM By Alicia H. Munnell* AN ISSUE IN BRIEF Introduction The retirement of the baby boom — those people born between CENTER FOR 1946 and 1964 — is almost upon us. The leading edge of this famous cohort turns 62 in 2008. With the aging of the baby RETIREMENT boom, the population of the nation is about to gray rapidly over the next three decades. The purpose of this brief is to put the RESEARCH baby boom and its impact on population aging in perspective. Specifically, the baby boom is not the reason for the aging of AT BOSTON COLLEGE the population; the aging is the result of long-term trends of APRIL 2004, ISSUE IN BRIEF 16 increasing longevity and declining fertility. The bust-boom- bust pattern in fertility rates that resulted in the baby boom inside simply changes the path to an older society. Thus, the baby boom is not “a pig in a python,” a somewhat graphic metaphor introduction ..............................................1 frequently used to suggest that the large cohort is just passing THE AGING OF THE POPULATION ......................1 through, and life will return to normal once the last member THE BUST-BOOM-BUST IN FERTILITY RATES ........5 dies. Rather, the nation is facing a permanent change in its CONCLUSION ....................................................7 demographic profile. REFERENCES ......................................................7 FIGURE 1. STATES WITH AT LEAST 18 PERCENT OF The Aging of the Population THE POPULATION 65 AND OVER, 2000 AND 2025................................................................2 In 2000, 12 percent of the U.S. population was age 65 or FIGURE 2. POPULATION DISTRIBUTION BY AGE older. By 2025, the share of the population aged 65 and over AND SEX IN THE UNITED STATES, 1880 AND is expected to rise to 19 percent, higher than the share of 2080 ..............................................................2 1 older residents in Florida today. This has led commentators FIGURE 3. IMMIGRATION TO THE UNITED STATES, to describe the future United States as “a nation of Floridas” 2 1900-2000......................................................3 (see Figure 1). FIGURE 4. FERTILITY RATES IN THE UNITED STATES, 1800-2080 ........................................3 FIGURE 5. FERTILITY RATES BY COUNTRY, 2000 ....3 * Alicia H. Munnell is the director of the Center for Retirement Research at FIGURE 6 . URBAN POPULATION AS PERCENT OF Boston College (CRR) and the Peter F. Drucker Professor of Management TOTAL POPULATION, 1790-2000......................4 Sciences at Boston College’s Carroll School of Management. The author FIGURE 7. INFANT MORTALITY IN THE UNITED would like to thank her colleagues at the CRR for data support and for STATES, 1850-2002 ........................................4 reviewing the draft. 1 U.S. Bureau of the Census (2000). FIGURE 8. LIFE EXPECTANCY AT AGE 65, 2 Peterson (1996). 1900-2080 ....................................................5 FIGURE 9. PERCENT OF THE U.S. POPULATION AGED 65 OR OLDER, 1860-2080......................5 2 Center for Retirement Research Figure 1. States with at Least 18 Percent of the Figure 2. Population Distribution by Age and Sex in Population 65 and Over, 2000 and 2025 the United States, 1880 and 2080 TODAY 1880 TOMORROW 2080 Source: Committee for Economic Development (1999). Updated with numbers from 2000 Census. At the same time, the nation is in the process of a dramatic long-term shift in the age structure of the population. Over the 200-year period between 1880 and 2080, the shape of the U.S. population will change from a pyramid to almost a rectangle as the Source: U.S. Bureau of the Census (1998). relative number of older people has increased and the relative number of children has declined (see Thus, while the population will age rapidly over Figure 2). Note that neither date that brackets this the next two decades, the aging of the population is 200-year period has anything to do with the baby not a new phenomenon. Indeed, the U.S. population boom generation; 1880 predates the first boomer by has been growing older since the dawn of the repub- more than six decades, and by 2080 virtually all the lic. This long-term trend is the inevitable result of boomers will have died. (The youngest boomers, two factors: (1) women have generally been having born in 1964, would be 116 years old in 2080.) fewer children than in previous generations; and (2) individuals have been living longer. These two trends reduce the number of young people in society and increase the number in older age groups. Issue in Brief 3 Immigration, the other key factor in U.S. demo- Figure 4. Fertility Rates in the United States, graphic history, also plays a role in long-term popula- 1800-2080 tion aging. Immigrants are disproportionately prime- age adults, and an influx of immigrants decreases the share of the population over age 65. More importantly, immigrants tend to have higher fertility rates than native born Americans. For example, the fertility rate among Hispanics in the U.S. today — a proxy for 3 immigrants — is 3.0 compared to 1.8 for whites. Thus, the decline in immigration after the First World War (see Figure 3) contributed to population aging in the mid- twentieth century. The recent rise will slow the process. Figure 3. Immigration to the United States, 1900-2000 Sources: Data prior to 1920: Coale and Zelnick (1963); 1920-1969: Bell (1997); 1970-2080: Social Security Administration (2003). Fertility rates, however, are hard to predict. Many European countries today have fertility rates well below two children (see Figure 5), which will result in declin- ing national populations and much older societies. U.S. fertility rates could look more like those in other developed countries if the number of immigrants, with their higher levels of fertility, falls short of projections. But for now, the consensus estimate is two children per woman, resulting in a stable population. Source: U.S. Department of Homeland Security (2002). Figure 5. Fertility Rates by Country, 2000 The Decline in Fertility The fertility rate measures, on average, the number of births for a woman throughout her childbearing years. The declining fertility rate is often perceived as a recent phenomenon — the baby bust that fol- lowed the post-war baby boom. In reality, as shown in Figure 4, the fertility rate in the United States has been falling for much of the past two centuries. In 1800, the average woman had 7.0 children. By the end of World War II, a century and a half later, the fertility rate was down to 2.4 children. The post-war baby boom — lasting from 1946 to 1964 — pushed Source: U.S. Bureau of the Census (2000). the rate back up to about 3.5 children. But it was a Note: These numbers from the census were obtained from the temporary phenomenon. By the mid-1960s, fertility censuses and surveys of other countries. began to head down sharply, dropping to a historic low of 1.7 children by the mid-1970s before bouncing Why was fertility so high, and why did it decline so back slightly and stabilizing at about two children, sharply? While individual couples make decisions which will keep the population roughly constant. about having children for a variety of reasons, many scholars who have examined the issue have found that economic factors often play a major role. At the beginning of the 19th century, fertility was much higher in the United States than in Europe and else- where. Most commentators attribute the high fertility rates to the availability of land in a frontier nation. 3 U.S. Bureau of the Census (1996). 4 Center for Retirement Research As long as the frontier existed, young families were Reinforcing the trend toward a declining demand continually taking over unimproved acreage, forcing for children was the drop in infant mortality. For farmers to spend most of their time clearing the land most of human history families had to produce a lot for planting. The need for this basic work to be done of children because many would not survive to adult- made children extremely valuable since any child hood. Significant improvements in public health over the age of six could participate in these tasks during the 19th and 20th centuries greatly increased year round. In a frontier nation, it was also easy for the chance of survival, which reduced the need for children to leave the family at an early age to develop additional births. For example, infant mortality was their own land. Thus, the need for basic labor and cut in half between 1880 and 1920 (see Figure 7). the likelihood that young adults would move on created a demand for lots of children. Indeed, studies Figure 7. Infant Mortality in the United States, show that the highest birth rates within the U.S. 1850-2002 occurred on the rural frontier, where land was cheapest 4 and labor scarcest. As the continent was settled and the frontier disap- peared towards the end of the 19th century, farmers had different needs. By 1920, owning and operating a family farm required more in the way of capital investment and relatively less labor, and the improved technology reduced the economic value of additional children. At the same time that the frontier was disappear- ing, people were moving off farms and into the cities. Between 1880 and 1900, the percent of the population living in urban areas increased from 26 percent to 40 percent (see Figure 6). By 1920, more than half Sources: Data prior to 1915: Montgomery and Cohen (1998); 1915- the population resided in cities. Men in towns and 1970: U.S. Bureau of the Census (1976); 1970-2002: Centers for cities faced the prospect of being unemployed, which Disease Control (2003). undermined security and discouraged childbearing. At the same time, the value of children as a resource During the late 19th century, the shift away from declined. They could not easily provide services to farming also had implications for the status of 5 the family, and meeting their need for food and women and the control they had over childbearing. clothing cost money. For a while, children could find In rural households, both husbands and wives worked employment outside of agriculture in textiles and at home, and the husband tended to control the house- other industries, but by 1920 most states had passed hold. As agriculture declined, men increasingly went laws that prohibited the employment of children outside the home to work and left women in charge under 14 and that required their attendance in school. of the household. Husbands frequently turned over their paychecks to their wives who managed the Figure 6. Urban Population as Percent of Total household budget. This shift in responsibility meant Population, 1790-2000 that women gained more control within the marriage, and presumably more control over reproduction. The rising educational attainment of women had a similar effect. Both increased the use of contraceptives, which reduced the number of births from the late 19th century forward. And in the early 20th century, advocates of family planning publicized the availability of birth control to new groups in the population. Thus, for a host of reasons — reduced economic need for children, higher likelihood of a child surviv- ing to maturity, and increased control by women over reproduction — fertility rates declined dramatically from the beginning of the 19th century. Viewed in this long-term context, the baby boom was a demo- Source: U.S. Bureau of the Census (1976, 2000). graphic blip that temporarily interrupted the decline in fertility, with the subsequent baby bust bringing fertility back to its long-term historic trend. 4 5 Easterlin (1968). Van Horn (1988). Issue in Brief 5 The Bust-Boom-Bust in The Increase in Life Expectancy The other demographic factor driving the aging of Fertility Rates the population is increased life expectancy. The gains The long-term trends in fertility and mortality, which over the last century have been as dramatic as the are causing the population to age, determine the pro- longer-term drop in fertility. They have shown less 6 jected ratio of the elderly to the working population fluctuation, however, as displayed in Figure 8. in, say, 2080. But the unprecedented speed at which In 1935, when Social Security was enacted and the the population will age over the next two decades is retirement age set at 65, life expectancy for individuals the result of the blip in fertility rates — the drop in at age 65 was about 12 years for men and 13 years for the 1920s and 1930s, the post-war boom, and the women. Today it is 16 years and 19 years, respectively. subsequent return to trend. By 2080, life expectancy at 65 is projected to be 20 years for men and 23 years for women. Moreover, the probability of a young worker surviving to retire- The Bust of the 1920s and 1930s ment has also risen dramatically. In the 1930s, the Right now, the nation is enjoying a “demographic probability of a 20-year old man surviving to age 65 holiday.” The over-65 population is growing very slowly, was only about 60 percent, while for women it was reflecting the low level of births during the 1920s and about 67 percent. By the mid-1990s, these fractions 1930s. At the same time, the non-elderly population is had increased to 77 and 87 percent respectively, and swelled by the ranks of the post-war baby boom genera- they are expected to continue rising in the future. 7 tion — those born between 1946 and 1964. The result is that the percent of the population aged 65 and over Figure 8. Life Expectancy at Age 65, 1900-2080 actually declined slightly in 2000, an occurrence unprecedented in U.S. history (see Figure 9). Figure 9. Percent of the U.S. Population Aged 65 or Older, 1860-2080 Source: Bell and Miller (2002). This combination of declining fertility and increased life expectancy will lead to a dramatic increase in the number of older people in our society Source: U.S. Bureau of the Census (2000). over the coming decades. If the fertility rate does remain around two children per woman as projected The usual story is that few people were born during and immigration trends remain constant, at some the Great Depression because times were difficult, and point the age structure of the population will stabi- couples were reluctant to take on increased responsi- lize, and any further aging will come solely from the bilities. In fact, the sharp downturn in fertility started increase in life expectancy. But these days, national in the early 1920s. In part, this was a continuation of attention is focused on the aging of the baby boom. the long-term trend toward declining fertility as the nation shifted away from a rural and agrarian economy to an urban and industrial society. Reinforcing this downward trend were economic pressures. While the 1920s look prosperous when measured in terms of aggregate statistics, virtually all the gains accrued 6 The Social Security Administration prepares two types of tables mortality improvements. All the numbers used in this brief are for measuring life expectancy. The first is a period life table that from the period life table (Bell and Miller 2002). shows how many people in each age group are expected to die in a 7 Council of Economic Advisers (1997). given year. The second type is a cohort life table that incorporates 6 Center for Retirement Research 8 The contention is that people’s desire for children to the wealthiest non-farm families. Wage rates for typical workers remained unchanged over the decade. depends on the consumption standards they enjoyed And farm families, who had the highest fertility when they were growing up. That is, would-be parents rates, faced low agricultural prices throughout the feel a need to ensure that their children will have every- 1920s and lost so much wealth in the 1920-21 crash thing they were provided by their parents. In the post- in land values that they did not recover for the rest of war period, young American families found that their the decade. In addition to economic pressures, the income prospects were much brighter than those of beginning of the mass production and advertising of their parents, and they decided they could afford to have luxury goods highlighted the tradeoff that workers more children. Some argue that their prospects were so much brighter because they belonged to a small cohort increasingly faced between children and consumption. due to the decline in fertility in the 1920s and 1930s. If The drop in fertility that began in the 1920s a cohort is relatively small, it has relatively few workers. continued in the 1930s, when the economy went into This increases their value in the labor market and allows freefall. One quarter of the workforce was unemployed them to enjoy steady employment and higher wages. at the nadir, and young adults suffered the most. They Thus, the favorable contrast of the post-war economy looked at an uncertain future, and the prospect of having with the Depression combined with the relatively small a child appeared fraught with risks. By 1940, 17 percent 9 size of the childbearing cohort contributed to a feeling of of married women aged 40-45 had never borne a child. economic security and a desire for children. As a result of the decline in fertility in the 1920s and Cultural factors may also have been important. Those 1930s, the population 65 and over grew half as rapidly born in the 1920s and 1930s, who were responsible for between the 1985 and 2005 period as it did between the baby boom, came of age during a period of amazing 1965 and 1985. economic and political achievement. Americans had overcome the stagnation of the Great Depression, won The Post World War II Baby Boom World War II, and emerged as the major world power. The fertility rate remained low until after the Second This feeling of success was reinforced by the high level World War. But then births increased at an unprecedent- of economic growth during the 1950s. Moreover, the ed rate between 1946 and 1964, reaching a peak in 1957 war-time spirit of engagement and patriotism lasted well when 4.3 million babies were born. At first, the baby beyond the immediate post-war period and strengthened boom seems like a natural response to husbands and social institutions such as the family, the church, and the wives reuniting after the war, but in fact 19 years is a military. The post-war period also re-focused attention long time for catching up. on gender differences as women, who had participated The upturn in fertility in the late 1940s is the actively in the war effort, returned to the home and most easily explained. A peak in marriages occurred in childbearing. And the baby boom persisted for 11 1946 shortly after the end of World War II. The follow- 19 years. ing year saw a spike in the number of first births, and these couples tended to have second, third, and fourth The Baby Bust children throughout the 1940s and into the 1950s. Eventually, the special circumstances that had Older women who had postponed childbearing during nurtured the baby boom began to fade. The baby boom the Depression and war also began to have children cohort that was born in the post-war period started to reach when the economy recovered in the 1940s. The greatest adulthood in the 1970s. Because this cohort was extremely increase in fertility occurred among the white urban large, many young adults began competing for jobs, driving middle class, particularly those with a college education. down their scarcity value as workers and their potential But people generally felt wealthy; they had put away earnings. Moreover, they did not envision a world in which money during the war and in the immediate post-war 12 they would be significantly better off than their parents. period had little on which to spend their savings other At the same time, the use of birth control became more than children. Thus, both the Depression and the war widespread which gave women much more control over may have increased the desire of women in the 1940s to their reproductive decisions. This control, in turn, made it have larger families. But if other factors had not come easier for women to delay marriage and motherhood in into play, the post-war baby boom would have been favor of advanced degrees and participation in the labor short-lived, as it was in Europe. 13 force. The net result was that fertility rates dipped sharply One explanation for the continued boom in babies is before rebounding to the level of replacement, where they that young couples in the post-war period felt that they 10 are projected to remain for the foreseeable future. could afford to have more children than their parents. 8 13 Van Horn (1988). The 1960s and 1970s saw two innovations in birth control: 9 Van Horn (1988). increasing access to the pill and the legalization of abortion. A 10 recent study examining these changes found that, of the two, the Easterlin (1968). pill was far more significant because it made sex safer and allowed 11 Van Horn (1988). women to invest in their professional education, which often led to 12 Lindhert (1977). full-time careers (Goldin 2004). Issue in Brief 7 The bust-boom-bust pattern is important in explain- ing the speed with which the population will age. The Conclusion U.S. population was aging long before the baby boomers Population aging is the result of very long-term came along and will probably continue well after they are trends in fertility and life expectancy. The post-war gone. But the boomers are the primary cause of the baby boom was a short-term departure from the trend rapid pace of aging in the coming decades. of lower fertility, and it was quickly followed by a baby bust that brought fertility back down to historically low levels. As the baby boom generation begins to reach traditional retirement ages, the pace of population aging will rapidly accelerate. By 2025, nearly 1 in 5 Americans will be age 65 or over compared to 1 in 8 today. But the outlook for 2080 is unaffected by the bust-boom-bust pattern and simply reflects the long- term trends in fertility and life expectancy. References Bell, Felicitie C. and Michael L. Miller. 2002. “Life Peterson, Peter G. 1996. Will America Grow Up Before Tables for the United States Social Security Area It Grows Old?: How the Coming Social Security Crisis 1900-2100.” Actuarial Study No. 116 (August). Threatens You, Your Family, and Your Country. Washington, D.C.: Social Security Administration. Random House. [Available at: http://www.ssa.gov/OACT/NOTES/ U.S. Bureau of the Census. 1976. Historical Statistics as116/as116TOC.html]. of the United States, Colonial Times to 1970. Bell, Felicitie C. 1997. “Table 3. Total Fertility Rates U.S. Bureau of the Census. 1996. Tables A-6 and A-7 by Calendar Year and Alternative.” Actuarial Study in “Population Projections of the United States by No. 112 (August). Washington D.C.: Social Age, Sex, Race, and Hispanic Origin: 1995 to 2050.” Security Administration. [Available at: Current Population Reports P25-1130. Washington, http://www.ssa.gov/ OACT/NOTES/AS112/tab3.html]. D.C.: U.S. Government Printing Office. Centers for Disease Control. 2003. Vital Health U.S. Bureau of the Census. 1998. “Summary Files.” Statistics. [Available at: http://www.cdc.gov/nchs/ National Population Projections. [Available at: products/pubs/pubd/hestats/infantmort/infant- http://www.census.gov/population/www/projec- mort.htm]. tions/ natsum-T3.html]. Coale, Ansley J. and Melvin Zelnick. 1963. New U.S. Bureau of the Census. 2000. “Projections of the Estimates of Fertility and Population in the U.S. Total Resident Population by 5-Year Age Groups, Princeton University Press. and Sex with Special Age Categories: Middle Committee for Economic Development. 1999. Series, 2001-2005, 2006-2010, 2016-2020, and New Opportunities for Older Workers. New York: 2025-2045.” National Population Projections, Research and Policy Committee of the Committee Summary Files. [Available at: for Economic Development. http://www.census.gov/population/ www/projec- tions/natsum-T3.html]. Council of Economic Advisers. 1997. Economic Report of the President. Washington, D.C.: U.S. U.S. Department of Homeland Security. 2002. Government Printing Office. Annual Yearbook of Immigration Statistics. [Available at: http://uscis.gov/graphics/shared/ Easterlin, Richard. 1968. Population, Labor Force, aboutus/statistics/IMM02yrbk/IMM2002list.htm]. and Long Swings in Economic Growth. Columbia University Press. U.S. Social Security Administration. 2003. The 2003 Annual Report of the Board of Trustees of the Federal Goldin, Claudia. 2004. “From the Valley to the Old-Age and Survivors Insurance and Disability Summit: The Quiet Revolution that Transformed Insurance Trust Funds. Washington, D.C.: U.S. Women's Work.” Working Paper No. w10335 Government Printing Office. [Available at: (March). Cambridge, MA: National Bureau of http://www.socialsecurity. gov/OACT/TR/TR03/ Economic Research. tr03.pdf ]. Lindhert, Peter H. 1977. “American Fertility Patterns Van Horn, Susan Householder. 1988. Women, Work, since the Civil War.” In Population Patterns and Fertility, 1900-1986. New York University Press. in the Past, edited by Ronald Demos Lee. Academic Press. Montgomery, Mark R. and Barney Cohen (Eds.). 1998. From Death to Birth: Mortality Decline and Reproductive Change. National Academy Press. CENTER FOR RETIREMENT RESEARCH AT BOSTON COLLEGE About the Center Affiliated Institutions The Center for Retirement Research at Boston American Enterprise Institute College, part of a consortium that includes parallel Massachusetts Institute of Technology centers at the University of Michigan and the Syracuse University National Bureau of Economic Research, was The Brookings Institution established in 1998 through a grant from the Social Urban Institute Security Administration. The goals of the Center are to promote research on retirement issues, to transmit new findings to the policy community and Contact Information the public, to help train new scholars, and to Center for Retirement Research broaden access to valuable data sources. Through Boston College these initiatives, the Center hopes to forge a strong Fulton Hall 550 link between the academic and policy communities Chestnut Hill, MA 02467-3808 around an issue of critical importance to the Phone: (617) 552-1762 nation’s future. Fax: (617) 552-1750 E-mail: crr@bc.edu Website: http://www.bc.edu/crr © 2004, by Trustees of Boston College, Center for Retirement The research reported herein was supported by the Cogan Family Research. All rights reserved. Short sections of text, not to exceed two Foundation. The opinions and conclusions expressed are solely those paragraphs, may be quoted without explicit permission provided that of the author and should not be construed as representing the opin- the author is identified and full credit, including copyright notice, is ions of the Center for Retirement Research at Boston College, or the given to Trustees of Boston College, Center for Retirement Research. Cogan Family Foundation.