Show Summary Details

Page of

PRINTED FROM OXFORD HANDBOOKS ONLINE (www.oxfordhandbooks.com). © Oxford University Press, 2018. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).

Subscriber: null; date: 01 April 2020

# US Fertility Rates and Childbearing in American Economic History, 1800–2010

## Abstract and Keywords

Over the past two centuries, the United States has witnessed dramatic changes in fertility rates and childbearing. This chapter describes shifts in childbearing and family size from 1800 to 2010 and describes the role of different factors in this evolution. Demand factors such as industrialization, urbanization, rising family incomes, public health improvements, and the growth in women’s wages generally reduced the benefits and raised the costs of having many children. Supply factors such as increases in infant and child survival and improvements in the technology of contraception and abortion have also altered parents’ decisions about their childbearing. This chapter summarizes the long-run trends in US fertility rates and completed childbearing, both overall and by mothers’ race/ethnicity and geography. Next, it evaluates evidence on the determinants of childbearing, including both economic and demographic explanations for these patterns. A final section weighs the evidence supporting the existence of two fertility transitions.

Over the past two centuries, the United States has witnessed dramatic changes in fertility rates and childbearing. Demand factors such as industrialization, urbanization, rising family incomes, public health improvements, and the growth in women’s wages generally have reduced the benefits and raised the costs of having many children. Supply factors such as increases in infant and child survival and improvements in the technology of birth control and abortion have also altered parents’ decisions about their childbearing.

This chapter discusses the long-run trends in US fertility rates and completed childbearing, both overall and by mothers’ race/ethnicity and region, within three broad periods: the 1800 to 1930 decline in fertility rates, the 1930 to 1960 stabilization in fertility rates followed by the baby boom, and the post-1960 decline and subsequent stabilization in fertility rates. We discuss the determinants of childbearing in each period, including both economic and demographic explanations for these patterns. A final section weighs the evidence supporting the existence of two fertility transitions: a first transition driven by shifts in the demand for children and a second transition catalyzed by changes in supply side factors.

# Changes in US Fertility Rates, 1800–2010

The standard fertility rate timeseries (fig. 3.1) shows that American women reaching childbearing age around 1800 averaged around seven to eight live births during their reproductive years and that this number fell to between two and three children by 1930. (p. 76) This decline also occurred in other industrialized countries and is considered part of a general demographic transition (Guinnane 2011). After the Great Depression, these patterns reversed, and the baby boom emerged. Between 1940 and 1960, the general fertility rate rose by around 60 percent, and cohort measures of completed fertility rates rose by 45 percent.1 Note that this is more than a World War II phenomenon: fertility rates began increasing before mobilization, and they remained high for fifteen years after V-Day. After 1960, the decline in fertility rates resumed, and rates stabilized by the mid-1970s to around present-day levels. For the past twenty-five years, completed childbearing has hovered around two children, even as the timing of marriage, the timing of first birth, and the composition of parents have shifted in important ways.

Figure 3.1. US crude birth rate (CBR) and total fertility rate (TFR), 1800–2000

Notes: The CBR is plotted on the left vertical axis and is measured as the number of births per 1,000 whites in the population. The TFR is plotted on the right vertical axis and measures the cumulative number of births a woman would be expected to have over her lifetime if she experienced the current period’s age-specific birth rates.

These broad facts are well established, but the exact timing of the onset of fertility decline is disputed. Until recently, US (marital) fertility decline was believed to have begun in the late eighteenth century, almost seventy-five years before marital fertility rates began declining in most other nations (Binion 2001; Guinnane 2011; Haines 2000). This would make the United States distinctive, more similar to France—where marital fertility began declining before industrialization—than to England, Germany, or Italy, where marital fertility began declining after industrialization.

Less well known is that US fertility rates before 1880 are estimated using indirect methods and implicitly rely on assumptions about mortality rates, immigration, and the quality of census enumeration. These assumptions are important. Hacker (2003) (p. 77) shows that fertility rate estimates can change in important ways with the use of earlier census microdata (1860 and 1870) and alternative methods. Figure 3.1 presents Hacker’s estimates, which account for increasing mortality rates in the early nineteenth century. Hacker’s revised series shows that the fertility rates of the US white population declined only slightly (if at all) between 1800 and 1840. The Hacker series estimates that the total fertility rate of white women was around 6.95 in 1843 versus 7.04 in 1800. Hacker also shows that marital fertility rates began to decline in earnest after 1860. The implication of these revised numbers is that little of the decline in fertility rate occurred until after the Civil War (ca. 1865). In short, this revised series would make the US experience look less exceptional relative to other industrialized countries. The sensitivity of Hacker’s estimates to earlier data and alternative assumptions remains an open research question.

Figure 3.2. General fertility rate (GFR) and completed childbearing, by year and cohort

Notes: The general fertility rate (GFR) is plotted on the left vertical axis and is measured as the number of births per 1,000 women ages 15–44 in the population (lower x-axis). Before 1910, this series is for white women only. Mean live births is plotted on the right vertical axis and is the mean self-reported number of children ever born for each birth cohort as measured between the ages of 41 and 70. This series is plotted by birth cohort (upper x-axis).

Sources: Fertility rates are from the National Center for Health Statistics, http://www.cdc.gov/nchs/data/statab/t001x01.pdf. Mean live births are computed using the 1900, 1910, and 1940–1990 Decennial Census IPUMS samples (Ruggles et al. 2010) and the 1995–2010 June Current Population Surveys.

An alternative method for examining changes in childbearing involves estimating cohort fertility rates based on census questions. The 1900 and 1910 censuses, as well as most later twentieth-century censuses, asked ever-married women to report their number of live births. Answers to this question among women ages forty-one to seventy—who should have completed their childbearing—also suggest that the decline in fertility began in the mid- to later nineteenth century. Figure 3.2 plots this series against the (p. 78) general fertility rate for reference. Women born around 1850 averaged about five births during their childbearing years, whereas women born twenty-five years later, in 1875, averaged 3.3 births over their lifetimes. This trend continued: women born in the early twentieth century averaged just 2.3 births over their lifetimes.

These census figures appear consistent with Hacker’s argument and inconsistent with the early decline in the crude birth rate (CBR) or general fertility rate (GFR) time series: completed childbearing changed little before the Civil War among ever-married women (cohorts born before 1850, reaching childbearing age before the 1870s) and only modestly among women born between 1850 and the conclusion of the Civil War (reaching childbearing age after 1870). However, reductions in childbearing appear to have gained momentum among women born in the 1870s and persisted until the Great Depression (cohorts born around 1910).2

Figure 3.2 makes apparent that each generation of women in the late nineteenth and early twentieth centuries substituted toward fewer children. This reduction resulted primarily from improvements in spacing and stopping, rather than changes in the age at first marriage (Bailey, Guldi, and Hershbein 2014). The distribution in the number of children born shifted from nearly uniform for ever-married women born in 1850 (childbearing age in the 1870s) to highly concentrated over the next fifty years. For instance, figure 3.3(a) shows that two-thirds of ever-married women in the birth cohort of 1910 bore two or fewer children; a striking 23 percent remained childless. Consequently, US women born in the aughts (1900–1910) are commonly referred to as the “low-fertility cohorts.”

The measurement of fertility rates in the twentieth century is more accurate and requires fewer assumptions. The voluntary adoption of national standards in vital statistics reporting by states in the early twentieth century facilitated the creation of national statistics on birth rates.3 The entry of all states into the birth registration area (BRA) by 1933 enabled the federal government to publish statistics based on actual birth records rather than census enumeration and assumptions about mortality rates based on small geographic areas. Entry into the BRA was based on “tests” of whether at least 90 percent of a state’s births were registered.

The availability of fairly complete birth registrations makes the size and duration of the baby boom much less controversial and even more impressive. This remarkable departure from longer-term trends was not a short-lived aberration reflecting postponed births from the Great Depression or World War II. The baby boom stretched over two decades and was driven by increases in completed childbearing (Rogers and O’Connell 1984; Ryder 1980). Between 1940 and 1960, the general and total fertility rates each rose by more than 50 percent, and cohort measures of completed fertility rose by 45 percent. Measured as the number of live births by cohort of women (reported at ages 41 to 70), completed childbearing increased by around one child per woman, from 2.3 to 3.3, between the low-fertility cohorts and the cohort of mothers born a generation later, around 1930. In a remarkable historical twist, women born in the mid-1930s had completed-fertility rates as high as their grandmothers born fifty years earlier (fig. 3.2).

The baby boom also reflected an increase in the fraction of women who ever had children. Figure 3.3(b) shows that, between the low-fertility cohorts and the mothers of the (p. 79) (p. 80) baby boom (1930s birth cohorts), childlessness among ever-married women fell from 23 percent to 8 percent, and the share of women having more than two children increased. The share of women ever marrying increased from around 90 percent to 95 percent, and the mean age at first marriage for the 1935 cohort was almost two years younger at 20.7 (Bailey, Guldi, and Hershbein 2014). In short, increases in the period fertility rates during the baby boom were achieved both by earlier and more universal marriage and by married women giving birth to more children. These changes were pervasive. Women born in different census regions, women living in both urban and rural areas, and women of different races and nativity experienced sharp upward changes in completed childbearing. Even the Amish experienced a dramatic baby boom (Bailey and Collins 2011).

Figure 3.3. The distribution of children ever born, by cohort

Notes: These figures plot the percent of women ages 41–70 who report having each number of children. We include never-married women in the 1970 to 1990 censuses and June CPS when available so that figures include the recent rise in nonmarital childbearing. Children ever born is top-coded at 12 in the census and 10 in the CPS.

Sources: 1850 to 1930 cohorts use the 1900, 1910, 1940–1990 Decennial Censuses (Ruggles et al. 2010); 1940 to 1969 cohorts use the 1981–2010 June Current Population Surveys (CPS).

Elevated fertility rates quickly subsided in the 1960s and early 1970s. Within roughly fifteen years of the baby boom’s peak, US fertility rates had fallen to around replacement levels, and below those of the low-fertility cohorts born around the turn of the twentieth century. Importantly, this post-1960 decline represented different shifts in behavior than those accompanying prior declines in the nineteenth and early twentieth centuries. Figure 3.3(b) shows that the low-fertility cohorts achieved low fertility rates with many more women remaining unmarried and large increases in childlessness among those who did marry. In contrast, a much larger share of women reaching childbearing age after 1960 had exactly two children (35 percent versus 22 percent for the previous generation), with the distribution of childbearing collapsing around a two-child mode.

The current stabilization in childbearing was reached in the mid-1970s among women born in the late 1940s and in the 1950s. In fact, both the distribution and mode of completed childbearing have changed very little over the past quarter century; figure 3.3(b) shows nearly identical distributions in completed childbearing among cohorts born between 1950 and 1970. An additional feature worth noting is that childlessness among women born around 1970 is lower (at 16 percent) than it was among the low-fertility cohorts. Today’s rates of childlessness are, however, higher than they were during the peak of the baby boom (10 percent).

Although broadly similar, these aggregate trends mask a great deal of heterogeneity in outcomes across racial and ethnic groups and geography. Figure 3.4(a) plots differences across time for both whites and nonwhites, as well as the US average.4 In the early 1800s, childbearing was around 1.5 births per woman higher for nonwhites than for whites. This difference fell to roughly half that for women born in the mid-1870s, with an average racial gap of around 0.8 births. The difference narrowed some 50 percent further for cohorts born around the turn of the twentieth century, approaching 0.4 births per woman. After expanding briefly during the baby boom, the gap returned to its pre–baby boom level.

The US experience also differed a great deal across regions, with the more economically developed regions generally having lower fertility rates. Figure 3.4(b) shows that the Northeast had significantly lower completed fertility (about four and one-half births) for women born as early as the 1830s, whereas other, less economically developed regions averaged closer to six births per woman. The Midwest, however, had converged to lifetime childbearing averages that were similar to those in the Northeast by the turn of the twentieth century (women born around 1880); lifetime childbearing (p. 81) (p. 82) in the West was actually lower than in the Northeast for women born from around 1860 to 1870. Childbearing in the South began falling for women born after the Civil War and continued until around 1910. Today, differences in fertility rates across regions are very small relative to historical standards, and women born in all regions of the country average around two births over their lifetimes. Figure 3.4(b) also plots fertility rates for women born outside of the United States and shows that fertility rates for this group followed a similar pattern to those for women in the South until the turn of the twentieth century.

Figure 3.4. Completed childbearing across race/ethnicity group and geographic area

Notes: See Figure 3.2. Due to small sample sizes, points between 1870 and 1879 in panel (a) for nonwhites and before 1900 in panel (b) for the west are averaged into one cell. In panel (a), this average is plotted as 1874, and in panel (b) it is plotted by decade.

Sources: Fertility rates are from the National Center for Health Statistics, http://www.cdc.gov/nchs/data/statab/t001x01.pdf. Mean live births are computed using the 1900, 1910, and 1940–1990 Decennial Census IPUMS samples (Ruggles et al. 2010).

# Explanations for Declining Fertility Rates, 1800–1960

The proximate causes of the nineteenth-century US fertility transition are well known. These fertility declines accompanied the disappearance of the Malthusian pattern of late marriage and the emergence of fertility control after the desired number of children had been reached. They took place in the absence of modern contraception and are widely believed to have been driven by changes in the demand for children due to industrialization, urbanization, increasing population density and land scarcity, public health improvements, rising educational attainment, and growth in women’s market wages. All of these forces tended to raise incomes and increase the implicit “price” of having children, which gradually drove fertility rates down. A complementary demand-side model emphasizes how rising incomes tended to increase investments in children’s health and education and to reduce fertility rates, a change that is often referred to as substitution between child quantity and quality.

In contrast, the Great Depression and the baby boom pose challenges to the simple demand-side narrative. As the economy entered the Depression, the downward trend in fertility rates changed little in pace. Even as per-capita disposable income plummeted by 23 percent between 1929 and 1932, the general fertility rate fell by a mere seven births per 1,000 women (55.9 to 48.6) between 1930 and 1933 (lagged one year because births occur nine months after pregnancies).5 But the fertility rate had declined by about six births per 1,000 women (61.4 to 55.9) between 1927 and 1930 (Linder and Grove 1947). Thus, the speed of fertility decline barely budged from its earlier trend. Additionally, fertility rates stabilized—but did not appreciably grow—as the economy slowly rebounded between 1933 and 1938. This is surprising, given recent evidence on the strong relationship between fertility rates and unemployment (Currie and Schwandt 2014).

Another challenge lies in explaining the complete and rapid reversal of these trends. Between 1940 and 1960, the baby boom took place in the context of increasing wages, urbanization, and educational attainment—all trends associated with declining fertility rates before 1930. Adding to the apparent puzzle is that fertility trends again reversed as these trends in incomes, urbanization, educational attainment, and women’s labor-force participation continued.

# (p. 83) Demographic Explanations for Declining Fertility Rates, 1800–1930

Much of the demographic literature has emphasized “demographic transition” theory as an explanation for the decline in nineteenth-century fertility rates. Based on the empirical observation that decreases in aggregate mortality rates were followed in many developed countries by declining birth rates, Notestein (1945) theorized a causal connection: high mortality rates lead married couples to have many births so that they can achieve their desired surviving family size. As mortality rates fall, couples have fewer births because fewer are necessary to achieve their desired family size. Interestingly, this story does not fit well with the US experience. Guinnane and others point out that the US fertility decline began before mortality rates began falling in the 1890s: “Fertility in the United States declined for decades before any noticeable decline in mortality” (Guinnane 2011, 599). This fact has provided additional motivation for claiming American exceptionalism.

The demographic literature has also focused on innovations in contraceptive methods, what economists might call the “supply side.”6 Generating systematic, empirical evidence on fertility control within marriage in the nineteenth and early twentieth centuries is difficult. Contraceptives and contraceptive practice were not often discussed openly and, therefore, were not well documented, except in special cases. As an alternative, historical demographers have studied “spacing” and “stopping” to identify effective contraceptive practice. For instance, being able to “stop” childbearing before one is no longer fecund is associated with highly effective contraceptive practice. Longer “spacing” between births, on the other hand, is generally associated with the consistent use of less-effective contraceptive practices.

For the sake of illustration, consider the following simple formula . Observe that the number of children one could expect to have, n, can be written as $n=EC+S$ where E is the number of months a woman is at risk of becoming pregnant (the interval from first coitus to when one is no longer fecund), and C + S represents the average number of months between births. Fertility rates could be reduced by delaying marriage (and thus reducing E, the Malthusian recommendation, in a paradigm where nonmarital fertility is rare) or by greater spacing, either by increasing the length of the sterile period during pregnancy and after birth, S, or the length of the fecund interval, C, by using different contraceptive techniques. Extended breastfeeding offers one natural method for increasing S, and coitus interruptus (withdrawal) and reduced frequency of coitus offer others. Women in the nineteenth and early twentieth centuries likely used other contraceptive methods as well (including abortifacients), though prevalence is difficult to document. Lahey (2014a, 2014b) shows that anti-obscenity statutes regulating primitive, premodern contraceptive techniques are associated with higher fertility rates. This indirect evidence suggests that (p. 84) these early contraceptives, when not restricted by laws, may have helped couples control childbearing within marriage in the late nineteenth century.

# Economic Explanations for Declining Fertility Rates, 1800–1930

The economics literature explains the nineteenth century’s negative association between rising income and falling fertility rates with two complementary models: the opportunity cost and child-quality models.7 The motivating questions in this literature relate to what led couples to desire fewer children (so that they used different contraceptive techniques or changed their sexual behaviors). The basic premise of both models is that children generate utility for parents, both because parents enjoy their children (children have “consumption value”) and because children bring many other benefits to their parents (children have “production value”) in addition to their costs. For instance, children might help with household chores, farm production, or care for parents in old age.

In standard economic formulations, parents weigh utility from children against their cost to the family. This cost includes both the costs of goods (such as food, clothing, and other necessities) as well as the opportunity costs (time of both parents away from other productive activities) of bearing and raising a child (Becker 1965). In this setting, benefits generated by children, such as their contribution to household production or their insurance value for parents in old age, can be modeled as reducing their net cost. Commensurately, the net cost of children rises if the value of these benefits falls or if bearing and raising children become more expensive. A key assumption of these models is that children are “normal” goods; that is, if parents’ incomes increase, these models generally assume that parents want more (“high-quality”) children.

In explaining the decline in fertility, the “opportunity cost” model emphasizes the importance of parents’ incomes and implicit (or shadow) prices of children. The cost of raising children (net of their benefits) did not rise because food prices or clothing prices increased. On the contrary, rising productivity in manufacturing and agriculture tended to reduce the cost of clothing and feeding children (Atack, Bateman, and Margo 2008; Craig and Weiss 2000; Gallman 1972). But increasing agricultural productivity and industrialization mattered in another way: by increasing potential wages, these forces increased the opportunity cost of having children.8 This could have been especially important to the extent that the value of women’s time increased with industrialization.

Increasing population density and urbanization should also have powerful effects on childbearing through both income and opportunity-cost channels. Three-quarters of the US labor force worked on farms in 1800 (Weiss 1992). But as Americans settled the interior of the country, farmland became increasingly expensive and agriculture began mechanizing. Reductions in the demand for family labor to work on farms led to lower fertility rates as well. Consistent with this idea, Wanamaker (2012) shows that (p. 85) fertility rates among Southern white households that had small slave workforces in the antebellum period rose after 1865, when the emancipation of their slaves increased their need for family farm labor.

Moreover, employment opportunities increasingly drew the young to cities and away from farms. This reflects both the fact that parents had smaller farms to bequeath to their children and that children would earn less if they remained on farms (Easterlin 1976). Sundstrom and David (1988) argue that these forces may have reduced the likelihood that children cared for their parents in old age, which further reduced the benefits of children to parents. Steckel (1992) and Basso, Bodenhorn, and Cuberes (2014) show that fertility rates were lower in Northeastern counties with greater availability of banking and financial services, which enabled parents to save directly for retirement outside of their investments in children.

Industrialization and urbanization also altered the locus of production in ways that increased the opportunity cost of childbearing. Both trends imply that the place of work and place of residence became increasingly separate. Moving to a city meant that jobs were more often outside the home and difficult to combine with child supervision. If parents worked in factories or outside the home, parents would have to pay for child care—either directly or by forgoing work—both of which tended to increase the opportunity cost of childrearing. A second implication of urbanization is the rising cost of space. Because adding more children requires more space (or increases crowding), which is more expensive in cities, urbanization also raised the implicit price of childrearing (Simon and Tamura 2009).

The U-shape in women’s labor force participation is also consistent with a rising opportunity cost of childbearing (Goldin 1990). Saxonhouse and Wright note that almost 60 percent of those employed in cotton textiles in the South in 1880 were women (Saxonhouse and Wright 1984, Table 1). Because many industries laid off women at marriage, industrialization may have also worked to limit completed childbearing by causing women to delay marriage, perhaps reducing the number of years exposed to the risk of childbearing (Goldin 1991).

The second model of the neoclassical school stresses the value of parental enjoyment of their children, or “child quality” (Becker and Lewis 1973; Willis 1973). Here, “quality” generally refers to the level of parents’ investments in their children in terms of time or economic resources. In Becker and Lewis’s canonical model of parental choice over child quantity and quality, an increase in household income is posited to induce substitution toward higher “quality” (but fewer) children. Consequently, increasing incomes (through rising productivity) leads parents to invest more in each child but perhaps have fewer children, which could also generate a negative association between income and fertility rates. Exogenous changes in the shadow price of child quality due to, for instance, improvements in public health could also lead parents to have fewer but healthier children. The neoclassical models allow both infant mortality and fertility rates to decline simultaneously or fertility rates to decline first (as was the case in the US), rather than in the sequence posited by demographic transition theory.

(p. 86) Evidence that parents invested more in their children is difficult to document empirically, but this model is consistent with other evidence. For instance, Moehling (1998) argues that the 75 percent decline in the employment of children ages ten to fifteen between 1880 and 1930 does not appear to be accounted for by changes in state labor legislation. Moreover, Stephens and Yang (2014) show that compulsory schooling laws explain very little of the shift in the quadrupling of American youth earning a high school diploma between 1910 and 1940.9 One interpretation of these findings is that changes in parents’ investments in children both increased children’s human capital and also caused complementary changes in public infrastructure benefiting children. This might involve the building of schools and investment in public health.10 Municipal sanitation, water filtration, and the expansion in water and sewer infrastructure, among other improvements, resulted in dramatic declines in infant and child mortality rates (Alsan and Goldin 2015; Cain and Rotella 2001; Cutler and Miller 2005; Ferrie and Troesken 2005; Troesken 2002). These investments continued with federal funding for maternity and child care through the 1922 Sheppard-Towner Act, which accounts for between one-tenth and one-fifth of the decline in infant mortality rates over the 1920s (Moehling and Thomasson 2014).

The quantity-quality model, therefore, suggests an important causal chain leading from reductions in fertility rates (or other changes in the implicit price of child quality), to increased investments in children and ultimately to faster economic growth; that is, dynamic versions of the quantity-quality model describe how economies transition from a “Malthusian” equilibrium of high fertility rates and slow growth to low fertility rates and rapid economic growth.11 This literature provides important links between childbearing, investments in children, and the dynamics of the economy.

Economists, therefore, explain the nineteenth- and early twentieth-century declines in childbearing as driven by changes in demand. Falling fertility rates are a byproduct of rising costs (due to rising adult wages and productivity, as well as the separation of production from the home), diminishing returns (due to falling farm sizes and changes in children’s care of their aging parents), and rising incomes (which induces greater investments in each child and a reduction in total number).

# Demographic and Economic Models of Childbearing, 1930–1960

Explanations for changes in childbearing during the 1930–1960 period are more disputed. Two schools of thought have emerged: one emphasizing shifting preferences and the other stressing the neoclassical paradigm of relative prices and incomes. More recently, authors differ even within their schools of thought.

Richard Easterlin’s “relative income hypothesis” represents one school of thought (Easterlin 1966, 1971, 1980). Easterlin’s model provides a unified explanation for the low (p. 87) fertility rates during the Great Depression and the baby boom and bust. It emphasizes the importance of a cohort’s perceived “earnings potential” relative to its “material aspirations” as a key input to adult preference formation. Children who grew up during the 1930s, according to the hypothesis, formed modest material aspirations due to the Great Depression. Rapid economic growth in the 1940s and 1950s, however, meant that this cohort’s actual experience as young adults much surpassed these aspirations. When these cohorts realized they could achieve a higher standard of living than expected, they also decided to have more children. On the other hand, children growing up in the more affluent, postwar period of the late 1940s and 1950s had the reverse experience. Their preferences for material goods were shaped by the booming postwar economy, but they entered the labor market in the late 1960s and early 1970s when the US economy was struggling. Consequently, these cohorts had fewer children, which led to the post-1960 fertility decline. One advantage of this theory is that it integrates an explanation of fertility change over three decades. Problematic for this theory, however, is that fertility rates have not cycled since the baby boom despite sizable economic fluctuations.

Becker’s neoclassical theory represents an alternative school of thought and emphasizes relative prices and incomes rather than preferences. In a dynamic reformulation of the static models of the 1960s and 1970s (including the opportunity cost and quantity-quality models discussed previously), Robert Barro and Becker stress how a one-time shock to the price of childbearing or childrearing could explain both low fertility rates during the Great Depression and the baby boom and bust. They are, however, noncommittal on the mechanism (Barro and Becker 1988, 1989).

Although the earlier neoclassical and Easterlin schools of thought in economics have moved closer together over time, a voluminous empirical literature has failed to reach consensus on the appropriate model (Sanderson 1976). In a comprehensive overview of studies of the Easterlin hypothesis, Pampel and Peters conclude that “different methods often result in conflicting findings” (1995, 165). Similarly, in their exhaustive review of tests of the neoclassical model, Hotz, Klerman, and Willis conclude that “the theory and econometric methods are much better developed than the empirical literature” (1997, 342).

Promising recent developments have extended both the theoretical and empirical literatures. In terms of theory, the Barro-Becker reformulation launched a new subfield in economics called “family macro.” Papers in this new subfield generally model the long-run downward trend in childbearing (in the nineteenth and early twentieth centuries) as the increasing opportunity cost of childrearing. Although unified in their emphases on prices and incomes, recent variations on the Barro-Becker formulation claim that different price mechanisms explain the baby boom. Greenwood, Seshadri, and Vandenbroucke (2005) argue that increases in household productivity due to mechanization reduced the shadow price of having children. Doepke, Hazan, and Maoz (2015) claim that the flood of older women into labor markets during World War II depressed younger women’s wages and therefore temporarily decreased the opportunity cost of childbearing. In response, younger women temporarily had more children until shifts in the demand for women’s labor again raised the opportunity cost. Albanesi and Olivetti (p. 88) (2016) assert that medical innovations that benefited the delivery of babies and artificial feeding (e.g., infant formula) reduced the price of childbearing and led to a temporary increase in childbearing. In a novel twist, they also argue that medical innovations contemporaneously altered parents’ investments in girls’ education (because these medical innovations raised their life expectancy), which increased the opportunity cost of childbearing for the next generation.

These theories illuminate how temporary shifts in the economy can influence a variety of outcomes for several generations. More problematic is that calibrations have difficulty predicting the baby boom and the speed of the post-1960 US fertility decline. Moreover, the proposed mechanisms are inherently difficult to test in empirical work. Two papers by Bailey and Collins (2011) and Lewis (2014) provide the only empirical tests of the electrification hypothesis. Using newly encoded, county-level data on appliance ownership and fertility for 1940, 1950, and 1960, Bailey and Collins show that the diffusion of modern household technologies—which began in earnest decades before the baby boom and continued afterward—does not align well with the burst in household productivity required to generate the baby boom. They also present regression evidence that measures of improved household productivity (such as electrification) are negatively correlated with fertility rates. Furthermore, the authors demonstrate that census data on completed fertility (the number of children ever born by ages forty-one to sixty) is negatively correlated with exposure to electrical service in early adulthood. Finally, Bailey and Collins show that the Old Order Amish—a group that on religious grounds strictly limited their use of modern appliances that increased household productivity—also had a relatively large and coincident baby boom. Lewis (2014) uses the expansion of the electrical grid between 1930 and 1960 to show that, consistent with electrification causing couples to substitute toward child quality, electrification caused decreases in both fertility and infant mortality rates.

This evidence provides a considerable challenge to arguments that modern household technology necessarily precipitated the US baby boom. It also challenges explanations that would not alter fertility rates among the Amish, including market wages for young women and the introduction of infant formula (Albanesi and Olivetti 2016; Doepke, Hazan, and Maoz 2015). In short, bursts in technologies boosting the productivity of women in the household do not appear, at least by themselves, to explain the baby boom. More theoretical and empirical research on the explanations for baby boom and bust is needed to vet existing theories and, perhaps, to develop new explanations.

# The Baby Bust and Childbearing after 1960: Changes in Demand, Supply, or Both?

Given that the baby boom interrupted the long-run decline in US fertility rates, it is sensible that the long-run decline eventually resumed. Yet the timing of this event is curious. By cause or coincidence, the end of the baby boom corresponded to revolutionary (p. 89) changes in contraception. The US Food and Drug Administration’s approval of the first oral contraceptive in 1960, which became known as “the Pill,” provided a more reliable and effective way to prevent pregnancy. The Pill required no attention or action at the time of intercourse and no cooperation or consent from male partners (Bailey 2006; Trussell 2004). And it was wildly popular: within five years of its release, almost one-quarter of married women had used the Pill (Bailey 2010).

Cursory empiricism attributed the post-1960 reduction in fertility rates to the causal role of the Pill. More rigorous survey research did as well. Ryder and Westoff (1971) heralded the 1960s as a period of “contraceptive revolution.” In Westoff’s Presidential Address to the Population Association of America, he argued that the effect of the Pill was so large that “the entire [emphasis added] decline in births within marriage across the decade of the ‘sixties’ can be attributed to the improvement in the control of fertility” (Westoff 1975, 579).12 In the parlance of economics, the Pill presented a large shift in the supply of births.

An obvious counterargument is that the post-1960 fertility decline appears more like a reversion to trends before the baby boom—a decline typically attributed to the demand-side factors (as outlined in the previous sections). Becker’s Treatise on the Family states plainly what became the dominant view in economics: “the ‘contraceptive revolution’ . . . ushered in by the Pill has probably not been a major cause of the sharp drop in fertility in recent decades” (1991, 143). But recent empirical work challenges this assertion with new causal evidence.

The challenge in testing these assertions about the role of the Pill and, more broadly, other supply-side developments (e.g., other modern contraceptives like the intrauterine device [IUD] or injectable or implantable contraceptives) is finding a credible empirical design. Theoretical reasoning argues that causal effects run in both directions. As Becker noted, women who desire fewer children are more likely to use the Pill. On the other hand, as Ryder and Westoff posit, users of the Pill are more likely to have fewer children due to lower failure rates.13 In the United States, access to the Pill has never been intentionally randomly assigned to a representative set of locations or group of participants, making it difficult to quantify the magnitude of the causal channels running in either direction.14

Recent empirical studies, therefore, use natural experiments that effectively randomized the availability of the Pill across locations to shed light on this question. One paper uses state-level Comstock-era restrictions on the sale of contraceptives. Bailey (2010) shows that when the Pill was introduced, anti-obscenity statutes (“Comstock laws”), which had existed for almost three-quarters of a century, varied significantly in their language regarding the sale of contraceptives. These legal restrictions in twenty-four states affected the diffusion of oral contraception and reduced the speed of fertility declines in restrictive states from 1958 to 1965. After the Griswold Supreme Court decision invalidated these sales bans for married women, however, fertility rates in states with sales bans dropped sharply relative to those without these bans. There is little reason to expect the demand for children to change with this pattern, but it is clear that the supply of contraceptives did. Counterfactual estimates imply that, without sales bans, the marital fertility rate could have been 8 percent lower in states with sales bans and 4 percent lower in the United States as a whole. Bailey uses a back-of-the envelope (p. 90) calculation to show that as much as 40 percent of the decline in the marital fertility rate from 1955 to 1965 might be attributable to the Pill.

A second empirical test of the relevance of the Pill and other supply-side developments uses the county-level expansion of federally funded family-planning programs in the 1960s and early 1970s to quantify the effects of subsidized contraception on the childbearing of lower-income women. Beginning with the 1964 Economic Opportunity Act (EOA) and continuing with the passage of Title X (“Title Ten”), over 650 family-planning programs started or expanded in US communities from 1964 to 1973. Bailey (2012) uses the idiosyncratic timing of the EOA and Title X granting process at the county level to estimate the program’s effects on fertility rates, using models that also account for the availability of abortion. The results show that family-planning programs, which reduced the cost of contraceptives and increased the availability of contraceptive-related services, led to substantial and sustained declines in fertility rates. The general fertility rate fell by roughly 2 percent within five years of the establishment of a federal family planning program and remained almost as low up to fifteen years after the program’s inception. Assuming that these programs were used only by poor women, they imply a reduction in fertility rates among treated women of 20–30 percent within a decade—magnitudes large enough to account for half of the 1965 gap in childbearing between poor and nonpoor women. Follow-up work by Bailey, Malkova, and McLaren (forthcoming) shows that children born after these family-planning programs began were significantly less likely to grow up in poverty or to reside in households collecting public assistance. In summary, family-planning programs reduced birth rates among poor women and increased economic resources in childhood.

In summary, a variety of studies using different empirical strategies document the important role of the Pill and other supply-side developments in changing childbearing outcomes. Although the empirical literature has been silent on the role of (p. 91) different demand-side factors in the post-1960 period, it is clear that continued shifts in the demand for children due to rising wages and incomes are key to understanding the post-1960 period, as well as the longer-term narrative of the US fertility decline. In the absence of important shifts in the demand for children, the contraceptive revolution may have mattered little. Also key to the story is that shifts in supply and demand can have strong influences on one another. As Becker (1991) argued, shifts in the demand for children were integral to the funding and development of innovations in contraception. Importantly, changes in contraceptive technology appear to have influenced the demand for children in the last fifty years.

# What Is the Case for a Second Demographic Transition?

A growing body of empirical work suggests that the Pill and other supply-side factors had quantitatively important effects on childbearing and women’s careers. While internally valid, the short-term comparisons in these studies shed little light on how changes on the supply side may have had larger, general equilibrium consequences. For instance, they could have shifted attitudes, changed gender roles and women’s bargaining power, and altered the culture of dating, cohabitation, and marriage. This means that recent empirical work documenting the short-term impact of the Pill is inherently silent on the bigger picture. We now weigh evidence related to whether the United States is in the midst of a second demographic transition, or SDT, and more broadly whether the same factors that led to the earlier fertility decline can satisfactorily explain the patterns observed in the post-1960 decline.

The concept of an SDT was introduced by Lesthaeghe and van de Kaa (1986), who hypothesized that the arrival of the contraceptive, sexual, and women’s rights revolutions of the 1960s engendered a distinct demographic era—a period exceptional enough to be called a transition in its own right. The distinctive characteristics of the SDT, they argue, are persistently low fertility rates and high rates of childlessness, substantially delayed marriage and childbearing, increases in nonmarital cohabitation and childbearing, and high divorce rates. (Some formulations also invoke increases in women’s labor-force participation.)

Other demographers have countered these claims. Cliquet argues that the SDT characteristics “already existed before the sixties [for Council of Europe member states]; in fact, most of them emerged with the . . . demographic transition around the turn of the [twentieth] century” (1991, 72). Others contend the SDT literature is ahistorical: “A graph truncated at [the 1950s and 1960s] gives a false impression of an inexorable downward slide coinciding with the onset of the [second demographic transition], while in fact in most countries the real decline was forty years earlier. The 1950s and the 1960s are a deceptive aberration in fertility history” (Coleman 2004, 18).

(p. 92) This critique of the SDT literature implicitly claims that the same forces (e.g., opportunity cost of childbearing, shadow price and income elasticity of child quality) drove fertility to decline in both periods. Under this hypothesis, the “right” test of recent exceptionalism is not an examination of the last fifty years, but rather the comparability of the early and later periods of fertility decline. By this metric, fertility rates are not much lower today than earlier in the twentieth century, nor did they decline much more quickly during the baby bust than they had some forty years earlier (Bailey, Guldi, and Hershbein 2014; see figs. 3.1 and 3.2 in this chapter). Although cohabitation rates have increased, age at first union (including both marriages and nonmarital cohabitations) is at its pre–baby boom level, as is age at first birth. Divorce rates are not appreciably different today from what their pre–baby boom trends would predict (Stevenson and Wolfers 2007).

Although this historical comparison of time series and trends appears inconsistent with the notion of an SDT, the current period is much more exceptional than is often credited. Generating time series spanning more than one hundred years of US history, Bailey, Guldi, and Hershbein (2014) show that the post-1960 fertility transition exhibits three features distinct from the nineteenth-century and early twentieth-century fertility decline.

Feature 1: The post-1960 period is distinguished by the emergence of a two-child norm, a reduction in the variance of completed childbearing, and a reduction in childlessness (see figs. 3.3a and 3.3b). Although women reaching childbearing age in the 1930s and 1970s had similar numbers of children on average, these cohorts achieved these means in very different ways. Most economic models simplify childbearing decisions to how many children to have and proxy for this theoretical concept in empirical work with the mean number of live births (or a measure of period fertility). This single moment of the childbearing distribution misses empirical regularities that enhance our understanding of the motivations and constraints that individuals faced over the twentieth century.

Consistent with demand-side explanations and patterns of increased spacing and stopping, the nineteenth-century and early twentieth-century fertility declines are associated with increasing variance in outcomes. That is, this period of fertility decline was characterized by high rates of childlessness, as well as still relatively high rates of women with high parity. On the other hand, consistent with supply-side models, the post-1960 fertility decline is associated with reductions in the variance in childbearing (Michael and Willis 1976). Rates both of childlessness and high parity (e.g., more than four live births) were relatively low.

Feature 2: The post-1960 period exhibits an unprecedented decoupling of marriage and motherhood. The age at first union (historically through marriage, more recently through cohabitation) and the age at first birth were strongly correlated in the early twentieth century, but this interrelationship began to disappear after 1960. Recent cohorts formed their first households at ages similar to cohorts born earlier in the century, but they more often cohabited before marriage and were more likely to switch partners. However, age at first union and age at first intercourse have become less predictive of motherhood timing, as many women give birth outside of marriage.

(p. 93) Feature 3: The post-1960 period has witnessed a transformation in the relationship between mothers’ education and childbearing. When comparing relatively more- and less-educated women, completed childbearing, childlessness, and the likelihood of marriage are all much more similar today than in the early twentieth century. Despite these similarities, age at first household formation, age at first birth, and nonmarital childbearing diverged after 1960 by mothers’ education, with more educated mothers more likely to delay household formation, motherhood, and childbearing within marriage.

These distinctive features should give pause to scholars who wish to argue that the fertility declines in the earlier and later parts of the twentieth century are part of the same, though interrupted, trend. After 1960, women were significantly more likely to have exactly two children and were less likely to remain childless. Patterns of completed childbearing and childlessness and the likelihood of eventual marriage are more similar across educational groups of mothers today than in the late nineteenth or early twentieth centuries. On the other hand, age at first union, age at first birth, and nonmarital childbearing have diverged sharply across educational groups. The decoupling of marriage and childbearing and the changing predictive importance of mothers’ education hint that a larger demographic and economic transition is underway. The fact that these patterns have not yet stabilized suggests that the current fertility transition, perhaps part of a larger gender and cultural revolution, is still ongoing (Goldscheider 2012). Future research should invest more work into understanding how these changes may impact the longer-term evolution of families and childbearing.

# Conclusion

Trends in fertility rates and the economic and demographic literature on the determinants of childbearing in three periods (1800–1930, 1930–1960, and 1960–2010) suggest that the first two periods of change in childbearing reflect demand-side factors. Fueled by economic development, the nineteenth- and early twentieth-century fertility decline reflected the rising cost of childbearing (due to rising adult wages and productivity, as well as the separation of production from the home), falling returns (due to falling farm sizes and changes in children’s care of their aging parents), and rising incomes (which induced greater investments in each child and a reduction in total number). The explanation for the stabilization of fertility rates and then baby boom (1930–1960) is disputed, but changes in childbearing in this period also appear consistent with changes in the demand for children due to rising incomes and possibly large reductions in the relative prices of factors related to childbearing. The explanations for changes in childbearing in the post-1960 period, however, are still evolving, and likely reflect both demand factors and supply factors. Ultimately, it remains unclear whether the post-1960 supply-side changes are significant enough to constitute a second demographic transition or are simply the most recent stage in the ongoing first demographic transition (Lee and Reher 2011).

## Bibliography

Albanesi, S., and C. Olivetti. (2016). “Gender Roles and Medical Progress.” Journal of Political Economy 124(3): 650–695.Find this resource:

Alsan, M., and C. Goldin. (2015). “Watersheds in Infant Mortality: The Role of Effective Water and Sewage Infrastructure, 1880 to 1915.” NBER Working Paper No. 21263.Find this resource:

Ananat, E. O., and D. Hungerman. (2012). “The Power of the Pill for the Next Generation: Oral Contraception’s Effects on Fertility, Abortion, and Maternal and Child Characteristics.” Review of Economics and Statistics 94(1): 37–51.Find this resource:

Atack, J., F. Bateman, and R. A. Margo. (2008). “Steam Power, Establishment Size, and Labor Productivity Growth in Nineteenth Century American Manufacturing.” Explorations in Economic History 45(2): 185–198.Find this resource:

Bailey, M. J. (2006). “More Power to the Pill: The Impact of Contraceptive Freedom on Women’s Lifecycle Labor Supply.” Quarterly Journal of Economics 121(1): 289–320.Find this resource:

Bailey, M. J. (2009). “More Power to the Pill: The Impact of Contraceptive Freedom on Women’s Lifecycle Labor Supply: Erratum and Addendum.” Available at http://www-personal.umich.edu/~baileymj/Bailey_Erratum.pdf.

Bailey, M. J. (2010). “‘Momma’s Got the Pill’: How Anthony Comstock and Griswold v. Connecticut Shaped U.S. Childbearing.” American Economic Review 100(1): 98–129.Find this resource:

Bailey, M. J. (2012). “Reexamining the Impact of U.S. Family Planning Programs on Fertility: Evidence from the War on Poverty and the Early Years of Title X.” American Economic Journal: Applied Economics 4(2): 62–97.Find this resource:

Bailey, M. J. (2013). “Fifty Years of U.S. Family Planning: New Evidence on the Long-Run Effects of Increasing Access to Contraception.” Brookings Papers on Economic Activity (Spring): 341–409.Find this resource:

Bailey, M. J., and W. J. Collins. (2011). “Did Improvements in Household Technology Cause the Baby Boom? Evidence from Electrification, Appliance Diffusion, and the Amish.” American Economic Journal: Macroeconomics 3(2): 189–217.Find this resource:

Bailey, M. J., M. Guldi, A. Davido, and E. Buzuvis. (2011). “Laws and Policies Governing Contraceptive Access, 1960–1980.” University of Michigan Working Paper.Find this resource:

Bailey, M. J., M. Guldi, and B. J. Hershbein. (2013). “Further Evidence on the Internal Validity of the Early Legal Access Research Design.” Journal of Policy Analysis and Management 32(4): 899–904. (p. 96) Find this resource:

Bailey, M. J., M. Guldi, and B. J. Hershbein. (2014). “Is There a Case for a ‘Second Demographic Transition’: Three Distinctive Features of the Post-1960 U.S. Fertility Decline.” In Human Capital and History: The American Record, edited by L. Boustan, C. Frydman, and R. A. Margo, 273–312. Cambridge, MA: National Bureau of Economics Research.Find this resource:

Bailey, M. J., B. J. Hershbein, and A. R. Miller. (2012). “The Opt-In Revolution? Contraception and the Gender Gap in Wages.” American Economic Journal: Applied Economics 4(3): 225–254.Find this resource:

Bailey, M. J., O. Malkova, and Z. McLaren. (Forthcoming). “Do Family Planning Programs Increase Children’s Opportunities? Evidence from the War on Poverty and the Early Years of Title X.” Journal of Human Resources.Find this resource:

Barro, R. J., and G. S. Becker. (1989). “Fertility Choice in a Model of Economic Growth.” Econometrica 57(2): 481–501.Find this resource:

Basso, A., H. Bodenhorn, and D. Cuberes. (2014). “Fertility and Financial Development: Evidence from US Counties in the 19th Century.” NBER Working Paper No. 20491.Find this resource:

Becker, G., K. Murphy, and R. Tamura. (1991). “Economic Growth, Human Capital and Population Growth.” Journal of Political Economy 98(5): S12–S137.Find this resource:

Becker, G. S. (1960). “An Economic Analysis of Fertility.” In Demographic and Economic Change in Developed Countries, 208–230. Princeton, NJ: Princeton University Press.Find this resource:

Becker, G. S. (1965). “A Theory of the Allocation of Time.” Economic Journal 75(299): 493–517.Find this resource:

Becker, G. S. (1991). A Treatise on the Family (Enlarged Edition). Cambridge, MA: Harvard University Press.Find this resource:

Becker, G. S., and R. J. Barro. (1988). “A Reformulation of the Economic Theory of Fertility.” Quarterly Journal of Economics 103(1): 1–25.Find this resource:

Becker, G. S., and H. G. Lewis. (1973). “On the Interaction between the Quantity and Quality of Children.” Journal of Political Economy 81(2): S279–S288.Find this resource:

Binion, R. (2001). “Marianne in the Home: Political Revolution and Fertility Transition in France and the United States.” Population: An English Selection 13(2): 165–188.Find this resource:

Cain, L. P., and E. J. Rotella. (2001). “Death and Spending: Urban Mortality and Municipal Expenditure on Sanitation.” Annales de Demographie Historique 1: 139–154.Find this resource:

Cliquet, R. L. (1991). The Second Demographic Transition: Fact or Fiction? Population Studies, Vol. 23. Strasbourg: Council of Europe.Find this resource:

Coleman, D. (2004). “Why We Don’t Have to Believe without Doubting in the ‘Second Demographic Transition’—Some Agnostic Comments.” Vienna Yearbook of Population Research 2: 11–24.Find this resource:

Craig, L. A., and T. Weiss. (2000). “Hours at Work and Total Factor Productivity Growth in Nineteenth Century U.S. Agriculture.” In Advances in Agricultural Economic History, Vol. 1, 1–30. Stamford, CT: JAI Press.Find this resource:

Currie, J., and H. Schwandt. (2014). “Short- and Long-Term Effects of Unemployment on Fertility.” Proceedings of the National Academy of Sciences 111(41): 14724–14739.Find this resource:

Cutler, D. M., and G. Miller. (2005). “The Role of Public Health Improvements in Health Advances: The 20th Century United States.” Demography 42(1): 1–22.Find this resource:

DiCenso, A., G. Guyatt, A. Willan, and L. Griffith. (2002). “Interventions to Reduce Unintended Pregnancies among Adolescents: Systematic Review of Randomized Controlled Trials.” British Medical Journal 324: 1426–1434.Find this resource:

Doepke, M., M. Hazan, and Y. D. Maoz. (2015). “The Baby Boom and World War II: A Macroeconomic Analysis.” Review of Economic Studies 82(3): 1031–1073. (p. 97) Find this resource:

Easterlin, R. A. (1966). “Economic-Demographic Interactions and Long Swings in Economic Growth.” American Economic Review 56(5): 1063–1104.Find this resource:

Easterlin, R. A. (1971). “Does Human Fertility Adjust to the Environment?” Family Economic Behavior 61(2): 399–407.Find this resource:

Easterlin, R. A. (1975). “An Economic Framework for Fertility Analysis.” Studies in Family Planning 6(3): 54–63.Find this resource:

Easterlin, R. A. (1976). “Factors in the Decline of Farm Family Fertility in the United States: Some Preliminary Research Results.” Journal of American History 63(3): 600–614.Find this resource:

Easterlin, R. A. (1980). Birth and Fortune: The Impact of Numbers on Personal Welfare. New York: Basic Books.Find this resource:

Easterlin, R. A., and E. M. Crimmins. (1985). The Fertility Revolution: A Supply-Demand Analysis. Chicago: University of Chicago Press.Find this resource:

Easterlin, R. A., R. A. Pollack, and M. L. Wachter. (1980). “Towards a More General Economic Model of Fertility Determination: Endogenous Preferences and Natural Fertility.” In Population and Economic Change in Less Developed Countries, edited by R. A. Easterlin, 81–150. Chicago: University of Chicago Press.Find this resource:

Ferrie, J., and W. Troesken. (2005). “Death and the City: Chicago’s Mortality Transition, 1850–1925.” NBER Working Paper No. 11427.Find this resource:

Gallman, R. E. (1972). “Changes in Total U.S. Agricultural Factor Productivity in the Nineteenth Century.” Agricultural History 46(1): 191–210.Find this resource:

Galor, O., and D. N. Weil. (2000). “Population, Technology, and Growth: From Malthusian Stagnation to the Demographic Transition and Beyond.” American Economic Review 90(4): 806–828.Find this resource:

Goldin, C. (1990). Understanding the Gender Gap: An Economic History of American Women. New York: Cambridge University Press.Find this resource:

Goldin, C. (1991). “Marriage Bars: Discrimination against Married Women Workers from the 1920s to the 1950s.” In Favorites of Fortune: Technology, Growth, and Economic Development since the Industrial Revolution, edited by H. Rosovsky, D. Landes, and P. Higonnet, 511–536. Cambridge, MA: Harvard University Press.Find this resource:

Goldin, C. (1998). “America’s Graduation from High School: The Evolution and Spread of Secondary Schooling in the Twentieth Century.” Journal of Economic History 58(2): 345–374.Find this resource:

Goldin, C., and L. F. Katz. (2002). “The Power of the Pill: Oral Contraceptives and Women’s Career and Marriage Decisions.” Journal of Political Economy 110(4): 730–770.Find this resource:

Goldin, C., and L. F. Katz. (2008). The Race between Education and Technology. Cambridge, MA: Belknap Press.Find this resource:

Goldscheider, F. (2012). “The Gender Revolution and the Second Demographic Transition: Understanding Recent Family Trends in Industrialized Societies.” Keynote Address, 2012 European Population Conference, Stockholm, Sweden, June 13.Find this resource:

Greenwood, J., A. Seshadri, and G. Vandenbroucke. (2005). “The Baby Boom and Baby Bust.” American Economic Review 95(1): 183–207.Find this resource:

Guinnane, T. W. (2011). “The Historical Fertility Transition: A Guide for Economists.” Journal of Economic Literature 49(3): 589–614.Find this resource:

Guldi, M. (2008). “Fertility Effects of Abortion and Pill Access for Minors.” Demography 45(4): 817–827.Find this resource:

Hacker, J. D. (2003). “Rethinking the ‘Early’ Decline of Marital Fertility in the United States.” Demography 40(4): 605–620. (p. 98) Find this resource:

Haines, M. (2000). “The White Population of the United States, 1790–1920.” In A Population History of North America, edited by M. R. Haines and R. H. Steckel, 305–369. New York: Cambridge University Press.Find this resource:

Haines, M. (2008). “Fertility and Mortality in the United States.” EH.Net Encyclopedia, edited by R. Whaples. Available at http://eh.net/encyclopedia/fertility-and-mortality-in-the-united-states/.

Haines, Michael R. (2006). “Fertility and Mortality by Race: 1800–2000.” In Historical Statistics of the United States: Millennial Edition, edited by Susan B. Carter, 1–391. New York: Cambridge University Press.Find this resource:

Hock, H. (2008). “The Pill and the College Attainment of American Women and Men.” Florida State University Department of Economics Working Paper.Find this resource:

Hotz, V. J., J. A. Klerman, and R. J. Willis. (1997). “The Economics of Fertility in Developed Countries.” In Handbook of Population and Family Economics, edited by M. R. Rosenzweig and O. Stark, Vol. 1A, 275–348. Amsterdam: Elsevier.Find this resource:

Hotz, V. J., and R. A. Miller. (1988). “An Empirical Analysis of Life Cycle Fertility and Female Labor Supply.” Econometrica 56(1): 91–118.Find this resource:

Kearney, M. S., and P. B. Levine. (2009). “Subsidized Contraception, Fertility, and Sexual Behavior.” Review of Economics and Statistics 91(1): 137–151.Find this resource:

Kearney, M. S., and P. B. Levine. (2015). “Media Influences on Social Outcomes: The Impact of MTV’s 16 and Pregnant on Teen Childbearing.” American Economic Review 105(12): 3597–3632.Find this resource:

Lahey, J. N. (2014a). “Birthing a Nation: The Effect of Fertility Control Access on the Nineteenth-Century Demographic Transition.” Journal of Economic History 74(2): 482–508.Find this resource:

Lahey, J. N. (2014b). “The Effect of Anti-Abortion Legislation on Nineteenth Century Fertility.” Demography 51(3): 939–948.Find this resource:

Lee, R. D., and D. S. Reher. (2011). “Introduction: The Landscape of Demographic Transition and Its Aftermath.” Population and Development Review 37(1): 1–7.Find this resource:

Lesthaeghe, R. J., and D. J. van de Kaa. (1986). “Twee demografische transities.” In Bevolking, groei en krimp, edited by R. Lesthaeghe and D. J. van de Kaa, 19–68. Deventer: Van Loghum Slaterus.Find this resource:

Lewis, J. (2014). “Fertility, Child Health, and the Diffusion of Electricity in the Home.” Department of Economics, University of Montreal Working Paper.Find this resource:

Linder, F. E., and R. D. Grove. (1947). Vital Statistics Rates in the United States 1900–1940. Washington, DC: National Office of Vital Statistics.Find this resource:

Mellor, J. M. (1998). “The Effect of Family Planning Programs on the Fertility of Welfare Recipients: Evidence from Medicaid Claims.” Journal of Human Resources 33: 866–895.Find this resource:

Michael, R. T., and R. J. Willis. (1976). “Contraception and Fertility: Household Production under Uncertainty.” In Household Production and Consumption, edited by N. E. Terleckyj, 25–98. Cambridge, MA: National Bureau of Economic Research.Find this resource:

Moehling, C. M. (1998). “Work and Family: Intergenerational Support in American Families, 1880–1920.” Journal of Economic History 58(2): 535–537.Find this resource:

Moehling, C. M., and M. A. Thomasson. (2014). “Saving Babies: The Impact of Public Education Programs on Infant Mortality.” Demography 51(2): 367–386.Find this resource:

Myers, C. K. (2017). “The Power of Abortion Policy: Re-examining the Effects of Young Women’s Access to Reproductive Control.” Journal of Political Economy 125(6): 2178–2224. (p. 99) Find this resource:

National Center for Health Statistics. (2003). “Table 1-1: Live Births, Birth Rates, and Fertility Rates, by Race: United States, 1909–2000.” Available at https://www.cdc.gov/nchs/data/statab/t001x01.pdf.

Notestein, F. W. (1945). “PopulationThe Long View.” In Food for the World, edited by T. W. Schultz, 36–57. Chicago: University of Chicago Press.Find this resource:

Pampel, F. C., and H. E. Peters. (1995). “The Easterlin Effect.” Annual Review of Sociology 21: 163–194.Find this resource:

Rogers, C. C., and M. O’Connell. (1984). “Childspacing among Birth Cohorts of American Women: 1905 to 1959.” In Current Population Reports, Series P-20 No. 385. Washington, DC: Bureau of the Census.Find this resource:

Rosenweig, M. R., and K. Wolpin. (1986). “Evaluating the Effects of Optimally Distributed Public Programs: Child Health and Family Planning Interventions.” American Economic Review 76(3): 470–482.Find this resource:

Ruggles, S., J. T. Alexander, R. Goeken, M. B. Schroeder, and M. Sobek. (2010). Integrated Public Use Microdata Series (Version 6.0) [Machine-readable database]. Minneapolis, University of Minnesota. Available at https://usa.ipums.org/usa/.

Ryder, N. B. (1980). “Components of Temporal Variations in American Fertility.” In Demographic Patterns in Developed Societies, edited by R. W. Hiorns, 15–54. London: Taylor and Francis.Find this resource:

Ryder, N. B., and C. F. Westoff. (1971). The Contraceptive Revolution. Princeton, NJ: Princeton University Press.Find this resource:

Sah, R. K., and J. A. Birchenall. (2012). “Coituses and Contraceptive Reliability: Some Results on Behavior and Welfare.” University of California at Santa Barbara Working Paper.Find this resource:

Sanderson, W. C. (1976). “On Two Schools of the Economics of Fertility.” Population and Development Review 2(3–4): 469–477.Find this resource:

Saxonhouse, G., and G. Wright. (1984). “Two Forms of Cheap Labor in Textile History.” In Technique, Spirit and Form in the Making of the Modern Economies: Essays in Honor of William N. Parker. Research in Economic History Supp 3, edited by G. Saxonhouse and G. Wright, 3–31. Greenwich, CT: JAI Press.Find this resource:

Shapiro, S. (1950). “Development of Birth Registration and Birth Statistics in the United States.” Population Studies 4(1): 86–111.Find this resource:

Simon, C. J., and R. Tamura. (2009). “Do Higher Rents Discourage Fertility? Evidence from US Cities, 1940–2000.” Regional Science and Urban Economics 39(1): 33–42.Find this resource:

Steckel, R. H. (1992). “The Fertility Transition in the United States: Tests of Alternative Hypotheses.” In Strategic Factors in Nineteenth Century American Economic History: A Volume to Honor Robert W. Fogel, edited by C. Goldin and H. Rockoff, 351–374. Cambridge, MA: National Bureau of Economic Research.Find this resource:

Stephens, M., Jr., and D.-Y. Yang. (2014). “Compulsory Education and the Benefits of Schooling.” American Economic Review 104(6): 1777–1792.Find this resource:

Stevenson, B., and J. Wolfers. (2007). “Marriage and Divorce: Changes and Their Driving Forces.” Journal of Economic Perspectives 21(2): 27–52.Find this resource:

Sundstrom, W. A., and P. A. David. (1988). “Old-Age Security Motives, Labor Markets, and Farm Family Fertility in Antebellum American.” Explorations in Economic History 25(2): 164–197.Find this resource:

Troesken, W. (2002). “The Limits of Jim Crow: Race and the Provision of Water and Sewerage Services in American Cities, 1880–1925.” Journal of Economic History 62(3): 734–772. (p. 100) Find this resource:

Trussell, J. (2004). “Contraceptive Efficacy.” In Contraceptive Technology, edited by R. A. Hatcher, J. Trussell, F. Stewart, A. Nelson, W. Cates, F. Guest, and D. Kowal, 18th rev. ed., 747–826. New York: Ardent Media.Find this resource:

U.S. Bureau of Economic Analysis. National Income and Product Accounts. Table 7.1, Selected Per Capita Product and Income Series in Current and Chained Dollars. Available at https://www.bea.gov/iTable/index_nipa.cfm. Accessed September 25, 2015.Find this resource:

U.S. Census Bureau. (1981–2010). Current Population Surveys, June Supplement. [Machine-readable data files]. Available at http://www.nber.org/data/current-population-survey-data.html.Find this resource:

Wanamaker, M. H. (2012). “Industrialization and Fertility in the Nineteenth Century: Evidence from South Carolina.” Journal of Economic History 72(1): 168–196.Find this resource:

Weiss, T. (1992). “U.S. Labor Force Estimates and Economic Growth, 1800–1860.” In American Economic Growth and Standards of Living before the Civil War, edited by R. E. Gallman and J. J. Wallis, 19–78. Chicago: University of Chicago Press.Find this resource:

Westoff, C. F. (1975). “The Yield of the Imperfect: The 1970 National Fertility Study.” Demography 12(4): 573–580.Find this resource:

Willis, R. J. (1973). “A New Approach to the Economic Theory of Fertility Behavior.” Journal of Political Economy 81(2): S14–S64.Find this resource:

## Notes:

(1.) The general fertility rate is the number of live births for every 1,000 women of childbearing age (typically ages 15 through 44) in a given calendar year. Completed childbearing is a cohort measure and represents the average number of live births to women born in a certain year.

(2.) The census figures are unadjusted for differential survival of women by their number of children born and for unmarried childbearing. Figure 3.2 shows a sharp decline in completed fertility for women born in the late 1860s, but this decline coincides with seaming issues between the 1910 and 1940 censuses (the 1920 and 1930 censuses did not ask about completed childbearing). This means that estimates for cohorts born in the late 1860s are drawn from the 1910 census (when these women were in their forties). Estimates for cohorts born in the early 1870s are drawn from the 1940 census, when these women were in their sixties, and may reflect age-based mortality rates that are correlated with completed childbearing. Investigating this is a subject for future research.

(3.) See Shapiro (1950) for a detailed history on the development of the national vital statistics registration areas.

(4.) “Nonwhite” is an unsatisfactory category for racial and ethnic group, but it is the only classification that can be made consistently over such a long period.

(5.) The income data can be found at U.S. Bureau of Economic Analysis, National Income and Product Accounts: Table 7.1. The data were accessed in September 2015.

(6.) Economics models innovation in contraceptive technology as a reduction in the marginal cost of averting births (Michael and Willis 1976).

(7.) Opportunity cost model: Becker (1965); child-quality model: Becker and Lewis (1973).

(8.) This is defined as the amount of time required to bear and raise a child valued by the dollar equivalent of the relevant adult’s marginal product in her most productive activity.

(9.) See Goldin (1998) for graduation rate estimates.

(10.) The building of schools is discussed in Goldin and Katz (2008).

(11.) See Becker, Murphy, and Tamura (1991) and Galor and Weil (2000). Galor and Weil call this “Unified Growth Theory.” Their formulation endogenizes population growth and human capital investment decisions to explain the transition dynamics from a Malthusian equilibrium to a post-Malthusian system in which income grows while population is stable.

(12.) Economists have also noted the potential importance of the supply side (Easterlin 1975; Easterlin and Crimmins 1985; Easterlin, Pollack, and Wachter 1980; Hotz and Miller 1988; Michael and Willis 1976). For empirical papers on the role of greater access to reliable medical contraceptives, see Bailey (2006, 2010, 2012); Bailey, Hershbein, and Miller (2012); Goldin and Katz (2002); Guldi (2008); Kearney and Levine (2009).

(13.) Theoretical models suggest that women who use family-planning services are different in many ways from those who do not. Sah and Birchenall (2012) show why women who use family-planning services may be expected to differ in terms of their unobserved preferences as well as in the price associated with a conception. Theory also suggests that cross-sectional associations in childbearing and family planning may reflect both greater local demand for services and the effects of those services.

(14.) The earliest studies on this question used multivariate regressions to adjust estimates of the relationship between access to family planning (whether areas had a program or individuals used them) and fertility rates. These largely cross-sectional studies were limited by well-known omitted variables and endogeneity problems (see Rosenweig and Wolpin 1986, and Hotz, Klerman, and Willis 1997) and it is not surprising that these studies led to mixed evidence on the effects of different types of family planning programs (see Mellor 1998 for a review). Recent studies use localized, randomized interventions that target teen pregnancies but generally conclude that these interventions had little effect (DiCenso et al. 2002 and Kearney and Levine 2015). See Bailey (2013) for a more detailed discussion of these studies.

(15.) See Bailey (2006, 2009); Bailey, Hershbein, and Miller (2012); Goldin and Katz (2002); Guldi (2008); Hock (2008). In a recent paper, Myers (2017) argues that the effects of changes in legal access to the Pill for younger women differs from Goldin and Katz’s and Bailey’s estimates when she changes the legal coding. While smaller, the magnitudes of her updated estimates are not statistically different from published estimates (Bailey, Guldi, and Hershbein 2013). Bailey et al. (2011) summarize the legal coding used in the different studies.