NBER WORKING PAPER SERIES
THE AMERICAN FAMILY AND FAMILY ECONOMICS
Shelly Lundberg
Robert A. Pollak
Working Paper 12908
http://www.nber.org/papers/w12908
NATIONAL BUREAU OF ECONOMIC RESEARCH
1050 Massachusetts Avenue
Cambridge, MA 02138
February 2007
We are grateful to the members of the MacArthur Network on the Family and the Economy, and to
Joanne Spitz, Dick Startz, Meredith Startz, James Hines, Michael Waldman, and especially Timothy
Taylor for their helpful comments. Pollak is grateful to the John D. and Catherine T. MacArthur Foundation
for their support and Lundberg to the Castor Professorship. The views expressed herein are those of
the author(s) and do not necessarily reflect the views of the National Bureau of Economic Research.
© 2007 by Shelly Lundberg and Robert A. Pollak. All rights reserved. Short sections of text, not to
exceed two paragraphs, may be quoted without explicit permission provided that full credit, including
© notice, is given to the source.
The American Family and Family Economics
Shelly Lundberg and Robert A. Pollak
NBER Working Paper No. 12908
February 2007
JEL No. D1,J1,J2
ABSTRACT
The twenty-fifth anniversary of the publication of Gary Becker's path-breaking "Treatise on the Family"
provides an occasion to reexamine both the American family and family economics. We begin by
discussing how families have changed in recent decades: the separation of sex, marriage, and childbearing;
fewer children and smaller households; converging work and education patterns for men and women;
class divergence in partnering and parenting strategies; and the replacement of what had been family
functions and home production by government programs and market transactions. After discussing
recent work in family economics that attempts to explain these changes, we point out some challenging
areas for further analysis, and highlight issues of commitment in two primary family relationships:
those between men and women, and those between parents and children. We conclude by discussing
the effectiveness of policies to target benefits to certain family members (e.g., children) or to promote
marriage and fertility.
Shelly Lundberg
Department of Economics
227 Savery Hall
Box 353330
University of Washington
Seattle, WA 98195-3330
LUNDBERG@U.WASHINGTON.EDU
Robert A. Pollak
Washington University in St. Louis
Faculty of Arts and Sciences
and the John M. Olin School
Campus Box 1133
1 Brookings Drive
St. Louis, MO 63130-4899
and NBER
pollak@wustl.edu
2
Gary Becker’s path-breaking Treatise on the Family subjected individuals’ decisions
about sex, marriage, childbearing, and childrearing to rational choice analysis. Becker’s aim was
to use the foundational assumptions of “maximizing behavior, market equilibrium, and stable
preferences” (1991, Enlarged edition, p. 5) to explain the basic empirical patterns of family life.
1
According to Becker (p. 135), “[T]he main purpose of marriage and families is the production
and rearing of own children . . .” and in his model of marriage, the gains to marriage depend on
specialization and exchange within the family. When Treatise was published in 1981, it was
already apparent that the American family had entered a period of rapid change: birth rates had
been falling for 20 years; cohabitation and childbirth without legal marriage had risen; divorce
had become commonplace; and women, including mothers of young children, were entering the
labor force in record numbers. Becker wrote in his “Introduction” that “the family in the Western
world has been radically altered—some claim almost destroyed—by the events of the last three
decades” (p. 1).
The American family has not been destroyed by these changes, but it has been radically
altered. Family structure has become more heterogeneous and less stable. Long-term marriage
combined with childrearing is no longer a near-universal adult experience, and the intense gender
specialization that characterized the traditional nuclear family of the 1950s now seems archaic.
In a continuation of a long-term trend noted by Becker, the economic role of the family
continued to decrease as the market and the state supplemented or replaced more and more
family functions, from food preparation to old-age support.
We begin by discussing how families have changed in recent decades: the separation of
sex, marriage, and childbearing; fewer children and smaller households; converging work and
education patterns for men and women; class divergence in partnering and parenting strategies;
and the replacement of what had been family functions and home production by government
programs and market transactions.
2
After discussing recent work in family economics that
1
Although Becker began with three foundational assumptions, in his more recent work—for example, Becker
(1996)—he relaxes the assumption of stable preferences. For a discussion, see Pollak (2003).
2
Betsey Stevenson and Justin Wolfers (2007) provide additional evidence on trends in marriage and divorce.
3
attempts to explain these changes, we point out some challenging areas for further analysis, and
highlight issues of commitment in two primary family relationships: those between men and
women, and those between parents and children. We then consider the effectiveness of policies
to target benefits to certain family members (like children) or to promote marriage and fertility.
Changes in the American Family, 1960–2006
The Separation of Sex, Marriage, and Childrearing
In the early days of family economics, models of marriage and fertility reflected widely
accepted social norms that were broadly consistent with social reality. Men were supposed to
finish school, get a job, marry, and have children. Educational attainment and employment were
less important for women, for whom the crucial steps were marriage, childbearing, and
childrearing. Economists assumed that decisions about goods purchases, labor supply, fertility,
and child investments maximized a family utility function subject to family resource constraints.
Single-parent families were the result of divorce, which economists assumed must be due to an
unhappy surprise in the realized value of marriage. Nonmarital childbearing was an anomaly.
Since the 1960s and 1970s, however, a framework that considered sex and childbearing
only within the context of a committed partnership has become increasingly disconnected from
reality. Reliable and convenient contraception and the availability of safe and legal abortion
permitted sex with minimal risk of childbearing and the concomitant long-term commitment to
parenting. These developments reduced the cost of premarital sex and, hence, the cost of
remaining single. Combined with rising income and the increasing independence of women, this
encouraged delayed marriage and delayed childbearing. As an alternative to marriage,
cohabitation provided increasing numbers of couples with the benefits of coresidence per se,
including economies of scale in consumption. In the 2000 Census, unmarried couple households
made up 9 percent of all coupled households, with 11 percent of these being same-sex couples
(Simmons and O’Connell, 2003, p. 1 and table 1).
Nonmarital fertility has risen dramatically in the past quarter century. According to the
CDC National Center for Health Statistics, 37 percent of U.S. births were out-of-wedlock in
4
2005.
3
Though fertility outside legal marriage has become commonplace in many developed
countries, especially in northwestern Europe, most unmarried European couples are living
together when their child is born. In contrast, in the United States, the majority of nonmarital
births are to lone mothers, although the proportion of nonmarital births to cohabiting couples is
increasing (Bumpass and Lu, 2000). This emergence of nonmarital childbearing reflects, to a
large extent, the decline of the social imperative that premarital pregnancy should lead to
marriage. For many men, paternity has been separated from parenting responsibilities other than
financial obligations, which are increasingly enforced by the state.
The increase in nonmarital childbearing and in divorce rates has altered the living
arrangements of children. The proportion of children under age 18 who are living with only one
parent rose from 9 percent in 1960 to 28 percent in 2005. These children also experience many
transitions in where and with whom they live, since most unmarried mothers eventually marry,
and the divorced remarry.
4
With multiple, serial partnerships between parents, families have
become more complex, with uncertain implications for parent–child relationships and
investments in children. Using a recent sample of births to unmarried women, Mincy (2002)
found that the majority of mothers with two or more children had at least one child whose
biological father was not the father of their most recent child. He found that this “multiple
partner fertility” compromised the marriage prospects of these new mothers.
All these factors have combined to increase the heterogeneity of families and to decrease
the stability of the living arrangements in which children are reared. Moreover, with increased
cohabitation, childbearing that frequently precedes the establishment of a long-term partnership,
and an increased propensity for adult children to move back in with parents during extended
education or early parenthood, “the sequencing of adult transitions has become increasingly
complicated” (Furstenberg et al., 2004).
Smaller Families and Longer Lives
3
The percentage of nonmarital births differs by race and ethnicity: 25 percent of non-Hispanic white births, 48
percent of Hispanic births, and 70 percent of non-Hispanic black births were nonmarital.
4
Before the twentieth century, serial partnerships and “blended” families arose from the remarriage of widows and
widowers with children. In the late twentieth century, they arose from the remarriage of divorced men and women
with children, or from the marriage of those who become parents in nonmarital relationships.
