Family cap provisions and changes in births and abortions

As part of welfare reform efforts in the 1990s, 23 states implemented family caps, provisions that deny or reduce cash assistance to welfare recipients who have additional births. We use birth and abortion records from 24 states to estimate effects of family caps on birth and abortion rates. We use age, marital status, and completed schooling to identify women at high risk for use of public assistance, and parity (number of previous live births) to identify those most directly affected by the family cap. In family cap states, birth rates fell more and abortion rates rose more among high-risk women with at least one previous live birth compared to similar childless women, consistent with an effect of the family cap. However, this parity-specific pattern of births and abortions also occurred in states that implemented welfare reform with no family cap. Thus, the effects of welfare reform may have differed between mothers and childless women, but there is little evidence of an independent effect of the family cap.     

Do family caps reduce out-of-wedlock births? Evidence from Arkansas, Georgia, Indiana, New Jersey and Virginia

During the 1990s, 23 states implemented family cap policies as a means to reduce the incidence of out-of-wedlock births among welfare recipients. Using Current Population Survey data from 1989 to 1999, we examine the impact of family cap policies on the birth rates of single, less-educated women with children. We use the first five states that were granted waivers from the Department of Health and Human Services to implement family caps as “natural experiments.” Specifically, we compare trends in out-of-wedlock birth rates in Arkansas, Georgia, Indiana, New Jersey and Virginia to trends in states that did not implement family caps or any other waivers prior to the Personal Responsibility and Work Opportunity Reconciliation Act. We employ several techniques to increase the credibility of results from our “natural experiment,” such as the inclusion of multiple comparison groups, controls for differential time trends, and “difference-in-difference-in-differences” estimators. Our regression estimates generally do not provide evidence that family cap policies reduce the incidence of out-of-wedlock births among single, less-educated women with children.     

State-level Welfare Policies and Nonmarital Subsequent Childbearing

Using discrete time event history analyses of data from the Panel Study of Income Dynamics (PSID), we examine the association between state-level welfare waiver policies implemented before the 1996 welfare reform legislation and the risk of a nonmarital subsequent birth. Our study makes a unique contribution to the existing literature by using a national-level sample of unmarried mothers who ever received welfare. This high-risk sample represents the women of most interest to policymakers, as it is the exact group to whom welfare reform is targeted—welfare mothers at risk of having nonmarital additional births. The state policies we study include: family cap, earnings disregard, work exemptions, work requirements, and sanctions. We conclude that, although reducing the number of nonmarital births is a key goal of welfare reform, state-established welfare waiver policies do not have any influence on women’s childbearing behaviors in this sample, net of women’s individual characteristics and state economic environments. Even the family cap policy, which was designed for the sole purpose of reducing additional births, has no significant association with nonmarital subsequent childbearing. Instead, personal characteristics, not public policies, are stronger determinants of women’s childbearing decisions. Age, race/ethnicity, marital status, number of previous children, education level, and welfare receipt are significantly associated with nonmarital subsequent births. Overall, this paper contributes to an expanding body of research that shows minimal effects of welfare waivers on fertility. Our work suggests that more targeted policies are necessary to be able to influence individual family formation behaviors.     

Do family caps reduce out-of-wedlock births? Evidence from Arkansas,Georgia, Indiana,New Jersey and Virginia

During the 1990s, 23 states implemented family cap policies as a means to reduce the incidence of out-of-wedlock births among welfare recipients. Using Current Population Survey data from 1989 to 1999, we examine the impact of family cap policies on the birth rates of single, less-educated women with children. We use the first five states that were granted waivers from the Department of Health and Human Services to implement family caps as natural experiments. Specifically, we compare trends in out-of-wedlock birth rates in Arkansas, Georgia, Indiana, New Jersey and Virginia to trends in states that did not implement family caps or any other waivers prior to the Personal Responsibility and Work Opportunity Reconciliation Act. We employ several techniques to increase the credibility of results from our natural experiment, such as the inclusion of multiple comparison groups, controls for differential time trends, and difference-in-difference-in-differences estimators. Our regression estimates generally do not provide evidence that family cap policies reduce the incidence of out-of-wedlock births among single, less-educated women with children.     

Blacks and the family cap: pregnancy, abortion, and spillovers

While reducing out-of-wedlock childbearing is a central goal of welfare reform, most policymakers prefer achieving this objective via a reduction in nonmarital pregnancy rates rather than through an increase in the incidence of abortion. Using aggregate state-level data from 1984 to 1998, I estimate fixed effects models that allow for autocorrelated and heteroskedastic disturbances to examine the association between the family cap and nonmarital birth, pregnancy, and abortion rates. I find robust evidence that the family cap is associated with a reduction in nonmarital birth rates, particularly among black women. This reduction is driven by a reduction in nonmarital pregnancy rates rather than through an increase in abortion or marriage rates. These findings suggest that that the stigmatizing effect of the family cap may influence the nonmarital pregnancy decisions of black women.

Exiting welfare in the 1990s: Did public policy influence recipients' behavior?

This paper evaluates the association between welfare policies implemented by states in the early to mid-1990s and the rate at which female household heads with children exit AFDC for work or for non-work reasons. The results show that waiver policies requiring work, such as the work requirement and the elimination of exemptions for mothers of very young children, are associated with more rapid exits through employment. Incentive policies that make work more attractive by allowing mothers to remain eligible as earnings rise extend the period of welfare receipt. Neither time limits, the family cap, nor sanctions were related to welfare exits during this early period of welfare reform. Results suggest that other policies implemented in the 1993-1996 period, such as the EITC, may have also influenced welfare exits. In addition, the economy played an important supportive role in facilitating recipient exits from public assistance. While relatively high unemployment rates did not deter recipients from working, state prosperity increased the positive impact of work requirements on work exits.     

Legal abortion and fertility in Maryland, 1960–1971

In the brief period between 1967 and 1971 about one-third of the state legislatures passed abortion reform bills, and in states such as Maryland the number of legal abortions soared. Maryland with its good reporting system for legal abortions, as well as its demographic representativeness, appears to offer an ideal "test situation" for assessing the impact on fertility of the new liberalization. Data on live births and reported induced abortions to residents of the state have been compiled and analyzed in an effort to interpret the recent changes in birth rates. Variables examined include maternal age, birth order, race, and legitimacy.Since 1968, Maryland, along with higher than national average abortion ratios, has experienced a rate of decline in fertility greater than that for the nation. In addition, most of the age and parity groups with high abortion ratios show fertility declines greater than those for groups not using abortion as extensively, Nevertheless, because a number of different factors simultaneously influence fertility, it is hazardous to make accurate cause-and-effect statements on the relationship of any single one of these to the observed change.     

An economic analysis of pregnancy resolution in Virginia: specific as to race and residence

This study analyses an economic model of pregnancy resolution; that is, a model of the choice by a pregnant woman to abort her fetus or carry it to term. This analysis, using an analytical model derived from the household utility framework, adds to previous research by presenting race and residence specific estimates of how individual characteristics, history of abortion, and the community-based factors determine women's choices of giving birth vs. abortion. The main data for estimating the model were drawn from the 1984 vital statistics of all induced abortions and live births in the Commonwealth of Virginia. The major findings indicate that low parental education, high maternal age, previous early abortions, and the availability of abortion providers all significantly reduce the probability of choosing the live birth option. Married status and the availability of family planning clinics significantly increase the probability of the live birth option. The findings also suggest that women's choices between abortion and live birth vary substantially with race (white vs. black) and residential (urban vs. rural) location.I am very grateful to Professors Michael Grossman and Theodore Joyce at CUNY/NBER for their advice and comments on earlier version of this paper, to Professors William Hsiao at Harvard and Richard Ernst at USC for their supportive encouragement and insightful comments, and finally to two anonymous referees for their constructive suggestions in revising this analysis.     

Welfare generosity, pregnancies and abortions among unmarried AFDC recipients

Even before the 1996 overhaul of the U.S. welfare system, a number of states had ended the practice of paying extra benefits to families who have additional children while receiving welfare. Proponents believe that this reform can reduce births to recipients, however many worry that it may encourage women to obtain abortions. Using a sample of unmarried AFDC recipients from the NLSY, we estimate a bivariate probit model of pregnancy and, conditional on becoming pregnant, the probability of abortion. Our results lend some support for the proposition that reducing incremental AFDC benefits will decrease pregnancies without increasing abortions. Received: 16 April 1998/Accepted: 11 March 1999     

Recent Trends in the Timing of First Births in the United States

We use vital registration data published since 1979 to update trends in the timing of first births. Two important trends are documented. First, the likelihood that childless women over age 30 will have a first birth has increased since the 1970s. This change shows that women born in the 1950s are "catching up" on fertility postponed at younger ages. Second, racial differences in the timing of first births are very large. For those born in the 1950s, nonwhites have first births much earlier, and far fewer nonwhite than white women will remain permanently childless. In the second part of the paper, we use these data for recent years to assess earlier projections of childlessness based on cohort and period approaches. We also assess the accuracy of stated intentions to have no children.     

Proximate determinants of Philippine fertility: 1983 levels and patterns

Data from the 1983 National Demographic Survey are used to analyze the proximate determinants of Philippine fertility in each of the 3 stages of family formation and to identify all of the direct and indirect factors affecting fertility levels and trends. 10,843 ever-married women and 12,771 children were included. The analysis pertains first to the starting patterns of family formation, the age at first birth, and the proximate determinants (age at menarche, age at first marriage/union, conception before first birth, fetal wastage first birth, interval between first marriage and first birth). Further analysis examines birth spacing patterns including the postpartum nonsusceptible period, the exposure interval and stopping patterns. Almost all births occur within marriage, and childbearing begins late at 22.5 years. However, 15.4% of first births are conceived premaritally. The mean age at first birth increases from younger to older cohorts. Urban women were slightly older (23.0 years) at the birth of their first child. Those with education below the 4th grade had first births 3.5 years earlier. Contraceptive use was low at 1.8% before first birth. Younger cohorts were more likely to use birth control and urban wives were more likely to use it than rural wives. 6.4% reported a first pregnancy ending in nonlive births, which were primarily spontaneous abortions (5.2%), stillbirths (1.0%), and induced abortions (.2%). 5.8% report never having been pregnant and 1.1% never having given birth to a live-born child. 20.4% were childless between the ages of 15-24 years, and 4.6% between 25-34 years. Childlessness was slightly higher among urban women (7.1%) than rural women (6.7%). A decreasing age at menarche has appeared; i.e., 13.6 years for the cohort 15-24 years, and 14.0 for the oldest cohort. By age 15, 82.9% had begun menstruating. The mean age at marriage is early at 20.7 years, and older cohorts tended to marry later at 21.4 years. Urban women marry a year later (21.4 years) than rural women. Lower educated women marry 4 years earlier. The mean length between first marriage and first birth was 18.4 months. In the younger cohorts, spacing patterns are shorter. Postpartum susceptibility is short. Return to sexual relations after a birth occurred at 2.8 months. The exposure time required to conceive is fairly long at 16.6 months and is attributed to contraceptive use, since coital frequency is high and temporary separation is infrequent. The average age at last birth is late at 37.6 years.     

Eastern Europe: pronatalist policies and private behavior

Fertility trends in the 9 Eastern European socialist countries (Albania, Bulgaria, Czechoslovakia, German Democratic Republic, Hungary, Poland, Romania, USSR, Yugoslavia) are reviewed. Official policy in all these countries but Yugoslavia is explicitly pronatalist to varying degrees. Attention is directed to the following areas: similarities and differences; fertility trends (historical trends, post World War 2 trends, and family size); abortion trends (abortion legislation history, current legislation, abortion data, impact on birth rates, abortion seekers, health risks, and psychological aftereffects); contraceptive availability and practice; pronatal economic incentives (impact on fertility); women's position; and marriage, divorce, and sexual attitudes. The fact that fertility was generally higher in the Eastern European socialist countries than in Western Europe in the mid-1970s is credited to pronatalist measures undertaken when fertility fell or threatened to fall below replacement level (2.1 births/woman) after abortion was liberalized in all countries but Albania, following the lead of the USSR in 1955. Fertility increased where access to abortion was again restricted (mildly in Bulgaria, Czechoslovakia, and Hungary at various times, and severely in Romania in 1966) and/or economic incentives such as birth grants, paid maternity leave, family and child care allowances, and low interest loans to newlyweds were substantially increased (Bulgaria, Czechoslovakia, Hungary, and Poland to some extent, in the late 1960s and early 1970s, and the German Democratic Republic in 1976). Subsequent declines in Bulgaria, Czechoslovakia, Hungary, and Romania suggest that policy induced increases in fertility are short-lived. Couples respond to abortion restrictions by practicing more efficient contraception or resorting to illegal abortion. It is evident that the region's low birth rate is realized mainly with abortion, for withdrawal remains the primary contraceptive method in all countries but Hungary and the German Democratic Republic. It seems that cash incentives have advanced the timing of 1st and 2nd births without substantially increasing the 3rd births required to keep national fertility above replacement level. Demographic factors alone will most likely keep birth rates low in several Eastern European countries during the 1980s and the 1990s. Due to the low birth rates in the 1960s, there will be fewer women in the prime childbearing ages of 20-29 in at least Poland, Czechoslovakia, Bulgaria, and Hungary. It becomes clear that policy efforts to influence private reproductive behavior can only be moderately successful if the living conditions are such that women are determined not to have more than 1 or 2 children.     

Baby boomlet for older mothers?

In recent years, a number of celebrities have begun childbearing after age 35. The phenomena of older first-time mothers has received a great deal of attention in the popular press. Are these celebrities indicative of a national trend? Does the increase in fertility portend a reversal of the declines in fertility which have been occurring since the baby boom? The present paper uses central and cumulative birth rates for cohorts of American white women born between 1882 and 1953 to investigate childbearing between ages 35 and 50. While there has been a noticeable upsurge in first birth rates for cohorts in their mid to late 30s in recent years, overall central birth rates for women in their 30s are among the lowest on record, with cumulative birth rates at record low levels. A major reason for this is that these women are having relatively few third and higher order births. These cohorts will need to have a relatively high proportion of births in their older years of childbearing in order to reach replacement level. However, attaining replacement level is unlikely because such a high proportion of women have remained childless at ages 35–40 and a relatively low proportion are having three or more children.     

Capping Kids: The Family Cap and Nonmarital Childbearing

An explicit goal of policymakers in drafting welfare reform policies was to reduce incentives for nonmarital childbearing. This paper estimates the extent to which state welfare reforms have lowered age and race-specific nonmarital fertility. Using state-level data from 1984 to 1999—a time period that includes the passage and implementation of national welfare reform—we estimate fixed effects models corrected for heteroscedasticity and autocorrelation. We find evidence that the family cap, a policy that decreases or eliminates the incremental increase in benefits for mothers who have an additional child while on welfare, is associated with a decline in nonmarital birth ratios. However, we also find that the family cap is associated with higher marital birth rates. Taken together with other research, our findings suggest evidence of policy endogeneity.

Family planning practice and women's fertility decision-making power

This paper investigates the effects of family planning practice on fertility decision-making power in South Korea. The log-linear analysis of the 1981 survey data by the Institute of Population and Health Services Research, Yonsei University, Seoul, Korea, shows that those urban and rural women who practice family planning or have experienced abortion exercise greater influence on a couple's fertility decision making than those who do not practice family planning or who have had no abortion experience. In addition, there is the interactive effect of abortion experience and contraceptive use on fertility decision making among urban women. This finding is significant because regardless of how birth control is available within a society, birth control use enhances women's decision making power where fertility is concerned.     

Does welfare reform affect fertility? Evidence from the UK

This paper provides evidence on the effect of welfare reform on fertility, focusing on UK reforms in 1999 that increased per-child spending by 50% in real terms. We use a difference-in-differences approach, exploiting the fact that the reforms were targeted at low-income households. The reforms were likely to differentially affect the fertility of women in couples and single women because of the opportunity cost effects of the welfare-to-work element. We find no increase in births among single women, but evidence to support an increase in births (by around 15%) among coupled women.     

The roles of child support enforcement and welfare in non-marital childbearing

This paper examines the effects of stronger child support enforcement and declines in welfare benefits on changes in non-marital childbearing between 1980 and 1996. Economic theory suggests that stricter child support enforcement will increase the costs of children for unwed fathers, making them less likely to have a child outside marriage. Reductions in welfare benefits also are expected to increase the costs of non-marital childbearing for both mothers and fathers. We examine these hypotheses, using aggregate state-level data and fixed effects regression models to identify the effects of policies on non-marital birth rates. We find that both stricter child support enforcement and declines in welfare benefits deter non-marital births. However, the estimated effects of child support enforcement are more robust and larger than those of welfare. The estimates imply that in the 1980–1996 period, increases in child support enforcement led to a decline in non-marital birth rates in the range of 6% to 9%, whereas decreases in welfare benefits led to a decline in the range of 2% to 4%. Received: 28 June 2000/Accepted: 17 September 2001     

Unmarried Motherhood: Recent Trends,Composition, and Black-White Differences

The National Survey of Family Growth (1982) is used to examine the extent to which racial differences in premarital birth rates can be explained by differences in parents' socioeconomic status, family structure, and residential characteristics. The findings document a large diversity in premarital births within both populations. Black women from high-risk backgrounds are three times more likely to have a premarital birth than black women from low-risk backgrounds. Racial differences in premarital births arise because (1) black women are more likely to come from high-risk backgrounds and (2) black women from low-risk backgrounds are more likely to have a premarital birth than white women with similar characteristics. There are similar rates of premarital births by race among persons from high-risk backgrounds.     

Does welfare reform affect fertility? Evidence from the UK

This paper provides evidence on the effect of welfare reform on fertility, focusing on UK reforms in 1999 that increased per-child spending by 50% in real terms. We use a difference-in-differences approach, exploiting the fact that the reforms were targeted at low-income households. The reforms were likely to differentially affect the fertility of women in couples and single women because of the opportunity cost effects of the welfare-to-work element. We find no increase in births among single women, but evidence to support an increase in births (by around 15%) among coupled women.     

A vital statistics-based procedure for estimating conception rates

A method for estimating conception rates, using vital statistics data, is developed and applied to data on five-year age groups of California women for 1971. The approach is deterministic and allocates total exposure time to the known pregnancy outcomes of live birth, spontaneous abortion, and induced abortion. The population at risk is defined to exclude women who are known to be sterile or sexually inactive. Early fetal loss, premarital conception, and contraceptive use are taken into account. Estimates are made of the fecundability which would obtain ifno contraception were used.