Parental Leave: Policy and Research

Journal of Social Issues. Vol. 52 , No. 3, 1996, pp. 91-109

Parental Leave: Policy and Research

Janet Shibley Hyde, Marilyn J. Essex, Roseanne Clark, Marjorie H. Klein, and Janis E. Byrd University of Wisconsin-Madison

Parental leave, a key aspect of family policy, is a relatively new concept in the United States; it is crucial that policy development be guided by social science research. In this article we first review policy issues associated with parental leave. Following that, we consider relevant social science research pertaining to parental leave and mothers’ and fathers’ functioning (e.g., mental health, relationship with spouse and infant). We focus particularly on results emerging from the Wisconsin Maternity Leave and Health Project, which suggest, for example, that short leave (6 weeks or less) for women is a risk factor, when combined with other risk factors such as marital concerns, for depression. We conclude with policy implications of this and related research.

In February 1993, President Clinton signed into law the federal Family and Medical Leave Act (FMLA). With that action, he removed from the United States its unique status as the only industrialized nation to have no national family leave policy. The idea of job-protected parental leave for mothers and fathers came relatively late to this country, and as policy has lagged, so has relevant social science research been sparse. This article reviews psychological and sociological research on parental leave, focusing particularly on the results

The Wisconsin Maternity Leave and Health Project was supported by National Institute of Mental Health Grant No. MH44340 to Janet S. Hyde and Marilyn J. Essex, with Roseanne Clark and Marjorie H. Klein. Additional funding was provided by the John D. and Catherine T. MacArthur Foundation Research Network on Depression, the University of Wisconsin Graduate School, and the Wisconsin Psychiatric Research Institute. Special thanks are extended to project staff, Will Shattuck, Laura Haugen, Nancy Smider, Ashby Plant, and Francine Horton, and to the Wisconsin Survey Research Laboratory and its staff. The researchers also thank the many independent study students who coded interviews. Rosalind Barnett and Joseph Pleck provided important consultations in the early stage of the project.

Correspondence concerning this article should be addressed to Janet Hyde, Department of Psychology, University of Wisconsin, 1202 W. Johnson St., Madison, WI 53706. Electronic mail may be sent via Internet to [email protected].

91

0022-4537/96/0900-01~3.0011 0 19% The Society for the Psychological Study of Social Issues

92 Hyde et al.

of our own study, the Wisconsin Maternity Leave and Health (WMLH) Project. We first consider policy issues associated with parental leave. Then we consider social science research relevant to parental leave for mothers and for fathers, including data on mental health, physical health, quality of relationships within the family, and economic factors. Finally, we explore the implications of this social science research for public policy.

Some terms require definition. In this article, “maternity leave” refers to the leave a woman takes from work at the time of the birth of a child; its chief purpose is recovery from childbirth. “Parental leave” is a broader term, referring to leave from work for purposes of caring for a new baby; it is a gender-fair concept, since both mothers and fathers may use this benefit. “Family leave” is a broader term still, referring to leave from work for purposes of caring for an infant or for an elderly or sick family member.

Parental Leave: Policy Issues

When framing a parental leave policy, policymakers-whether legislators or corporate executives-must make decisions regarding a number of issues:

1. Is the leave job-guaranteed? That is, does the employee have a right to return from leave to the same job or a job that is comparable in terms of pay and responsibilities? Job guarantee is the most fundamental provision in parental leave policy. Without it, the employee is not taking a leave. Rather, he or she is quitting or being fired. The FMLA requires that employers provide job-guaranteed leave.

2. Do benefits, especially health insurance and life insurance, continue during the leave? Who bears the cost-the employer, the employee, or is there sharing? The time when a family has a new baby is definitely not the time when they want to be without health insurance. The FMLA stipulates that the employer must continue any employer-provided group health plan for the employee under the same conditions as if the employee were not on leave.

3. What is the length of leave allowed? The FMLA requires employers to provide a minimum of 12 weeks leave.

4. Is there provision for income replacement? Our data indicate that economic factors are a crucial consideration in determining the length of leave that women and men take, regardless of the length of leave provided by legislation. The FMLA provides a minimum standard of unpaid leave. That is, employers are not required to provide paid leave, although they are free to be more generous than the minimum standard and provide partial or complete paid leave.

5. May employees use accumulated sick days or vacation days instead of or

Parental Leave 93

in addition to parental leave days? The use of vacation and sick days is one method that new parents use to gain a paid or partially paid leave. The FMLA permits employees to substitute accumulated paid vacation and personal leave for parental leave (Malin, 1994).

6. Is the policy gender fair? That is, are fathers given the same rights to leave as mothers? Such a provision is crucial if a more equitable division of childcare responsibilities is to emerge. The FMLA provides parental leave for both mothers and fathers. (See Malin, 1994, for an extended discussion of legal issues surrounding fathers and parental leave.)

7. Which employees are excluded from coverage? The political negotiating that led to the passage of the FMLA resulted in some compromises, among them the exclusion of small businesses, defined as those with 50 or fewer employees. Women work disproportionately for small businesses, leaving a large number of women uncovered by the legislation.

Parental Leave Policies in Cultural Context

To put parental leave policies into perspective and to see how cultural forces help shape these policies, it is helpful to compare the current U.S. policy with the policies of other nations. Space does not permit-nor would it be feasible-to enumerate policies in all countries. Therefore, we will review policies in three countries about which there are published analyses: Sweden, Italy, and the People’s Republic of China.

Sweden has the most generous leave policies of any country (Haas, 1991; Haas & Hwang, 1995). Cultural context is a key factor in understanding these policies, which were motivated as part of a broader pronatalist policy designed to restore the Swedish population following severe losses during World War 11. These policies were also part of an effort at social engineering designed to ensure gender equity and to increase the participation of fathers in childrearing. Swedish couples have 15 months of job-guaranteed paid leave to share between them. For the first 12 months, the pay is approximately 90% of normal wages; for the last 3 months it is reduced further. Mothers also have the right to work at 75% time until their child is 8 years of age.

In Italy, maternity leave (not parental leave) legislation was passed easily, beginning in 1950, because it was consistent with that culture’s values about mothers and infants (Saunders, 1991). The current law provides women with 2 months of employer-paid leave prior to the birth of a child and 3 months after the birth, with pay at 70% of normal wages, as well as the option of an additional 6 months of leave at 30% of wages paid by a national social insurance system. In 1977 a parity law was passed that allowed fathers a right to job-guaranteed leave equivalent to mothers’ after 3 months postpartum; the leave is available only to fathers whose wives are employed.

94 Hyde et al.

As early as 1953, Chinese women were guaranteed 8 weeks of maternity leave at full pay (Li Min, 1991). A 1988 law extended maternity leave to 3 months and added 15 days of prenatal leave. All parts of the leave are job- guaranteed and paid. In the baby’s first year of life, the mother is allowed to nurse twice a day while she is at work, and nurseries are provided at work sites.

These three cases illustrate the complex ways in which culture shapes public policy, particularly in an area such as parental leave, which involves key cultural concepts such as the nature and proper functioning of the family. In a country such as Sweden, which prizes gender equity, there are strong inducements for fathers to take parental leave. A relatively poor developing nation such as China, with a culture based on collectivist values, can “afford” these policies, whereas a wealthy nation such as the United States, with individualistic values, has some constituencies that argue that the U.S. cannot.

The Wisconsin Maternity Leave and Health Project

Before reviewing the social science data relevant to parental leave, we will first describe the WMLH Project, the source of many of the results discussed below. The WMLH Project is a longitudinal study of women, their hus- banddpartners, and their infants. The study was motivated by a desire to under- stand how the conditions of parental leave (e.g., length of leave) were related to a variety of outcomes, including mental health, physical health, and the quality of relationships within the family.

We initially recruited 570 women, all in the second trimester of pregnancy, and 550 of their husbands/partners to participate in the study (see Hyde, Klein, Essex, & Clark, 1995, for additional details on the sample and methods). Ap- proximately 78% of the sample resided in the Milwaukee, W1, Standard Metro- politan Statistical Area (SMSA) and the remaining 22% resided in the Madison, WI, SMSA. Women were interviewed in their homes by a female interviewer on four occasions: (a) during the second trimester of pregnancy (Time l), (b) 1 month postpartum (Time 2), (c) 4 months after birth (Time 3), and (d) 12 months after birth (Time 4). Fathers were interviewed by telephone at the same four times. In addition, mother-infant interactions were videotaped at Time 3 and Time 4. At the time of the first interview, 81.5% of the women were employed and 95% were married to their partner.

The first wave of data collection extended from June 1990 to September 199 1. Therefore, the data were collected before enactment of the FMLA. How- ever, the state of Wisconsin had a parental leave law in force at the time that provided 6 weeks of job-guaranteed leave to new mothers and fathers, excluding employees in small businesses.

This is the first large-scale study of this type, investigating maternity leave and outcomes such as mental health, physical health, and relationship quality.

Parental Leave 95

There are some related, smaller studies, and these will be reviewed in the sections that follow.

Parental Leave: Issues for Mothers

Our research addresses four key issues associated with maternity leave for mothers: (a) mental health and stress, (2) physical health and recovery from childbirth, (3) relationship with the husband/partner and with the infant, and (4) economic issues.

Mental Health

In one set of analyses, we examined the women’s mental health at Time 3 , 4 months after the birth, as a function of employment status (employed full time, employed part time, or homemaker) and length of leave from work (see Hyde et al., 1995, for details of statistical analyses and results).

There is abundant empirical evidence that employment for women is associated with improved physical health and, less consistently, improved mental health (see reviews by Crosby, 1991; Repetti, Matthews, & Waldron, 1989). These conclusions are based, however, on studies of women at various times in the life span, and may include single and married women, and women with and without children. Our concern is that the year after a birth may be a time of particular stress stemming from multiple sources, including physical recovery from childbirth, finding suitable child care, caring for an infant with around-the- clock needs, and associated sleep disruption and deprivation for the mother.

We compared women at 4 months postpartum in three employment status groups: (a) employed full time (32 or more hours per week), (b) employed part time (between 6 and 31 hours per week), and (c) homemakers. The outcome measures were depression (measured by the Center for Epidemiological Studies-Depression Scale, CES-D, Radloff, 1977) and anxiety (Spielberger, 1983). There were no significant differences among the three groups for depression. There were, however, differences among the groups in anxiety: women employed full time showed significantly higher levels of anxiety than did women employed part time and homemakers (Hyde et al., 1995).

We conducted hierarchical multiple regression analyses to examine the relationship between length of leave and the outcome measures of depression and anxiety. Of course, we did not believe that the only influence on depression and anxiety would be length of leave. We therefore included other relevant variables such as work-role quality, marital quality, infant temperament, and equity of division of household labor (see Hyde et al., 1995, for details on measures and statistical analyses). Baseline (Time 1) levels of depression and anxiety were controlled; therefore, in the following discussion, “depression” is used as a

96 Hyde et al.

shorthand for “Time 3 depression scores with Time 1, baseline, depression scores controlled .” The analysis for depression indicated that work rewards were negatively associated with depression-i.e., the more rewarding the woman’s work, the lower her depression score. High levels of marital concerns were associated with higher levels of depression. Length of leave did not have a simple main effect. However, there was a significant interaction between length of leave and marital concerns.

Women who took a short leave (6 weeks or less) and who were high in marital concerns showed elevated levels of depression compared with women who took long leaves (12 weeks or more) or with women who took a short leave and who were low in marital concerns. These findings have led us to conceptual- ize short maternity leave as a risk factor, when combined with another risk factor such as a troubled marriage, for increased symptoms of depression.

The regression analysis predicting anxiety showed a positive association between anxiety and hours worked per week, consistent with the analysis com- paring the three employment groups discussed earlier. Difficult infant temperament also was associated with higher levels of anxiety, as were high scores on work concerns. Length of leave did not show a main effect, nor did it interact significantly with any of the other variables.

In summary, the WMLH data provide evidence that short maternity leave (6 weeks or less) is a risk factor, when combined with other risk factors such as a troubled marriage, for depression. Length of leave does not seem to be the crucial variable predicting anxiety; rather, hours worked per week, and especially full-time employment, were associated with higher anxiety scores.

Physical Health

In a period of a brief two generations we have shifted, in the United States, from a belief that childbirth is arduous and requires a long period of recovery, including two weeks in the hospital, to a belief in the “in-control woman” who, using the Lamaze method, spends only one night in the hospital and is ready for work within a week. Meanwhile, there has been little research on the actual time course of various aspects of physical recovery from childbirth.

In one of the few relevant studies, Tulman and Fawcett (1991) interviewed 100 women 3 weeks, 6 weeks, 3 months, and 6 months after delivery. Although traditional obstetrics textbooks indicate that physical recovery is accomplished in 6 weeks (8 weeks for cesarean deliveries), this in fact refers to the length of time for the uterus to return to its pre-pregnant size, and tells us nothing about other aspects of recovery (Tulman & Fawcett, 1991). Tulman and Fawcett’s results indicated that only 27% of women felt they had regained their usual level of energy by 3 weeks postpartum, and that figure rose to 57% by 6 weeks, to 71% by 3 months, and to 87% by 6 months after delivery. Moreover, 6 weeks after

Parental Leave 91

delivery, women under 30 years of age showed significantly greater recovery than women over 30.

Women’s functioning also is influenced by interrupted and lost sleep due to an infant awakening in the night. The Tulman and Fawcett data indicated that 5% of the infants slept through the night at 3 weeks, 40% did so at 6 weeks, 82% did so by 3 months, and 86% slept all night by 6 months of age.

Tulman and Fawcett found no association between a woman’s timing of return to work and her level of recovery, suggesting that factors other than a feeling of physical recovery played the crucial role in determining length of leave. Most of the women respondents said that they returned when they did either because of financial considerations (a theme that is echoed by the partici- pants in the WMLH Project) or because their leave time was over.

Based on the results of another research project, Gjerdingen and Froberg (1991) dubbed the 6 weeks after birth the “fourth stage of labor.” Their design included three groups of women: married, first-time biological mothers; married, first-time adoptive mothers; and controls (married women with no children). At 6 weeks, both adoptive and biological mothers reported more fatigue and less readiness to work at a job, compared with the control group. Biological mothers reported more health problems (specifically, genitourinary problems) than adoptive mothers did. Gjerdingen and Froberg concluded, in agreement with Tulman and Fawcett, that some aspects of health recovery after childbirth continue beyond the sixth week postpartum.

Self-ratings of overall health by women in the WMLH Project indicated worse health at 1 and at 4 months postpartum than during pregnancy or at 12 months postpartum (Byrd, Essex, Hyde, & Klein, 1994). Interestingly, the differ- ence in ratings from 1 month to 4 months was not significant; not surprisingly, women were recovering at 1 month, but they still felt their health was diminished at 4 months. The women reported a mean number of hours of sleep of 7.54 prior to pregnancy, 7.90 during pregnancy, 6.00 at 1 month postpartum, 6.83 at 4 months postpartum, and 6.96 at 12 months postpartum. Clearly the greatest loss of sleep occurs at 1 month postpartum, but at 4 months women are still getting less sleep, on the average, than they were before pregnancy.

In summary, Tulman and Fawcett’s data, like Gjerdingen and Froberg’s and our own, indicate that full recovery from childbirth takes more than 6 weeks and may take 6 months or even more for some women. At the same time, individual differences are great. Some women feel back to their normal level of energy just 3 weeks postpartum-.

Relationships

In this section we consider the quality of women’s relationships with their partner and with their infant in the first year postpartum, and we examine how

98 Hyde et al.

the quality of these relationships may be related to work status and length of leave.

Research consistently has shown a decline in marital satisfaction over the transition to parenthood, i.e., with the birth of the first child (e.g., Belsky, Lang, & Rovine, 1985; Cowan & Cowan, 1992; Cowan et al., 1985). This effect is especially true for women (Belsky, Spanier, & Rovine, 1983), and the decline may last for 18 months or more following the birth (Cowan & Cowan, 1992). Many factors may contribute to this decline, including the reduction in time the couple spend together, the husband’s sense of pressure to support his family, the wife’s worry about what to do about her job or career, the wife’s dissatisfaction with the division of labor in the family, some husbands’ disappointment with their level of involvement with the infant, a decline in the quality of the sexual relationship, and in some cases, ambivalence about having the baby in the first place (Cowan & Cowan, 1992).

Importantly, many items on this list are closely related to issues and policies concerning gender and parental leave. For example, job-guaranteed leave is crucial in reducing women’s fears about loss of job (note that the Cowan and Cowan study was conducted before the enactment of the FMLA). A gender-fair parental leave policy-allowing men and women equal access to leave-is important for equalizing the division of family labor and for giving fathers the opportunity for more involvement with their baby.

We turn now to the mother’s relationship with her infant. A number of noted specialists in child development have offered expert opinions about the issue of length of leave and its importance for optimal infant development. Brazelton (1985) has argued that mothers should spend the first 4 months at home with the infant full time. Zigler and Muenchow (1983) advocated a 6-month infant-care leave, based on considerations of the needs of infants.

Such recommendations are driven by consideration of a number of processes in infant development. First among these concerns the importance of the infant forming a secure attachment to a stable caregiver during the first months of life (Weinraub & Jaeger, 1991). Indeed, Bowlby (1973) believed mothers and infants should be in an exclusive, one-to-one relationship until the child is 3 years old, a recommendation that seems almost quaint given current patterns of employment (55% of mothers with children under age 3 are in the labor force; U.S. Department of Labor, 1993). On the other hand, Weinraub and Jaeger (1991) advanced the interesting hypothesis that it is better for the infant if the mother returns to work somewhat before the child reaches 8 months of age. Beginning around 8 months, the infant develops a wanness of strangers and distress at separation from the mother. Weinraub and Jaeger hypothesized that delaying the time of return to work until after the onset of stranger and separation anxiety may make it more difficult for the infant than it would have been earlier.

Empirical data on the relationship between the infant’s attachment and the

Parental Leave 99

timing of the mother’s return to work (or length of leave) have been conflicting and controversial. Some studies have found less secure attachment among infants whose mothers returned to work during the first year (e.g., Belsky & Rovine, 1988), whereas other studies have found no disturbances in attachment, even for infants whose mothers returned to employment in the first 6 months (e.g., Chase- Lansdale & Owen, 1987).

Another concern arises from the infant’s need to stabilize biological rhythms of sleeping, waking, and feeding, which depends on a responsive synchronicity with the primary caregiver (Koester, 1991). This regulation process occurs gradually over the first 3 months of life. It may be advisable therefore to introduce a new, additional caregiver only after these processes are stabilized.

In the WMLH Project, we examined the videotaped interactions of 198 employed women and their infants at 4 months of age (Clark, Hyde, Essex, & Klein, in press). Multiple regression analyses were used to predict 6 components of the quality of mother-infant interactions rated from the videotapes. The results indicated that shorter leaves were associated with more negative maternal affect and behavior. Length of leave interacted significantly with other possible risk factors such as mother’s depressive symptoms, mother’s physical health symptoms, and difficult infant temperament in predicting the quality of mother- infant interactions. Among those experiencing vulnerabilities that are common in the postpartum period (e.g., elevated depressive symptoms), longer leave (12 weeks or more) can provide a buffer effect for women and their infants. Policies must appreciate these individual differences and allow for choices based on the emotional, physical, and temperamental needs of women and their families.

This discussion, of course, begs the question of why fathers could not be the primary caregiver and attachment figure for the infant, or the joint primary caregiver. Such questions lead us to take the issue of fathers’ parental leave seriously.

Economic Issues

Economic considerations surrounding parental leave for women are many, involving basic issues of survival and the family’s standard of living. Women today do not work to provide a few luxuries for the family; rather, they work out of economic necessity, to maintain the family’s standard of living. Women’s contributions to family income are substantial. For example, the median income of married-couple families with the wife in the paid labor force is $48,169, compared with $30,075 for families without the wife in the paid labor force (U . S . Department of Labor, 1993).

Historically, the most basic economic issue has been loss of job, leading to complete loss of income. For this reason, the job-guarantee provision of the FMLA is crucial. Although some have called parental leave a yuppie issue,

100 Hyde et al.

allowing successful career women to keep their glamorous jobs, in fact job- guaranteed parental leave is far more important for the working class and working poor. The brilliant economic analyses of Spalter-Roth and Hartmann (1991) showed that, before the FMLA, black women and poor women suffered the most because they were more likely to lose their jobs following the birth of a child. Moreover, these women are more likely to contribute a large percentage to family income or to be the sole earner.

Even among women who are fortunate to have job-guaranteed parental leave, as most of the women in the WMLH Project did, economic issues still exert a powerful force. At Time 4, 12 months after the birth, we asked the WMLH women to reflect back on their maternity leave and whether they would do it the same or differently if they had it to do over again. In response to the question-"How do you feel about the amount of time you took 0@?"-31% said it was just right, but 50% said it was a little too short and 16% said it was much too short (Hyde, 1995). Only 2% said it was too long.' The data indicate, then, that the majority of women wished they had taken a longer leave than they actually did. In another question, we asked the women in what ways they would have done things differently; 50% said they would have taken a longer leave, and an additional 17% said they would have worked fewer hours per week in the year after the birth.

Why didn't these women take longer leaves? By far the most common reason, given by 46% of the women, was that it was because they needed the money or could not afford a longer leave. Many of the women in the study were able to gain pay for some of their leave by using accumulated sick days and vacation days. What the women could not afford was additional unpaid leave. Insofar as women of color are disproportionately represented at lower income levels, these financial issues have an even greater impact on them.

These findings call into question whether women will really be able to use the full 12 weeks of leave guaranteed by the FMLA, if that leave is unpaid. To address this issue, we conducted interviews with 25 women in the WMLH Project who gave birth to an additional child after the passage of the FMLA (unpublished data). Their mean length of leave was 10.69 weeks. These women took slightly shorter leaves than the WMLH Project women did before the FMLA. Moreover, these 25 women who took leaves after the FMLA took an average of 12.91 weeks with the earlier child, before the FMLA. That is, they

'The median length of leave taken by the women was 9 weeks; the mean was 11.49, and 71% of the women had returned to work by 12 weeks. This average length of leave differs from our earlier report of a mean of 8.49 weeks (Hyde et al., 1995). That earlier report was based on data only through the Time 3 interview, 4 months after birth. Therefore, women who were still on leave at that time did not contribute to the data. The median reported here includes all women who had returned to work in the first 12 months. It is also noteworthy that this mean length of leave is noticeably longer than the 6 weeks guaranteed by the Wisconsin law in effect at the time.

Parental Leave 101

took a shorter leave, by more than 2 weeks, with the later child born after the FMLA. There was no evidence therefore that women took longer leaves as a result of the FMLA, and we believe this is a result of the fact that it guarantees only unpaid leave.

Economic factors clearly are at the very core of parental leave issues for women.

Parental Leave: Issues for Fathers

Because of normative assumptions that mothers should be the primary care- givers for infants and therefore should be the ones to take parental leave, the topic of fathers’ parental leave has received even less research attention than mothers’. Here we review the scant existing research.

Haas’s (199 1) research on the Swedish experience is of particular interest because no other country has worked as vigorously to encourage fathers to take extended parental leaves. In 1974, the first year that parental leave policies were instituted in Sweden, only 3% of eligible fathers actually took leave; by 1976, that number was only 5%. In 1978, parental leave benefits became more generous; by 1980, 23% of eligible fathers took parental leave, and by 1989, 44% were doing so (Pleck, 1993). These results indicate that, over a period of 15 years and with changes in legislation, remarkable social changes can occur. According to Haas’s (1991) data for the 1980s, those fathers who took leave took an average of 53 days, compared with 225 for mothers.

Haas (1991) argued that four categories of factors limit the amount of parental leave taken by Swedish fathers: biological factors, social-psychological factors, lack of social support, and economics. In regard to the first influence, biological factors, Swedes traditionally have assumed, as have their U.S. coun- terparts, that mothers are better able to care for infants than fathers are. Further- more, 90% of Swedish mothers breastfeed for 5 months or more, again favoring women over men in taking leave in the early months of the child’s life. Social- psychological factors include men’s belief that they should be the primary bread- winner and the men’s lack of exposure to role models of men caring for infants. In regard to social support, fathers fear that colleagues at work will form a negative opinion of them for taking parental leave. Economic factors are involved as well, since the system of pay reimbursement favors leave taking by the lower paid parent, usually the mother. These dynamics are similar to those in the United States.

Two previous U .S. studies have investigated fathers’ parental leaves. Based on interviews with 142 fathers before passage of the FMLA, Pleck (1993) found that 87% took at least some time off work, with an average of 5.3 days. The men took these days as sick, vacation, and discretionary days, and tended not to think of them as parental leave.

102 Hyde et al.

The Work and Families Institute (Bond, Galinsky, Lord, Staines, & Brown, 1991) surveyed both employers and employees, prior to implementation of the FMLA, in four states with parental leave legislation. Data were collected from employees who had babies before the legislation and employees who had babies after the legislation. They found that 70% of fathers took at least some leave prelegislation, compared with 75% postlegislation. Fathers took a mean of 3.7 days prelegislation and 4.7 days postlegislation (note that this is state legislation, not the FMLA), generally taking the leave as paid vacation and sick days.

Analyses of the WMLH data indicated that 91% of fathers took at least some leave, with a mean of 5 days (Hyde, Essex, & Horton, 1993). This average of 5 days matches closely the comparable figures from the Pleck and the Work and Families Institute studies. Unlike these other studies, we had a wealth of psychosocial measures that allowed us to explore correlates of fathers’ leave taking. We based predictions on two theoretical considerations. Identity theory (Stryker, 1980; Stryker & Serpe, 1982), derived from symbolic interactionism, predicts that men who are highly committed to the family role, relative to the work role, should manifest this commitment in behavior, and so take a longer leave. The second theoretical consideration had to do with changes in expecta- tions for the father role over the last few decades. In the 20th century, the good provider role has been an important component of male identity (Pleck, 1981). Men are expected to support their families well by working diligently and earn- ing high wages. For men, then, a high degree of family commitment might not manifest itself in taking a long leave, but rather in taking a short leave so as to get back to work quickly, do well on the job, and earn money to support his wife and new child, at least under conditions of unpaid leave. There is an additional degree of complexity, though, because there has been a shift over the last two decades toward expecting fathers to be more involved with their children (Jump & Haas, 1987), at least among dual-earner couples and those who hold liberal ideologies about gender roles (Volling & Belsky, 1991). We call this a shift from the good provider model to the father involvement model. Therefore, we pre- dicted that the longest leaves would be taken by men who were both high in family commitment and held liberal gender role ideologies. That is precisely the result we obtained (Hyde et al., 1993).

Malin (1994) has argued that the FMLA will provide fathers with the opportunity for parental leave in a way that has not previously been possible in the U.S. Based on the reports of the small sample of WMLH mothers (n = 25), discussed earlier, who had an additional child after the FMLA went into effect, fathers took only an average 4.21 days of leave post-FMLA (unpublished data), which is approximately the same length of leave that fathers took before the legislation (but in a state that had parental leave legislation). Fathers do not seem to be leaping at the opportunity to take the longer leaves they are guaranteed. However, the legislation was quite recent at the time the data were collected. As

Parental Leave 103

the provisions of the FMLA become more widely known, and as attitudes in the workplace change, fathers in the U.S., like those in Sweden, may gradually shift to taking longer leaves.

Further analyses of the WMLH Project data will allow us to examine many of the same questions that we have for mothers. We can ask, Is the length of fathers’ leave related to their mental health? Alternatively, is the length of mothers’ leaves related to fathers’ mental health? Is the length of the father’s leave related to the quality of his relationship with his wife or partner? Is it related to the quality of his relationship with his new baby? These are exciting new avenues for exploration. The research of others hints at interesting possibilities. For example, Ferketich and Mercer (1989) assessed fathers during their spouse’s pregnancy and 1, 4, and 8 months after the birth. They found that fathers’ self- reported health showed a decline at all three postpartum testing times. Moreover, they concluded that more involvement with the baby had a positive effect on health. This suggests that longer parental leaves for fathers, providing them with the opportunity for more involvement with their baby and for the satisfactions such involvement brings, may have positive effects on physical and mental health.

Policy Implications

In this section we consider policy implications of the research reviewed in the earlier sections.

The Case for a Minimum of 12 Weeks of Parental Leave

As noted earlier, researchers such as Brazelton (1985) and Zigler and Muenchow (1983) have advocated leaves for mothers of 4-6 months. Do we know more now to inform this discussion?

A number of research findings and theoretical considerations argue for 12 weeks of leave as a minimum for new mothers. First, there are considerations of the mother’s health and well-being. Tulman and Fawcett’s data (1991) indicate that most women do not feel they have regained their usual level of energy until 3 months postpartum. It may be to employers’ advantage to allow women these longer leaves in order to return to normal energy levels more quickly. By 3 months, 82% of babies sleep through the night, allowing mothers relatively uninterrupted sleep before rising to go to a job. Furthermore, the WMLH Project data indicate that short leaves of 6 weeks, in combination with another risk factor such as marital concerns, are associated with elevated depression levels, whereas longer leaves of 12 weeks or more are associated with lower levels of depression, regardless of whether women score high or low on a risk factor such as marital concerns (Hyde et al., 1995).

104 Hyde et al.

Second, there are important considerations regarding optimal infant development. infants need to stabilize their biological rhythms of sleeping, waking, and feeding, processes that seem to depend on having a responsive relationship with a primary caregiver; this process has stabilized by 3-4 months. These processes may progress well in high-quality, stable care, but they may well be disturbed under conditions of shifting care or poor-quality care. In addition, there is a general consensus that infants need to form an attachment to a stable caregiver in the first few months after birth. Research indicates that there is an increase in the likelihood of insecure attachment when the infant is placed in nonparental care very early and there are significant life stresses for the family (Gamble & Zigler, 1988).

These arguments are couched in terms of the well-being of mothers and infants. What research does not tell us-and cannot tell us until sufficient num- bers of fathers take longer leaves-is whether longer leaves of 6 weeks or 12 weeks might benefit fathers, perhaps by reducing depression or improving physical health as a result of more involvement with the infant and the satisfaction that derives from such involvement. These benefits to the father may in turn yield benefits to the infant. If future research does indeed document such effects, policy changes will be necessary to encourage fathers to take longer leaves.

The Case for Paid Leave

Currently, fathers limit themselves to about one week of leave, which they take as paid leave financed by using accumulated vacation and personal days. Mothers seem to limit themselves to 9-12 weeks of leave, part of which is paid through the use of accumulated vacation, personal, and sick days. According to our data, the most frequent reason for mothers not taking a longer leave is that they could not afford it. In short, two conclusions seem clear: (a) fathers will not take much more than one week of leave unless the leave is paid; and (b) many mothers prefer longer leaves than they take and they may not be able to take the full 12 weeks guaranteed by the FMLA, largely due to financial concerns.

Why should we care whether fathers take leaves of more than a week? Feminist theorists have long contended that the inequality of division of labor in the family, and particularly women’s disproportionate responsibility for childrearing, is the cornerstone of gender inequality (e.g., Chodorow, 1978). These patterns may well be set in the first days and months after a birth. One might argue, then, that only with serious parental leave for fathers will there be an alteration in these patterns. Certainly there are considerations of benefits to fathers as well, through increased involvement with their children and the satisfaction this brings.

One caveat is in order. Men’s interest in and willingness to take longer leaves is probably not solely an economic issue. Supervisors’ and co-workers’ attitudes also exert a powerful influence, as men worry about whether they will

Parental Leave 105

be viewed as serious about their work. When the men in the WMLH sample were questioned at Time 1 about what their supervisor’s and co-workers’ reactions would be if they took a substantial amount of time off (a month or more), 43% anticipated that their co-workers would be negative and 63% thought their supervisor would be negative. If men are to have longer leaves, policy change must be accompanied by attitude change.

How, then, might the U.S. finance a policy of paid leave? Several models are available and have proved to be feasible in other countries. For example, in Canada there is a fund much like the U.S. Social Security fund into which employers and employees pay a small tax. Parents on leave then draw pay from this fund, and the burden does not fall on employers. The state of New Jersey has a program that provides pay during leave through a temporary disability fund, drawn on both by mothers with new babies and many others, such as middle- aged men temporarily disabled by heart attacks. Mandatory contributions are shared equally between employer and employee, amounting to .05% of the first $10,700 in wages; in 1987 the average employee paid $56.50 per year into the fund (Bookman, 1991). The average weekly benefit was $169, and pregnancy claims constituted 17% of all claims. In short, there is no mystery about how to provide paid leave. The country and our legislators simply need to develop the will to do so.

Options for Part-Time Employment

Those concerned with issues of women balancing work and family have long argued for employment options such as part-time work and flextime. The importance of such options becomes especially clear when we consider the issue of women returning to work in the first year after a birth. The data from the WMLH Project indicate that there are some stresses associated with full-time employment-at least at 4 months postpartum-that are not associated with part-time employment, as evidenced by the elevated levels of anxiety among the full-time workers compared with the part-time workers and full-time homemakers. When asked what they would have done differently, the most common response from the women is that they would have taken a longer leave, and the second most frequent response was that they would have worked fewer hours. All of these findings point to the importance of policies-whether legislated or voluntary policies-allowing part-time employment for mothers in the first year after a birth.

The Importance of Individual Diferences

Many of the policy implications that we have drawn based on social science research have revolved around group means, thereby running the risk of ignoring often dramatic individual differences. For example, Tulman and Fawcett (1991)

106 Hyde et al.

found that 7 1 % of women reported that they had regained their usual level of energy by 3 months postpartum, but 27% of women said they had by 3 weeks postpartum. Women in the WMLH Project who had a short leave (6 weeks or less) and had marital concerns showed elevated levels of depression, but women who had short leaves and few marital concerns did not show elevated depression. The benefits of longer leaves must be balanced against the health-enhancing effects of employment for women that have been documented in many studies. As we seek to provide health-promoting options for mothers, fathers, and infants, we must recognize that individual needs and circumstances vary widely. A job-guaranteed, paid leave of 12 weeks is an important policy provision for women so that they can choose to take a leave of that length, but some women and their partners may function well with shorter leaves.

Conclusion

In this article we have considered an important component of family policy, namely parental leave policy. It is clear that proponents of legislated parental leave policies aim at strengthening the family, while at the same time recognizing the realities of women’s and men’s work demands in this last decade of the 20th century. These policies seek to foster the nurturing of children by mothers and fathers, while at the same time ensuring continuing economic support for these families.

There has been considerable tension between social science research- particularly psychological research-on the one hand, and politics and financial considerations on the other, in framing parental leave policy. For example, in the debate over the FMLA, psychologists such as Ed Zigler advocated a 6-month parental leave, whereas business interests, represented especially by the Cham- ber of Commerce, argued that the costs of any legislated parental leave, even an unpaid leave, would be prohibitively expensive to American business. Social scientists believe outcomes such as optimal child development and women’s mental health should be important considerations in framing policy, whereas some business interests argue that the chief consideration should be the bottom line.

As social scientists seek to have their voices and their data heard, it will be important to reframe this debate so that it is no longer viewed as oppositional; rather, it can become a win-win dialogue. For example, research indicates that depression is associated with decreased work productivity (Klerman & Weiss- man, 1992). It may actually be financially advantageous to employers to allow women longer parental leaves so that they return to their normal energy level more quickly and there is reduced risk of depression and the costs associated with it. Moreover, very early nonparental care increases somewhat the likelihood of insecure parent-child attachment, which in turn makes the child more vulnerable

Parental Leave 107

to later stressful life events, increasing the risk of pathology (Gamble & Zigler, 1988). Violent crime costs taxpayers and victims $426 billion per year (Miller, Cohen, & Wiersema, 1996). It is crucial that social scientists’ policy recommendations be seen not only as the right thing to do, but as cost effective as well.

References

Belsky, J., & Rovine, M. J. (1988). Nonmaternal care in the first year of life and the security of infant-parent attachment. Child Development. 59. 157- 167.

Belsky, J., Lang, M., & Rovine, M. (1985). Stability and change across the transition to parenthood: A second study. Journal of Marriage and the Family, 47, 855-865.

Belsky, J., Spanier, G., & Rovine, M. (1983). Stability and change in marriage across the transition to parenthood. Journal of Marriage and the Family, 45, 567-577.

Bond, J. T., Galinsky, E., Lord, M., Staines, G . L., & Brown, K. R. (1991). Beyond the parental leave debate: The impact of laws in four slates. New York: Families and Work Institute.

Bookman, A. (1991). Parenting without poverty: The case for funded parental leave. In J. S. Hyde & M. J. Essex (Us.), Parental leave and child care: Setting a research and policy agenda (pp. 66-90). Philadelphia, PA: Temple University Press.

Bowlby, J. (1973). Attachment and loss. New York: Basic Books. Brazelton, T. B. (1985). Working and caring. New York: Basic Books. Byrd, J. E., Essex, M. J., Hyde, J. S., & Klein, M. H. (1994, October). Women’s health, childbirth,

and employment. Paper presented at the meeting of the North American Primary Care Re- search Group, Toronto, Ontario, Canada.

Chase-Lansdale, P. L. & Owen, M. T. (1987). Maternal employment in a family context: Effects on infant-mother and infant-father attachments. Child Development, 58, 1505-15 12.

Chodorow, N. (1978). The reproduction of mothering. Berkeley: University of California Press. Clark, R., Hyde, J. S., Essex, M. J., & Klein, M. H. (in press). Length of maternity leave and

Cowan, C. P., & Cowan, P. A. (1992). When partners become parents. New York: Basic Books. Cowan, C. P., Cowan, P. A., Heming, G., Garrett, E., Coysh, W. S., Curtis-Boles, H., & Boles,

A. J. (1985). Transitions to parenthood: His, hers, and theirs. Journal of Family Issues, 6 , 45 1-48 I .

Crosby, F. J. (1991). Juggling: The unexpected advantages of balancing career and home for women and their families. New York Free Press.

Ferketich, S. L., & Mercer, R. T. (1989). Men’s health status during pregnancy and early fatherhood. Research in Nursing and Health. 12, 137-148.

Gamble, T. J., & Zigler, E. (1988). Effects of infant day care: Another look at the evidence. In E. F. Zigler & M. Frank (Eds.), The parental leave crisis: Toward a national policy (pp. 77-99). New Haven, CT: Yale University Press.

Gjerdingen, D. K., & Froberg, D. G. (1991). The fourth stage of labor: The health of birth mothers and adoptive mothers at six weeks postpartum. Family Medicine, 23, 29-35.

Haas, L. (1991). Equal parenthood and social policy: Lessons from a study of parental leave in Sweden. In J. S. Hyde & M. J. Essex (Eds.), Parental leave and child care: Setting a research and policy agenda (pp. 375-405). Philadelphia, PA: Temple University Press.

Haas, L., & Hwang, P. (1995). Company culture and men’s usage of family leave benefits in Sweden. Family Relations. 44, 28-36.

Hyde, J. S. (1995). Women and maternity leave: Empirical data and public policy. Psychology of Women Quarterly. 19, 299-313.

Hyde, J. S., Essex, M. J., & Horton, F. (1993). Fathers and parental leave: Attitudes and experi- ences. Journal of Family Issues. 14. 616-641.

Hyde, J. S., Klein, M. H., Essex, M. J., & Clark, R. (1995). Maternity leave and women’s mental health. Psychology of Women Quarterly. 19, 257-285.

Jump, R., & Haas, L. (1987). Dual-career fathers participating in child care. In M. Kimmel (Ed.),

quality of mother-infant interactions. Child Development.

108 Hyde et al.

Changing men: New directions in research on men and masculinity (pp. 98-1 14). Newbury Park, CA: Sage.

Klerman, G. L., & Weissman, M. H. (1992). The course, morbidity, and costs of depression. Archives of General Psychiatq 49, 831-834.

Koester, L. S. (1991). Supporting optimal parenting behaviors during infancy. In J. S. Hyde & M. J. Essex (Eds.), Parental leave and child care: A research and policy agenda (pp. 323-336). Philadelphia, PA: Temple University Press.

Li Min (1991). Parental leave and child care in China. In J. S. Hyde & M. J. Essex (Eds.), Parental leave and child care: Setting a research and policy agenda (pp. 424-434). Philadelphia, PA: Temple University Press.

Malin, M. M. (1994). Fathers and parental leave. Texas Law Review, 72, 1047-1095. Miller, T. R., Cohen, M. A., & Wiersema, B. (1996). Victim costs and conseqrences: A new look.

Pleck, J. H. (1981). The myth of masculinity. Cambridge, MA: MIT Press. Pleck, J. H. (1993). Are “family-supportive” employer policies relevant to men? In J. C. Hood (Ed.),

Men, work, &family (pp. 217-237). Newbury Park, CA: Sage. Radloff, L. S. (1977). The CES-D Scale: A self-report depression scale for research in the general

population. Applied Psychological Measurement, I , 385-401. Repetti, R. L., Matthews, K. A, , & Waldron, I. (1989). Employment and women’s health: Effects of

paid employment on women’s mental and physical health. American Psychologist. 44, 1394- 1401.

Saunders, G. R. (1991). Cultural values, child care, and parenting: The Italian experience in anthro- pological perspective. In J. S . Hyde & M. J. Essex (Eds.), Parental leave and child care: Setting a research and policy agenda (pp. 435-460). Philadelphia, PA: Temple University Press.

Spalter-Roth, R. M., & Hartmann, H. I. (1991). Science and politics and the “dual vision” of feminist policy research: The example of family and medical leave. In J. S. Hyde & M. J. Essex (Eds.), Parental leave and child care: Setting a research and policy agenda (pp. 41- 65). Philadelphia, PA: Temple University Press.

Spielberger, C. (1983). Manual for the State-Trait Anxiery Inventory (Form Y). Palo Alto, CA: Consulting Psychologists Press.

Stryker, S. (1980). Symbolic interactionism: A social structural version. Menlo Park, CA: Benjamin Cummings.

Stryker, S., & Serpe, R. T. (1982). Commitment, identity salience, and role behavior: Theory and research example. In W. lckes & E. S. Knowles (Eds.), Personality, roles, and social behavior (pp. 199-218). New York: Springer-Verlag.

Tulman, L., & Fawcett, J. (1991). Factors influencing recovery from childbirth. In J. S. Hyde & M. J. Essex (Eds.), Parental leave and child care: Setting a research and policy agenda (pp. 294-303). Philadelphia, PA: Temple University Press.

U.S. Department of Labor (1993). 20 facts on women workers. Washington, DC: U.S. Government Printing Office.

Volling, B. L.. & Belsky, J. (1991). Multiple determinants of father involvement during infancy in dual-earner and single-earner families. Journal of Marriage and the Family, 53, 461 -474.

Weinraub, M., & Jaeger, E. (1991). Timing the return to the workplace: Effects on the mother-infant relationship. In J. S. Hyde & M. J. Essex (Eds.), Parental leave and child care: Setting a research and policy agenda (pp. 307-322). Philadelphia, PA. Temple University Press.

Zigler, E. F., & Muenchow, S. (1983). Infant day care and infant care leaves: A policy vacuum. American Psychologist, 38, 91-94.

Washington, DC: National Institute of Justice.

JANET SHIBLEY HYDE is Professor of Psychology and Evjue-Bascom Pro- fessor of Women’s Studies at the University of Wisconsin-Madison. She earned her Ph.D. in psychology in 1972 from the University of California, Berkeley. Her major research area is the psychology of women. The research

Parental Leave 109

project on psychosocial aspects of maternity leave, begun in 1987, grew out of her interest in doing research on psychology of women that would be relevant to public policy. The author of a highly regarded human sexuality text, she has served as a member of JSI’s editorial board and been active in assisting Division 35, The Psychology of Women, in their publication activities.

MARILYN J. ESSEX is Senior Scientist in the Wisconsin Psychiatric Research Institute at the University of Wisconsin-Madison. She earned her Ph.D. in sociology in 1978 from the University of Maryland. Her specialty is the sociology of mental health. Previous research focused on the mental health of elderly women [e.g., Geriatric Medicine Today (1988) vol. 7, pp. 50-651. She has been a member of the Wisconsin Maternity Leave and Health Project since its beginning in 1987.

ROSEANNE CLARK is Assistant Professor of Psychiatry, and Director of the Parent-Infant Clinic at the University of Wisconsin-Madison. She earned her Ph.D. in 1983 from Northwestern University. Her research interests include maternal depression and mother-infant relationships (e.g., Zero to Three, 13, 16-23). She developed the Parent-Child Early Relational Assessment (ERA), which is the major measure of the quality of the mother-infant and father-infant relationship in the Wisconsin Maternity Leave and Health Project. She has been a member of the project since its beginning in 1987.

MARJORIE KLEIN is Professor of Psychiatry at the University of Wisconsin- Madison. She earned her Ph.D. in personality and clinical psychology in 1964 from Harvard University. She has published extensively on issues of women and mental health, and has focused particularly on depression and personality disor- ders (e.g., Personality and Depression: A Current View, Guilford, 1993). She has been a member of the Wisconsin Maternity Leave and Health Project from its beginning in 1987.

JANIS M. BYRD practices family medicine in Columbus, Wisconsin. She earned her M.D. in 1976 from the University of Wisconsin-Madison, and has served on the faculty of the University of Wisconsin. Her special interests are obstetrics, perinatal care, and women’s health. She has served as advisor to the Wisconsin Maternity Leave and Health Project on medical issues.

Parental Leave: Policy and Research

Documents