Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Part I Work, Family, and Child Care Trends Part I reviews evidence concerning rates of employment, family respon- sibilities, and the care of children and adolescents. Chapter 2 reviews current trends in work patterns and family functioning among working families in the United States. It highlights various dimensions of work and family trends, including work schedules, parenting patterns, and family management. Chapter 3 describes the diverse patterns of child care use, as well as details family expenditures on child care, the child care supply, and child care quality.
2 Work and Family Trends T he changing context of working families in America includes de- layed marriage, more childbearing outside marriage, high levels of divorce, lengthening life expectancies, and changing work pat- terns of parents, especially mothers (Casper and Bianchi, 2002). Perhaps the most profound change, which has influenced the reallocation of time in working families over the past few decades, is the dramatic rise in the labor force participation of women--particularly married mothers. For families, this has meant a shift away from maternal time spent in housework and other nonmarket activities to paid market work. More families juggle childrearing with paid market work without benefit of the services of an adult in the home full-time. Historically, economic need has compelled single mothers to enter the paid labor force in larger proportions than married mothers. In recent decades, married mothers--a group tradi- tionally thought to have more freedom to curtail market work to rear children than single mothers--have become increasingly likely to remain in the labor force throughout their childrearing years. The change in employ- ment rates has been most dramatic for mothers, while employment of fa- thers has remained high and stable. Our focus, therefore, is primarily on the implications of maternal employment on the well-being of children and adolescents in working families. In this chapter, we first review evidence on the timing of childbearing and work patterns of mothers surrounding the first birth. Then we describe the labor force patterns of women in the most intense years of childbearing and rearing, ages 25 to 44. We next examine data on mothers' and fathers' 23
24 WORKING FAMILIES AND GROWING KIDS involvement in housework and child care. We conclude with a look at nonstandard work schedules and family functioning. WORK AND CHILDBEARING PATTERNS Most women in the United States (more than 80 percent) become moth- ers by age 40, averaging two children per woman. As shown in Table 2-1, among cohorts of women born in the 1950s who have now completed their childbearing, 83 percent became mothers by age 40, and the cumulative births (per 1,000) of these cohorts were just under two per woman (O'Connell, 2002: Table 3.1). In the 1958 birth cohort, 22 percent had already become mothers by age 20, 50 percent by age 25, and 83 percent by age 40. The percentage of mothers with births by age 20 hover in the range of 20-23 percent for the cohort of women born up through the mid-1970s; the percentage who are mothers by age 25 is a little under 50 percent (47- 49 percent) for women born in the mid-1960s to the mid-1970s, women still in their childbearing years and for whom we do not yet know their completed fertility by age 40. Although fertility is being delayed by many women, more than half are likely to become mothers by their early to mid- 20s, and another third will enter parenthood between age 25 and age 40. One significant change has been the increase in the workforce partici- pation of new mothers. Table 2-2 shows the employment patterns before and after a first birth for the 1961-1995 period. In the 1990s, more women Permission to post Table 2-1 on the web denied. Table is printed in the book but not available online.
WORK AND FAMILY TRENDS 25 Permission to post Table 2-2 on the web denied. Table is printed in the book but not available online. worked continuously for at least six months before their first birth. In addition, in the 1990s, two-thirds of women worked during their preg- nancy compared with 44 percent in the early 1960s. Over one-third of pregnant women at the later time point worked right up to the birth and, by one year after the birth, 61 percent had returned to employment compared with only 17 percent of women who became mothers in the early 1960s. These data suggest a growing demand for infant and toddler care while mothers work. Shifting attention to slightly older children, children ages 3 to 5, regard- less of their mother's employment status, increasingly spend some hours per week in an early education or care setting (e.g., nursery schools, child care centers with an educational curriculum, prekindergartens), as shown in Figure 2-1. Interestingly, although mothers who are in the labor force are more likely to enroll children in preprimary education settings, the increase in the use of nonparental care for at least some hours per week for young children has been as great among mothers not in the labor force as among employed mothers. This suggests that, along with the increase in maternal employment, preferences may be changing more generally toward the desir- ability of at least some nonparental care and organized educational experi- ences in early childhood. TRENDS IN MOTHERS' EMPLOYMENT Mothers' workforce participation has risen rapidly over the past three decades, as suggested by the statistics on mothers who return to work after
26 WORKING FAMILIES AND GROWING KIDS 60 50 40 30 Percent 20 10 Mother in labor force Mother not in labor force 0 1967 1969 1971 1973 1975 1982 1987 1989 1991 1993 1995 1997 FIGURE 2-1 Preliminary enrollment status and attendance in early education or care settings for children 3 to 5 years old, by labor force status of mother, selected years. SOURCE: Casper and Biancchi (2002). the birth of a child. Table 2-3 shows the percentage of mothers employed by age of youngest child. Data are shown for both single and married mothers, ages 25 to 44, for selected years between 1970 and 2001. Given the timing of births, this is the age group of mothers who for the most part have completed schooling but are intensively involved in childrearing. Employment rates for all women of this age group increased over the period, from 43 to 71 percent (first row of the table). The levels of employ- ment showed similar sharp increases for mothers, rising from 38 to 67 percent for all mothers with children under age 18 living in the home.1 Among those with children under age 3, 24 percent were employed in 1970 compared with 57 percent in 2001 (see Cohen and Bianchi, 1999, for similar findings). The trends for single mothers and married mothers are distinctive in the following ways. First, single mothers in most years had higher levels of participation than married mothers. The employment gap by marital status shrank between 1970 and 1995, when, in fact, married mothers had slightly higher employment rates than single mothers. Among married mothers, the most rapid increases occurred prior to 1990, and increases were quite modest in the 1990s. For single mothers, in contrast, employment rates 1During this same period of time, employment of fathers of a comparable age remained high and stable, fluctuating between 86 and 91 percent.
WORK AND FAMILY TRENDS 27 for 2001 71.4 56.8 59.3 67.4 63.8 66.8 74.6 55.7 57.9 65.5 adjust to 2000 71.6 57.5 60.2 67.7 67.0 69.6 75.3 56.1 58.6 65.7 and Child sample 1995 68.9 55.4 57.5 64.6 48.4 52.2 63.2 56.6 58.6 65.0 the into Youngest 1990 68.1 51.6 54.0 62.5 45.4 51.0 64.1 52.6 54.6 62.1 of selection Age of 1985 63.6 45.9 48.2 57.6 39.4 44.9 61.4 46.7 48.7 56.8 by 1980 59.3 40.0 42.5 53.7 44.1 51.1 62.9 39.5 41.4 51.9 probabilities Participation 1975 47.2 28.7 31.8 41.9 32.9 40.9 54.0 28.4 30.7 40.1 differential for 1970 42.5 23.8 27.5 37.9 33.0 41.5 55.1 23.2 26.6 36.4 correct Workforce 1970-2000 to (2002). 25-44) 25-44) 25-44) 0-3 0-5 0-17 25-44) 0-3 0-5 0-17 0-3 0-5 0-17 weighted Bianchi Mothers' (ages are ages ages ages ages ages ages ages ages ages employed), (ages (ages (ages and 2-3 mothers numbers Casper children children children women mothers mothers children children children children children children :All with with with with with with with with with TABLE (percentage Total Total Single Married NOTE nonresponse. SOURCE:
28 WORKING FAMILIES AND GROWING KIDS TABLE 2-4 Marital Status of Unmarried Mothers Living with Children Under Age 18 for 1978, 1988, and 1998 Percentage Point 1978 1988 1998 Change Total (number) 6,194 8,160 9,845 Total (percent) 100 100 100 Never married 22.4 33.2 42.1 19.7 Separated/spouse absent 27.6 21.9 18.7 -8.9 Divorced 39.0 38.3 34.9 -4.1 Widowed 11.0 6.7 4.4 -6.6 NOTE: All numbers are weighted. SOURCE: Casper and Bianchi (2002).: jumped rather dramatically in the second half of the 1990s, a period of strong economic growth and restructuring of welfare support for single mothers and their children (see Chapter 7 for overview of welfare reform). By 2001, single mothers again had employment levels that were substan- tially higher than those of married mothers. Single-mother families increased most rapidly in the latter 1960s and 1970s; the increase then slowed in the 1980s. During the latter half of the 1990s, the percentage of families with a single mother remained steady. The composition of these families shifted over time, with never-married mothers accounting for a growing share of such families. For example, in 1978, 22 percent of single mothers with children under age 18 had never married, whereas this percentage was almost twice as large by the end of the 1990s (see Table 2-4). This compositional shift tended to dampen labor force participation rates of single mothers at the same time that higher average education levels of all women, including single mothers, increased the likelihood of employment. Overall employment rates of single mothers, which did not change greatly during the 1980s or even the early 1990s, did rise considerably after the mid-1990s (Casper and Bianchi, 2002:Table 4.4). Single-father families also increased rapidly after 1970, although they remain a relatively small proportion of all single-parent families (Casper and Bianchi, 2002:Table 1.1; Garasky and Meyer, 1996). The trend data on marital status of single mothers does not adjust for the fact that the proportion of unmarried mothers who live with a cohabit- ing partner has been rising. Table 2-5 shows that the proportion of single mothers with children who were living with a cohabiting partner increased from 5 to 13 percent between the late 1970s and late 1990s. The propor- tion of these mothers who lived with a parent also increased slightly. Due to these trends, the proportion of single mothers who had no other adult present in the household declined.
WORK AND FAMILY TRENDS 29 TABLE 2-5 Living Arrangements of Unmarried Mothers Living with Children Under Age 18 for 1978, 1988, and 1998 Percentage Point 1978 1988 1998 Change Total (number) 6,194 8,160 9,845 Total (percent) 100 100 100 Never married 59.8 56.4 54.0 -5.9 Separated/spouse absent 4.7 9.6 12.7 8.0 Divorced 14.3 15.2 16.7 2.7 Widowed 21.2 18.8 16.6 -4.6 SOURCE: Casper and Bianchi (2002).: With respect to the increase in married women's employment, some argue that the increase has been propelled by the stagnation in men's wages (Jacobs and Gerson, 2001), but market opportunities have also expanded for women. Between 1978 and 1998, increases in annual hours of employ- ment were greatest for highly educated women, but the positive correlation of a husband's earnings with market participation of wives diminished (Cohen and Bianchi, 1999). Hence, although in the aggregate the increase in wives' employment would seem to compensate for the decline in men's earning power, this need-based interpretation is at odds with the fact that labor force gains have been largest for wives married to highly educated, high-earning husbands (Juhn and Murphy, 1997). The implication of these demographic and labor force changes is that a larger fraction of children live in families in which all available parents are in the labor force--either they live with a single parent who is employed or they live with two parents, both of whom work at least some hours for pay each week. In 1997, 68 percent of children had all parents with whom they lived working for pay, compared with 59 percent of children in the mid- 1980s (Bianchi, 2000a). Work Hours Table 2-6 shows the average hours of employment per week for work- ing mothers (ages 25 to 44) for the 1970-2001 time period. The most striking feature of these data is that weekly work hours rose only slightly, certainly far less sharply than employment rates, over the period. Em- ployed mothers with children averaged 33 hours of employment per week in 1970, and this rose by 3 hours per week (to 36) hours in 2001. Hence, balancing work with child care is not a new issue for working women. What is new is that the pool of women who work outside the home, who
30 WORKING FAMILIES AND GROWING KIDS for 2001 37.7 34.5 34.9 36.0 37.4 38.0 34.0 34.3 35.3 adjust to 2000 37.5 33.3 34.1 35.6 37.3 37.5 32.6 33.4 34.9 and Youngest of sample 1995 36.7 33.4 33.6 35.0 36.4 36.0 32.9 33.2 34.3 the Age by into 1990 37.1 33.1 33.5 35.1 36.3 37.0 32.7 33.0 34.2 selection of 1985 36.2 32.6 33.0 34.5 37.0 36.7 32.2 32.5 33.6 Participation 1980 35.4 32.0 32.3 34.0 37.3 37.4 31.3 31.4 33.0 probabilities 1975 34.4 30.9 31.4 32.9 34.2 34.9 30.6 30.9 32.3 Workforce differential 1970-2001 for 1970 34.1 30.1 31.1 32.8 34.8 34.8 29.7 30.7 32.3 correct Mothers' week), to per (2002). 25-44) weighted hours 0-3 0-5 0-17 25-44) 0-3 0-5 0-3 0-5 0-17 Bianchi Employed 25-44) 25-44) are (ages and ages ages ages ages ages ages ages ages (ages (ages (ages 2-6 numbers Casper (average mothers children children children All women mothers mothers children children children children children : with with with with with with with with TABLE Child Total Total Single Married NOTE nonresponse. SOURCE:
WORK AND FAMILY TRENDS 31 face issues of balancing work and family needs when their children are young, has expanded. More generally, scholars disagree on the extent to which individuals' work hours have actually increased in recent years, if they have at all. Juliet Schor (1992), in The Overworked American, argues that work hours have increased, whereas John Robinson and Geoffrey Godbey (1999), using time diary evidence, argue that the trend in leisure activities is not consistent with the notion that work hours are expanding. Leisure time has not decreased. Careful examination of employment data from the Current Population Survey, by Rones et al. (1997), shows that, on average, work hours have not changed much, but the unchanging average disguises the fact that there is increasing heterogeneity among workers, with some un- able to work as many hours as they would like and others working very long workweeks. In addition, with the rise in mothers' employment, house- holds as a unit are allocating more of their "available adult time" to paid work and there is evidence that the share of dual-earner couples working very long workweeks (exceeding 100 hours for the couple, husband's and wife's hours combined) has increased over the past three decades (Jacobs and Gerson, 2001). Wives as Primary Earners Beyond the increase in the overall time spent in paid work in dual- earner and single-parent households, nontraditional breadwinning patterns in married couple households have increased. Households in which wives earn more than their husbands and are the primary earners have grown from 15.9 percent of married-couple families in 1981 to 22.5 percent in 2000. Not surprisingly, such households are especially common among dual-earner couples with low-wage husbands (Winkler, 1998). Despite the growth in the number of wives who earn more than their husbands, it remains more common for wives rather than husbands to adjust their labor force attachment downward when children are born. While a majority of married mothers with young children (under age 6) were employed in 1998, only a little over one-third of married mothers of preschoolers were full-time, year-round workers (Cohen and Bianchi, 1999). This suggests that, at least when young children are at home, mar- ried mothers continue to reduce their hours of employment or even drop out of the labor force altogether for a short time. In-depth interviews with over 100 middle-class dual-earner couples confirmed that, relative to their husbands, wives disproportionately reduce and restructure their commit- ment to paid work over the life course to protect the family from work encroachments (Becker and Moen, 1999). In addition, young mothers in America incur a wage penalty for motherhood of approximately 7 percent
32 WORKING FAMILIES AND GROWING KIDS per child, which is unexplained by measured productivity factors, such as years of past job experience and seniority, and which remains after correc- tions for unobserved heterogeneity (Budig and England, 2001). HOUSEWORK AND TIME WITH CHILDREN Virtually the only data that exist with which to measure trends in nonmarket activities, such as housework and child care, are time diaries that capture all uses of time over a reporting period, typically a 24-hour period. Time diaries have been conducted at roughly 10-year intervals since 1965, using consistent coding categories over time. Labor market surveys, such as the Current Population Survey (CPS), only capture market activities because they are included in the system of national accounts. When respondents are asked direct questions in surveys about how much time they spend on nonmarket activities, such as housework, reports tend to be inflated (Bianchi et al., 2000). Often survey reports of time use sum to more than 24 hours in a day or more than the possible 168 hours in the week (Bianchi et al., in press). Time diaries for which respondents are walked through the previous day's activities have proven to be a more reliable way to obtain estimates of time use, especially in nonmarket do- mains in which there is no other source of comparable estimates (Juster 1985; Robinson and Godbey 1999). What effect has the change in market work had on the allocation of time to nonmarket activities? With respect to housework, Table 2-7 shows that mothers have dramatically curtailed the time they spend in housework tasks. Mothers' hours of housework (exclusive of child care) fell from an average of 32 hours per week in the mid-1960s to about 19 hours per week in 2000. Fathers' participation in housework chores increased from 4 to around 10 hours per week. This increase occurred in the 1965-1985 pe- riod, with little change afterward. In 2000, mothers of children under age 18 averaged about twice as much time as fathers in household chores. Time diary evidence on fathers' participation in child care suggests greater change than in housework, at least among married fathers. Such data allow for three measures of father's participation in child care: the time they spend primarily engaged in a direct child care activity, the time they spend either directly focusing on child care or doing a child care activity in conjunction with something else, and finally, the overall time they spend with their children whether engaged in child care or not (the most inclusive category). Figure 2-2 shows estimates of married fathers' hours per day with their children. For comparison purposes, comparable figures for mothers are shown in Figure 2-3 and Figure 2-4 shows the ratio of married fathers' to married mothers' time with children. On average, mothers' time with children has not decreased and fathers' time, at least among married fathers, has increased appreciably. Whereas
WORK AND FAMILY TRENDS 33 2000 2.0 3.4 2.5 4.9 2.9 9.4 0.7 0.3 0.7 1.6 1.1 Time Father's 1995 1.7 3.9 3.5 6.8 4.0 3.8 0.6 0.3 0.3 0.4 1.2 to Time 1985 2.0 4.4 4.2 4.6 3.7 8.7 0.5 0.3 0.2 0.7 0.8 3.1 0.3 0.3 0.2 1.0 0.6 Mother's 1975 12.3 10.0 11.4 14.8 17.6 of 7.3 1.0 0.5 0.3 2.4 2.4 Ratio 1965 18.5 13.8 14.0 23.0 34.5 9.5 4.4 2.2 0.3 1.6 0.4 5.1 2.0 1.4 0.4 1.3 2000 472 week) 3.9 1.5 0.1 1.6 0.6 6.9 2.3 1.9 1.1 1.6 10.8 1995 181 per 4.0 1.8 0.4 1.5 0.3 6.3 1.3 2.3 0.9 1.8 10.3 1985 699 hours 7.5 1.8 0.9 0.2 0.5 0.2 5.8 1.0 3.2 0.4 1.1 1975 480 (average Fathers 4.4 1.6 0.7 0.3 0.4 0.2 2.8 0.6 1.3 0.2 0.7 All 1965 337 Parent 5.4 1.3 4.8 3.3 3.8 0.7 1.0 0.7 1.5 18.6 14.8 of 2000 728 5.3 0.8 6.5 2.4 3.8 0.7 0.7 0.5 2.0 18.8 15.0 1995 312 Gender by 7.5 1.9 5.5 2.7 2.9 0.4 0.5 0.6 1.4 20.5 17.6 1985 913 8.7 2.5 7.1 3.4 2.0 0.4 0.6 0.4 0.7 23.7 21.7 1975 607 Housework Mothers 4.8 8.1 6.2 3.0 0.3 0.4 0.5 1.8 32.1 29.2 10.1 in All 1965 405 care Trends ironing animal financial 2-7 meals and chores and cleanup other size housework housework housework Cooking Meal Housecleaning Laundry Outdoor Repairs Garden Bills, TABLE Total Core Other Sample xxx
34 WORKING FAMILIES AND GROWING KIDS 4 3.8 * 3.5 y 3 2.8 Da 2.5 per 2 * 1.5 1.3 Hours 1* 1 0.7 0.4 0.5 0 Child Care Child Care Time Spent with Children (Primary Activity) (Primary or Secondary Activity) (Any Activity) 1965 1998 FIGURE 2-2 Change in married fathers' hours of child care and time with chil- dren. NOTES: Estimates based on one-day, "yesterday" time diaries collected from 326 married fathers in 1965-1966, 194 married fathers in 1998-1999, all with children under age 18 at the time of the interview. Child care includes: child and baby care, helping/teaching children, talking/reading to children, indoor/outdoor play with children, medical/travel/other child related care. "*" indicates that test of 1965- 1998 difference in means is statistically significant, p < 0.05. SOURCE: Bianchi (2000a). in 1965 fathers reported only one quarter of the time mothers reported in direct child care, they reported 55 percent as much time as mothers in the late 1990s. And whereas fathers were only with their children half as much time as mothers were in the 1960s, in the late 1990s they were spending 65 percent as much time with children as were mothers. Note that some of this time is double counted, in that both mother and father can be present. Fathers remain much more likely to have their spouse present when with their children, whereas mothers spend more solo time with children (Sayer et al., 2002). Other research shows parallel findings of fathers' increased time with children, at least among married fathers, and no substantial decline in mothers' time with children, on average (Sandberg and Hofferth, 2001). While most research shows that employed mothers spend less time with their children relative to nonemployed mothers, the difference is not dra- matic, except perhaps for very young children. The differences between employed and nonemployed mothers in time spent with children may be minimized because working mothers curtail work hours when children are young, try to synchronize work hours with children's school schedules when children are older, "tag-team" work hours with a spouse so as to maximize parental availability to children, and curtail time spent in other
WORK AND FAMILY TRENDS 35 6 5.5 5.3 5 y Da 4 per 2.8* 3 2.2 2 1.7* Hours 1.5 1 0 Child Care Child Care Time Spent with Children (Primary Activity) (Primary or Secondary Activity) (Any Activity) 1965 1998 FIGURE 2-3 Change in mothers' hours of child care and time with children. NOTES: Estimates based on one-day, "yesterday" time diaries collected from 417 mothers in 1965-1966, 273 mothers in 1998-1999, all with children under age 18 at the time of the interview. Child care includes: child and baby care, helping/ teaching children, talking/reading to children, indoor/outdoor play with children, medical/travel/other child related care. "*" indicates that test of 1965-1998 differ- ence in means is statistically significant, p < 0.05. SOURCE: Bianchi (2000a). 0.7 0.65* 0.6 0.55 * 0.51 0.5 0.45* Mother/ 0.4 0.3 0.3 0.24 Father 0.2 0.1 0 Child Care Child Care Time Spent with Children (Primary Activity) (Primary or Secondary Activity) (Any Activity) 1965 1998 FIGURE 2-4 Ratio of married fathers' to married mothers' hours with children, 1965 and 1998. NOTES: Estimates based on one-day, "yesterday" time diaries collected from 358 married mothers in 1965-1966, 326 married fathers in 1965-1966, 194 married mothers in 1998-1999, 141 married fathers in 1998-1999. Ratios are averages across married men and women with children under age 18, not couples married to each other. "*" indicates that test of 1965-1998 difference in means is statistically significant, p < 0.05. SOURCE: Bianchi (2000a).
36 WORKING FAMILIES AND GROWING KIDS activities, such as housework other than child care, volunteer work, per- sonal care, and free-time pursuits (Bianchi, 2000b). With respect to fathers, one large gap in knowledge is the involvement of nonresidential fathers in their children's lives. For fathers who live with children, there is evidence that stepfathers spend less time with children than biological fathers. A father who cohabits with a partner and her children spends more time with those children than stepfathers, on average, but still less than biological fathers (Hofferth et al., 2002). A stepfather's involvement is greater when he also has a biological child in a remarriage (Hofferth, 2002). Whether this is because biological children enhance the likelihood of a father's involvement or whether more "child-oriented" fa- thers have children in a second marriage is not clear. The end result, however, is that fathers who are married to a child's mother and biologi- cally related to a child spend more time rearing those children than fathers whose linkages are other than biological. Consequently, some children benefit from much higher involvement, financial and otherwise, from their fathers than other children. NONSTANDARD WORK SCHEDULES AND FAMILY FUNCTIONING While the number of hours that people work has not changed much over the past few decades, distribution of these hours has become more diverse (Bureau of Labor Statistics, 2002). The economy in the United States is increasingly operating on a "24/7" basis, with more people work- ing late hours and on weekends. The demand for such around-the-clock employment has been attributed to growth in the service economy, interre- lated with demographic and technological changes, including, but not lim- ited to, the growth in women's employment and in dual-earner couples (Presser, 1999). The result appears to be that more workers are employed at nonstandard times--and, indeed, that nonstandard hours are becoming less so. Although good trend data are not available, it is estimated from the May 1997 Current Population Survey (the most recent year of available data) that one-fifth of all employed people in the United States work most of their hours in the evenings, during nights, on weekends, on a rotating schedule, or have highly variable hours (see Box 2-1 for shift definitions). Moreover, one-third of all those employed work weekends (Presser, 1999). Presser (1999) has estimated that of the 134 million people employed in the United States in 1997, only 29.1 percent worked 35-40 hours a week, Monday through Friday, and during daytime hours. Without the full-time restriction of 35 to 40 hours a week, only 54.4 percent work Monday through Friday and most of their hours during the daytime. One in five people works mostly during the evenings, nights, or on rotating schedules,
WORK AND FAMILY TRENDS 37 BOX 2-1 Shift Definitions Fixed day shift: at least half the hours worked during the reference week fall between 8:00 a.m. and 4:00 p.m. Fixed evening shift: at least half the hours worked during the reference week fall between 4:00 p.m. and midnight. Fixed night shift: at least half the hours worked during the reference week fall between midnight and 8:00 a.m. Rotating shift: work hours change periodically (e.g., from daytime to evening to nighttime). Hours vary: an irregular schedule that cannot be classified in above categories. SOURCE : Adapted from Presser (1999). and one in three works Saturdays or Sundays or both. Furthermore, it is more common for married couples with at least one spouse working non- standard hours to have children (see Table 2-8). There is a large body of literature on the health consequences for individuals who work nights and rotate shifts because of its effect on circa- TABLE 2-8 Percentage of Married Couples with at Least One Spouse Who Works Nonstandard Hours At least one earner 23.8 At least one earner and Child < age 14 25.8 Child < 5 30.6 Two earners only 27.8 Two earners and Child < age 14 31.1 Child < 5 34.7 NOTES: Based on data from the May 1997 Current Population Survey. Nonstandard work hours are work hours most days of reference week being between 4:00 p.m. and 8:00 a.m., rotating hours, and those too variable to classify. Couples with at least one employed spouse on the job during the week, including all rotators. Both spouses age 18 and over. Couples with both spouses on the job during the reference week, including all rotators, both in nonagricultural occupations, and both age 18 and over. SOURCE: Presser (in press).
38 WORKING FAMILIES AND GROWING KIDS A mother living in an inner-city neighborhood in Chicago cares for her children during the day. However, she leaves them on their own during the evening hours as she works the graveyard shift: Before she left for work at 11 P.M., she made sure the children had completed their homework and gone to bed. The children, ages five and seven, had her work number posted on a telephone between their beds. Because there had been three fires on their block that winter, she and her children would stage a fire drill each night after supper (Edin and Lein, 1997:134-135). dian rhythms, which in turn affect body temperature, hormone levels, and sleep. Behaviorally, there is evidence of greater gastrointestinal disorders, higher rates of cardiovascular disease, preterm births, and low birthweight (U.S. Congress, Office of Technological Assessment, 1991; Boggild and Knutsson, 1999; Wedderburn, 2000; Schernhammer et al., 2001). While employment at nonstandard hours is widespread among all those employed, it is disproportionately found among those with low incomes. For example, the top five occupations of nonday workers are cashiers, truck drivers, waiters and waitresses, cooks, and janitors/cleaners (Presser, in press); these jobs are often low paying. Until recently there has been limited research on how low-income working families cope with work responsibilities while caring for their children (Bogen and Joshi, 2002). Data from the 1997 Current Population Survey on reasons for working such hours indicate that most employees do so for job-related reasons (e.g., because it is a requirement of their job or they cannot find another job) rather than for personal reasons (e.g., child care or other family reasons), even when looking specifically at parents with young children (Presser, in press). Thus, while some parents may prefer such schedules, employment at nonstandard times is driven by demand and those with limited job possi- bilities are generally recruited. This includes mothers who move from welfare to work, who often experience a lack of correspondence between their required hours and days of employment and the availability of formal child care arrangements (Presser and Cox, 1997). The limited data available indicate that working nonstandard work schedules has some some negative consequences for the quality and stability of marriages and both positive and negative consequences for other aspects of family functioning.
WORK AND FAMILY TRENDS 39 Marital Quality Staines and Pleck (1983) have shown that shift work affects the quality of family life and leads to greater marital conflict. Updated research based on data from the National Survey of Families and Households on families shows that evening and rotating shifts allow for less quality time together for spouses, compared with working daytime shifts (Presser, in press). Marital Stability Research conducted by White and Keith (1990) suggests that nonstand- ard work schedules have a negative effect on marital stability. Further- more, Presser (2000) looked at the National Survey of Families and House- holds to consider whether nonstandard work schedules increase the likelihood of martial instability. This research looked at the effects of different shifts, as well as the gender of the spouse and the spouse's respec- tive gender ideologies, conditioned by the duration of the marriage and the presence of children. This research showed that night and rotating shifts significantly increased the odds of marital instability for couples with chil- dren. Specifically, separation or divorce over a five-year period is about six times higher among couples in which fathers work nights (and have been married less than five years). In similar couples in which mothers work nights, separation or divorce is three times as high. One could speculate that individuals who are already in troubled marriages choose nonstandard work, but, in fact, the data do not suggest this direction of causality (Presser, 2000). Rather, some types of nonstandard shifts may place stress on rela- tionships that then heightens the likelihood of divorce. Two mothers describe the time crunch that work entails: [My job] was too far . . . I would have to get up at four o'clock in the morning in order to be at work at seven. [I] leave work at three- thirty and still wouldn't make it home until eight o'clock. And it was too far when I wasn't making anything. . . . I didn't have no time for my kids, no time for myself (Jarrett, 1994). You see I work ten hours a day. . . . I take two hours to travel to work, two hours to travel back. . . . So that's like, fourteen hours a day I'm out of the house (Cook and Fine, 1995:127).
40 WORKING FAMILIES AND GROWING KIDS Family Functioning Nonstandard work has an effect on the ways in which families function and interact. Parents working nonstandard hours often suffer from irregu- lar and limited sleep, contributing to increased parental stress, greater fam- ily conflict, and poorer relationships with their children (Simon, 1990; Rahman and Pal, 1994). One indicator of family involvement is having dinner together: it often allows for consistent family interaction and an opportunity for family organizing. Presser (2001) found that parents who work evenings and rotating shifts are significantly less likely to have dinner with their children than parents who work day shifts. Working nights, however, has not shown the same effects on full family involvement in dinner. In an examination of employment schedules and the division of house- hold labor by gender among married dual-earner households, Presser (1994) found that when husbands work a nondaytime or rotating shift and their wives work a day shift, the men are significantly more likely to do tradition- ally female household tasks than couples in which both spouses work day shifts. These tasks include preparing meals, washing dishes and cleaning up after meals, cleaning house, and washing, ironing, and mending clothes. Based on this research, it appears that a husband's being home alone, particularly during the day, is an important factor in increasing his share of traditionally female household tasks. SUMMARY One fundamental and very important finding drawn from the literature is that more parents, particularly mothers, are working. American family life has changed dramatically as the result of this increased labor force participation of mothers. Mothers are now more often employed right up to the birth of their child and are increasingly likely to return to paid work in the first year of their child's life. In addition, the proportion of pre- school-age children who spend at least some hours per week in nonparental care settings has risen rapidly for employed and nonemployed mothers alike. An increasing proportion of children live in households in which all available adults are in the labor force, either in single-parent families or in two-parent, dual-earner families. The result of this trend is a heightened demand for nonmaternal care for children, particularly very young chil- dren. We might have expected less maternal time with children, on average, as mothers' employment has increased. However, there have been offset- ting trends. In two-parent families, fathers appear to have increased their time with children. Mothers often do not work full time, although a
WORK AND FAMILY TRENDS 41 growing number of mothers are the main economic provider in their house- hold, either because they are a single parent or because they are the primary earner in a two-parent family. Mothers appear to have decreased time spent in housework, leisure, and personal care in order to maintain a given level of involvement in their children's lives. These types of choices may have consequences for the well-being of children and adolescents in these families. Another important finding is that a sizable number of families with young children work nonstandard hours. Working nonstandard hours complicates child care arrangements and may place additional strains on the family. Marital quality is lower and marital disruption is higher when couples work evening or rotating shifts. Evening shifts also reduce time together as a family; families are less likely to eat dinner together on a regular basis when one of the parents works evenings or a rotating shift. However, nonstandard work shifts may be part of the explanation for greater father involvement in family life, as fathers are more involved in housework and child care when they work rotating or nonday shifts.
