Effects of Intermittent Labor Force Attachment on Women's Earnings

Women who interrupt their careers and leave the labor market for family responsibilities often return to find that their wages lag behind those of women at comparable stages in their careers who did not leave the labor force. Many reasons account for this lag. First, women who leave the labor force and later re-enter do not build up seniority, with associated higher wages. Second, women who return to the labor force are less likely to receive on-the-job training to increase their productivity and thereby raise their pay. Third, when women are not in the work force, their job skills may decline. Finally, employers may view gaps in work history as an indication that women may leave again. Some employers would, therefore, hire them for less important, lower paying jobs to limit the impact of a future leave.

This study calculates the cost of taking a break from work in terms of the wage difference between women who work continuously and women who take one or more leaves. Because those who do not leave the work force tend to be younger and better educated than those who do, a straight-forward comparison cannot be made--foregone earnings would be overestimated.

Previous studies of gaps in labor force participation have found that these gaps affect earnings. One hypothesis is that women returning to the work force who find their wages lower than they had expected are quite likely to leave again; and over time only the relatively high-earning women who have had a break in labor force participation will be left in the work force. Another hypothesis is that earnings will rebound soon after women re-enter the work force.

This study tests for the rebound effect by restricting the sample of women with labor force breaks to those women who display continuous labor force attachment for an extended period after a break. Findings indicate that when women re-enter the labor market, their earnings are much lower than those of a comparable group of women who did not leave the labor market. Over time, that difference diminishes (due to the rebound effect) but never disappears, even after as long as 20 years.

Data used were from the 1984 panel of the Survey of Income and Program Participation. Each individual in the data set was interviewed eight times at 4-month intervals. Participants were asked in each interview about their labor force participation in the previous 4 months. Thus, data for 32 consecutive months for each individual (June 1983 to April 1986) were collected. Only women ages 30 to 64 at the start of the sample period were included.

Only women who worked relatively continuously during the 32 months of the sample were included in the "no gap" group; women must have reported earnings in the 1st, 6th, 12th, 18th, 24th, and 32d months of the sample. Thus, women were included only if their gaps were shorter than 6 months. In this way, the majority of women who have seasonally intermittent work schedules, such as teachers, participated in the "no gap" group.

To be included among the sample of women with labor force breaks ("gaps"), a woman must have taken at least one break from work, lasting 6 months or longer, between the year she received her last educational degree and the beginning of the survey.(1)

Total work experience was the same for the two groups (see table), which reflects the higher age and lower educational attainment of the women who left the work force. These women were much more likely to be working part time and were more heavily represented in less skilled and service occupations--both blue- and pink-collar positions.

Characteristics of women who remained in the labor force (no gaps) and women who left the labor force (1 or more gaps)

                               Women who remained    Women who left
                               in the work force     the work force
Item                               (no gaps)        (1 or more gaps)

Number of people                      696                1,730
Age                                    39                   45
Years of education                     14                   12
Total years worked                     17                   17

                                                 Percent
Education
  No high school diploma                6                  21
  High school diploma                  33                  47
  Some college                         27                  19
  College degree                       15                   7
  Graduate work                        19                   6

Occupation
  Professional/executive               38                  21
  Service occupations                  10                  17
  Craft occupations                     2                   3
  Pink collar/blue collar              50                  59

Marital status
  Married                              58                  70
  Widowed                               3                   5
  Divorced                             21                  21
  Never married                        18                   4

Number of children ever born
  None                                 39                   9
  1                                    18                  14
  2                                    24                  33
  3 or more                            19                  44

Source: Jacobsen, J.P. and Levin, L.M., 1995, Effects of intermittent labor force attachment on women's earnings, Monthly Labor Review 118(9):14-19.

Women who left the work force were more likely than their counterparts who remained in the work force to be married and to have children. The reason mentioned most often for taking leave from the work force was family reasons (85 percent gave this response). Other reasons included poor health and inability to find a job; leaving work to attend school was not counted as a gap.

Regression analysis was used to show the direct effects on wages of gaps occurring at different times in the past and to calculate wage ratios that control for differences in age, education, work experience, and other factors between those who had left the work force and those who remained at work. The regression equation was estimated at three different times during the sample period: the 1st, 18th, and 32d months. The dependent variable was the natural logarithm of the hourly wage. Independent variables controlled for individual characteristics (age, geographic location, occupation class, and human capital) and also a set of dummy variables for number of years since a worker ended her last absence from the labor force (measured from the beginning of the survey). Thus, the wages of the same group of women could be measured and examined to determine what changes occurred over the duration of the survey.

A lasting negative effect and a gradual rebound effect resulted from the period out of the labor force. For any particular length of time out of the labor force, 2-1/2 years of continuous labor force attachment will, on average, diminish the difference in wages between those who have left the work force and those who remained. For example, in the initial period, women whose gaps ended less than 1 year ago had wages that were 33 percent lower than those of women who did not leave the labor force. By the third year (when they would have returned to the work force more than 3 years ago), these women's wages were only 20 percent lower than those of women who remained in the labor force.

Although there is strong evidence for a partial rebound effect, the wages of women who have taken a leave from the labor market never catch up to the wages of women who never left. Even women whose labor force gap occurred more than 20 years ago still earn between 5 and 7 percent less than women who never left the labor force and have comparable levels of experience.

The effect of a gap on a woman's lifetime earnings is significantly larger than just her foregone wages during the time away from work. This finding has significant implications for the way in which compensation between husband and wife is calculated in divorce proceedings.

To illustrate the cost of taking an employment gap for a particular case, assume a woman with the following characteristics: graduates college at age 21, immediately begins full-time work (40 hours a week, 50 weeks a year) in a pink-collar occupation in a city outside the South. She leaves work at age 25 for 7 years and re-enters full-time work in 1984 at age 32. The difference between her earnings for the 20 years after she re-enters and what they would have been had she remained constantly employed is $52,000. Part of this is caused by her fewer years of experience: part is due to her decision to leave the labor force. This amount is equal to 15 percent of her prospective earnings had she worked constantly--or about 3 years of wages. Thus, the cost of taking a 7-year gap is 10 years of earnings.

(1) Includes women who worked before taking a break, and women who had an initial gap between the year of their last degree and the year in which they started working.

Source: Jacobsen, J.P. and Levin, L.M., 1995, Effects of intermittent labor force attachment on women's earnings, Monthly Labor Review 118(9):14-19.

COPYRIGHT 1996 Superintendent Of Documents
COPYRIGHT 2004 Gale Group