Printer Friendly

Education and the work histories of young adults.

Data from the National Longitudinal Survey of Youth show significant differences by sex and race in youth labor market experience; however, many of these differences become smaller or disappear completely with increases in educational attainment

Data from the National Longitudinal Survey of Youth allow for fairly precise determination of measures of labor market activity not available from any other data source.[1] These data indicate that work experience between the ages of 18 and 27 vanes substantially by sex, race, and educational level, and reveal patterns of work behavior that are somewhat surprising. For example, by age 27, individuals with 1 to 3 years of college education have, on average, worked more weeks than have high school graduates. Also, college graduates average more total weeks worked than do high school dropouts at all ages, even between the ages of 18 and 22, when many college graduates are attending school full time. This finding reflects the fact that young female high school dropouts acquire very httle work experience.

In addition, the data indicate that education accounts for much of the sex and race differentials in labor market activity. Gender and race differentials are particularly conspicuous among high school dropouts. However, with increased educational attainment, most of these differentials become smaller, and are particularly small or nonexistent among college graduates. For example, college educated women hold more jobs and work more weeks from age 18 to 27 than do college educated men, whereas female high school dropouts hold approximately 1-1/2 fewer jobs and work only about half as many weeks as do their male counterparts. Also, by age 27 there is very little difference among college educated whites, blacks, and Hispanics in the number of weeks worked. In general, the data indicate that greater educational attainment allows young workers to spend more time employed, and thus to acquire work experience more readily.

This article analyzes the work histories of young workers, focusing on differences in work experience by educational level. The results permit comparison by educational level of work patterns by years of age for persons aged 18 to 27 over the 197 8-90 period.


Individuals demonstrate a great deal of job mobility during their early years in the labor market. Brief and transitory periods of employment are common among young workers. Previous research indicates that the first 10 years of a young worker' s career account for about two-thirds of all lifetime job changes, and for nearly two-thirds of lifetime wage growth. During the first 10 years in the labor market, an individual works for an average of eight employers. Research also suggests that only 1 of 20 male workers remains at his first job over a 10-year period.[2] However, as workers age, employment patterns tend to stabilize, and the probability of leaving or losing a job eventually declines.

Young workers change jobs and employment status for a variety of reasons. Information about how to find a job and the nature of employment often are difficult to acquire, particularly for young workers. Some individuals accept a job offer and remain in that job so long as the wage paid exceeds alternative wage offers.[3]

Information about the quality of the match between a worker and a firm reveals itself over time. Workers who are well matched remain on the job, and those who are poorly matched are most likely to leave. In addition, many young workers acquire firm-specific on-the-job training, which increases their productivity and their wage in a particular job. This training may reduce the probability of a job separation, because a trained employee might be less likely to quit a job for which he or she has specific training that might not be useful in other employment. Similarly, an employer might be less likely to fire an employee whom he or she has trained to meet the special needs of the firm.

Individuals may also move into and out of the labor market because of decisions relating to schooling, marital status, childbearing, or other factors. In particular, decisions relating to education affect labor market experience for young workers. Individuals who drop out of high school or stop schooling after high school graduation may be expected to acquire more work experience at a younger age than would those who go on to college. However, college graduates may acquire more experience per year once they complete college. More-educated workers also may experience fewer spells of unemployment and fewer weeks unemployed than would less-educated workers.

Longitudinal data provide an avenue through which the dynamics of labor force transitions can be better understood. Studies of labor market behavior using data from the Current Population Survey (cPs) allow for an examination of an individual's employment situation at a particular point in time. Yet, the number and duration of employment and unemployment spells over an extended period cannot be determined using cross-sectional data from the CPS, because CPS respondents are interviewed only over a short timespan.

In this analysis, four dimensions of the labor force behavior of young workers are analyzed: number of jobs held, number of weeks worked, number of spells of unemployment, and number of weeks unemployed.4 Very little is known about these four aspects of labor market activity. For instance, an individual's work experience is a commonly desired variable in labor market analyses. However, because most data series contain little information on labor market history, work experience often is roughly approximated by using an individual's age and education. Similarly, very little is known about the other dimensions of labor market behavior over extended periods.

In this study, the employment and unemployment histories of young workers are examined using data from the Youth cohort of the National Longitudinal Surveys. These data describe a sample of young men and women who were between the ages of 14 and 22 in 1979, and who have been interviewed annually since that year. In 1990, the sample, which includes an overrepresentation of blacks, Hispanics, and economically disadvantaged whites, had 10,436 respondents. In all computations, weights are used to adjust for different sampling rates and nonresponse rates, so that the data are nationally representative of the age group.5

A key feature of this survey is that it garners information in an event history format, in which dates are collected for the beginning and ending of important life events. In the case of employment, the starting date for every job is recorded, and if a person stops working for an employer, the ending date is recorded as well.6 For multiple jobholders, information is gathered for each job, with starting and ending dates. Periods of nonwork within a period of employment--such as layoffs, or absences from work due to illness, pregnancy, and so forth--also are recorded. By recording the dates of all jobs and all periods of nonwork, the survey provides a nearly complete employment history for each individual in the sample.

This study of young workers examines the average number of jobs held, weeks worked, number of unemployment spells, and weeks unemployed over the 10-year age span, 18 to 27.7 In particular, this analysis focuses on how the work histories of these young people vary by their educational attainment levels in 1990, when nearly the entire sample had completed their schooling. The sample is restricted to those who were aged 18 or younger as of January 1, 1978, and traces their experiences through January 1, 1990.

Table 1 displays the percentage of workers in this cohort in each educational category. A little less than half (45.4 percent) are high school graduates; about a fifth (20.7 percent) have some college education (1 to 3 years of college completed); and another fifth (21.2 percent) are college graduates. Approximately 1 of every 8 individuals did not complete high school. A disproportionate percentage of blacks and Hispanics are high school dropouts. Also, the percentage of white college graduates is more than twice that of blacks and nearly 3 times that of Hispanics.

Jobs held and weeks worked

Analysis of the numbers of jobs held and weeks worked by young workers provides several interesting results. In particular, employment patterns vary significantly when the data are broken down by categories such as race and ethnicity, sex, and educational attainment, with the education categories showing the most striking differences. In general, the differences in the number of jobs held and weeks worked at each age by race and ethnicity tend to be more substantial than are differences by sex. (See chart 1 .) However, both differences tend to become much smaller as the individual's level of education increases.

For workers as a whole, the rate of increase in the number of jobs held decreases slightly with age. Individuals average about 6-1/2 jobs from age 18 to age 27.8 At each age, the number of jobs held by men since the initial interview is actually quite similar to the number of jobs held by women. Between ages 18 and 27, men on average have held about half a job more than have women.9 However, a more pronounced pattern is evident between racial and ethnic groups. Whites average more jobs at each age than do Hispanics, and Hispanics average more jobs at each age than do blacks? By age 27, whites, on average, have held about half a job more than Hispanics, and about 1 - 1/2 more jobs than blacks.

In contrast, the number of weeks worked rises at an increasing rate for young workers, ranging from an average of about 30 weeks at age 18 to approximately 370 weeks by age 27. At early ages, there is only a small difference between men and women in the number of weeks worked. However, the differential gradually increases to the extent that men average about 50 more weeks of work than do women by age 27. This difference is probably due to the fact that women gradually leave the work force to start families as they age. And, as in the case of the number of jobs held, whites average more weeks worked than Hispanics, and Hispanics average more weeks than blacks. By age 27, whites average about a year more of work experience than do Hispanics, and about a year and three-quarters more than do blacks.

Education and the number of jobs. Results on the number of jobs held since age 18 are broken down by educational category in table 2. The differentials by educational category basically follow the hierarchy of education, in that college graduates average the greatest number of jobs by age 27 and high school dropouts the fewest, while those with some college and high school graduates fall in between. By age 27, the average college graduate has held almost 7-1/2 jobs, approximately two more than a high school dropout. Furthermore, a person with some college has held about seven jobs, compared with nearly six jobs for the high school graduate.

The data reveal substantial variation in the differentials by sex and race/ethnicity within each educational grouping. For example, among high school dropouts, the differential between men and women in the number of jobs held at each age is larger than that of any of the other educational groups. Among high school graduates and those with some college, men average only slightly more jobs than do women at each age. Among college graduates, women, on average, have held more jobs at each age and approximately one more job than men by age 27. This gradual equalization between men and women as education increases is the result of prominent educational differences among women. By age 27, college educated women average more than 2-1/2 more jobs than women who are high school dropouts. In contrast, college educated men average less than half a job more than male high school dropouts.

The differences by race/ethnicity within educational categories also are very interesting. Among high school dropouts and high school graduates, whites and Hispanics are very similar in the number of jobs held at each age, and both groups have held more jobs than black high school dropouts and high school graduates at each age. Among those with some college, whites average slightly more jobs than Hispanics, and Hispanics slightly more than blacks. Yet, for college graduates, blacks and whites are very similar in average number of jobs held, whereas Hispanic college graduates have held slightly fewer jobs. Once again, there is an equalization in the number of jobs held as education increases. By age 27, a black worker with a college education has held about two more jobs than a black high school graduate, and about three more jobs than a dropout.

Education and weeks worked. Table 3 provides information by educational category on the cumulative number of weeks worked since age 18. Because college graduates were attending school while others were working, it might be expected that college graduates would average fewer weeks worked by age 27 than would others. However, the data indicate that individuals with some college register the most weeks worked by age 27, followed by high school graduates, college graduates, and high school dropouts. This finding for those with some college suggests that these individuals worked a great deal while attending classes.

Also, it is fairly remarkable to see that, even though many college graduates would have been in school between the ages of 18 and 22, they averaged more weeks of work at these ages than did high school dropouts. However, this is entirely due to the large disparity in the pattern of work experience for women. Among men, high school dropouts work more than college graduates from ages 18 to 22.

As in the case of the data on number of jobs held, there is considerable variation in weeks worked by demographic groups within educational categories. The largest difference in weeks worked between men and women occurs among dropouts. By age 27, male high school dropouts average about 200 more weeks worked (nearly 4 years) than do female high school dropouts. This differential between men and women is sharply reduced with each advance in educational level to about 74 weeks for high school graduates, and to about 12 weeks among those with some college. This differential actually reverses itself among college graduates, with female college graduates having worked about 20 weeks more than male graduates by age 27, and having worked consistently more over all ages as well. Hence, it appears that women work more than men while in college, as well as during the first few years after college.

The differences by race/ethnicity within each educational category exhibit a great deal of variability as well. Among high school dropouts, whites and Hispanics work a similar number of weeks at each age, whereas blacks average nearly 3 years less than the other groups by age 27. Among high school graduates and those with some college, whites average more weeks worked than Hispanics, and Hispanics more than blacks. Yet among college graduates, whites and blacks average nearly the same number of weeks worked at each age, and Hispanics average nearly the same as whites and blacks until age 25. This pattern is explained by the fact that, by age 27, college educated blacks on average have worked about 4-1/2 more years since age 18 than have black high school dropouts.

These results suggest that advancement in education equalizes many of the sex and race differentials in work experience among young adults. The demographic differences in the number of weeks worked considered here are erased almost completely with increases in education. In fact, women with a college education work more weeks and hold more jobs than college educated men. Still, the most striking comparison is probably between high school dropouts and the other educational attainment groups. In particular, college graduates work more than high school dropouts, even at ages when college graduates are most likely to be in school full time. This differential exists because female high school dropouts acquire substantially less work experience than do female college graduates at these ages.


The patterns in the number of unemployment spells and number of weeks unemployed are similar to the trends seen in the analysis of employment. Again, the differentials by race are more substantial than are differentials by sex. There is only a small difference between men and women in the number of unemployment spells. Still, there is considerable variation in the number of unemployment spells and weeks unemployed as educational attainment changes. On the whole, individuals average 4.4 unemployment spells from age 18 to age 27. The rate of increase in the number of spells decreases with age and stays fairly constant after age 24. (See chart 2.) Men average slightly more spells than women, but this difference is only about half a spell at its greatest. The numbers of unemployment spells of blacks and Hispanics are fairly similar at each age, whereas whites average significantly fewer spells at all ages. By age 27, whites have had about two fewer unemployment spells than blacks and about 1-1/2 fewer spells than Hispanics.

Likewise, the number of weeks unemployed since age 18 increases at a decreasing rate, and individuals average about 31 weeks unemployed by age 27. There is somewhat of a differential by sex, as men average about 8 more weeks unemployed by age 27 than do women. Moreover, there is a substantial differential by race, as whites average about half the number of weeks unemployed experienced by blacks and Hispanics at all ages.

Education and unemployment spells. Table 4 provides data on the number of unemployment spells by educational attainment. The average number of spells appears to vary inversely with education, as high school dropouts average the highest number and college graduates the lowest. By age 27, college graduates experience 3.7 spells of unemployment, while high school dropouts average 6.2 spells. The number of spells also has a tendency to level off after age 24 in each educational category. In other words, it appears that a large percentage of unemployment spells occur at early ages. This finding is consistent with information from the Current Population Survey, which typically shows that individuals in their late teens and early twenties experience higher unemployment rates than do those in other age groups.[11]

The differential between men and women decreases as education increases because of a sharp drop in the number of unemployment spells for men. Among high school dropouts, men, by age 27, average 7 unemployment spells, while women average only 5.3 spells. However, among high school graduates, the differential is much smaller, with men experiencing 4.7 spells and women 4.3 spells by age 27. For individuals with some college and college graduates, the gender differential also is small.

In contrast, the differential among race and ethnic groups does not vary significantly with education. The actual number of unemployment spells does decrease as education increases. However, there is not an explicit pattern among the groups as education changes. Still, the smallest racial differentials occur among college graduates. Overall, blacks experience the greatest number of unemployment spells, Hispanics experience fewer, and whites average the smallest number. By age 27, whites average one to two fewer spells than blacks and 1/2 to 1-1/2 fewer spells than Hispanics.

Education and weeks unemployed. The information in table 5 indicates that, like the number of unemployment spells, the average number of weeks unemployed since age 18 varies inversely with education, as high school dropouts average the most weeks unemployed and college graduates the fewest. While the average high school dropout spends about a year unemployed from ages 18 to 27, the college graduate spends about 4-1/2 months unemployed. For each educational level, most of the weeks unemployed occur at the earlier ages, and then taper off significantly. This leveling off with age, and the similar pattern seen in number of unemployment spells, may be due to a variety of factors, including the greater familiarity with job search and reduced difficulty in finding a job match among more experienced workers. It also may be due to a lack of on-the-job training among young workers.

Variations in the number of weeks spent unemployed by men and women are seen both across and within educational attainment groups. The differential between men and women in weeks unemployed is greatest for high school dropouts: by age 27, male high school dropouts spend about 20 more weeks unemployed than do female high school dropouts. However, the gender differential is much smaller at other educational levels. For those with some college and those who are college graduates, men and women spend approximately the same number of weeks unemployed at all ages.

For high school graduates, the differential is only about 3 weeks greater than that for the more highly educated groups. The most striking contrast is illustrated by the fact that a 27-year-old male high school graduate has been unemployed for about the same number of weeks as a 21-year-old male high school dropout.

The data also reveal that blacks and Hispanics spend more weeks unemployed than whites at all educational levels. Although this differential does not appear to have a clear relationship with education, it is smallest among college graduates. In addition, college graduates are the only group in which blacks spend fewer weeks unemployed than Hispanics. (At all other educational levels, the opposite is true.) By age 27, a black high school dropout has spent almost a year and a half unemployed, whereas a black college graduate has spent only about 7 months unemployed.

While not directly computed here, the differentials in both weeks employed and weeks unemployed show the strong variation in labor force participation by educational level. Individuals who are not classified as employed or unemployed are considered to be out of the labor force. The data reveal that high school dropouts participate in the labor force at a much lower rate than others. By age 27, high school dropouts have spent 1-1/2 years less in the labor force than high school graduates. Most of the difference can be attributed to female high school dropouts, because they have spent about 3 years less in the the labor force by age 27 than have female high school graduates. In contrast, male high school dropouts and male high school graduates participate in the labor force for approximately the same number of weeks. This finding indicates that female high school dropouts, many of whom probably are young mothers, lose an extremely large amount of labor market experience at early ages.

DATA FROM the National Longitudinal Survey of Youth used to examine the work histories of young workers indicate that there are significant differences by sex and race in the number of jobs held, weeks worked, number of unemployment spells, and number of weeks unemployed. However, comparisons by educational levels show that many of the differences become smaller or disappear completely with an increase in educational attainment. For instance, differences in work experience by sex or race and ethnicity among college graduates are extremely small.

The data also clearly show that young high school dropouts, particularly young female high school dropouts, acquire substantially less work experience between the ages of 18 and 27 than do groups with higher educational attainment.


1 The National Longitudinal Surveys (NLS) are sponsored by the Bureau of Labor Statistics (BLS), with data collection undertaken for BLS by the U.S. Bureau of the Census and the National Opinion Research Center, a social science research center affiliated with the University of Chicago. The BLS contracts with the Center for Human Resource Research of the Ohio State University to manage the surveys and provide user services.

The NLS are repeated interviews of groups of individuals over time. Four nationally representative samples were drawn in the 1960's: Young Men 14 to 24 years old in 1966, Young Women 14 to 24 years old in 1968, Older Men 45 to 59 years old in 1966, and Mature Women 30 to 44 years old in 1967. A fifth sample of Youth--young men and women aged 14 to 21--was drawn in 1979. The Young Women, Mature Women, and Youth surveys still are collected. For a detailed explanation of the NLS, see NLS Handboook 1992 (The Ohio State University, Center for Human Resource Research).

2 See Robert E. Hall, "The Importance of Lifetime Jobs in the U.S. Economy," American Economic Review, vol. 72, no. 4, 1982, pp. 716-24; Robert H. Topel and Michael P. Ward, "Job Mobility and the Careers of Young Men," Quarterly Journal of Economics, vol. 107, no. 2, 1992, pp. 439-79; and Work and Family: Jobs Held and Weeks Worked by Young Adults, Report 827 (Bureau of Labor Statistics, August 1992).

3 For theoretical economic models of job turnover, see Steven Lippman and John McCall, "The Economics of Job Search: A Survey, Part I," Economic Inquiry, vol. 14, June 1976, pp. 155-89; Boyan Jovanovic, "Job Matching and the Theory of Turnover," Journal of Political Economy, vol. 87, no. 5, 1979, pp. 972-90; and Jacob Mincer, Schooling, Experience and Earnings (New York, Columbia University Press, 1974).

4 Unemployed persons are those who had no employment during the survey week, were available for work at that time, and had made specific efforts to find employment sometime during the prior 4 weeks.

5 The sample is restricted to those interviewed in 1990, and the 1990 sample weight is used.

6 A job is defined as a period of work with a particular employer. For self-employed workers, each "new" job is defined by the individuals themselves.

7 Using 1990 data, it is actually possible to examine the complete work history since age 18 for individuals from age 18 to 29. However, the sample sizes among those aged 28 and 29 are too small to make meaningful comparisons by education, race, and sex.

8 For numbers on the overall totals on jobs held and weeks worked, see Work and Family, Report 827 (Bureau of Labor Statistics, August 1992).

9 All inferences drawn in the text are statistically significant at the 90-percent confidence level. Standard errors are available from the authors upon request.

10 These groups are mutually exclusive. The Hispanic category includes both blacks and whites who categorize themselves as Hispanic.

11 Employment and unemployment figures from the CPS are published monthly by the Bureau of Labor Statistics in Employment and Earnings.

Government's place in the labor market

Politics and government have no necessary place in economic thinking on the labor market. For mainstream and critical economic theorists alike, the labor market operates through decisions--private decisions, that is---of suppliers and purchasers of labor acting singly or collectively in their specific best interests. That government initiatives affect transactions in the labor market is admitted---after all, government policies on, for example, immigration, schooling, old-age pensioning, trade unionism, and assistance to the disadvantaged shape total labor supply and demand and the respective leverage of workers and employers. Yet government interventions are envisioned by economists as ancillary (and, some would also argue, as meddlesome) to the system or the natural workings of the labor marketplace.

Walter Licht,

Getting Work: Philadelphia, 1840-1950

(Cambridge, MA, Harvard University Press, 1992), p. 174.
COPYRIGHT 1993 U.S. Bureau of Labor Statistics
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1993 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Veum, Jonathan R.; Weiss, Andrea B.
Publication:Monthly Labor Review
Date:Apr 1, 1993
Previous Article:Employment effects of the rise and fall in defense spending.
Next Article:Occupational employment in commercial banking, 1987-90.

Related Articles
School of hard knocks.
`Tighten tests for teachers'.
Switched-on school seeks radio licence.
Giacomo Meyerbeer; a guide to research.
Improving learning, skills and inclusion; the impact of policy on post-compulsory education.
James busy with his studies - at 82; EDUCATION: Maths course gave former construction worker the learning bug.

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters