Click for next page ( 35


The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 34
2 Youth Employment and Unemployment THE YOUTH EMPLOYMENT PROBLEM: Unemployment Rates ITS NATURE AND DIMENSIONS The United States recently experienced its most serious unemployment problems since the Great Depression of the 1930s. In the depths of this recession, in December 1982, the overall unemployment rate reached a postwar high of 10.8 percent. But the unemployment rate for teenagers was 24.5 percent--more than twice the overall rate--and the unemployment rate for black teenagers was 49.5 percent. Since the trough of the recession, the national employment situation has improved somewhat, so that during 1984 the unemployment rate averaged 7.5 percent. The rate for teenagers was still substantially higher--18.9 percent--and the rate for black teenagers had improved only slightly, to 42.7 percent. The youth employment problem is not due merely to the greater vulnerability of young workers to the swings of the business cycle. There has been a long-term upward trend in youth unemployment rates over the last several decades (Congressional Budget Office, 1982~. Table 2.1a provides statistics for four periods from 1957 to 1984: 1957, 1964, and 1978 were chosen because they were years of relatively high economic activity and had identical unemployment rates for adult white men aged 35-44. Over the period spanned by these statistics, the unemployment rate for all youths climbed steadily. ~ ~ ~ ~ . . . . . . . . . In addition, the gap between Wh' te and nonwhite youths that was evident in 1957 became much larger over these decades. Thus, even among the more "settled" 20- to 24-year-old youths, the 1957 unemployment rate for white males was 7.1 percent while the rate for nonwhites was 12.7 percent; by 1984, this gap had expanded to 9.8 percent for whites and 24.5 percent for nonwhites. For women aged 20-24, the unemployment gap had expanded similarly, from 5.1 percent for whites and 12.2 percent for nonwhites in 1957 to 8.8 percent for whites and 23.5 percent for nonwhites in 1984. Table 2.lb, which compares the unemployment rates for young white males with other youths, shows that nonwhite females aged 20-24 were 1.7 times as likely as white males to be unemployed in 1957; by 1984 they were 2.4 times as likely to be unemployed. In contrast, white females have in most years been less likely to be unemployed than white 34

OCR for page 34
35 TABLE 2.1a Youth Unemployment Rates in the Civilian Population for Selected Years (in percentages) Year Group 1957 1964 1978 1984 Adult white males 35-44 years old2.52.52.54.6 All youths 16-17 years old12.517.819.321.2 18-19 years old10.914.914.217.4 20-24 years old7.18.39.611.5 White males 16-17 years old11.916.116.919.7 18-19 years old11.213.410.815.0 20-24 years old7.17.47.79.8 Nonwhite males 16-17 years old16.325.939.839.8 18-19 years old20.023.130.738.5 20-24 years old12.712.620.024.5 Hispanic males 16-17 years oldaa27.530.5 18-19 years oldaa13.921.6 20-24 years oldaa9.412.1 White females 16-17 years old11.917.117.117.8 18-19 years old7.913.212.413.6 20-24 years old5.17.18.38.8 Nonwhite females 16-17 years old18.336.541.542.2 18-19 years old21.329.236.336.6 20-24 years old12.218.321.323.5 Hispanic females 16-17 years oldaa29.925.2 18-19 years oldaa16.621.4 20-24 years oldaa13.012.5 NOTE: _ The vears 1957. 1964, and 1978 were selected because in each of these years the unemployment rate for white males aged 35-44 was an identical 2.5 percent and the business cycle was about at its peak; 1984 was selected to provide a view of recent youth unemployment. aNo data for persons of Hispanic origin are available for 1957 or 1964. SOURCE: Data from U.S. Department of Labor (1982, 1985b) .

OCR for page 34
36 TABLE 2.lb Ratios Between Unemployment Rates for Young White Males and Other Groups Year _ 1957 1964 1978 1984 Group White males 16-17 years old1.01.0 1.01.0 18-19 years old1.01.0 1.01.0 20-24 years old1.01.0 1.01.0 Nonwhite males 16-17 years old1.411.61 2.362.02 18-19 years old1.741~72 2.842.57 20-24 years old1.791.70 2.602.50 Hispanic males 16-17 years oldaa 1.631.52 18-19 years oldaa 1.291.44 20-24 years oldaa 2.601.23 White females 16-17 years old1.031.06 1.01.90 18-19 years old.69.99 1.15.91 20-24 years old.72.96 1.08.90 Nonwhite females 16-17 years old1.592.27 2.462.14 18-19 years old1.852.18 3.362.44 20-24 years old1.712.47 2.772.40 Hispanic females 16-17 years oldaa 1.771.28 18-19 years oldaa 1.541.43 20-24 years oldaa 1.691.28 NOTE: The years 1957, 1964, and 1978 were selected because in each of these years the unemployment rate for white males aged 35-44 was an identical 2.5 percent and the business cycle was about at its peak; 1984 was selected to provide a view of recent youth unemployment. aNo data for persons of Hispanic origin are available for 1957 or 1964. SOURCE: Data from U.S. Department of Labor (1982, 1985b).

OCR for page 34
37 males, but between 1957 and 1984 this ratio approached parity: for 20- to 24-year-old white females, the ratio of unemployment rates was 0.7 in 1957 and 0.9 in 1984. (Although comparable historical data are not available for Hispanic youths, the available data indicate that Hispanic males aged 20-24 were 1~2 times as likely as white males to be unemployed in 1984, and Hispanic females were 1.3 times as likely to be unemployed as white males.) These continuing trends in the relative unemployment rates of young Americans were a primary motivation for the launching in the late 1970s of federally funded programs designed to provide employment and training services to disadvantaged youths. Yet, as the last column of Table 2.1a indicates, the gap between white and nonwhite unemployment rates has persisted: in 1984 unemployment among white youths aged 20-24 was 9.8 percent for males and 8.8 percent for females; for nonwhite youths the rates were 24.5 and 23.5 percent, respectively. While the unemployment rates and ratios shown in Tables 2.1a and 2.lb demonstrate that young people's problems have been increasing, the unemployment rate can sometimes be a misleading indicator, particularly when applied to the youngest segment of the labor force (Hahn and Lerman, 1983:2~. To be counted as unemployed a person must indicate in answer to a survey question that (1) she or he is not currently employed and (2) she or he is currently looking for work. People who are not working and who say they are not actively looking for work are counted as "out of the labor force" rather than unemployed. The unemployment rate is calculated by dividing the number of people who are unemployed by the number of people in the labor force (defined as the sum of the employed [E] and unemployed [Ul): unemployment rate = U/(U + E) It can be seen that the unemployment rate can rise even though the number of employed (E) stays constant. And, given the way in which one is defined as being "in the labor force," it is not necessary that there be any change in the number of people who are not working. The unemployment rate may rise simply because more people begin looking for work (or at least say they are looking for work), thereby increasing the size of the labor force. [See Bailar and Rothwell (1984) and National Commission on Employment and Unemployment Statistics (1979) for discussions of this and other aspects of unemployment measurements.] The unemployment rate is particularly ambiguous as an indicator of employment problems in the youth population because it becomes entangled with school attendance. When young people say that they are looking for work even though they are also enrolled in school, they are none- theless counted as unemployed. This method of counting raises serious questions of interpretation since full-time students, it can be argued, already have a full-time though unpaid occupation, attending school. This component of youth unemployment statistics is not insubstantial: for example, almost half of the 1978 teenage unemployment shown in Table 2.1a is generated by youths who were enrolled in school. It is thus necessary to examine other measures to better understand the nature and scope of youth employment problems.

OCR for page 34
38 TABLE 2.2a Civilian Employment-to-Population Rates for Selected Groups (in percentages) Year Group 1957 1964 1978 1984 Adult white males 35-44 years old 95.6 95.1 93.9 91.6 All youths 16-19 years old 43.9 37.3 48.5 43.7 20-24 years old 59.5 60.9 69.6 68.7 White males 16-19 years old 52.4 45.0 56.3 49.0 20-24 years old 80.5 79.3 76.0 78.0 Nonwhite males 16-19 years old 48.0 37.8 29.8 25.2 20-24 years old 78.2 78.1 61.1 58.3 White females 16-19 years old 38.3 32.2 48.7 47.0 20-24 years old 43.4 45.3 60.6 66.1 Nonwhite females 16-19 years old 26.5 21.8 23.5 21.8 20-24 years old 40.9 43.7 45.4 46.3 NOTE: The years 1957, 1964, and 1978 were selected because in each of these years the unemployment rate for white males aged 35-44 was an identical 2.5 percent and the business cycle was about at its peak. In 1984 the rate of unemployment among white males aged 35-44 was 4.6 percent. SOURCES: Data from U.S. Department of Labor (1979, 1980a, 1985b); Bureau of Labor Statistics (1983~. Employment-to-Population Rates Table 2.2a presents the employment-to-population rates (the number of employed divided by the total civilian population) for youths in the same years for which the unemployment rates are presented. Over the period 1957 to 1978, the employment rate in the youth population actually ~ slightly increase increased from 52.0 to 59.9 percent, although it then declined to 58.3 percent in 1984 (not shown in the table). The in employment rates between 1957 and 1978 was more marked for

OCR for page 34
39 TABLE 2.2b Ratio of Civilian Employment-to-Population Rates for Young White Males to Other Young Groups Year Group 1957 1964 1978 1984 White males 16-19 years old 1.0 1.0 1.0 1.0 20-24 years old 1.0 1.0 1.0 1.0 Nonwhite males _ 16-19 years old .92 .84 .53 .51 20-24 years old .97 .98 .80 .75 White females 16-19 years old .73 .72 .87 .96 20-24 years old .54 .57 .80 .85 Nonwhite females 16-19 years old .51 .48 .42 .44 20-24 years old .51 .55 .60 .59 NOTE: The years 1957, 1964, and 1978 were selected because in each of these years the unemployment rate for white males aged 35-44 was an identical 2.5 percent and the business cycle was about at its peak. In 1984 the rate of unemployment among white males aged 35-44 was 4.6 percent. SOURCES: Data from U.S. Department of Labor (1979, 1980a, 1985b); Bureau of Labor Statistics (1983~. the older youth group, aged 20-24, than for the younger group, and the decline from 1978 to 1984 was steeper for the younger group. At a more detailed level, the trends for various demographic groups are not homogeneous. For example, there was a large increase in the employment rates of white women aged 20-24 (from 43 percent in 1957 66 percent in 1984), but there was also a substantial decline in the to employment rates for nonwhite men of the same age group (from 78.2 to 58.3 percent). Table 2.2b presents the ratios of the employment rates of each group to the employment rate for white males. From 1957 to 1984 this ratio declined markedly for nonwhite males. For nonwhite females the ratios declined for the younger group, while they increased somewhat for 20- to 24-year-olds. Nonetheless, in all years for both age groups, the likelihood that nonwhite females would be employed was less than 0.6 times the likelihood that white males would be employed. For white females, the ratios showed steady increases from 1964 to 1984, with the ratio for the most recent year approaching parity for 16- to 19-year-olds; however, it was somewhat lower (0.85) for 20- to 24-year-olds.

OCR for page 34
40 Employment of In-School and Out-of-School Youths Any discussion of employment-to-population rates runs the risk of confusing trends in school attendance with trends in employment. In the present case, this is a particularly worrisome possibility. While the employment rate for nonwhite youths has declined over the last 3 decades (as shown in Table 2.2a), the school enrollment rate for nonwhite youths has increased during these same decades. The rate of high school completion among black men and women aged 25-29 rose from 47.7 percent in 1960 to 65.4 percent in 1970 and to 79.4 percent in 1983. The employment patterns of youths who are enrolled in school are, of course, considerably different from those who are out of school. Table 2.3 provides a breakdown by school enrollment of the employment rates for 1964, 1978, and 1981 for all youths aged 16-24.2 As one would expect, in-school youths are less likely to be employed than out-of-school youths. However, there are significant differences in these rates over time for different groups. For white males, the employment rates increased for in-school youths from 34.0 percent in 1964 to 43.4 percent in 1981, while the rate for out-of-school youths was stable at approximately 87 percent between 1964 and 1978 and then declined slightly during the economic downturn in 1981. In contrast, the employment rates of black males have shown a marked decline for both in-school and out-of-school youths: the rate for those out of school was 80.5 percent in 1964, 67.8 in 1978, and 57.8 in 1981; the rate for those in school dropped from 30 percent in 1964 to 20 percent in 1978 and was still at 20 percent in 1981. 3 iIdeally, one would like to examine trends in employment status broken down by school enrollment, race, sex, age, presence of dependents, and living arrangements. Unfortunately, tabulations of employment statistics (e.g., the Employment and Earnings series and the Handbook of Labor Statistics) do not provide the appropriate detail. Indeed, even with the 60,000+ sample size of the Current Population Survey, we suspect it would be difficult to obtain reliable estimates for all the cells of such a cross-tabulation. Consequently we use the strategy of examining the employment status of older, out-of-school youths as a crude substitute. 2 The years 1964 and 1978 were selected to provide consistency with other tables in this chapter. Appropriate data were not published in 1957 (or earlier years). No data are currently available for 1984; consequently, we have used the most recent published statistics, for 1981. 3 In this discussion of Tables 2.3 and 2.4 we used statistics for black youths rather than for nonwhite youths. This reflects the categorization used in the published statistics. Federal statistics for recent years generally divide the population by black and white and include counts for the total population (so nonwhite statistics can be computed). For earlier years it is often

OCR for page 34
41 TABLE 2.3 Employment-to-Population Rates for In-School and Out-of-School Youths Aged 16-24 by Sex and Race: 1964-1981 Emp Out-of Group School Youths In-School Youths White males 1964 86.7 34.0 1978 86.9 46.9 1981 81.1 43.4 Black males 1964 80.5 30.0 1978 67.8 20.3 1981 57.8 20.1 White females 1964 47.3 23.3 1978 66.2 45.7 1981 68.3 43.0 Black females 1964 48.0 15.4 1978 46.9 20.6 1981 43.0 17.2 SOURCE: Bureau of Labor Statistics (1982:Table C-42. Comparing the data for young males, one finds that in 1964 the employment rates of both in-school and out-of-school black males were roughly 90 percent as large as those of white males. However, by 1981 this gap had widened enormously: in-school black males were less than 50 percent as likely to be employed as white males, and out-of-school black males were only 71 percent as likely to be employed as white males. For young females, the data for blacks and whites also show very different trends. Both in-school and out-of-school white females registered roughly a 20 percentage point increase in their employment the case that only statistics for whites and nonwhites were published. It is thus impossible to produce long time series (e.g., 1950-1980) that describe the black youth population. Nonetheless, the nonwhite statistics, while less than ideal, do capture much of what is important since blacks constitute the vast majority of the nonwhites in the United States. In 1980 the nonwhite population included: 26.5 million blacks; 3.5 million Asians and Pacific Islanders; 1.4 million American Indians, Eskimos, and Aleuts; and 6.8 million persons whose race was classified as "other."

OCR for page 34
42 TABLE 2.4 Employment-to-Population Rates for Out-of-School Youths by Age, Race, and Sex: 1964-1981 Age Group 16-17 18-19 20-24 White males 1964 65.6 80.9 90.0 1978 52.2 85.0 89.5 1981 54.3 75.3 84.0 Black males 1964 43.8 73.4 86.6 1978 19.4 44.7 59.4 1981 22.2 39.8 62.8 White females 1964 30.7 51.9 47.3 1978 47.3 64.7 67.7 1981 34.1 62.4 68.3 Black females 1964 35.9 45.1 50.2 1978 21.4 46.7 63.4 1981 5.5 29.5 48.5 SOURCE: Bureau of Labor Statistics (1982:Table C-42. to-population rates between 1964 and 1981. For the in-school group, their employment rate in 1981 was virtually identical to that of white males. For out-of-school females, their employment rates were con- sistently below those of young white males, although they increased significantly between 1964 and the later two years. For black females, there was a slight upward trend in employment while in school during the period 1964 to 1981; for the out-of-school group, the rate declined slightly over the period, from 48 percent in 1964 to 43 percent in 1981. Table 2.4 disaggregates the employment-to-population rates of out-of-school youths by age. This breakdown shows that the aggregate results hold for all age groups of out-of-school youths, including the oldest. Many researchers argue that unemployment among these older, out-of-school~youths is of particular concern because they are more likely to have dependents to support and to be living outside their parental home. This group shows the familiar pattern of rather high employment rates among white males (90.0 percent in 1964, 84.0 percent in 1981) and consistently lower rates for blacks and females. For out-of-school black males aged 20-24, the employment-to-population rate in 1964 (86.6 percent) approaches that of white males, but the rate declines by over 20 percentage points in the following decade. White females in this age group show increasing rates of employment, but they are still less likely to be employed than white males (47.3 percent of

OCR for page 34
43 TABLE 2.5a Civilian Labor Force Participation Rates by Age, Race, and Sex (in percentages) Year Group 1957 1964 1978 1984 All Youths 16-17 years old 40.2 35.1 48.6 42.4 18-19 years old 60.4 57.2 67.3 64.9 20-24 years old 64.0 66.3 76.8 77.6 White males 16-17 years old 49.6 43.5 55.3 47.0 18-19 years old 71.6 66.6 75.3 70.8 20-24 years old 86.7 85.7 87.3 86.5 Nonwhite males 16-17 years old 47.5 37.3 33.2 27.0 18-19 years old 72.0 67.2 58.9 55.4 20-24 years old 89.6 89.4 77.5 77.2 White females 16-17 years old 32.1 28.5 48.8 44.8 18-19 years old 52.6 49.6 64.6 65.2 20-24 years old 45.8 48.8 69.3 72.5 Nonwhite females 16-17 years old 24.1 19.5 27.7 24.7 18-19 years old 42.S 46.5 48.4 45.8 20-24 years old 46.6 53.6 62.6 60.5 NOTE: The years 1957, 1964, and 1978 were selected because in each of these years the unemployment rate for white males aged 35-44 was an identical 2.5 percent and the business cycle was at about its peak. In 1984 the rate of unemployment among white males aged 35-44 was 4.6 percent. SOURCE: Data from U.S. Department of Labor (1982, 1985b). white females were employed in 1964 and 68.3 percent in 1981~. For black females aged 20-24, the employment rate rises from 50.2 percent in 1964 to 63.4 percent in 1978 and then declines to 48.5 percent in 1981. Labor Force Participation Rates and Summary of Employment Data Tables 2.5a and 2.5b provide complementary information on the aggregate civilian labor force participation rates of youths by age, sex, and race. (The civilian labor force participation rate is the ratio of employed and unemployed people to the total nonmilitary

OCR for page 34
44 TABLE 2.Sb Ratio of Civilian Labor Participation Rates for Selected Groups to Rate for White Males Group Year 1957 1964 1978 1984 White males 16-17 years old1.01.0 1.0 1.0 18-19 years old1.01.0 1.0 1.0 20-24 years old1.01.0 1.0 1.0 Nonwhite males 16-17 years old.96.86 .60 .57 18-19 years old1.001.00 .78 .78 20-24 years old1.031.04 .89 .89 White females 16-17 years old.65.66 .88 .95 18-19 years old.73.74 .86 .92 20-24 years old.53.57 .79 .84 Nonwhite females 16-17 years old.43.45 .50 .53 18-19 years old.60.70 .64 .65 20-24 years old.54.63 .72 .70 NOTE: The years 1957, 1964, and 1978 were selected because in each of these years the unemployment rate for white males aged 35-44 was an identical 2.5 percent and the business cycle was at about its peak. In 1984 the rate of unemployment among white males aged 35-44 was 4.6 percent. SOURCE: Data from U.S. Department of Labor (1982, 1985b). population; as noted above, the labor force statistics exclude persons who are unemployed and not looking for work.) A comparison of the rates in Tables 2.1a through 2.2b and Tables 2.5a and 2.5b suggest that youth unemployment is generated by somewhat different underlying trends for males and females and whites and nonwhites. For that reason it is useful to discuss each group separately. White Males The labor force participation rates for white males fluctuated over the 1957-1984 period. They were up in 1957 and 1978 (particularly for 16- to 17-year-olds) but declined in 1964 and 1984. There was also an upward movement in the employment-to-population rate between 1957 and

OCR for page 34
58 approximately 224,000 is quite substantial when viewed in perspective with the number of nonwhite male youths of the same age who were employed in civilian occupations (1.1 million). Changing patterns of military service by different racial groups masked some of the differ- ential in civilian employment between black and white youths. During the 1970s white youths' participation in the military declined sub- stantially while black youths' participation remained approximately constant (Ellwood and Wise, 1983~. If, after 1972, the proportion of black youths in the military had declined in proportion to the white decline, the proportion of black youths without work in 1982 would have have risen by about 3 percent (Mare and Winship, 1983~. (Participation in military service by females involved only 0.6 percent of white females and 1.4 percent of nonwhite females in 1984.) Factors Affecting the Supply of Labor Demographic Trends During the 1970s several demographic trends might have affected youth employment. First, and most prominent, was the entry of the massive baby-boom generation into the labor force. Theory suggests that as the supply of young workers rises relative to the supply of both older workers and other factors of production, youth wages or employment will fall relative to that of older workers. Indeed, in a cross-sectional analysis of standard metropolitan statistical areas, Freeman (1982) found that as the youth share of the population increased, employment prospects declined by a moderate amount, particularly for those aged 16-17. However, analysis by Wachter and Kim (1982) suggests that, at a national level and over time, the primary effect appears to have been on wages. For example, as shown in Tables 2.2a and 2.2b, during the period of rapid expansion of the youth labor force, 1957 to 1978, the employment-to-population rate for white youths stabilized or actually increased. In contrast, Table 2.8 shows that during the 1970s the wages of white youths declined relative to adult wages. Whatever the effects of this large demographic bulge, it did not overcome other factors tending to raise the employment-to- population rate for white youths, but it may have played a role in lowering their relative wages. Among black youths, however, the pattern appears different: relative wages over those years declined by less than those of white youths, but unemployment rose and employment rates fell substantially. Two other factors increased the supply of labor during the same period: the sharp and continuing rise in the labor force participation of adult women (see Hahn and Lerman, 1983) and the influx of immigrant workers into the United States. Each of those groups might draw jobs away from youths if they enter the labor market in part-time or low- skill jobs (particularly if employers prefer to discriminate in their favor or can pay lower wages to these groups). It is possible that increased numbers of women in the labor force may have worsened the employment prospects and lowered the wage rates of youths (Borjas,

OCR for page 34
59 1983), although there have been few studies of such effects. Estimating the employment interactions between youths and immigrants has been difficult because of the lack of reliable data on the illegal component of the immigrant work force. However, Freeman and Holzer (1985) report that there is no evidence to support the view that increases in the Hispanic population (which accounts for a substantial number of immigrants) have hurt job opportunities for black youths, since black youth unemployment rates are similar in cities with large and small Hispanic populations. A fourth demographic development of considerable importance is the change in childbearing and marital patterns in the youth population. During the 1970s there was both a decline in the rate of marriage among youths and an increase in divorce among those who did marry. It is possible that these changes might increase the supply of female labor. While childbearing declined sharply among young married women, it did not decline among unmarried women. Births to unmarried women tripled as a share of all births between 1960 and 1979 (although their number did not rise). In 1983 among married and unmarried women aged 18-24, there were 965 births per 1,000 female high school dropouts and 506 births per 1,000 female high school graduates who did not attend college (U.S. Bureau of the Census, 1984:Table 43. Since young, unmarried women with children have disproportionately lower incomes and consequently may have difficulty obtaining affordable child care, they may have more difficulty in finding and holding jobs than other young people. However, while the magnitude of this effect has not been estimated for youths, there is evidence that lack of satisfactory child care is a restraint on women's employment (Presser and Baldwin, 1980~. (As noted in Chapter 1, research on this important topic should be encouraged.) As we noted above, there has been a substantial expansion of the youth labor supply over the last several decades, resulting from changes in birth rates during the immediate postwar period and sub- stantial increases in the number of young women who entered the labor market. To the extent that an excess of "supply" is (by definition) a prerequisite for unemployment, it is prudent to ask whether this growth in the supply of young workers will continue in the next decade. Since one aspect of such a forecast involves making assumptions about the future decisions of millions of young women, any answer would be quite speculative. It may not be unreasonable to expect the rate of female participation in the labor force to approximate that of men, but it may also not be unreasonable to speculate that traditional patterns will die hard, thus restraining further large jumps in the rate of female participation in the labor market. There is, however, one aspect of a forecast about which we do have some "hard" evidence: the l990s "supply" of teenagers has already been born, and barring massive changes in death rates or patterns of migra- tion, one can venture a prospective count of their numbers. Figure 2.1 shows the actual and projected size of the older teenage population for 1960-2000. For the period 1960 through 1982, this segment of the population grew from roughly 13 million to more than 21 million in 1980 and then began to decline. In 1982 the population aged 15-19 totaled

OCR for page 34
60 20 ~ 15 o . _ . _ . _ - O o cat 5 o 15-19 Year Olds f ~ - 1 I I 1 1 1 1 1960 1970 1980 YEAR 1 990 2000 FIGURE 2.1 Actual and projected trends in the youth population aged 15-19, 1960-2000. NOTE: The 1970 and 1980 figures are population counts from decennial censuses reported in U.S. Department of Commerce (1985:Table 30~. The 1960 figures are population counts from the 1960 decennial census as reprinted in Bureau of the Census {1973:Table 189~. The 1975, 1981, and 1982 figures are estimates based on Current Population Survey sample surveys of population as reported in Bureau of the Census (1979, 1982~. The 1985-2000 figures are population projections (middle series) made by and reported in Bureau of the Census (1982~. 19.8 million. When the size of this population group is projected to later years, it shows continuing declines through 1995; at that time it is roughly 80 percent of its peak (1980) size. Thus, on the supply side, the demographic projections indicate that there will be a steady decline in the number of potential participants in the labor market through 1995. Enrollment in School Changes in school enrollment patterns could have a direct effect on the measured extent of unemployment problems among youths (Hahn and Lerman, 1983~. During the 1960s and 1970s, the enrollment rates for white males declined somewhat and those for white women and blacks of

OCR for page 34
61 TABLE 2.9 Percentage of Persons Aged 16-24 Enrolled in School, by Race and Sex Group 1964 1974 1983 White males 51.0 45.8 44.7 Nonwhite males 39.4 48.5 45.4 White females 36.4 39.1 40.7 Nonwhite females 34.1 38.6 40.9 SOURCE: Data from Bureau of Labor Statistics (1982, 1984). both sexes increased (see Table 2.9~. The difference in the patterns of school enrollment between blacks and whites contributed in part to the growing differential in employment-to-population ratios between black and white youths (Freeman, 1980~. The declining school enroll- ment rate of whites would tend to increase their employment rates since the employment rate for those out of school is generally higher than that for those in school. However, most of the increased employment for young whites in the past two decades came from rising employment rates for in-school youths. More interesting than the question of enrollment status is the degree to which educational attainment has an effect on the labor market experiences of youths. Many studies indicate that dropouts have a more difficult time in the labor market than do high school graduates. These effects are seen in difficulties in obtaining the first job, in duration of unemployment between jobs, and in wage rates. Academic performance appears to be positively related to both number of weeks employed and wage rates for youths. Other studies find that vocational training in high school appears to be unrelated to employment and wage rates, while there are some indications that vocational training after nlgn school may nave some positive ettects (Freeman and Wise, 1982). These findings on the effects of education on employment experi- ences may help to explain the distribution of employment and unemploy- ment among youth groups, but they do not appear to help to explain the growing differential between black and white youths. One explanation that has been put forward is that differences in the quality of educa- tion are responsible for the growing differential. The validity of this explanation is difficult to test. Studies of functional literacy than among whites, do show that literacy rates are lower amona ~4 arks but there is no indication that this gap has widened over recent years. Similarly, while there has been some overall decline in the Scholastic Aptitude Test (SAT) and other test scores, there is no indication that racial differentials in test scores have increased over time {Hahn and Lerman, 1983~. _ ~

OCR for page 34
62 Youths' Expectations and the Reservation Wage The "reservation wage" of a person is defined as the lowest wage at which that person would be willing to take a job. It has been suggested that some of the employment problems of youths may be related to a reservation wage that is too high. In addition, some analysts argue that increasing incomes throughout the society have caused the level of the reservation wage to rise over time to a greater degree than warranted by the increasing skills of the labor force. Data on reservation wages have not been collected over long enough periods of time for conclusions to be drawn as to whether rising reservation wages have been a major cause of increased unemployment for younger workers. On the whole, recent studies find (e.g., Freeman and Holzer, 1985) that the reservation wages of unemployed younger workers appear on average to be quite realistic: both white and black youths appear to have reservation wages that are quite close to the prevailing federal minimum wage. While the reservation wages of white and black male youths are about the same, Freeman and Holzer suggest that the fact that the employment prospects for blacks are worse means that their reservation wages are higher relative to the actual wages they are likely to be able to obtain. And reservation wages for specific low-wage jobs are generally lower for blacks than for whites. Reservation wages of young blacks appear to have the effect of lengthening the period of nonemploy- ment but also of increasing subsequent wages. The reservation wages of young whites have somewhat less effect on the duration of nonemployment but greater effects on their subsequent wages. Summing Individual Effects Thus far we have been serially reviewing possible causes of the trends in youth employment problems within a framework of demand and supply factors. Two sets of researchers, Ellwood and Wise (1983) and Mare and Winship (1983), have independently sought to bring together most of the factors covered above in a consistent accounting framework in order to see what proportion of the growth in the gap in black/white youth employment rates can be explained by the sum of the individual effects of all the factors. Though their accounting frameworks are quite different, both sets of researchers conclude that they can account for only about 50 percent of the diverging racial employment patterns among youths in the 1970s. In discussing each factor separately we have also not touched upon possible (nonadditive) interactions among factors; such interactions might yield results that are different from the simple sum of each individual factor. Two hypothetical examples can illustrate such interactions. It was previously noted that increases in the supply of young workers seem to be related to increases in employment rates and decreases in wages (relative to adults) for young white males, but they seem to be related to sharp decreases in employment rates and smaller relative wage decreases for young black males. These differences might -

OCR for page 34
63 be due to the interaction of the increased supply of labor and the existence of minimum wage rates and increased civil rights enforcement and affirmative action programs. The wages of young black males were already closer to the minimum wage than were those of young whites, so when the youth labor supply increased employers had less room to compress black wages than white wages. [Hall (1982) suggests this possibility in his commentary on the research of Wachter and Kim (1982~.] A second possible interaction is between the demographic increase in supply and employer discrimination. Even if the desire to discrimi- nate on the part of employers was not increasing during recent decades, the increase in the supply of both young whites and blacks may have increased the scope for the exercise of discriminatory hiring by employers.5 This theoretical possibility was emphasized in the earliest exposition of an economic theory of discrimination by Becker (1957) While these higher-order interactions generate interesting hypotheses, they are extraordinarily difficult to assess empirically, particularly when they involve such factors as the minimum wage or discrimination, which have proved challenging to assess even as singular first-order factors. Other Influences on Youth Employment Several research findings do not fit neatly into the supply and demand framework we have used in the preceding sections of this chapter. We note several of these briefly and then turn to a dis- cussion of social context. Family Influences and Teenage Experiences Family background has a positive relationship to the probability that a young person is employed, and Meyer and Wise (1982) find that an increase of $5,000 in parental income is associated with an increase of more than three weeks in the number of weeks worked by teenagers. Other family structure factors do seem to affect employment probabilities (see Rees and Gray, 1982; Corcoran, 1982~. Youths with siblings working are more likely to be working themselves, suggesting the importance of family connections for information or role models. sSimilarly, in the increasing concern with civil rights and affirmative action there may have been greater pressure for equal wage treatment, leading employers to make more of their adjustment to increased supply by decreased hiring of blacks. Freeman {1985) rejects this hypothesis, arguing that affirmative action increases relative employment of blacks by punishing discrimination in employment as well as in wage setting.

OCR for page 34
64 Youths from female-headed families or families on welfare have slightly lower probabilities of being employed. A somewhat surprising and potentially important finding in several studies (e.g., Meyer and Wise, 1982; Stevenson, 1980) is that there is a strong relationship between hours worked while in high school and later employment and wage rates. Whether the relationship is really causal or simply correlative (i.e., due to a common underlying factor such as motivation) remains unclear. Obviously, for those interested in the potential benefits from employment and training programs for in-school youths, this finding is intriguing. A final finding that has drawn the attention of many analysts is that the long-term (i.e., 4-5 years later) effects of unemployment during younger years appear to be rather less than had been previously suggested. Once individual characteristics have been controlled for, the experience of early unemployment does not appear to raise the probability of unemployment in the following 4-5 years. This result appears to hold for both young men and young women (Ellwood, 1982; Corcoran, 1982~. In addition, once individual characteristics are held constant, initial wage levels seem to have little relationship to wage levels 4-5 years later. These relatively encouraging findings about the limited effects of early unemployment and wages are, however, counterbalanced by another finding: early unemployment experience does seem to affect wage levels 4-5 years later, and this effect appears to be stronger and more substantial for youths with lower levels of education. Social Context We conclude by noting a final factor that may strongly influence the employment experiences of young minority youths: the social context that has formed their perceptions and responses. We have chosen to discuss this issue of social context in the final part of this section because it affects both the supply and the demand for labor and because the effects may be strong. The residue of past and current discrimination finds its expression on the demand side in diminished opportunities for minority youths in the labor market (because of the attitudes of employers); and, to the extent that the social context affects the perceptions, attitudes, and responses of youths, it can have a quite fundamental impact on the supply of labor. The long history of the exclusion of blacks from social and economic power, government, and prestigious occupations affects youths in many ways. As Ogbu (1985a, 1985b) has observed in his study of minority youths in Stockton, California, there is a racial or castelike stratification between blacks and whites that historically found expression in such things as job ceilings for black workers. A pilot project by Culp and Dunson (1983) finds evidence of such stratification in the treatment of matched young black and white "auditors n who applied for jobs at firms in the Newark, New Jersey, area. The auditors were recent high school graduates who were trained to make systematic observations of their treatment. Although the study was

OCR for page 34
65 only a pilot project and the samples were too small for statistical testing, the results suggested that black youths may be treated with less courtesy and may be less likely to be informed of job prospects (Culp and Dunson, 1983~. Other independent anthropological studies (e.g., Ogbu, 1985b) have found evidence of negative stereotyping of low-income blacks. The collective adaptation of black youths to this and other features of a stratification system may be a source of the disproportionate rates of black school failure and unemployment. In the face of bleak future prospects, diligence in school may not appear to be adaptive to social reality, but rather may be seen as "doing the white man's thing" (Ogbu 1985a; Anderson, in this volume; and Foster, 1974~. The castelike stratification of minorities has effects beyond those of youths' perceptions. The historic exclusion of minorities from some occupations deprives them of the chance to learn the requirements of such employment and to undertake the necessary preparation. Minority children will be limited in their opportunity to observe role models pursuing such occupations, and parents, having been excluded by past discrimination, will often be unable to guide and advise their children on the preparations required for such occupations. This may result in a dearth of knowledge on the part of the child and entirely inappro- priate preparation for desired "mainstream" occupations. In one study (Ogbu, 1985a), it was reported that black high school students desiring to become doctors, engineers, and teachers were as likely to take shop courses as those desiring office work. Similarly, minority youths who aspired to be engineers took no more mathematics courses in high school than youths wishing to become physical education teachers. What such findings make clear is not only that children did not learn about the requirements of those occupations in their home environment, but also that the schools did little, if anything, to convey crucial information. Anderson (in this volume) emphasizes the increasing significance of class factors in determining the social context in which black and other minority inner-city youths are raised. The substantial increase in the size of the black middle and upper classes in recent years has resulted in greater residential dispersion of higher-income blacks within the metropolitan area: black inner-city communities have experienced a loss of leadership and important role models that has contributed to the problems faced by the remaining youths. DEVELOPMENTS S INCE 19 8 0 As the Youth Employment and Demonstration Projects Act (YEDPA) programs ended in 1981, the U.S. economy had begun its descent into the worst recession since the 1930s. The economy bottomed out at the end of 1982 with overall unemployment at a post-World War II high of 10.8 percent. The unemployment rate of youths aged 16-19 was more than double that at 24.5 percent. The greater sensitivity of youth employment to the business cycle noted previously can be seen for this period as well in the data on employment-to-population rates given in Table 2.10. In 1978 the

OCR for page 34
66 TABLE 2.10 Employment-to-Population Rates for Total Civilian Population and for All Civilian Youths Aged 16-19, 1977-1984 Employment-to-Population Rate All Civilian All Youths Workers Aged 16-19 Ratio Year (1) (2) (1~/~2) 1977 1978 1982 1984 57.9 59.3 57.8 59.5 46.1 48.3 41.5 43.7 .796 .814 .718 .734 SOURCE: U.S. Department of Labor (1985:Table B-12~. workers (column 3). . , _ employment-to-population rates for all civilian workers (column 1) was 59.3, while the employment-to-population rate for youths aged 16-19 was 48.3 (column 2~; hence, the youth rate was 81.4 percent of that for all ' ~ ~' By 1982 the employment-to-population rate for all wormers nag fallen sharply, to 57.8, but the rate for youths had fallen even more precipitously, to 41.5, so that the youth rate was only 71.8 percent of the rate for all workers. It is also of some interest to note that in 1982 the employment-to-population rate for all workers was at about the same level as in 1977 (57.8 and 57.9), but the youth rate was considerably lower in 1982 than in 1977 (41.5 compared with 46.1~. For black youths the employment situation in 1982 was disastrous: their employment-to-population rate was only 19.0 percent. The economic recovery began in 1983 and continued through 1984. On the upswing youth employment again showed greater sensitivity so that by 1984 the youth employment-to-population rate had recovered more than that for all workers: it was 73.4 percent of the rate for all workers (compared with 71.8 percent at the bottom of the recession in 1982~. However, if one compares the situation in 1984 with that in 1978, it is clear that, despite the recovery, the youth employment-to-population situation has deteriorated both absolutely--from 48.3 in 1978 to 43.7 in 1984--and relative to the rest of the labor force--from 81.4 percent of the rate for all workers in 1978 to 73.4 percent in 1984. If one looks back to Tables 2.2a and 2.2b, it is apparent that the employment-to-population rate for nonwhite males (both those aged 16-19 and 20-24) is not only worse in 1984 than it was in 1978 but has further deteriorated relative to white male youths, while the employment-to- population rates of nonwhite females remain the lowest of the youth groups. These very summary data indicate both that, as would be expected, the recession hurt youth employment seriously and also that even with the economic recovery youth employment problems remain very serious.

OCR for page 34
67 Compared either with 1977, just before YEDPA started, or 1978, the first year of the program, the youth employment problem in 1984 was as bad, or worse. Even more disturbing, the employment situation of black youths, particularly males, has worsened even more relative to white youths, apparently continuing the long-term trend observed up to 1978. While our committee has not tried to assess systematically the economic outlook for the future and its implications for youth employ- ment problems, we do wish to comment on one feature that has sometimes been pointed to as a possibly important sensitive development, namely, the decline in the absolute size of the youth cohort. In the previous section, it was noted that one of the possible causes of youth employ- ment problems was the massive, unprecedented rise in the size of the youth cohort, both absolutely and relative to the adult worker popula- tion (shown in Figure 2.1~. Over the 15 years from 1980, when the size of the youth population reached its absolute peak, to 1995, the youth cohort will decline from 21 million to about 17 million. It has been suggested that this decline will significantly improve the employment situation for youths. We have two observations to make about this suggestion. First, while there are some indications that the youth demographic bulge may have contributed to youth employment problems, the evidence is by no means overwhelming. If it is hard to find the effects of this dramatic bulge in relative supply on youth employment problems, it seems unwise to count on the decline in relative supply of youths to have over- whelming effects in reducing youth employment problems over the next decade. Second, by 1985 two-thirds of the total projected decline in the size of the youth population will have occurred. The figures just reviewed above give no indication that this relative supply effect is currently having a substantial impact on youth employment problems. If reductions in relative supply of youths have been having some positive effect, they have not been substantial enough to overcome other negative factors. POSTSCRIPT When focusing on the youth unemployment problem, there is a tendency to lose sight of the fact that the majority of teenagers find jobs relatively easily and that, when they leave or lose one job, they often find another without a long period of unemployment. As Freeman and Wise (1982) observe: Constant references to the youth employment problem, as if all or the majority of young persons had trouble obtaining jobs, appear to misinterpret the nature of the difficulty. Youth joblessness is in fact concentrated among a small group who lack work for extended periods of time. n The vexing problem about the "youth unemployment problem" is that for some groups of youths--disproportionately black youths--finding any job, remaining employed, and finding a new job when necessary is a major and continuing difficulty. Throughout this chapter, it has become apparent that blacks suffer inordinately from unemployment. But while table after table has shown a widening gap between white and

OCR for page 34
68 black unemployment, inactivity, etc., it is not only the black population among whom unemployment is concentrated. In 1978 Hispanics experienced long-term unemployment at 1.3 times the rate of the population as whole, children from poverty families at 1.6 times the national rate, and those living in inner cities at 1.4 times the national rate (Congressional Budget Office, 1982~. It was against this background that Congress enacted YEDPA in 1977. This act instructed the Secretary of Labor "to establish a variety of employment and training programs to explore the methods of dealing with the structural unemployment problems of the Nation's youth. n In the following chapters we review the implementation and effects of these programs. In Chapter 3 we describe the YEDPA programs and their implementation. In Chapters 4 through 9 (and related appendixes), we review the effectiveness of those programs and the scientific adequacy of the research which was conducted to evaluate them.