Accounting for the non-employment of U.S. men, 1968-2010.

Kudlyak, Marianna ; Lubik, Thomas ; Tompkins, Jonathan 等

Men in their prime working age, defined as men between the ages of 25 and 64, constitute 33 percent of the civilian non-institutionalized population in the United States. At the trough of the 1969-1970 recession, 6.5 percent of this group (henceforth, "population") were out of the labor force (OLE), 90.8 percent were employed, and 2.7 percent were unemployed. Since then, the employment-to-population ratio has trended persistently downward, while the OLF-to-population ratio has increased substantially. (1) In 2010, the aftermath of the 2007-2009 recession, the employment-to-population ratio of this same group declined to an all-time low of 76.3 percent, while the OLF-to-population ratio increased to an all-time high of 14.7 percent (see Figure 1, Panels A C).

In this article, we investigate the extent to which the change in the sociodemographic composition of the population (by age, educational attainment, marital status, and race) has contributed to the changes in the aggregate labor market outcomes. Our emphasis on the compositional changes in the sociodemographic characteristics of the population is motivated by a literature rife with correlations between sociodemographic factors and labor market outcomes. In particular, older workers typically experience lower rates of labor force participation and, conditional on participating, older workers are less likely to be unemployed than younger workers (see, for example, Shinier 1999). The literature also finds that (i) more highly educated workers have a higher opportunity cost of not working; (ii) married men are more likely to participate in the labor force and, conditional on participation, more likely to be employed; and (iii) non-white persons are usually underrepresented in the labor force and employment. Thus, one expects a strong association between labor market outcomes and the demographic composition of the labor force, which serves as a reduced-form representation of underlying structural relationships.

[FIGURE 1 OMITTED]

In this article, we decompose the observed changes in aggregate labor market outcomes into changes in the sociodemographic composition of the population and changes in the labor market outcomes of different sociodemographic groups. For each yea, we generate two sets of counterfactual aggregate labor market outcomes. The first set is generated by using the sociodemographic composition of the population from all the years in the sample and holding the labor market outcomes of different sociodemographic groups constant at the actual level of the reference year. The second set is generated by holding the sociodemographic composition constant instead. We then use these counterfactuals to perform the decomposition of the changes in the aggregate labor market outcomes. Finally, we use the most recent sociodemographic composition of the population to forecast the aggregate OLF-to-population ratio in 2015.

Given the similarities between the 1980-1982 and 2007-2009 recessions in terms of severity, we emphasize, throughout this article, comparisons between the labor market outcomes in 1983 and 2010. We find that the changes in the demographic composition of the population explain much of the historical upward trend in the OLF ratio. The OLF ratio increased from 11.1 percent in 1983 to 14.7 percent in 2010. Of this increase, 1.9 percentage points (49 percent of the total change) are attributable to changes in the sociodemographic composition of the population. The employment-to-population ratio fell from 80.2 percent in 1983 to 76.3 percent in 2010. We find that changes in the sociodemographic composition of the population account for 1.7 percentage points (44 percent) of this decline. The unemployment-to-population ratio increased from 8.7 percent in 1983 to 8.9 percent in 2010, but none of this increase can be accounted for by changes in the demographic composition of the population. Finally, using predicted changes in the age distribution of the population, we estimate that the OLF-to-population ratio will increase to more than 16 percent in 2015 as compared to 14.7 percent in 2010.

When interpreting our results we need to be wary that changes in the sociodemographic composition might cause changes in the labor market outcomes of different sociodemographic groups. Alternatively, changes in the labor market outcomes of some sociodemographic groups might cause changes in the sociodemographic composition of the population. For example, an increase in the employment probability for higher educated workers relative to other education levels might contribute to an increase in the educational attainment of the population. Alternatively, an increase in educational attainment of the population can change the employment probabilities of different groups. Our accounting exercise does not account for these effects.

This article is related to the existing literature that documents a secular decline in the labor force participation of prime working age men. Autor and Duggan (2003) document a substantial fall in labor force participation among men. Using data from the Current Population Survey (CPS), Juhn, Murphy, and Topel (1991, 2002) find that falling unemployment rates among men in the 1990s greatly exaggerated the improvements of the labor outcomes for this population because the period was also characterized by a fall in the labor force participation rate. We update their analyses by focusing on the decomposition of the changes in the labor outcomes into changes in the sociodemographic characteristics of the population and by adding data from the most recent decade. Our work is also closely related to Little and Bradley (2007), who use a multinomial logistic model to study the sociodemographic determinants of labor market outcomes, distinguishing employment, unemployment, peripheral inactivity (marginally attached to labor force), and OLF, in Great Britain. Finally, our work expands on that of Fallick and Pingle (2006), who decompose movements in U.S. labor force participation into aging of the population and labor force trends within age groups. Whereas these authors focus solely on changes in the OLF-to-population ratio caused by aging of the population, we consider changes in each employment status caused by changes in four different sociodemographic factors.

The article is structured as follows. Section 1 describes the data. Section 2 summarizes the changes in the demographic composition of the population of working age men between 1968-2010 and documents trends in labor market outcomes by sociodemographic groups. Section 3 describes the decomposition exercise and presents results of the decomposition of changes in labor outcomes between 2010 and the earlier years. Section 4 presents the forecast of the 2015 OLF-to-population ratio. Section 5 concludes.

1. DATA

We use data from the Integrated Public Use Microdata Series CPS (IPUMS-CPS), which comprises data from the March Supplement of the Current Population Survey (hereafter referred to as the March CPS). The CPS is a monthly survey of U.S. households' activities, conducted by the Census Bureau and the Bureau of Labor Statistics and designed to measure unemployment. The basic survey is conducted every month; over time various supplementary surveys have been conducted to study different social and economic questions. The March CPS contains in-depth information on sociodemographic characteristics of the population and income. The variables in IPUMS-CPS are coded identically or "harmonized" over the years.

The CPS is a collection of individual-level data obtained from the interviewed households. We focus on males between the ages of 25 and 64.2 Throughout the analysis we use the March CPS sampling weights that account for a complex survey design. The aggregate annual statistics that we report thus correspond to March of a respective year.

It should be noted that in 1994 the CPS underwent a major redesign both in the wording of its questions and in the methodology of the data collection process, which led to some discrepancies between the aggregate series constructed from the microdata prior to the redesign and post-redesign. However, it is not a concern of our analysis because the inconsistencies associated with the aggregate labor statistics for the sample of 25-64-year-old men are minor and not statistically significant (see Polivka and Miller [1995]).

2. SOCIODEMOGRAPHIC COMPOSITION OF THE POPULATION AND LABOR OUTCOMES BY DIFFERENT GROUPS

Between 1968-2010 there have been considerable changes in the distribution of 25-64-year-old civilian, non-institutionalized men by age and education, and some noticeable changes by marital status and race (Figure 2). Figures 3-5 display the time series of labor outcomes by different sociodemographic groups. In general, across different groups the employment-to-population ratio has been trending down, while the OLF-to-population has been gradually increasing. We now describe each figure in detail.

Sociodemographic Composition

Panel A of Figure 2 shows the changes in the shares of 25-34, 35-44, 45-54, and 55-64-year-old men in the population. From 1968-1986, the share of 25-34-year-old men grew steadily, reaching its largest fraction of 35.6 percent in 1986, and declined thereafter. The share of 55-64-year-old men fell from 1968-7995, reaching its smallest fraction of 15 percent in 1995, and has increased steadily since. From 2000 to present, the shares of older workers (45-54 and 55-64-year-olds) have been increasing, while the shares of younger workers (25-34 and 35 44-year-olds) have been decreasing. This shift in the age distribution toward older workers is largely because of the aging of the Baby Boom generation. In 2010, the shares of 25-34, 35-44, 45-54, and 55-64-year-olds were 25.8, 25.2, 27.5, and 21.5, respectively.

[FIGURE 2 OMITTED]

Panel B of Figure 2 shows the upward trend in the educational attainment of the population and reveals that the shares of men with some college, college, or higher than college education have been increasing at the expense of men with at most a high school degree. The share of the latter has declined from 74.6 percent in 1969 to 44.0 percent in 2010. Panel C shows that the population distribution by marital status has shifted toward divorced or separated and single men at the expense of married men. In 1968, 84.8 percent of the 25-64-year-old men were married, while only 61.1 percent were married in 2010. Finally, Panel D shows that the share of white men in the population has been steadily falling over the last 40 years while the shares of black men and men of other races have been increasing.

[FIGURE 3 OMITTED]

Employment-to-Population Ratio

Employment-to-population ratios by age are shown in Figure 3, Panel A, The figure shows that the proportion of employed workers for the 25-34,35-44, and 45-54 age groups move in sync over the years, declining from roughly .95 in 1968 to approximately .80 in 2010. The 55-64 age group has displayed markedly different behavior, declining from approximately .82 in 1968 to an all-time low of .60 in 1994 before trending back up to .65 in 2010.

Figure 3, Panel B displays the employment-to-population ratio by educational attainment. For each group there is a clear decline in the employment-to-population ratio over time, with the severity of this decline decreasing in years of schooling. From 1968-2010, those with more than a college education and those with just a college degree have seen moderate declines of .06 and .1 1, respectively, while those with some college experience and those with at most a high school degree have experienced larger declines of .18 and .21, respectively.

The employment-to-population ratio trends by marital status are displayed in Figure 3, Panel C. Though there have been decreases in the employment-to-population ratio across marital status groups from 1968-2010, we observe that much of these declines came prior to 1980 and after 2007. Between these two dates, the employment-to-population ratio across marital status groups shows little or no trend.

Panel D of Figure 3 shows the employment-to-population ratio for each race. We observe declines in the employment-to-population ratio for whites, blacks, and others over the sample period, with this decline being most pronounced for blacks. The employment-to-population ratio fell by approximately .24 for this group from 1968-2010, whereas the employment-to-population ratio for whites and others declined by .14 and .10, respectively.

Unemployment-to-Population Ratio

The unemployment-to-population ratio shows a clear cyclical pattern, rising during economic contractions and falling during expansions. As can be seen from Panel A of Figure 4, the unemployment-to-population ratio is decreasing with age. The rise in the unemployment-to-population ratio across age groups from 2007-2010 is comparable to the increase from 1980-1983, though the increase for the 25-34 age group was more pronounced in the 1980-1983 recession, and the increases for the 35-44 and 45-54 age groups are more pronounced in the 2007-2009 recession.

The unemployment-to-population ratios for educational attainment groups are displayed in Figure 4, Panel B, which shows that more years of schooling are associated with lower unemployment. Figure 4, Panel C shows the unemployment-to-population ratio by marital status and shows that single, never married and separated/divorced individuals have consistently higher unemployment-to-population ratios than those who are married. Finally, as can be seen from Panel D of Figure 4, the unemployment-to-population ratio for blacks is consistently higher than the unemployment-to-population ratio for whites and others. Between 2007 and 2010, the unemployment-to-population ratio for blacks increased by .08, whereas the ratios of whites and others increased by approximately .04.

OLF-to-Population Ratio

[FIGURE 4 OMITTED]

Figure 5, Panel A displays the OLF-to-population ratio for each age group. We note that each time series has a distinct upward trend from 1968-2010, with the upward trend for the 55-64 age group from 1968-1985 being particularly severe. However, the OLF-to-population ratio for the 35-44 age group has been relatively stable since 1994 and that of the 55-64 age group has actually declined moderately since 1994. The most notable feature of this figure is the disparity between the 55-64 age group and the other age groups. Historically, the OLF-to-population ratio of the 55-64 age group has dwarfed that of the younger age groups. In 2010, the OLF-to-population ratio for the 55-64 age group was .30, whereas the ratios of the other age groups ranged from .09 to .13. We also observe that the OLF-to-population ratio of the 45-54 group tends to be higher than the 25-34 and 35-44 age groups by approximately .02.

[FIGURE 5 OMITTED]

The OLF-to-population ratios for different educational attainment groups are shown in Panel B of Figure 5. There are upward trends since 1968 for each group, with the largest increases occurring for those with at most a high school education and those with some college education. The figure shows that those with a college degree and those with more than a college education have experienced similar OLF-to-population ratios across time, though since 2005 these series have diverged, with the OLF-to-population ratio for those with a college degree continuing to trend upward while the ratio for those with more than a college education has fallen. Finally, we observe that fewer years of education are associated with a larger OLF-to-population ratio: In 2010, the OLF-to-population ratios were .19, .15, .09, and .07 for those with at most a high school degree, some college education, a college degree, and more than a college degree, respectively.

Panel C of Figure 5 gives the OLF-to-population ratio by marital status. There is a notable upward trend in the OLF-to-population ratio for each marital status group. Those in the married group have the lowest OLF-to-population ratio throughout the observation period, while the separated/divorced and single, never married groups have an OLF-to-population ratio that is between .05 and .1 higher. The widowed group has had the highest OLF-to-population ratio historically, occasionally exceeding .40.

Figure 5, Panel D breaks down the OLF-to-population ratio by race. The most notable feature of this figure is the large difference in the growth of the OLF-to-population ratio between blacks and the other groups. The OLF-to-population ratio for blacks has increased from .09 in 1968 to .24 in 2010, whereas the OLF-to-population ratios for the other and white groups have increased from .09 and .06 to .15 and .14, respectively. Thus, while each series has trended upward, that of blacks has done so more rapidly.

3. ACCOUNTING FOR CHANGES IN AGGREGATE LABOR OUTCOMES

Method

In this subsection we discuss the methods by which we create counterfactual labor outcomes and decompose the changes in actual labor market outcomes.

The aggregate share of persons with labor outcome L O, where LO = {Employed, Unemployed, OLF}, in year t can be described by the following equation:

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (1)

where i [member of] A x E x M x R corresponds to a vector of demographic charac-teristics consisting of age (A), educational attainment (E), marital status (M), and race (R); [g.sub.i,t] is the number of persons in group i; L[O.sub.i,t] is the number of persons with the labor status LO in group i; and pop, is the size of the population.

In essence, we divide the population into mutually exclusive groups (e.g., married, college-educated white males between the ages of 25-34). Equation (1) describes the aggregate proportion as the sum of the labor outcomes by group (the first term in the equation) weighted by the size of the groups in the population (the second term in the equation). For example, fixing either term for all i at its 2010 level while allowing the other to take on historical values allows us to construct counterfactual aggregate labor outcomes for 2010. By creating and comparing time series of these counterfactuals, we can observe the degree to which these two terms are driving changes in aggregate labor outcomes in 2010.

One concern when creating these counterfactuals is that the changing sociodemographic composition of the population could affect the labor outcomes of different groups. Alternatively, changes in the labor outcomes of a sociodemographic group could change the sociodemographic composition of the population. For example, if college-educated individuals are more likely to be employed relative to other educational attainment subgroups, there will likely be an influx of individuals into the college-educated demographic group. By simply fixing either of these terms while varying the other, we do not account for these endogeneity effects.

To analyze the change in labor outcomes, we perform the following decomposition:

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (2)

where [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII].

The component in the first set of brackets in equation (2) measures the effect of changes in the labor outcomes of different groups from year [t.sub.1] to [t.sub.2], given the sociodemographic composition of the population in year [t.sub.1]. The second term captures the effect of changes in the sociodemographic composition of the population, given the labor outcomes of different groups in year [t.sub.2].

Alternatively, we can write

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (3)

The component in the first set of brackets in equation (3) measures the effect of changes in the sociodemographic composition of the population, given the labor outcomes of different groups in year [t.sub.1]. The second term captures the effect of changes in the labor outcomes of different groups from year [t.sub.1] to [t.sub.2], given the sociodemographic composition of the population in year [t.sub.2]. The difference between these two decompositions is the base year, i.e., the year at which the component, other than the component of interest, is held constant. For example, the change as a result of a change in the labor outcomes of different groups in equation (2) is calculated using [t.sub.1] as the base year, while in equation (3) it is calculated using [t.sub.2] as the base year. Because of the endogeneity issues mentioned above, these two decompositions do not necessarily deliver the same results. It is also unclear that one is theoretically better than the other. We perform both decompositions and, despite some small quantitative differences, find the qualitative conclusions from the two decompositions to be the same. For this reason, we report only the results of the decomposition corresponding to equation (2). (3)

Changes in Aggregate Labor Outcomes

Tables 1-3 show the predicted labor outcomes--proportion employed, unemployed, and OLF, respectively--calculated using the labor outcomes of different groups from year [t.sub.1] and the demographic composition of the population from year [t.sub.2], where [t.sub.1],[t.sub.2][member of][1968, 2010].

The diagonal elements in Tables 1-3 (darkly shaded) show the actual labor outcomes for their respective year. The off-diagonal elements show the counterfactual labor outcomes. Thus, for each year we have two sets of counterfactual predictions. Moving along a row gives the predicted labor outcome for a fixed demographic composition, while moving down a column gives the predicted labor outcome for fixed labor outcomes of different groups. For example, the (1983, 2010) entry of Table 1 gives the predicted proportion of employed individuals in 2010 given the 1983 demographic composition (77.9 percent), while the (2010, 1983) entry gives the predicted proportion of employed individuals in 2010 given the 1983 labor outcomes of different groups (78.6 percent).

Table 1 Counterfactual Predictions of the Employment-to-Population
Ratio by Percent, 25-64-Year-Old Males

Dema      1968  1969   1970   1971  1972  1973  1974  1975  1976  1977
graphics
1968       919   917   90.7   89.0  88.7  88.7  88.5  84.5  84.3  85.1
1969       918   919   90.8   89.0  88.8   889  88.6  84.6  84.3  85.2
1970       918   918  90.81   89.0  88.8  88.7  88.6  88.6  84.9  85.2
1971       918   917   90.7   98.1  88.7  88.5  88.5  84.5  84.8  85.2
1972       918   917   90.7   89.0  88.8  88.6  88.5  84.5  84.9  88.2
1973       918   917   90.7   89.1  88.8  88.7  88.6  84.7  84.9  85.4
1974       918   915   90.5   88.9  88.8  88.5  88.8  84.6  84.5  85.2
1975       916   916   90.5   88.7  88.5  88.5  88.4  84.6  84.8  85.2
1976       917   917   90.6   88.9  88.6  88.6  88.5  84.7  84.9  85.3
1977       917   916   90.5   88.8  88.5  88.5  88.5  84.7  84.9  85.4
1978       915   915   90.4   88.8  88.4  88.4  88.3  84.5  84.7  85.2
1979       916   915   90.4   88.7  88.4  88.3  88.4  84.6  84.8  85.5
1980       915   914   90.3   88.6  88.3  88.2  88.3  84.5  84.7  85.5
1981       914   913   90.2   88.5  88.2  88.3  88.2  84.4  84.6  85.2
1982       915   914   90.3   88.5  88.2  88.2  88.2  84.5  84.7  85.2
1983       916   915   90.2   88.6  88.3  88.2  88.3  84.7  84.6  85.3
1984       915   914   90.3   88.6  88.3  88.3  88.2  84.5  84.6  85.3
1985       914   913   90.1   88.5  88.2  88.3  88.1  84.5  84.5  85.3
1986       914   913   90.1   88.4  88.1  88.2  88.1  84.8  84.7  85.2
1987       916   914   90.1   88.4  88.1  88.3  88.2  84.5  84.8  85.3
1988       914   9l3   90.9   88.3  88.0  88.5  88.2  84.5  84.8  85.3
1989       913   913   90.9   88.3  88.0  88.1  88,1  84.5  84.7  85.2
1990       914   913   90.9   88.3  88.0  88.2  88.2  84.5  84.7  85.3
1991       913   912   89.9   88.2  87.9  88.0  88.0  84.5  84.8  85.1
1992       912   912   89.9   88.2  87.9  88.0  88.0  84.6  84.8  85.2
1993       912   912   89.9   88.2  87.9  88.0  88.0  84.7  84.7  85.2
1994       912   912   89.9   88.1  87.9  88.0  88.0  84.8  84.8  85.3
1995       912   913   90.0   88.2  88.0  88.0  88.0  85.0  85.0  85.5
1996        91   911   89.8   88.0  87.9  87.0  87.0  84.7  84.8  85.3
1997       909    91   89.7   87.8  87.9  87.9  87.8  84.6  84.7  85.1
1998       909   911   69.7   87.8  87.7  87.9  87.6  84.6  84.6  85.2
1999       908    91   89.7   87.9  87.7  87.9  87.7  84.7  84.6  85.2
2000       909   911   89.6   87.9  87.8  87.8  87.6  84.5  84.6  85.2
2001       906   909   89.5   87.9  87.6  87.8  87.6  84.5  84.4  85.0
2002       905   903   89.3   87.9  87.5  87.9  87.4  84.5  84.0  84.8
2003       903   903   89.0   87.4  87.2  87.4  87.2  84.2  83.8  84.7
2004       903   905   88.9   87.2  87.1  87.3  87.0  84.0  83.8  84.5
2005        90   903   88.6   88.6  88.9  87.1  86.8  87.0  83.5  84.2
2006      89.9   903   88.5   88.9  88.7  88.9  88.6  86.8  83.3  84.0
2007      69.9   903   88.4   88.9  88.3  88.8  86.7  86.6  83.4  84.0
2008      69.6  89.9   88.2   86.6  88.4  86.3  86.3  86.7  83.0  83.7
2009       693  89.7   88.0   86.3  86.1  88.4  86.1  86.3  82.8  83.6
2010      99.1  89.4   87.9   86.1  85.1  88.2  85.8  86.1  83.8  83.3
           919   919   90.0   89.1  88.8  88.7  88.8  84.8  84.9  85.4

Dema       1978  1979   1980   1981  1982  1983  1984  1985  1986  1987
graphics

1968       85.9  86.4   85.2   83.8   817  79.5   819  82.4  82.0  82.5
1969       85.9  88.4   85.3   83.9   818  79.6  82.0  82.5  82.1  82.6
1970       85.9  86.4   85.3   83.9   818  79.6  82.0  82.5  82.1  82.7
1971       85.9  86.4   85.3   83.9   818  79.6  82.0  82.5  82.1  82.7
1972       85.9  86.4   85.3   84.0   818  79.7  82.0  82.5  82.2  82.7
1973       86.1  86.5   85.4   84.1  82.0  78.8  82.2  82.8  82.4  82.9
1974       86.0  86.4   85.4   84.0   819  79.7  82.1  82.7  82.3  82.8
1975       86.0  86.5   85.4   84.1   818  79.8  82.2  82.7  82.4  82.8
1976       86.1  85.5   85.5   84.3  82.1  80.0  82.4  82.9  82.6  83.1
1977       86.2  86.5   85.5   84.3  82.1  80.0  82.4  83.0  82.6  83.1
1978       86.1  86.5   85.4   84.2  82.1  80.0  82.3  82.9  82.5  83.0
1979       86.2  86.6   85.5   84.3  82.1  80.0  82.4  82.9  82.6  83.1
1980       86.1  86.6   85.5   84.3  82.1  80.0  82.4  83.0  82.6  83.1
1981       86.1  86.5   85.5   84.2  82.0  79.9  82.3  82.9  82.6  83.1
1982       86.1  86.6   85.5   84.3  82.2  80.0  82.5  83.0  82.7  83.2
1983       86.2  86.7   85.6   84.4  82.3  80.1  82.6  83.2  82.9  83.3
1984       86.2  86.7   85.6   84.3  82.2  80.1  82.6  83.1  82.8  83.3
1985       86.2  86.7   85.5   84.3   822  80.1  82.6  83.1  82.8  83.3
1986       86.1  86.7   85.5   84.3  82.2  80.1  82.6  83.1  82.9  83.3
1987       86.2  86.8   85.7   84.4  82.4  80.2  82.7  83.2  83.0  83.5
1988       86.2  86.8   85.6   84.4  82.3  80.1  82.6  83.2  82.9  83.4
1989       88.2  88.8   85.7   84.4  82.4  80.2  82.7  83.3  83.0  83.4
1990       88.3  88.8   85.7   84.4  82.4  80.2  82.8  83.3  83.0  83.5
1991       86.3  86.9   85.8   84.2  82.2  80.0  82.8  83.1  83.1  8.33
1992       86.3  88.7   85.7   84.4  82.4  80.2  82.8  83.2  83.2  83.5
1993       85.9  86.9   85.9   84.4  82.5  80.3  83.0  83.3  83.0  83.5
1994       85.4  85.9   85.9   84.5  82.5  80.3  83.1  83.3  83.0  83.5
1995       85.6  87.0   86.0   84.6  82.7  80.5  83.1  83.5  83.2  83.7
1996       85.4  87.2   85.8   84.5  82.6   804  82.9  83.3  83.1  83.5
1997       86.3  85.0   85.7   84.4  82.5  80.3  82.5  83.2  83.0  83.4
1998       86.3  85.8   85.7   84.4  82.5  80.3  82.0  83.2  83.0  83.3
1999       86.3  86.9   85.8   84.4  82.6  80.3  83.0  83.2  82.9  83.4
2000       86.5  87.0   85.9   84.5  82.8  80.5  83.1  83.2  83.0  83.5
2001       86.3  86.7   85.7   84.4  82.5  80.4  82.9  83.1  82.9  83.3
2002       86.1  86.6   85.5   84.1  82.3  80.2  82.5  82.9  82.7  83.1
2003       85.9  86.4   85.4   84.0  82.2  80.0  82.4  82.6  82.5  82.8
2004       85.7  88.2   85.3   83.8  82.0  79.7  82.1  82.4  82.2  82.6
2005       85.5  88.0   85.0   83.5   817  79.5   819  82.2   819  82.3
2006       85.3  85.5   84.8   83.2   816  79.3   816   819   817  82.2
2007       85.3  85.5   84.8   83.3   816  79.3   816   819   816  82.1
2008       85.9  85.5   84.6   83.0   813  79.1   813   816   813   818
2009       84.9  85.4   84.4   82.8   812  76.9   811   815   812   818
2010       84.6  85.1   84.1   82.5  80.9  78.6  80.8   812  80.9   814
           86.1  86.6   85.5   84.2  82.2  80.2  82.6  83.1  82.9  83.5

Dema      1988  1989  1990  1991  1992  1993   1994   1995  1996  1997
graphics

1968      83.3  83.7  83.3   816  80.5  80.5   79.9    812   817  82.4
1969      83.4  83.7  83.3   818  80.6  80.6   80.1    813   818  82.5
1970      83.4  83.8  83.3   818  80.6  80.6   80.1    814   818  82.5
1971      83.4  83.8  83.4   818  80.7  80.6   80.1    814   819  82.6
1972      83.5  83.8  83.4   819  80.7  80.7   80.2    815   819  82.6
1973      83.7  84.0  83.6  82.0  80.9  80.9   80.4    817  82.1  82.8
1974      83.6  83.9  83.5   819  80.8  80.8   80.4    818  82.0  82.8
1975      83.6  83.9  83.5  82.0  80.9  80.9   80.4    818  82.0  82.8
1976      83.8  84.1  83.7  82.2   811   811   80.6    815  82.2  83.0
1977      83.8  84.1  83.8  82.2   812   812   80.7    818  82.4  83.1
1978      83.7  84.0  83.7  82.1   811   811   80.7    819  82.3  83.0
1979      83.7  84.1  83.8  82.3   813   812   80.8    819  82.4  83.1
1980      83.8  84.1  83.9  82.3   813   812   80.9   82.0  82.4  83.2
1981      83.7  84.2  83.9  82.3   813   812   80.8   82.1  82.4  83.2
1982      83.8  84.2  84.0  82.3   814   814    810   82.1  82.5  83.3
1983      84.0  84.4  84.2  82.5   816   815    811   82.4  82.7  83.5
1984      84.0  84.3  84.2  82.5   816   815    811   82.4  82.7  83.4
1985      83.9  84.3  84.2  82.5   816   815    811   82.4  82.7  83.4
1986      83.9  84.4  84.2  82.5   816   815    812   82.5  82.7  83.4
1987      84.1  84.5  84.4  82.7   818   817    814   82.6  82.9  83.5
1988      84.0  84.4  84.4  82.6   817   815    813   82.5  82.8  83.5
1989      84.1  84.8  84.4  82.7   819   816    814   82.7  82.8  83.5
1990      84.1  84.6  84.5  82.8  82.0   817    815   82.7  82.9  83.6
1991      33.8  84.3  84.3  82.9  82.1   815    813   82.5  82.6  83.4
1992      84.0  84.6  84.5  82.9  82.3   819    815   82.7  82.8  83.6
1993      84.0  84.6  84.6  83.0  82.1   B16    815   82.7  82.3  83.6
1994      84.1  84.7  84.8  83.2  82.1   817    816   82.8  82.8  83.7
1995      84.2  84.8  84.6  83.1   025   819    818   83.0  83.0  83.6
1996      84.0  84.7  84.5  83.0  82.1   617    818  82.71  82.8  83.5
1997      83.9  84.5  84.4  82.9  82.0   919    815   82.6  82.7  83.5
1998      83.8  84.5  84.4  82.8   819   819    814   82.5  82.8  83.6
1999      83.8  84.5  84.5  82.5   819   B14    814   82.5  82.8  83.4
2000      83.0  84.5  84.5  83.0  82.1   816    815   82.6  82.6  83.1
2001      83.7  84.4  84.3  82.9   819   813    813   82.4  82.2  83.4
2002      83.5  84.1  84.0  82.8   817    80    811   82.1  82.0  83.1
2003      83.3  83.9  83.9  82.5   816   609   80.8   82.0  82.0  83.0
2004      83.0  83.7  83.7  82.2   813   607   80.7    818   817  82.8
2005      82.7  63.4  83.3  82.0   811   604   80.4    814   814  82.4
2006      82.5  83.1  83.1   818  80.9   602   80.1    812   812  82.2
2007      82.5  83.1  83.1   817  80.8   602   80.1    812   812  82.1
2008      82.2  82.8  82.8   814  80.5   799   79.8   80.9  80.8   818
2009      82.0  82.6  82.6   813  80.4  79.7   79.7   80.7  80.7   817
2010       817  82.4  82.3   810  80.1   794   79.3   80.4  80.3   814
          84.0  84.6  54.5  82.6   819   816    816   83.0  82.3  83.6

Dema      1998  1999  2000  2001  2002   2003  2004  2005  2006  2007
graphics

1968      83.0  83.4  83.3  82.8   812    811   813   810  82.7  82.3
1969      83.1  83.4  83.4  82.9   813    812   814  82.1  82.8  82.5
1970      83.1  83.5  83.4  82.9   814    812   814  82.1  82.8  82.6
1971      83.2  83.5  83.5  82.9   814    812   814  82.1  82.8  82.6
1972      83.2  83.5  83.5  83.0   815    813   815  82.1  82.9  8207
1973      83.4  83.6  83.7  83.2   8l6    814   817  82.3  83.0  82.9
1974      83.3  83.6   836  83.1   816    813   816  82.2  82.9  82.8
1975      83.4  83.7  83.7  83.2   817    814   818  82.2  83.0  82.9
1976      83.6  83.9  83.9  83.3   819    816   818  82.4  83.2  83.1
1977      83.7  83.9  84.0  83.4   819    816   817  82.5  83.2  83.1
1978      83.6  83.9  83.9  83.3   819    815   818  82.4  83.1  83.1
1979      83.7  84.0  84.0  83.5  82.0    816   818  82.5  83.2  83.2
1980      83.6  84.0  84.0  83.5  82.0    817   818  82.5  83.3  83.2
1981      83.6  84.0  84.1  83.6  82.0    817   815  82.6  83.3  83.3
1982      83.9  84.2  84.3  83.7  82.2    818  82.0  82.8  83.4  83.4
1983      84.1  84.3  84.4  83.8  82.3    819  82.0  82.8  83.5  83.6
1984      84.1  84.3  84.4  83.9  82.3    819  82.0  82.7  83.5  83.6
1985      84.1  84.3  84.4  83.8  82.3    819  82.0  82.7  83.5  83.6
1986      84.1  84.3  84.4  83.8  82.3   82.1   829  82.8   835  83.6
1987      84.3  84.5  84.7  84.1  82.5   82.0  82.2  82.9  83.7  83.5
1988      84.2  84.4  84.6  84.0  82.6   82.1  82.1  82.8  83.6  83.6
1989      84.3  84.5  84.6  84.1  82.4   82.1  82.1  82.9  83.7  83.7
1990      84.4  84.5  84.7  84.2  82.6   82.3  82.2  83.0  83.7  83.8
1991      84.2  84.3  84.5  84.0  82.4   82.1   819  82.8  83.5  83.6
1992      84.3  84.5  84.7  84.2  82.6    819  82.1  82.9  83.6  83.8
1993      84.3  84.5  84.7  84.2  82.6   82.1  82.1  82.9  83.6   839
1994      84.4  84.6  81.8  84.3  82.7   82.1  82.1  83.0  83.7  83.9
1995      84.5  84.6  84.9  84.5  82.5   82.1  82.3  83.2  83.9  84.1
1996      84.3  84.6  84.7  84.3  82.7   82.1  82.1  83.3  83.7  83.9
1997      84.3  84.4  84.8  84.2  82.6    819   819  82.9  83.6  83.8
1998      84.2  84.4  84.5  84.1  82.5    819   819  82.8  83.4  83.7
1999      84.1  84.4  84.5  84.1  82.5    819   819  82.8  83.2  83.7
2000      84.2  84.4  84.6  84.1  82.6    819  82.0  82.9  83.0  83.8
2001      84.0  84.2  84.3  84.0  82.4    818   818  82.8  83.8  83.8
2002      83.8  84.0  84.1  83.7  82.2    815   816  82.5  82.5  83.4
2003      83.6  83.8  83.9  83.5  82.0    814   815  82.4  82.3  83.2
2004      83.4  83.6  83.7  83.3   818    812   812  82.2  82.8  82.3
2005      83.1  83.4  83.5  83.1   816   80.9   810   819  82.5  82.0
2006      82.9  83.1  83.2  82.8   813   80.7  80.8   817  82.3  82.8
2007      82.8  83.1  83.2  82.7   813   80.6  80.8   817  82.3  82.8
2008      82.6  8.28  82.9  82.5   816   80.3  80.5   815  82.0  82.3
2009      82.4  82.7  82.7  82.3  80.9   80.2  80.3   813   819  82.1
2010      82.1  82.4  82.4  82.1  80.5   79.9  80.0   810   815   818
          84.2  84.5  84.6  84.0  82.2    814   812   819  82.3  8.25

Dema      2008  2009  2010
graphics

1968       816   772  76.9
1969       818   773  77.9
1970       818   774  77.1
1971       818   774  77.1
1972       819   774  77.2
1973      82.0   776  77.3
1974      82.0   778  77.3
1975      82.0   778  77.4
1976      82.2   776  77.8
1977      82.3   779  77.7
1978      82.2   773  77.8
1979      82.3   789  77.9
1980      82.4   785  77.9
1981      82.3   779  77.9
1982      82.4   789  77.9
1983      82.6   782  77.9
1984      82.6   782   773
1985      82.6   782  77.9
1986      82.6   782  78.0
1987      82.6   783  78.0
1988      82.6   782  77.8
1989      82.6   783  77.9
1990      82.8   783  78.0
1991      82.8  78.1  78.1
1992      82.8   783  78.3
1993      82.9  78.4   769
1994      83.0  78.5  78.1
1995      83.2  76.7  78.3
1996      83.0  78.5  78.1
1997      82.9  78.4   769
1998      82.9  78.4  78.0
1999      82.9  78.5  78.1
2000      83.0  78.8  78.2
2001      82.5  78.4  78.1
2002      82.6  78.2  77.9
2003      82.5  78.1  77.7
2004      82.5  78.0  77.6
2005      82.1  77.7  77.3
2006       819  77.4  77.1
2007       818  77.4  77.1
2008       816  77.2  76.8
2009       814  77.0  76.6
2010       811  76.7  76.3
           816  77.0  76.3

Notes: Authors' own calculations from the IPUMS-CPS data. The columns
correspond to the year of the labor outcomes used. The rows correspond
to the year of the sociodemographic composition used. Lightly shaded
columns correspond to the NBER-dated contractions (from peak to trough).
Darkly shaded elements correspond to actual values for given year.

Table 2 Counterfactual Predictions of the Unemployment-to-Population
Ratio by Percent, 25-64-Year-Old Males

Oemooiaciliica  1968  1969  1970  1971  1972  1973  1974  1975  1976

1968             2.1    17   2.7   3.8   3.5   3.0   2.9   6.0   4.9
1969             2.0    17   2.7   3.8   3.5   3.0   2.9   6,0   4.8
1970             2.0    17   2.7   3.8   3.5   3.0   2.9   6.0   4.9
1971             2.0    17   2.7   3.8   3.5   3.0   2.9   6.0   4.9
1972             2.0    17   2.7   3.9   3.5   3.0   2.9   6.0   4.9
1973             2.0    17   2.7   3.8   3.5   3.0   2.9   5.9   4.9
1974             2.0    17   2.7   3.0   3.6   3.0   2.9   5.9   4.9
1975             2.0    17   2,7   3.0   3.6   3.0   2.9   5.9   4.9
1976             2.0    17   2.7   3.0   3.6   3.1   2.9   5.9   4.9
1977             2.0    17   2.7   3,0   3.6   3.1   3.0   5.9   4.9
1978             2.0    17   2.7   4.0   3.6   3.1   3.0   5.9   4.9
1979             2.0    17   2.7   4.0   3.6   3.1   3.0   5.9   5.0
1980             2.0    17   2.7   4.0   3.6   3.1   3.0   5.9   5.0
1981             2.1    17   2.8   4.1   3.7   3.2   3.1   6.0   5.1
1982             2.1    17   2.8   4.1   3.7   3.2   3.1   6.1   5.1
1983             2.1    17   2.8   4.1   3.7   3.2   3.2   6.0   5.1
1984             2.1    17   2.8   4.1   3.8   3.2   3.1   6.1   5.1
1985             2.1    17   2.8   4,1   3.8   3.2   3.2   6.1   5.2
1986             2.1    17   2.8   4.2   3.8   3.3   3.2   6.1   5.2
1987             2.1    18   2.8   4.2   3.9   3.3   3.3   6.1   5.2
1988             2.2    18   3.0   4.2   3.9   3.3   3.3   6.2   5.3
1989             2.2    17   3.0   4.2   3.8   3.3   3.3   6.1   5.2
1990             2.2    17   3.0   4.3   3.9   3.3   3.3   6.1   5.2
1991             2.3    18   3.0   4.3   3.9   3.3   3.3   6.2   5.3
1992             2.3    17   3.0   4.3   3.9   3.3   3.3   6.1   5.3
1993             2.2    17   3.0    43   3.9   3.3   3.3   6.0   5.3
1994             2.3    17   3.0    43   3.9   3.3   3.3   6.0   5.2
1995             2.2    17   3.0   4.3   3.9   3.2   3.3   5.9   5.1
1996              23    17   3.0   4.3  3..9   3.3   3.3   5.9   5.2
1997             2.3    17   3.0   4.3   3.9   3.2   3.3   5.8   5.1
1998             2.3    17   3.0   4.3   3.0   3.2   3.3   5.8   5.1
1999             2.3    17   3.0   4.2   3.9   3.2   3.3   5.7   5.0
2000             2.2    17   3.0   4.1   3.9   3.1   3.3   5.7   5.0
2001             2.2    17   3.0   4.2   3.9   3.1   3.2   5.5   5.0
2002             2.2    17   3.0   4.3   3.9   3.1   3.3   5.6   5.0
2003             2.2    17   3.0   4.3   3.9   3.1   3.3   5.5   5.0
2004             2.2    17   3.0   4.3   3.9   3.1   3.3   5.6   5.0
2005              22    17   3.0   4.3   4.0   3.1   3.3   5.6   5.0
2006             2.2    17   3.0   4.4   3.9   3.1   3.3   5.6   5.0
2007             2.2    16   3.0   4.4   4.0   3.1   3.3   5.6   4.5
2008             2.2    17   3.1   4.4   4.0   3.1   3.3   5.5   5.0
2009             2.2    17   3.1   4.4   4.0   3.1   3.4   5.5   5.0
2010             2.2    17   3.1   4.4   4.0   3.2   3.4   5.6   5.1

                 2.1    17   2.7   3.8   3.5   3.0   2.9   5.9   4.9

Oemooiaciliica  1977  1978  1979  1980  1981  1982  1983  1984  1985

1968             4.6   3.7   3.5   4.3   5.2   6.9   8.8   6.1   5.3
1969             4.6   3.7   3.4   4.2   5.2   6.8   8.7   6.1   5.3
1970             4.6   3.7   3.4   4.3   5.1   6.8   8.7   5.0   5.3
1971             4.7   3.7   3.5   4.2   5.2   6.8   8.7   5.0   5.3
1972             4.6   3.7   3.4   4.2   5.1   6.8   8.7   5.0   5.2
1973             4.6   3.7   3.4   4.2   5.1   6.8   8.7   5.9   5.2
1974             4.7   3.7   3.4   4.2   5.1   6.8   8.6   5.9   5.2
1975             4.7   3.7   3.4   4.2   5.0   6.8   8.6   5.9   5.2
1976            4..7   3.7   3.4   4.2   5.0   6.7   8.5   5.9   5.2
1977             4.7   3.7   3.4   4.2   5.0   6.7   8.5   5.9   5.2
1978             4.7   3.7   3.4   4.2   5.0   5.7   8.5   5.9   5.2
1979             4.6   3.7   3.4   4.2   5.0   6.8   8.6   5.9   5.2
1980             4.9   3.7   3.5   4.3   5.1   6.8   8.7   5.9   5.2
1981             5.0   3.8   3.6   4.4   5.2   7.0   8.8   6.0   5.3
1982             5.0   3.8   3.6   4.4   5.2   7.1   8.7   6.1   5.3
1983             5.0   3.8   3.6   4.4   5.2   7.0   8.8   6.1   5.3
1984             5.1   3.9   3.6   4.4   5.3   7.1   8.8   6.1   5.4
1985             5.1   3.9   3.6   4.4   5.3   7.1   8.9   6.1   5.4
1986             5.2   4.0   3.6   4.5   5.3   7.2   8.9   6.2   5.5
1987             5.2   4.0   3.6   4.5   5.3   7.2   8.0   6.2   5.5
1988             5.3   4.0   3.7   4.5   5.4   7.3   8.9   6.2   5.8
1989             5.3   4.0   3.7   4.5   5.4   7.3   8.9   6.2   5.5
1990             5.3   4.0   3.7   4.5   5.4   7.3   8.0   6.2   5.5
1991             5.3   4.0   3.7   4.5   5.4   7.4   8.9   6.3   5.6
1992             5.2   4.0   3.6   4.5   5.3   7.3   8.8   6.2   5.6
1993             5.2   3.9   3.6   4.4   5.2   7.2   8.8   5.1   5.5
1994             5.2   3.9   3.6   4.4   5.3   7.2   8.7   5.1   5.6
1995             5.1   3.9   3.6   4.4   5.2   7.2   8.7   5.1   5.5
1996             5.2   3.9   3.6   4.4   5.2   7.2   8.7   5.1   5.5
1997             5.1   3.9   3.5   4.4   5.2   7.2   8.7   6.1   5.5
1998             5.1   3.8   3.5   4.3   5.1   7.1   8.6   6.0   5.4
1999             5.0   3.8   3.5   4.2   5.0   7.0   8.5   5.9   5.4
2000             4.9   3.7   3.5   4.2   4.9   6.9   8.4   5.9   5.4
2001             4.9   3.7   3.5   4.2   4.9   6.8   8.4   5.8   5.3
2002             4.9   3.7   3.5   4.2   4.9   6.8   8.3   5.8   5.3
2003             4.9   3.8   3.5   4.1   4.0   6.9   8.3   5.8   5.4
2004             4.9   3.8   3.5   4.1   4.0   6.8   8.3   5.9   5.3
2005             5.0   3.8   3.5   4.1   4.9   6.8   8.4   5.9   5.4
2006             5.0   3.8   3.5   4.1   4.9   6.8   8.4   5.9   5.4
2007             5.0   3.8   3.5   4.1   4.8   6.8   8.3   5.9   5.4
2008             5.0   3.8   3.6   4.1   4.8   6.8   8.3   5.8   5.4
2009             5.0   3.8   3.6   4.1   4.8   6.8   8.3   5.9   5.4
2010             5.1   3.9   3.6   4.1   4.8   6.8   8.3   5.9   5.4

                 4.7   3.7   3.4   4.3   5.2   7.1   8.7   5.1   5.4

                                   Labor  Outcomos
Oemooiaciliica  1986  1987  1988    1989      1990  1991  1992  1993

1968             5.5   5.2   4.5     3.9       4.0   5.7   6.5   6.2
1969             5.5   5.1   4.4     3.9       4.0   5.6   6.5   6.2
1970             5.5   5.1   4.4     3.9       4.0   5.6   6.5   6.2
1971             5.5   5.1   4.4     3.0       4.0   5.7   6.5   6.2
1972             5.4   5.1   4.4     3.9       4.0   5.6   6.4   6.1
1973             5.4   5.1   4.4     3.9       4.0   5.6   6.4   6.1
1974             5.4   5.0   4.4     3.9       4.0   5.6   6.4   6.1
1975             5.3   5.0   4.3     3.8       3.9   5.6   6.4   6.1
1976             5.3   5.0   4.3     3.8       3.9   5.5   6.3   6.1
1977             5.3   5.0   4.3     3.8       3.9   5.5   6.3   6.0
1978             5.3   5.0   4.3     3.9       3.9   5.5   6.3   6.0
1979             5.3   5.0   4.3     3.9       3.9   5.5   6.3   6.1
1980             5.3   5.1   4.3     3.9       4.0   5.6   5.4   6.1
1981             5.4   5.1   4.4     4,0       4.1   5.7   5.4   6.2
1982             5.4   5.2   4.4     4.0       4.1   5.7   5.4   6.2
1983             5.4   5.2   4.4     4.0       4.1   5.7   5.4   6.2
1984             5.5   5.2   4.5     4.1       4.1   5.8   6.5   6.3
1985             5.5   5.2   4.5     4.1       4.1   5.6   6.5   6.3
1986             5.5   5.3   4.6     4.1       4.2   5.8   6.5   6.3
1987             5.5  5.3|   4.6     4.1       4.2   5.9   6.6   6.4
1988             5.6   5.4   4.6     4.2       4.2   5.9   8.6   6.4
1989             5.6   5.4   4.6     4.2       4.2   5.9   6.6   6.4
1990             5.6   5.4   4.6     4.2       4.2   6.0   6.6   6.5
1991             5.7   5.5   4.7     4.3        43   5.9   6.7   6.5
1992             5.6   5.4   4.7     4.2       4.3   5.9   6.7   6.5
1993             5.5   5.4   4.6     4.2       4.2   5.9   6.6   6.5
1994             5.6   5.5   4.6     4.2       4.2   5.9   6.6   6.5
1995             5.5   5.4   4.6     4.2       4.2   5.9   6.5   6.5
1996             5.5   5.4   4.6     4.2       4.2   5.9   6.6   6.5
1997             5.5   5.4   4.6     4.2       4.2   5.8   6.5   6.5
1998             5.5   5.4   4.6     4.1       4.2   5.8   6.6   6.5
1999             5.4   5.3   4.5     4.1       4.1   5.8   6.5   6.4
2000             5.4   5.3   4.5     4.1       4.1   5.7   6.4   6.4
2001             5.4   5.3   4.5     4.0       4.1   5.7   6.4   6.3
2002             5.4   5.3   4.5     4.0       4.0   5.7   6.4   6.3
2003             5.3   5.3   4.5     4.0       4.1   5.7   6.4   6.3
2004             5.3    53   4.5     4.0       4.1   5.7   6.4   6.3
2005             5.4    53   4.6     4.1       4.1   5.7   6.5   6.4
2006             5.4   5.3   4.6     4.1       4.1   5.7   6.5   6.4
2007             5.4   5.3   4.6     4.1       4.0   5.7   6.5   6.4
2008             5.3    53   4.5     4.1       4.0   5.7   6.5   6.3
2009             5.4   5.3   4.6     4.0       4.1   5.7   6.5   6.4
2010             5.4   5.3   4.6     4.0       4.1   5.7   6.5   6.4
                                  Actual
                 5.5   5.3    46     4.2       4.2   6.0   6.7   6.5

Oemooiaciliica  1994  1995  1996  1997  1998  1999  2000  2001  2002

1968             5.1   4.2   4.5   3.8   3.5   3.0   2.8   3.3   4.8
1969             5.1   4.2   4.4   3.8   3.5   3.0   2.8   3.3   4.8
1970             5.1   4.1   4.4   3.8   3.5   3.0   2.7   3.3   4.8
1971             5.1   4.2   4.4   3.8   3.5   3.0   2.7   3.3   4.8
1972             5.1   4.1   4,4   3.8   3.5   3.0   2.7   3.3   4.8
1973             5.0   4,1   4.4   3.8   3.5   3.0   2.7   3.2   4.8
1974             5.0   4.1   4.4   3.8   3.5   3.0   2.7   3.3   4.8
1975             5.0   4.1   4.4   3.8   3.4   2.9   2.7   3.2   4.7
1976             5.0   4.1   4.4   3.8   3.4   2.9   2.7   3.2   4.7
1977             5.0   4.1   4.3   3.8   3.4   2.9   2.7   3.2   4.7
1978             5.0   4.1   4.4   3.8   3.4   2.9   2.8   3.2   4.7
1979             5.0   4.1   4.4   3.8   3.4   2.9   2.8   3.2   4.7
1980             5.0   4.1   4.4   3.8   3.4   2.9   2.8   3.3   4.7
1981             5.1   4.2   4.5   3.8   3.5   3.0   2.9   3.3   4.8
1982             5.1   4.2   4.5   3.9   3.5   3.0   2.9   3.3   4.9
1983             5.1   4.2   4.5   3.9   3.5   3.0   2.9   3.3   4.9
1984             5.2   4.3   4.5   4.0   3.5   3.0   2.9   3.3   4.9
1985             5.2   4.3   4.6   4.0   3.5   3.0   2.9   3.4   4.9
1986             5.2   4.3   4.6   4.1   3.6   3.1   3.0   3.4   5.0
1987             5.3   4.3   4.6   4.1   3.6   3.1   3.0   3.4   5.0
1988             5.3   4.4   4.6   4.1   3.6   3.1   3.0   3.4   5.0
1989             5.3   4.4   4.7   4.1   3.7   3.1   3.0   3.4   5.0
1990             5.3   4.4   4.7   4.2   3.7   3.1   3.0   3.4   5.0
1991             5.4   4.3   4.7   4.2   3.7   3.2   3.1   3.4   5.0
1992             5.3   4.4   4.7   4.2   3.7   3.1   3.0   3.4   5.0
1993             5.3   4.4   4.6   4.1   3.8   3.1   3.0   3.4   4.9
1994             5.4   4.4   4.7   4.2   3.7   3.2   3.0   3.4   4.9
1995             5.3   4.4   4.6   4.1   3,7   3.1   3.0   3.4   4.9
1996             5.3   4.5   4.7   4.1   3.7   3.1   3.0   3.4   4.9
1997             5.3   4.4   4.6   4.1  _3.6   3.1   3.0   3.4   4.9
1998             5.3   4.4   4.6   4.1   3.6   3.1   3.0   3.4   4.9
1999             5.3   4.4   4.6   4.1   3.6   3.1   3.0   3.4   4.9
2000             5.2   4.4   4.5   4.0   3.5   3.1   2.9   3.3   4.9
2001             5.2   4.3   4.5   4.0   3.5   3.1   2.9   3.3   4.8
2002             5.2   4.4   4.5   4.0   3.5   3.1   2.9   3.3   4.8
2003             5.2   4.4   4.5   4.0   3.5   3.1   2.9   3.3   4.7
2004             5.2   4.4   4.5   4.0   3.5   3.1   3.0   3.3   4.8
2005             5.2   4.4   4.5   4.1   3.6   3.1   3.0   3.4   4.8
2006             5.2   4.4   4.5   4.1   3.6   3.1   3.0   3.4   4.8
2007              52   4.4   4.5   4.0   3.6   3.1   3.0   3.3   4.7
2008             5.2   4.4   4.5   4.0   3.6   3.1   3.0   3.4   4.7
2009             5.2   4.4   4.5   4.9   3.6   3.1   3.0   3.4   4.7
2010             5.2   4.4   4.5   4.9   3.6   3.1   3.0   3.4   4.8

                 5.4   4.4   4.5   4.1   3.6   3.1   2.9   3.3   4.8

Oemooiaciliica  2003  2004  2005  2006  2007  2008  2009  2010

1968             4.7   4.3   3.9   3.3   3.5   4.3   8.2   9.1
1969             4.7   4.3   3.9   3.3   3.5   4.3   8.2   9.0
1970             4.7   4.3   3.9   3.3   3.5   4.3   8.1   9.0
1971             4.7   4.3   3.9   3.3   3.5   4.3   8.1   9.0
1972             4.7   4.3   3.9   3.3   3.5   4.2   8.1   9.0
1973             4.6   4.2   3.9   3.2   3.5   4.2   8.1   8.9
1974             4.6   4.3   3.9   3.3   3.5   4.2   8.1   8.9
1975             4.6   4.2   3.9   3.3   3.5   4.2   8,1   8.8
1976             4.6   4.3   3.8   3.3   3.5   4.1   8.0   8.8
1977             4.6   4.3   3.8   3.3   3.5   4,1   8.0   8.7
1978             4.6   4.3   3.8   3.3   3.5   4.1   8.0   8.7
1979             4.7   4.3   3.8   3.3   3.5   4.1   8.0   8.7
1980             4.7   4.4    39   3.3   3.5   4.2   8.0   8.8
1981             4.8   4.4   4.0   3.4   3.6   4.3   8.2   8.9
1982             4.8   4.5   4.0   3.4   3.6   4.3   8.2   8.9
1983             4.8   4.5   4,0   3.4   3.6    43   8.1   8.9
1984             4.8   4.5   4.0   3.5   3.6   4.3   8.2   9.1
1985             4.8   4.6   4.0   3.5   3.6   4.3   8.2   9.1
1986             4.9   4.6   4.1   3.5   3.7   4.4   8.3   9.1
1987             4.9   4.6   4.1   3.5   3.7   4.4   8.3   9.1
1988             4.9   4,7   4.1   3.5   3.7   4.4   8.4   0.1
1989             4.9   4.7   4.1   3.6   3.7   4.4   8.4   9.1
1990             5.0   4.7   4.1   3.6   3.7   4.4   8.4   9.2
1991             5.0   4.8   4.1   3.7   3.8   4.5   8.5   9.3
1992             5.0   4.8   4.1   3.7   3.8   4.4   8.4   9.2
1993             5.0   4.8   4.1   3.7   3.7   4.3   8.3   9.1
1994             5.1   4.8   4.1   3.7   3.7   4.4   8.4   9.2
1995             5.1   4.8   4,1   3.7   3.7   4.3   8.3   9.1
1996             5.1   4.8   4,1   3.7   3.7   4.3   8.3   9.1
1997             5.1   4.8   4.1   3.7   3.7   4.3   8.3   9.1
1998             5.1   4.7   4.1   3.7   3.7   4.3   8.3   9.1
1999             5.0   4.7   4.1   3.7   3.7   4.2   8.2   9.0
2000             5.0   4.7   4.0   3.8   3.7   4.2   8.1   8.9
2001             5.0   4.6   4.0   3.6    35   4.2   8.1   8.9
2002             5.0   4.7   4.0   3.7   3.7   4.2   0.1   8.9
2003             5.0   4.7   4.0   3.7   3.7   4.2   8.1   8.9
2004             5.0   4.7   4.0   3.7   3.7   4.2   8.1   8.9
2005             5.0    47   4.0   3.7   3.7   4.2   8.1   8.9
2006             5.0   4.7   4.0   3.7   3.7   4.2   8.1   8.9
2007             5.0   4.7   4.0   3.7   3.7   4.2    81   8.9
2008             5.0   4.7   4.0   3.7   3.7   4.2   8.1   8.9
2009             5.0   4.7   4.0   3.7   3.7   4.2   8.1   8.9
2010             5.0   4.7   4.1   3.7   3.8   4.2   8.1   8.9

                 5.0   4.7   4.0   3.7   3.7    42   8.1   8.9

Notes: Authors' own calculations froms the IPUMS-CPS data. The
columns correspond to the year of the labor outcomes used. The
rows correspond to the tear of the sociodemographic composition
used. Lightly shaded columns correspond to the NBER-dated
contractions (from peak to trough). Darkly shaded elements
correspond to actual values for given year.

Table 3 Counterfactual Predictions of the OLF-to-Population Ratio
by Percent, 25-64-Year-Old Males

Demo      1968  1968  1970  1971  1972  1973  1974  1975  1976  1977
graphics
1968       2.1    17   2.7   3.8   3.5   3.0   2.9   6.0   4.9   4.6
1969       2.0    17   2.7   3.8   3.5   3.0   2.9   6.0   4.8   4.6
1970       2.0    17   2.7   3.8   3.5   3.0   2.9   6.0   4.9   4.6
1971       2.0    17   2.7   3.8   3.5   3.0   2.9   6.0   4.9   4.7
1972       2.0    17   2.7   3.9   3.5   3.0   2.9   6.0   4.9   4.6
1973       2.0    17   2.7   3.8   3.5   3.0   2.9   5.9   4.9   4.6
1974       2.0    17   2.7   3.9   3.6   3.0   2.9   5.9   4.9   4.7
1975       2.0    17   2.7   3.9   3.6   3.0   2.9   5.9   4.9   4.7
1976       2.0    17   2.7   3.9   3.6   3.1   2.9   5.9   4.9   4.7
1977       2.0    17   2.7   3.9   3.6   3.1   3.0   5.9   4.9   4.7
1978       2.0    17   2.7   4.0   3.6   3.1   3.0   5.9   4.9   4.7
1979       2.0    17   2.7   4.0   3.6   3.1   3.0   5.9   5.0   4.8
1980       2.0    17   2.7   4.0   3.6   3.1   3.0   5.9   5.0   4.9
1981       2.1    17   2.9   4.1   3.7   3.2   3.1   6.0   5.1   5.0
1982       2.1    17   2.9   4.1   3.7   3.2   3.1   6.1   5.1   5.0
1983       2,1    17   2.9   4.1   3.7   3.2   3.2   6.0   5.1   5.0
1984       2.1    17   2.9   4.1   3.8   3.2   3.1   6.1   5.1   5.1
1985       2.1    17   2.9   4.1   3.8   3.2   3.2   6.1   5.2   5.1
1986       2.1    17   2.9   4.2   3.8   3.3   3.2   6.1   5.2   5.2
1987       2,1    18   2.9   4.2   3.9   3.3   3.2   6.1   5.2   5.2
1988       2.2    18   3.0   4.2   3.9   3.3   3.3   6.2   5.2   5.3
1989       2.2    17   3.0   4.2   3.8   3.3   3.2   6.1   5.2   5.3
1990       2.2    17   3.0   4.3   3.9   3.3   3.3   6.1   5.2   5.3
1991       2.3    18   3.0   4.1   3.9   3.3   3.3   6.2   5.3   5.3
1992       2.3    17   3.0   4.1   3.8   3.3   3.2   6.1   5.3   5.3
1993       2.2    17   3.0   4.3   3.9   3.3   3.3   6.0   5.3   5.2
1994       2.3    17   3.0   4.3   3.8   3.3   3.3   6.0   5.3   5.2
1995       2.3    17   3.0   4.3   3.9   3.2   3.3   5.9   5.2   5.1
1996       2.3    17   3.0   4.3   3.9   3.3   3.3   5.9   5.2   5.2
1997       2.3    17    3d   4.3   3.9   3.2   3.3   5.8   5.1   5.1
1998       2.3    17    3d   4.3   3.9   3.2   3.3   5.8   5.1   5.1
1999       2.3    17   3.0   4.2   3.9   3.2   3.3   5.7   5.0   5.0
2000       2.2    17    3d   4,1   3.9   3.1   3.3   5.7   5.0   4.9
2001       2.2    17   3.0   4.2   3.9   3.1   3.2   5.6   5.0   4.9
2002       2.2    17   3.0   4.3   3.9   3.1   3.3   5.6   5.0   4.9
2003       2.2    17   3.0   4.3   3.9   3.1   3.3   5.6   5.0   4.9
2004       2.2    17   3.0   4.3   3.9   3.1   3.3   5.6   5.0   4.9
2005       2.2    17   3.0   4.3   4.0   3.1   3.3   5.6   5.0   5.0
2006       2.2    17   3.0   4.4   3.9   3.1   3.3   5.6   5.0   5.0
2007       2.2    16   3.0   4.4   4.0   3.1   3.3   5.6   4.9   5.0
2008       2.2    17   3.1   4.4   4.0   3.1   3.3   5.5   5.0   5.0
2009       2.2    17   3.1   4.4   4.0   3.1   3.4   5.5   5.0   5.0
2010       2.2    17   3.1   4.4   4.0   3.2   3.4   5.6   5.1   5.1

           2.1    17   2.7   3.8   3.5   3.0   2.9   5.9   4.9   4.7

Demo       1978  1979  1980  1981  1982  1983  1984  1985  1986  1987
graphics
1968        3.7   3.5  4.36   5.2   8.9   8.8   6.1   5.3   5.5   5.2
1969        3.7   3.4   4.2   5.2   8.8   8.7   6.1   5.3   5.5   5.1
1970        3.7   3.4   4.3   5.1   8.8   8.7   6.0   5.3   5.5   5.1
1971        3.7   3.5   4.3   5.2   8.8  8.77   6.0   5.3   5.5   5.1
1972        3.7   3.4   4.2   5.1   8.8   8.7   6.0   5.2   5.4   5.1
1973        3.7   3.4   4.2   5.1   8.8  8.77   5.9   5.2   5.4   5.1
1974        3.7   3.4   4.2   5.1   8.8   8.8   5.9   5.2   5.4   5.0
1975        3.7   3.4   4.2   5.0   8.8   8.8   5.9   5.2   5.3   5.0
1976        3.7   3.4   4.2   5.0   8.7   8.8   5.9   5.2   5.3   5.0
1977        3.7   3.4   4.2   5.0   8.7   8.5   5.9   5.2   5.3   5.0
1978        3.7   3.4  4.25   5.0   8.7   8.5   5.9   5.2  5.36   5.0
1979        3.7   3.4   4.2   5.0   5.8   8.5   5.9   5.2   5.3   5.0
1980        3.7   3.5   4.3   5.1   5.8   8.8   5.9  5.22   5.3   5.1
1981        3.8   3.5   4.4   5.2   7.0   8.7   5.9   5.3  5.45   5.1
1982        3.8   3.6   4.4   5.2   7.1   8.8   6.0   5.3   5.4   5.2
1983        3.8   3.6   4.4   5.2   7.0   8.7   6.1   5.3   5.4   5.2
1984        3.9   3.6   4.4   5.3   7.1   8.8   6.1   5.4   5.5   5.2
1985        3.9   3.6   4.4   5.3   7.1   8.8   6.2   5.4   5.5   5.2
1986        4.0   3.6   4.5   5.3   7.2   8.9   6.2   5.5   5.5   5.3
1987        3.9   3.8   4.5   5.3   7.2   8.9   6.2   5.5   5.5  5.3.
1988        4.0   3.7   4.5   5.4   7.3   9.0   6.2   5.6   5.6   5.4
1989        4.0   3.7   4.5   5.4   7.3   8.9   6.2   5.5   5.6   5.4
1990        4.0  3.74   4.5   5.4   7.3   8.9   6.3   5.5   5.7   5.4
1991        4.0   3.7   4.5   5.4   7.4   9.0   6.2   5.6   5.6   5.5
1992        4.0   3.6   4.5   5.3   7.3   8.9   6.1   5.6   5.5   5.4
1993        3.9   3.6   4.4   5.2   7.2   8.8   6.1   5.5   5.6   5.4
1994        3.9   3.6   4.4   5.3   7.2   8.8   6.1   5.6   5.5   5.5
1995        3.9   3.6   4.4   5.2   7.2   8.8   6.1   5.5   5.5   5.4
1996        3.9   3.6   4.4   5.2   7.2   8.7   6.1   5.5   5.5   5.4
1997        3.9   3.6   4.4   5.2   7.2   8.7   6.0   5.5   5.4   5.4
1998        3.8   3.6   4.3   5.1   7.1   8.7   5.9   5.4   5.4   5.4
1999        3.8   3.5   4.2   5.0   7.0   8.8   5.9   5.4   5.4   5.3
2000        3.7   3.5   4.2   4.9   8.9   8.4   5.8   5.4   5.4   5.3
2001        3.7   3.5   4.2   4.9   8.8   8.4   5.8   5.3   5.3   5.3
2002        3.7   3.5   4.2   4.9   8.8   8.3   5.8   5.3   5.3   5.3
2003        3.8   3.5   4.1   4.9   8.8   8.3   5.8   5.4   5.4   5.3
2004        3.8   3.5   4.1   4.9   8.8   8.3   5.9   5.3   5.3   5.3
2005        3.8   3.5   4.1   4.9   8.9   8.4   5.9   5.4   5.4   5.3
2006        3.8   3.5   4.1   4.9   8.8   8.4   5.9   5.4   5.4   5.3
2007         39   3.5   4.1   4.8   8.8   8.3   5.9   5.4   5.4   5.3
2008         32   3.6   4.1   4.8   8.8   8.3   5.8   5.4   5.3   5.3
2009         33   3.6   4.1   4.8   8.8   8.3   5.9   5.4   5.4   53.
2010         39   3.6   4.1   4.8   8.8   8.3   5.9   5.4   5.4   5.3

            3.7   3.4   4.3   5.2   7.1   8.7   6.1   5.4   5.5   5.3

                 Labor  Outcomos
Demo      1988    1989      1990  1991  1992  1993  1994  1995  1996
graphics

1968       4.5     3.9       4.0   5.7   5.5   6.2   5.1   4.2   4.5
1969       4.4     3.9       4.0   5.6   5.5   6.2   5.1   4.2   4.4
1970       4.4     3.9       4.0   5.6   5.5   6.2   5.1   4.1   4.4
1971       4.4     3.9       4.0   5.7   5.5   6.2   5.1   4.2   4.4
1972       4.4     3.9       4.0   5.6   5.4   6.1   5.1   4.1   4.4
1973       4.4     3.9       4.0   5.6   5.4   6.1   5.0   4.1   4.4
1974       4.4     3.9       4.0   5.6   5.4   6.1   5.0   4.1   4.4
1975       4.3     3.9       3.9   5.6   5.4   6.1   5.0   4.1   4.4
1976       4.3     3.8       3.9   5.5   8.3   6.1   5.0   4.1   4.4
1977       4.3     3.8       3.9   5.5   8.3   6.0   5.0   4.1   4.3
1978       4.3     3.8       3.9   5.5   8.3   6.0   5.0   4.1   4.4
1979       4.3     3.9       3.9   5.6   8.3   6.1   5.0   4.1   4.4
1980       4.3     3.9       4.0   5.7   6.4   6.1   5.0   4.1   4.4
1981       4.4     4.3       4.1   5.7   6.4   5.2   5.1   4.2   4.5
1982       4.4     4.0       4.1   5.7   6.4  5.20   5.1   4.2   4.5
1983       4.4     4.0       4.1   5.8   6.4   5.2   5.1   4.2   4.5
1984       4.5     4.1       4.1   5.8  6.55   5.3   5.2   4.3   4.5
1985       4.5     4.1       4.1   5.8   6.5   5.3   5.2   4.3   4.6
1986       4.6     4.1       4.2   5.9  6.52   5.3   5.2   4.3   4.6
1987       4.6     4.1       4.2   5.9   6.6   5.4   5.3   4.3   4.6
1988       4.6     4.2       4.2   5.9   6.6   5.4   5.3   4.4   4.6
1989       4.6     4.2       4.2   5.9   6.6   5.4   5.3   4.4   4.7
1990       4.6     4.2       4.2   6.0   6.7   5.5   5.3   4.4   4.7
1991       4.7     4.3       4.3   5.9   6.7   5.5   5.4   4.5   4.7
1992       4.7     4.2       4.3   5.9   6.6   5.5   5.3   4.4   4.7
1993       4.6     4.2       4.2   5.9   6.8   6.5   5.3   4.4    46
1994       4.6     4.2       4.2   5.9   6.5   6.5   5.4   4.4    47
1995       4.6     4.2       4.2   5.9   6.6   6.5   5.3   4.4   4.6
1996       4.6     4.2       4.2   5.9   6.6   6.5   5.3   4.5   4.7
1997       4.6     4.2        42   5.8   6.6   6.5   5.3   4.4   4.4
1998       4.6     4.1       4.2   5.8   6.8   6.5   5.3   4.4   4.6
1999       4.6     4.1       4.1   5.8   6.5   6.5   5.3   4.4   4.6
2000       4.5     4.1       4.1   5.7   6.4   6.4   5.2   4.4   4.5
2001       4.5     4.0       4.1   5.7   6.4   6.4   5.2   4.3   4.5
2002      4.55     4.0       4.0   5.7   6.4   6.3   5.2   4.4   4.5
2003       4.5     4.0       4.1   5.7   6.4   6.3   5.2   4.4   4.5
2004       4.5     4.0       4.1   5.7   6.4   6.3  5.25   4.4   4.5
2005       4.5     4.1       4.1   5.7   6.5   6.4   5.2   4.4   4.5
2006       4.6     4.1       4.1   5.7   6.5   6.4   5.2   4.4   4.6
2007       4.6     4.1       4.0   5.7   6.5   6.4   5.2   4.4   4.6
2008       4.5     4.1       4.0   5.7   6.5   6.3   5.2    44   4.5
2009       4.6     4.0       4.1   5.7   6.5   6.4   5.2   4.4   4.6
2010       4.6     4.0       4.1   5.7   6.5   6.4   5.2   4.4   4.5
                Actual
           4.6     4.2       4.2   6.0   6.7   6.5   5.4   4.4   4.6

Demo      1997  1998  1999  2000   2001  2002  2003  2004  2005   2006
graphics

1968       3.8   3.5   3.0   2.8    3.3   4.8   4.7   4.3   3.9    3.3
1969       3.8   3.5   3.0   2.8    3.3   4.8   4.7   4.3   3.9    3.3
1970       3.8   3.5   3.0   2.7    3.3   4.8   4.7   4.3   3.9    3.3
1971       3.8   3.5   3.0   2.7    3.3   4.8   4.7   4.3   3.9    3.3
1972       3.8   3.5   3.0   2.7    3.3   4.8   4.7   4.3   3.9    3.3
1973       3.8   3.5   3.0   2.7    3.2   4.8   4.5   4.2   3.9    3.3
1974       3.8   3.5   3.0   2.7    3.3   4.8   4.6   4.3   3.9   3.23
1975       3.8   3.4   2.9   2.7    3.2   4.7   4.6   4.2   3.9    3.3
1976       3.8   3.4   2.9   2.7    3.2   4.7   4.6   4.3   3.8    3.3
1977       3.8   3.4   2.9   2.7    3.2   4.7   4.6   4.3   3.8    3.3
1978       3.8   3.4   2.9   2.8    3.2   4.7   4.6   4.3   3.8    3.3
1979       3.8   3.4   2.9   2.8    3.2   4.7   4.7   4.3   3.8    3.3
1980       3.8   3.4   2.9   2.8    3.3   4.7   4.7   4.4   3.9    3.3
1981       3.9   3.5   3.0   2.9    3.3   4.7   4.8   4.4   4.0    3.3
1982       3.9   3.5   3.0   2.9    3.3   4.8   4.8   4.5   4.0    3.4
1983       3.9   3.5   3.0   2.9    3.3   4.9   4.8   4.5   4.0    3.4
1984       4.0   3.5   3.0   2.9    3.3   4.9   4.8   4.5   4.0    3.4
1985       4.0   3.5   3.0   2.9    3.4   4.9   4.8    46   4.0    3.5
1986       4.1   3.6   3.1   3.0    3.4   4.9   4.8   4.6   4.1    3.5
1987       4,1   3.6   3.1   3.0    3.4   5.0   4.9   4.6   4.1    3.5
1988       4.1   3.6   3.1   3.0    3.4   5.0   4.9   4.7   4.1    3.5
1989       4.1   3.7   3.1   3.0    3.4   5.0   4.9   4.7  4.12    3.5
1990       4.2   3.7   3.1   3.0    3.4   5.0   4.9   4.7   4.1    3.5
1991       4.2   3.7    32   3.1    3.5   4.9   4.9   4.8   4.1    3.7
1992       4.2   3.7   3.1   3.0    3.4   4.9   5.0   4.8   4.1    3.7
1993       4.1   3.6   3,1   3.0    3.4   4.9   5.0   4.8   4.1    3.7
1994       4.2   3.7    32   3.0    3.4   4.9   5.1   4.8   4.1    3.7
1995       4.1   3.7   3.1   3.0    3.4   4.9   5.1   4.8   4.1    3.7
1996       4.1   3.7   3.1   3.0    3.4   4.9   5.1   4.8   4.1    3.7
1997       4.1   3.6   3.1   3.0    3.4   4.8   5.1   4.8   4.1    3.7
1998       4.1   3.6   3.1   3.0    3.4   4.8   5.1   4.7   4.1    3.7
1999       4.1   3.6   3.1   3.0    3.4   4.8   5.0   4.7   4.1    3.7
2000       4.0   3.5   3.1   2.9    3.3   4.8   5.0   4.7   4.0    3.5
2001       4.0   3.5   3.1   2.9    3.3   4.8   5.0    45   4.0    3.6
2002       4.0   3.5   3.1   2.9    3.3   4.8   5.0   4.7   4.0    3.7
2003       4.0   3.5   3.1   2.9    3.3   4.7    Sd   4.7   4.0    3.7
2004       4.0   3.6   3.1   3.0    3.3   4.8   5.0   4.6   4.0    3.7
2005       4.1   3.6   3.1   3.0    3.3   4.8   5.0   4.7   4.0    3.7
2006       4.1   3.6   3.1   3.0    3.4   4.8   5.0   4.7   4.0    3.7
2007       4.0   3.6   3.1   3.0    3.4   4.7   5.0   4.7   4.0    3.7
2008       4.0   3.6   3.1   3.0    3.3   4.7   5.0   4.7   4.0    3.7
2009       4.0   3.6   3.1   3.0    3.4   4.7   5.0   4.7   4.0    3.7
2010       4.0   3.6   3.1   3.0    3.4   4.8   5.0   4.7   4.1    3.7

           4.1   3.6   3.1   2.9    3.3   4.8   5.0   4.7   4.0    3.7

Demo      2007  2008  2009  2010
graphics

1968       3.5   4.3   8.2   9.1
1969       3.5   4.3   8.2   9.0
1970       3.5   4.3   8.1   9.0
1971       3.5   4.3   8.1   9.0
1972       3.5   4.2   8.1   9.0
1973       3.5   4.2   8.1   8.9
1974       3.5   4.2   8.1   8.9
1975       3.5   4.2   8.1   8.8
1976       3.5   4.1   8.0   8.8
1977       3.5   4.1   8.0   8.7
1978       3.5   4.1   8.0   8.7
1979       3.5   4.1   8.0   8.7
1980       3.5   4.2   8.0   8.8
1981       3.6   4.3   8.2   8.9
1982       3.6   4.3   8.2   8.9
1983       3.6   4.2   8.1   9.0
1984       3.6   4.3   8.2   9.0
1985       3.6   4.3   8.2   9.1
1986       3.7   4.4   8.3   9.1
1987       3.7   4.4   8.3   9.1
1988       3.7   4.4   8.4   9.1
1989       3.7   4.4   8.4   9.2
1990       3.7   4.4   8.4   9.3
1991       3.9   4.5   8.5   9.2
1992       3.8   4.4   8.4   9.1
1993       3.7   4.3   8.3   9.2
1994       3.7   4.4   8.3   9.1
1995       3.7   4.3   8.3   9.1
1996       3.7   4.3   8.3   9.1
1997        37   4.3   8.2   9.0
1998       3.7   4.3   8.1   8.9
1999       3.7   4.2   8.1   8.9
2000       3.7   4.2   8.1   8.9
2001       3.6   4.2   8.1   8.9
2002        37   4.2  8.12   8.9
2003       3.7   4.2   8.1   8.9
2004       3.7   4.2   8.1   8.9
2005       3.7   4.2   8.1   8.9
2006       3.7   4.2   8.1   8.9
2007       3.7   4.2   8.1   8.9
2008       3.7   4.2   8.1   8.9
2009       3.7   4.2   8.1   8.9
2010       3.8   4.2   8.1   8.9

           3.7   4.2   8.1   8.9

Notes: Authors' own calculations froms the IPUMS-CPS data. The
columns correspond to the year of the labor outcomes used. The
rows correspond to the tear of the sociodemographic composition
used. Lightly shaded columns correspond to the NBER-dated
contractions (from peak to trough). Darkly shaded elements
correspond to actual values for given year.

Table 4 Decomposition of the Change in theEmployment
-to-Population Ratio Between 2010 and Earlier
Years, 25-64-Year-Old Males

       Total  Change Because of   Change Because of
Year  Change  Labor Outcones      Socidemographics

              Percentage  % of    Percentage  % of
              Points      Total   Points      Total
                          Change              Change

1968  -15.52      -14.96   96.39       -0.56    3.61
1969  -15.52      -14.87   95.81       -0.65    4.19
1970  -14.45      -13.70   94.78       -0.76    5.22
1971  -12.75      -12.03   93.32       -0.72    5.68
1972  -12.46      -11.61   93.18       -0.85    6.82
1973  -12.40      -11.40   91.92       -1.00    8.08
1974  -12.22      -11.26   92.11       -0.96    7.89
1975   -8.26       -7.23   87.47       -1.04   12.53
1976   -8.16       -7.33   85.20       -1.27   14.80
1977   -9.08       -7.75   85.36       -1.33   14.64
1978    9.79       -8.54   87.21       -1.25   12.79
1979  -10.03       -8.90   86.41       -1.40   13.59
1980   -9.19       -7.77   84.59       -1.42   15.41
1981   -7.91       -6.57   83.10       -1.34   16.90
1982   -5.83       -4.37   74.85       -1.47   25.15
1983   -3.86       -2.25   58.29       -1.61   41.71
1984   -6.25       -4.65   74.44       -1.60   25.56
1985   -6.81       -5.25   77.03       -1.56   22.97
1986   -6.56       -5.03   76.73       -1.53   12.27
1987   -7.17       -5.50   16.59       -1.67   12.31
1988   -7.66       -6.07   70.23       -1.59   20.77
1989   -8.26       -6.59   79.68       -1.68   20.32
1990   -8.20       -6.49   79.15       -1.71   20.85
1991   -6.28       -4.80   76.45       -1.48   23.55
1992   -5.61       -4.01   71.44       -1.60   28.56
1993   -5.32       -3.61   67.81       -1.71   32.19
1994   -5.28       -3.53   66.78       -1.76   33.22
1995   -6.62       -4.61   69.59       -2.01   30.41
1996   -6.48       -4.69   72.39       -1.79   27.61
1997   -7.25       -5.56   76.65       -1.69   23.35
1998   -7.82       -6.17   78.87       -1.65   21.13
1999   -8.04       -6.29   78.27       -1.75   21.73
2000   -8.22       -6.38   77.57       -1.84   22.43
2001   -7.62       -5.90   77.39       -1.72   22.61
2002   -5.89       -4.33   73.61       -1.55   26.39
2003   -5.07       -3.66   72.14       -1.41   27.86
2004   -4.92       -3.69   75.03       -1.23   24.97
2005   -5.59       -4.62   82.75       -0.96   17.25
2006   -6.00       -5.24   87.39       -0.76   12.61
2007   -6.20       -5.46   88.10       -0.74   11.90
2008   -5.28       -4.80   91.00       -0.47    9.00
2009   -0.68       -0.39   56.79       -0.29   43.21
2010    0.00        0.00    0.00        0.00    0.00

Notes:  Authors' own calculations from the IPUMS-CPS data.
Bold rows correspond to the NBER-dated contractions (from
peak  to trough). Columns 3-6 correspond to the decomposition
as described in (2).

Table 5 Decomposition of the Change in the Unemployment
-to-Population Ratio Between 2010 and Earlier Years

       Total   Change Because of Change Because of
Year  Change  Labor Outcomes    Socidemographics

              Percentage    % of  Percentage   % of
                 Points    Total      Points   Total
                          Change               Change

1968    6.87        7.04  102.42       -0.17    -2.42
1969    7.20        7.31  101.46       -0.10    -1.46
1970    6.25        6.28  100.57       -0.04    -0.57
1971    5.10        5.18  101.62       -0.08    -1.62
1972    5.41        5.44  100.53       -0.03    -0.53
1973    5.90        5.88   99.61        0.02     0.39
1974    6.02        5.97   99.14        0.05     0.86
1975    3.00        2.89   96.20        0.11     3.80
1976    3.99        3.83   95.98        0.16     4.02
1977    4.22        4.05   95.79        0.18     4.21
1978    5.24        5.07   96.64        0.18     3.36
1979    5.48        5.28   96.33        0.20     3.67
1980    4.66        4.49   96.41        0.17     3.59
1981    3.73        3.72   99.76        0.01     0.24
1982    1.86        1.86   99.75        0.00     0.25
1983    0.18        0.18  101.02        0.00    -1.02
1984    2.82        2.86  101.41       -0.04    -1.41
1985    3.50        3.56  101.66       -0.06    -1.66
1986    3.40        3.55  104.50       -0.15    -4.50
1987    3.62        3.78  104.28       -0.15    -4.28
1988    4.32        4.53  105.07       -0.22    -5.07
1989    4.72        4.95  104.72       -0.22    -4.72
1990    4.69        4.96  105.69       -0.27    -5.69
1991    2.90        3.25  112.03       -0.35   -12.03
1992    2.26        2.58  114.43       -0.33   -14.43
1993    2.45        2.67  108.79       -0.22    -8.79
1994    3.55        3.79  106.79       -0.24    -6.79
1995    4.47        4.63  103.61       -0.16    -3.61
1996    4.28        4.49  105.03       -0.21    -5.03
1997    4.78        5.00  104.59       -0.22    -4.59
1998    5.32        5.49  103.31       -0.18    -3.31
1999    5.83        5.90  101.11       -0.06    -1.11
2000    5.99        6.00  100.09       -0.01    -0.09
2001    5.57        5.53   99.21        0.04     0.79
2002    4.16        4.12   98.96        0.04     1.04
2003    3.93        3.87   98.53        0.06     1.47
2004    4.24        4.20   98.87        0.05     1.13
2005    4.89        4.87   99.63        0.02     0.37
2006    5.24        5.23   99.98        0.00     0.02
2007    5.25        5.19   98.87        0.06     1.13
2008    4.76        4.71   99.04        0.05     0.96
2009    0.83        0.80   96.35        0.03     3.65
2010    0.00        0.00    0.00        0.00     0.00

Notes: Authors' own calculations from the TPUMS-CPS data.
Bold rows correspond to the NBER-datcd contractions (from
peak  to trough). Columns 3-6 correspond to the decomposition
as described in (2).

Table 6 Decomposition of the Change in the OLF-to-Population
Ratio Between 2010 and Earlier Years

              Change Because of  Change Because of
              Labor Outcomes     Socidemographics

       Total  Percentage   % of   Percentage   % of
Year  Change    Points     Total     Points    Total
                          Change              Change

1968    8.65        7.92   91.51        0.73     8.49
1969    8.32        7.55   90.73        0.77     9.27
1970    8.21        7.41   90.25        0.80     9.75
1971    7.65        6.84   89.40        0.81    10.60
1972    7.05        6.15   87.27        0.90    12.73
1973    6.50        5.51   84.88        0.98    15.12
1974    6.20        5.28   85.16        0.92    14.84
1975    5.26        4.31   81.82        0.96    18.18
1976    4.61        3.50   75.80        1.12    24.20
1977    4.85        3.70   76.15        1.16    23.85
1978    4.55        3.46   76.03        1.09    23.97
1979    4.83        3.60   74.62        1.22    25.38
1980    4.53        3.27   72.15        1.26    27.85
1981    4.18        2.85   68.10        1.33    31.90
1982    3.97        2.50   62.90        1.47    37.10
1983    3.68        2.06   55.89        1.62    44.11
1984    3.43        1.77   51.70        1.66    48.30
1985    3.31        1.68   50.69        1.63    49.31
1986    3.16        1.46   46.24        1.70    53.76
1987    3.55        1.72   48.30        1.84    51.70
1988    3.35        1.53   45.69        1.82    54.31
1989    3.54        1.63   46.01        1.91    53.99
1990    3.51        1.51   43.12        2.00    56.88
1991    3.38        1.54   45.67        1.84    54.33
1992    3.36        1.42   42.26        1.94    57.74
1993    2.87        0.93   32.32        1.94    67.68
1994    1.73       -0.28  -16.19        2.02   116.19
1995    2.15       -0.03   -1.41        2.18   101.41
1996    2.21        0.18    7.96        2.03    92.04
1997    2.47        0.54   21.98        1.93    78.02
1998    2.51        0.63   25.06        1.88    74.94
1999    2.21        0.37   16.76        1.84    83.24
2000    2.23        0.36   15.93        1.88    84.07
2001    2.05        0.35   16.92        1.70    83.08
2002    1.72        0.19   10.93        1.53    89.07
2003    1.14       -0.24  -20.80        1.38   120.80
2004    0.67       -0.53  -78.90        1.20   178.90
2005    0.70       -0.27  -38.66        0.97   138.66
2006    0.76        0.00   -0.16        0.76   100.16
2007    0.95        0.27   28.02        0.69    71.98
2008    0.52        0.08   14.69        0.44    85.31
2009   -0.15       -0.43  284.63        0.28  -184.63
2010    0.00        0.00    0.00        0.00     0.00

Notes: Authors' own calculations from the IPUMS-CPS data.
Bold rows correspond to the NBER-dated contractions (from
peak to trough). Columns 3-6 correspond to the decomposition
as described in (2)

Table 7 Predicted Aggregate OLF-to-Population
Ratio Among 25-64-Year-Old Men, 2015

Panel A: Based on 2010 Labor Outcomes

                            Actual  Predicted
                              2010       2015

Employment-to-Population     76.33          X
Unemployment-to-Population    8.92          X
OLF-to-Population            14.75      16.27

Panel B: Based on 2007 Labor Outcomes

                           Actual  Predicted
                             2007       2015

Employment-to-Population   82.53           X
Uncmployment-to-Population  3.67           X
OLF-to-Population          13.79       16.04

Panel C: Age Composition (Percent)

Age                         Actual  Simulated   Census
                              2010       2015     2015

25-34                        25.80      26.36    26.74
35-44                        25.20      23.93    24.63
45-54                        27.50      25.87    25.43
55-64                        21.50      23.85    23.20

Employment

The actual 2010 value of the employment-to-population ratio for 25-64-year-old men and the two series of counterfactual predictions for 2010 are shown in Figure 6, Panel A. The dashed line shows the predicted employment-to-population ratio from holding the demographic composition of the population constant at its 2010 level but varying the labor outcomes of different groups. The dotted line shows the predicted employment-to-population ratio from holding the labor outcomes of different groups at their 2010 level but varying the sociodemographic composition of the population. The actual employment-to-population ratio in 2010 is lower than any point of the predicted counterfactual series. This implies that changes in both the sociodemographic composition and labor outcomes of different groups in 2010 contribute to the historically low employment-to-population ratio among men.

To examine the contribution of the change in the labor outcomes of different groups and the sociodemographic composition to changes in the employment-to-population ratio, we construct each term of equation (2) for [t.sub.1] [member of] [1968,2010] and [t.sub.2] = 2010. Table 4 reports the change in the employment-to-population ratio, the total change as a result of change in each of the two terms, and the percentage of the total change that is accounted for by changes in each of the two terms. Figure 7, Panel A plots the change in the employment-to-population ratio and the total change as a result of each of the two terms. We find that the change in the labor outcomes of different groups accounts for the majority of the change in the employment-to-population ratio. Comparing the aftermath of the 2007-2009 and 1980-1982 recessions, we see that between 1983-2010 the employment-to-population ratio fell by 3.86 percentage points, of which 41.7 percent is a result of the change in the demographic composition.

[FIGURE 6 OMITTED]

We conclude that the decline in the employment-to-population ratio is a result of both changes in the sociodemographic composition and changes in the labor outcomes of different groups.

Unemployment

Panel B of Figure 6 plots the two counterfactual series of the 2010 unemployment-to-population ratio against its actual 2010 value. We draw two key observations from the figure: (1) the actual 2010 unemployment-to-population ratio is higher than any point of the predicted counterfactual series that holds the sociodemographic composition constant at its 2010 level; and (2) the actual 2010 unemployment-to-population ratio is lower than the predicted counterfactual for some periods when we hold the labor outcomes of different groups constant at their 2010 level.

These observations suggest that (1) the labor outcomes of different sociodemographic groups in 2010 contribute to a higher unemployment-to-population ratio than the labor outcomes of different groups in all previous years; and (2) the sociodemographic composition of the population in 2010 actually contributes to a lower unemployment-to-population ratio as compared to the sociodemographic composition in some earlier years.

Table 5 breaks down the total change in the unemployment-to-population ratio between a given year and 2010 into change caused by developments in the demographic composition of the population and change caused by developments in the labor outcomes of different groups. Figure 7, Panel B plots the total change in the unemployment-to-population ratio and the change as a result of each of the two terms. The results of this table and graph corroborate our above claims. Changes in the demographic composition of the population have contributed a small, and often negative, amount of the increase in the unemployment-to-population ratio, whereas changes in the labor outcomes of different groups have been responsible for approximately 100 percent of the increase. We observe that, relative to 1983, the 2010 unemployment-to-population ratio is 0.18 percentage points higher, of which 101.0 percent of the change is a result of a change in the labor outcomes of different groups. Thus, the rise in the 2010 unemployment-to-population ratio relative to its 1983 level is driven entirely by changes in the labor outcomes of different groups.

Out of Labor Force

Panel C of Figure 6 plots the two counterfactual series of the 2010 OLF-to-population ratio against its actual 2010 value. We highlight two features of the figure: (1) The actual 2010 OLF-to-population ratio is always higher than the counterfactual series calculated by holding the labor outcomes of different groups at their 2010 level; and (2) prior to 1994, the actual 2010 OLF-to-population ratio is always higher than the counterfactual series that is calculated by holding the sociodemographic composition constant at its 2010 level, although after 1994 the counterfactual is sometimes higher. Thus, we infer that the demographic composition contributes substantially to the high OLF-to-population ratio in 2010.

Table 6 and Figure 7, Panel C formalize this result, showing that the total change in the incidence of the OLF-to-population ratio between any year prior to 2009 and 2010 is positive. It also shows that the change in the sociodemographic composition contributes to a higher OLF-to-population ratio. The contribution from the change in the sociodemographic composition has been increasing since 1968. The contribution from the change in the labor outcomes of different groups has been significantly smaller, and has even lowered the OLF-to-population ratio in more recent years. Turning our attention once again to the 1983 and 2010 comparison, we see that the OLF-to-population ratio increased 3.68 percentage points from 1983-2010, of which 1.66 percentage points can be attributed to the change in the demographic composition. Thus, changes to the demographic composition of the population have played a large role in increasing the 2010 OLF-to-population ratio relative to its 1983 level.

4. FORECAST OF THE OLF-TO-POPULATION RATIO

Our findings show that changes in the employment-and OLF-to-population ratios of 25-64-year-old men during the last four decades are, to a large degree, associated with changes in the sociodemographic composition of the population. Using the labor outcomes of different sociodemographic groups from 2010 and a projected sociodemographic composition of the population in 2015, we are able to create projections of the aggregate labor outcomes in 2015 using equation (1).

As Figure 1 shows, there is a large cyclical component in the employment and unemployment-to-population ratios. Consequently, the forecasts of these ratios depend heavily on the business cycle phase of the year of our decomposition. In addition, the changes in unemployment (and employment to a lesser extent) are mostly dominated by changes in the labor outcomes of different sociodemographic groups rather than changes in the sociodernographic composition. In contrast, the OLF-to-population ratio has a much smaller cyclical component. Consequently, we focus on forecasting the OLF-to-population ratio.

[FIGURE 7 OMITTED]

To perform this forecasting exercise, we project the sociodemographic composition of the population of 25-64-year-old men in 2015. In projecting this composition, we focus on the changes in age of the individuals in the 2010 sample while holding education, race, and marital characteristics constant at their 2010 levels.

To simulate a sample of 25-64-year-old male workers in 2015, we use the 2010 sample of 20-59-year-old male workers and construct the age variable for 2015. We use the agespecific annual male mortality rates from the Social Security Administration (4) and accordingly choose which workers of a particular age survive from 2010-2015. (5) Each worker in the simulated sample is aged five years, but has the same education, race, and marital status as in the 2010 sample. We use the projected 2015 population and the CPS sampling weights to construct the sociodemographic composition terms in equation (1). Then we use these forecasted demographic composition terms and the labor outcomes of the corresponding sociodemographic groups from 2010 to construct the predicted aggregate 2015 OLF-to-population ratio using equation (1).

Note that this exercise assumes that the mortality rates for each age remain unchanged from 2007-2015. Also, we use the sampling weights from 2010, which may not deliver a representative population for our simulated 2015 sample (for example, because we do not adjust the weights to reflect the demographic composition of the surviving individuals). However, given the relatively short forecast horizon, these weights provide a good approximation for aggregation. Finally, when aging the 2010 population, we do not accordingly adjust demographic factors other than age. For example, aging a male from 20 to 25 might alter both his educational attainment and marital status. Our forecasting exercise does not take these effects into account.

The results of our forecast are displayed in Table 7. Panel C of Table 7 contains our forecast and the U.S. Census forecast for the age distribution of 25-64-year-old men in 2015. As the forecast shows, the shares of 55-64 and 25-34-year-old males are projected to increase, while the share of 35-54-year-old males is projected to decrease.

Panel A of Table 7 displays the results of the forecast of the OLF-to-population ratio based on the labor status outcomes of different groups in 2010. For comparison, Panel B contains the results based on the labor status outcomes of different groups in 2007, i.e., the year of a recent business cycle peak. The results show that under both sets of labor status outcomes of different groups, the OLF-to-population ratio is predicted to reach more than 16 percent in 2015 as compared to the actual rate of 14.7 percent in 2010.

5. CONCLUSIONS

The OLF-to-population ratio among 25-64-year-old men has increased from 6.5 percent in 1970 to 14.7 percent in 2010. In the aftermath of the 1969-1970 recession, the employment-to-population ratio among this group was 89.1 percent, while in the aftermath of the 2007-2009 recession, the ratio is nearly 13 percentage points lower. Throughout this article we have examined the degree to which these changes can be explained by changes in the composition of the population by age, race, education, and marital status, and the degree to which they can be attributed to changes in the labor market outcomes of different sociodemographic groups.

We find that the rise in the OLF-to-population ratio since the early 1980s is primarily a result of changes in the demographic composition of the population. Changes in the demographic composition account for about 25 percent of the increase in the employment-to-population ratio during the same period, and changes in the unemployment-to-population ratio are almost entirely driven by changes in the employment status composition. Finally, simulating the 2010 sample five years forward and using labor outcomes of different sociodemo-graphic groups from 2010, we project that the OLF-to-population ratio among 25-64-year-old men will rise to 16 percent in 2015.

REFERENCES

Autor, David H., and Mark G. Duggan. 2003. "The Rise in the Disability Rolls and the Decline in Unemployment." The Quarterly Journal of Economics 118 (February): 157-205.

Fallick, Bruce, and Jonathan Pingle. 2006. "A Cohort-Based Model of Labor Force Participation." Federal Reserve Board of Governors Finance and Economics Discussion Series 2007-2009.

Juhn, Chinhui, Kevin M. Murphy, and Robert H. Topel. 1991. "Why Has the Natural Rate of Unemployment Increased Over Time?" Brookings Papers on Economic Activity 22 (2): 75-142.

Juhn, Chinhui, Kevin M. Murphy, and Robert H. Topel. 2002. "Current Unemployment, Historically Contemplated." Brookings Papers on Economic Activity 33 (1): 79-136.

Little, Allan, and Steve Bradley. 2007. "The Determinants of Changes in the Risk of Unemployment and Inactivity in Britain: A Decomposition Analysis." Labour Economics Research Group Discussion Paper No. 17.

Polivka, Anne E., and Stephen M. Miller. 1995. "The CPS After the Redesign: Refocusing the Economic Lens." Manuscript, Bureau of Labor Statistics.

Shimer, Robert. 1999. "Why is the U.S. Unemployment Rate so Much Lower?" In NBER Macroeconomics Annual 1998, Volume 13, edited by Ben S. Bernanke and Julio Rotemburg. Chicago, The University of Chicago Press, 11-61.

U.S. Census Bureau. 2008. "2008 National Population Projections." Washington, D.C.: U.S. Census Bureau. www.census.gov/population/www/projections/2008projections.html.

(1.) The focus of this article is on the employment-, unemployment-and OLF-to-population ratios. These are defined as the proportion of individuals in the entire population with a given labor status. They are thus distinct from rates (e.g., the unemployment rate), which are defined as the proportion of the labor force (i.e., the sum of unemployed and employed persons) with a given labor status outcome.

(2.) This article focuses on the male population. The OLF-to-population ratio for women fell drastically from 1968 to the mid-1990s as females entered employment, while that of males trended upward. Since the mid-1990s, the OLF-to-population ratios for males and females have experienced similar trends, though the OLF-to-population ratio for females remains approximately 10 percentage points higher. While studying aggregate employment outcomes for both genders would be an interesting exercise, doing so is beyond the scope of this article.

(3.) Results of the alternative decomposition are available upon request.

(4.) Data available at: www.ssa.gov/oact/STATS/table4c6.html#ss.

(5.) For example, the probability that a 20-year-old worker in 2010 survives to 2015 is (1 - [p.sup.m.sub.20]) (1 - [p.sup.m.sub.21])(1 - [p.sup.m.sub.22](1 - [p.sup.m.sub.23])(1 - [p.sup.m.sub.24), where [p.sup.m.sub.a] is the annual mortality rate of a worker at age a.

We would like to thank Huberto Ennis, Arantxa Jarque, Nadezhda Malysheva, and Alexander Wolman for their invaluable comments. The views expressed here do not necessarily reflect those of the Federal Reserve Bank of Richmond or the Federal Reserve System. E-mail: marianna.kudlyak@rich.frb.org; thomas.lubik@rich.frb.org; jonathan.tompkins@rich.frb.org.