Employment in recession and recovery: a demographic flow analysis Essay

As in earlier downturns, the impacts of recession during 1981-82 were
not evenly distributed among the many demographic groups in the labor
force. For example, the rise in the unemployment rate was greatest, in
relative terms, for men. The decrease in labor force participation was
most pronounced among teenagers, while the labor force participation
rate for women actually increased during this period of general economic

To what extent were these and other differential impacts of the
recession the result of differences in the behavior of the labor force
participants? To what extent were they instead the result of differing
labor market opportunities? These, of course, are very difficult
questions to answer, particularly when dealing with aggregate data. To
illustrate, a decrease in labor force participation can be the result of
two factors–an increase in the rate at which individuals leave the
labor force, or a decrease in the rate at which workers enter the labor
force. Because these and other types of labor force transitions can
have different behavioral interpretations (that is, they may have
“different kinds of sources”), it is important to identify
which transitions generate demographic differences in labor force
participation and unemployment experience. To address these issues, I
examine, by age, sex, and race, the monthly flows into and out of the
labor force and between employment and unemployment from January 1981 to
January 1984, well into the current recovery period.

Distribution of economic impacts

Race, sex, and age differences in the levels of unemployment and
labor force participation rates can be seen in table 1. The entries are
averages over the period December 1980 to December 1983 of data from
Current Population Survey “Gross Change Tabulations,” which
give monthly estimates of the numbers of people employed, unemployed,
and out of the labor force during the preeceding month. The entries in
the table therefore are not based on or equivalent to the unemployment
and participation rates published by the Bureau of Labor Statistics.

Inspection of the table indicates, however, that the well-known race, sex, and age differences found in the published estimates are also
found here. Blacks and members of other races, on average, have higher
unemployment rates than whites, and lower levels of labor force
participation, regardless of sex or age. Women have slightly lower
unemployment rates than men (a relatively recent phenomenon), and lower
labor force participation rates, regardless of age or race.
Unemployment rates are seen to decrease with age for all sex/race
groups, while labor force participation rates increase and then decrease
with age, peaking in the 25- to 59-year-old “prime-age”
category. Although the point estimates from the gross change data may
differ from the published BLS estimates, the age, race, and sex
relationships seem to be the same.

The focus of this study is not on differences in the levels of
unemployment and participation, however, but rather on differences in
their behavior over the most recent business cycle. The National Bureau
of Economic Research has identified the peak of that cycle as July 1981
and the trough as November 1982. The corresponding changes in the
unemployment rates during the period for each demographic group are
presented in table 2, along with changes since the recovery began, for
the November 1982 to December 1983 period. During the downturn, the
unemployment rate increased more on average for men than for women, more
for whites than for blacks and others, and more for older (over 59)
workers than for teenagers (age 16 to 19), youth (20 to 24), or
prime-age workers (24 to 59). The greatest increases were felt among
older women, who experienced growth in their unemployment rate of more
than 158 percent. The sex difference was reversed for nonwhite teens,
youth, and older workers, with nonwhite women experiencing greater
relative unemployment increases than nonwhite men. The racial
difference was reversed for teenagers.

Of course, the lags in the impacts of an economic downturn can vary
across demographic groups, so that the “official” definition
of the timing of the downturn may not be the appropriate timeframe for
this type of analysis. For example, the unemployment rate for black men
did not peak until July 1983. To account for this, I computed the
percentage change in the unemployment rate for each demographic group
between the month the group’s rate was at its minimum and the month
it reached its maximum. These estimates are presented in the following
tabulation, for the “all ages” and “teens”

Most of the qualitative conclusions noted above do not change. The
relative increases in the unemployment rate were still worse for men
than for women and worse for whites than for members of other races
(except among teens). One differences is that, by this measure, teens
suffered greater than average unemployment rate increases, while one
might conclude the opposite using the measure in table 2.

Referring again to table 2, we see that the pattern in the recovery
period differs somewhat from that of the recession. For instance, the
effect of the recovery was relatively stronger for women than for men,
while the opposite was true of the recession. The racial difference
remained the same: the effect of the recovery was felt more, on average,
by whites than by nonwhites. The sex difference is primarily due to the
fact that the unemployment rate continued to rise for nonwhite men well
into the recovery period. Again, these observations are consistent with
those based on the published unemployment rates.

Many explanations have been offered for these differences. For
example, the effect of the downturn has been said to have been greater
for men than for women because the economic decline affected primarily
the goods-producing, as opposed to the service-producing, sector.
Construction and auto-related industries, including steel manufacturing,
were especially hard hit. In contrast, some service industries actually
increased employment (although at a decreasing rate) throughout most of
the recession. Along the same lines, blue-collar workers suffered worse
employment losses than white-collar workers. Because men and women are
distributed differently among industries and occupations, with men in
the more cyclically sensitive ones, men would be expected to suffer
relatively greater increases in their unemployment rates. The fact that
the industries and occupations that incurred the greatest losses in
demand are also those with traditionally higher than average layoff rates could have further aggravated their employment declines.

The contribution this makes to the sex difference in the employment
declines is unclear, however. We know that men have higher layoff rates
than women, but that is probably primarily because of the sex difference
in the occupational distribution. Any sex differences in the cyclic sensitivity of layoff rates are also probably due to the industrial or
occupational distributions. To fully understand the role of layoff
rates in explaining the sex differences in the cyclic behavior of
unemployment rates, we need to know whether the responsiveness of the
layofff rate is less for women than for men in the same industry and
occupation. Evidence presented by Norman Bowers suggests that in the
three previous recessions the responsiveness of the layoff rate was
actually greater for women than for men, both on average and by industry
and occupation. Findings by Francine Blau and Lawrence Kahn, however,
seem to show that there is little, if any, sex difference in the
cyclical component of layoffs after controlling for industry,
occupation, and other worker characteristics.

Differences in cyclical variations in layoff rates also fail to
explain the racial difference in changes in the unemployment rate.
Nonwhites suffered relatively smaller unemployment rate increases than
whites during the last recession, yet their layoff rates have
historically been more cyclically responsive, even after controlling for
worker and job characteristics. Instead of layoff rate disparities, the
racial difference in the unemployment response is probably due, at least
in part, to the fact that members of racial minorities never fully
recovered from the 1980 recession. Their unemployment rates were
already high when the most recent downturn began, so that the increases
it brought about were relatively small.

One other factor that could be important in explaining the
differential unemployment rate impacts both by race and by sex is the
propensity, as unemployment rates increase (or, put differently, as
employment opportunities decline), for labor participation rates to
decrease. If women and nonwhites tend to drop out of the labor force at
a greater rate than white males in response to a given change in
employment opportunities, then their unemployment rates will not rise by
as much as those for white males. The “economic impact” for
men and women could therefore be the same–women could suffer as much as
men–but it would not be reflected in the unemployment rate. It is for
this reason that many analysts argue that unemployment rates are not
appropriate measures of the welfare of a demographic group, and prefer
to study the “employment to population ratio” instead. I
prefer to examine the problem directly and look at the behavior of both
the unemployment and labor force participation rates. In particular, we
need to examine the relationships between the two.

Estimates of the percentage changes in (seasonally adjusted) labor
force participation rates for the July 1981-November 1982 period and the
November 1982-December 1983 period are presented in table 2. As with
the cyclic behavior of the unemployment rate, differences exist
according to age, race, and sex. Note that the participation rate
decreased for men during the economic decline, while it increased for
women. The rate rose for whites, but the increase was small relative to
the increase for blacks and others. Referring to the previous
discussion, we find these results suggest that the unemployment rate
measure actually overstates the burden of the recession for women and
members of racial minorities relative to white men, rather than
understating it as had been hypothesized above.

Certainly, these changes may be due to recent trends more than to
the business cycle. To correctly interpret changes in the unemployment
rate, we need to look at its relationship with participation rates net
of trend. I do this by examining the coefficient on the unemployment
rate variable in the following equation: (1) log (LFPR).sub.t =
[BETA].sub.0 + [BETA].sub.1.TIME.sub.1 + [BETA].sub.2.URATE.sub.t.-1 +
*(seasonal dummies) + u.sub.1 where LFPR is a given group’s labor
force participation rate in period t, and URATE.sub.t.-1 is the
unemployment rate (for that group, for the entire population, or for
some reference group, such as prime-age men), lagged one period. Lagging the unemployment rate is one way to eliminate the problems created by
the fact that sampling errors in URATE and LFPR may be highly correlated at any point in time. Estimates of BETA.sub.1 and BETA.sub.2 are
presented in table 3, by age, race, and sex. (The estimates are derived using the Cochrane-Orcutt technique, assuming first-order serial
correlation. The unemployment rate variable is here defined as the
average unemployment rate for the population as a whole.)

The results indicate that the relationship between the unemployment
rate and the labor force participation rate (as measured by the
coefficient on URATE) did not differ much by race, except for male
teenagers. For nonwhite male teens, a 1-percent increase in the
unemployment rate (that is, from 10.0 to 10.01) is associated with a
.3483-percent decrease in their labor force participation rate. That
response is almost four times the response exhibited by whites. For the
population as a whole, however, the magnitudes of the responses vary
little by race. Some differences do exist by sex, with males exhibiting
a strong tendency to decrease their participation as unemployment rate
rise. This is true for all groups except white teens. The coefficients
on TIME indicate that the increases in the participation rates of women
during the period (recall the results in table 2) were indeed largely
the effect of a trend component rather than a cyclic one. Relating
these results back to our interpretation of the “burdens” of
the recession, the fact that declines in aggregate demand seem to
generate relatively larger decreases in participation for men and teens,
and especially minority male teens, suggests that the unemployment rates
for those groups may understate the true relative burden of the

Explanations for the differing participation rate responses include
the notion that teens and men exhibit greater than average decreases in
participation as unemployment rates rise because they suffer greater
than average decreases in demand for their labor. A decrease in demand
can have two effects: first, assuming some degree of wage rigidity,
there is a direct effect on employment, and hence a direct effect on
participation–if the number of people employed declines then, other
things equal, the participation rate will decline. Second, there is the
“discouraged worker effect,” the decline in participation
because persons think they will have little success finding a job. An
alternative explanation is that demand does not decrease more for teens
or men, but rather that, given a change in the demand for their labor,
teens and men simply respond more. Results from another study have
shown that resources of differences in participation responses include
differential costs of search, differential wage rates, and differential
levels of (not changes in) labor demand, in addition to differential
“preferences” for work.

Possible explanations for the relatively small decreases in
participation exhibited by women may therefore include the following:
(1) demand for women’s labor does not decline much as unemployment
rates rise; (2) women have stronger preferences for work and lower costs
of search; or (3) women will enter the labor force as unemployment rates
rise to compensate for income lost because of the unemployment of other
family members (the “added worker effect”). Evidence of the
validity of each of these hypothesis is presented later in this study.

In sum, using relative changes in the unemployment rate as a
measure of the impact of the recent recession, the evidence indicates
that the heaviest burdens were placed on male, white, and prime-aged and
older workers. The magnitude of the burdens is open to question,
however, if one keeps in mind that changes in labor force participation
rates affect measured unemployment rates, and that the participation
rate is endogenously determined. Inspection of the relationship between
labor force participation rates and aggregate demand suggests that the
unemployment rate variable probably understates the recession’s
relative impact on men and on teens.

The nature of differential impacts

According to the gross change data, 3,293,000 workers became
unemployed during December 1983. Some 1,837,000 entered unemployment
from employment, while 1,456,000 entered unemployment from outside the
labor force. During the same month, 3,576,000 workers left
unemployment–1,745,000 into employment and 1,831,000 into the
non-participation state. At this example illustrates, the labor market
is in continual motion. The goal of the following discussion is to
examine the cyclical variations in unemployment and labor force
participation noted earlier in the context of such labor market flows.

Let us denote the number of workers who make a transition from
state I to state J (for example, from employment (E) to unemployment
(U), or from unemployment to non-participation (N) during month t as IJ.
Define the probability of making such a transition, given that one is in
state I in month t-1, as lambda.sub.IJ = IJ.sub.t./I.sub.t.-1., where
I.sub.t.-1 is the number of people in state I in period t-1. It can
then be shown that unemployment rates and labor force participation
rates can be expressed as explicit functions of the six transition
probabilities lambda.sub.NE., lambda.sub.NU., lambda.sub.EN.,
lambda.sub.EU., lambda.sub.UE., and lambda.sub.UN.. The relationships
are such that the unemployment rate increases with increases in
lambda.sub.NU and lambda.sub.EU and decreases with increases in
lambda.sub.UE and lambda.sub.UN. The effects of changes in
lambda.sub.NE and lambda.sub.EN depend on the relative magnitudes of the
other transition probabilities. The participation rate will increase
with increases in lambda.sub.NE and lambda.sub.NU., and decrease with
increases in lambda.sub.EN and lambda.sub.UN.. The efects of
lambda.sub.UE and lambda.sub.EU depend on the relative magnitudes of
lambda.sub.UN and lambda.sub.EN. Whatever their size or direction,
changes in these transition probabilities are the sources of changes in
unemployment and labor force participation rates. We can therefore
analyze cyclical changes in unemployment and participation rates in
terms of cyclical variations in transition probabilities.

Before proceeding to that analysis, however, it may be useful to
examine age, race, and sex differences in levels of transition
probabilities. The average over the December 1981-December 1983 period
are presented in table 4 for the population as a whole, and for the
teenage and prime-aged groups. Given the race, sex, and age differences
in unemployment and participation rates, the differences in transition
probabilities are not surprising. Women have lower probabilities of
making the transitions from N-to-E and N-to-U, and much higher
probabilities of moving from E-to-N and U-to-N. All of these
differences contribute to the lower labor force participation rates for
women. Members of racial minorities have much lower rates of transition
from U-to-E than do whites, and slightly higher transition rates from
E-to-U, which contribute to their higher unemployment rates. Racial
differences also exist in the N-to-U and U-to-N transition rates, with
nonwhites more likely to enter unemployment on the one hand, more likely
to leave it on the other. These differences tend to cancel one another
out. A significant racial difference also exists for the N-to-E
transition for teenagers, with nonwhites much less likely to make the
transition. On average, teenagers are much more volatile than other
labor force groups, with higher than average probabilities for the
N-to-E, N-to-U, E-to-N, E-to-U, and U-to-N transitions. The U-to-E
transition rate does not differ much by age. Prime-aged workers differ
from others primarily in their lower E-to-N and U-to-N transition

The hypothetical relationship between aggregate demand and each of
the transition probabilities are relatively straight-forward for some
flows ad very complex for others, depending on one’s model and
assumptions. In a fairly general model, all of the effects of a change
in demand are indeterminant. A decline in aggregate demand will tend to
decrease lambda.sub.UE and lambda.sub.NE because the number, frequency,
and attractiveness of job offers will decline. A decrease in the
frequency of job offers can cause worker’s reservation wages to
fall, however, which would tend to increase lambda.sub.UE and
lambda.sub.NE. A decline in aggregate demand can increase the flows
from E-to-U and E-to-N due to an increase in layoffs nd terminations,
but it can decrease the same flows if it lowers workers’ propensity
to quit a job. As aggregate demand falls, we might expect lambda.sub.UN
to increase and lambda.sub.NU to decrease as a result of declining job
offers, but this conclusion depends critically on the relative
magnitudes of the levels of changes in job offer rates to people in the
U and N states. In addition, lambda.sub.UN may decrease and
lambda.sub.NU may increase when aggregate demand falls, as individuals
respond to the unemployment of other family members. The actual
realtionships between aggregate demand and transition probabilities are,
at best, empirical issues.

Using the lagged population-average unemployment rate as a measure
of aggregate demand, I have explored these relationships by estimating
the parameters of the following equation for each transition rate and
for the entire population, teens, and the prime-aged group: (2)
log(lambda.sub.IJ.).sub.T = beta.sub.0 + beta.sub.1.TIME.sub.t +
beta.sub.2.URATE.sub.t-1 + *(seasonal dummies) + u.sub.t

These estimates of beta.sub.1 and beta.sub.2 are presented in table
5. The

results indicate that some transition probabilities were much more
cyclically responsive than others and that the responsiveness varied
significantly across demographic groups. First, the N-to-E transition
rate declined with aggregate demand, for the population as a whole and
for each of the subgroups except nonwhite teenage females. The decline
is especially large for nonwhite males. Nonwhite male teenagers
exhibited the strongest response, which would contribute to their
stronger participation rate response. (See table 3.) Overall, the
N-to-E transition rate seems more responsive for racial minorities than
for whites, and more responsive for men than women. The responsiveness
of the N-to-U transition rate differs primarily by race, not only in
magnitude but also in direction. The N-to- transition rate tends to
increase for whites as aggregate demand falls, but decreases for blacks
and others (though the effect is often statistically insignificant).
The effect of this difference is to decrease labor force participation
among nonwhites and boost it among whites. The E-to-N transition rate
declines as aggregate demand falls, for all age, race, and sex groups.
The effect is stronger for nonwhites, with little difference by sex.
The U-to-N transition rate also decreases with aggregate demand for the
population on average, although it increases for female teens. Both of
these transition rate responses (for E-to-N and U-to-N) are counter to
standard views of the effects of declines in aggregate demand. In
particular, they tend to increase rather than decrease labor force
participation. The strong negative relationship between the unemployment
rate and participation rates exhibited by many of the demographic groups
therefore is not the result of an increased tendency to drop out of the
labor force. Rather, the relationship is the result of a decrease in
the tendency to enter the labor force, particularly directly into

The E-to-U and U-toE transition rates increase and decrease,
respectively, as aggregate demand falls. There is little difference in
the E-to-U response by race or by sex, excpet for teens and perhaps
prime-age men. Large race and sex differences do exist for the U-to-E
transition rate, however, which are probably the primary source of the
differential unemployment rate responses noted earlier. As aggregate
demand fell during the recession, the U-to-E transition rate declined
more for whites than for racial minorities (except prime-age men), and
more for males than for females except, again, among teens. These
differences may be the result of the disproportionate distribution of
the sexes and races across occupations and industries.

All of these differences in the responsiveness of transition
probabilities can be related to race, sex, and age differences in the
cyclic responsiveness of unemployment and labor force participation
rates, and can help identify their sources. The fact that the
unemployment rate increased more for men than for women during the
recession seems to be the result of the sex differences in the
responsiveness of the U-to-E transition probability. This may be
interpreted as support for the hypothesis that the demand for labor
declined relatively more for men. The fact that the participation rate
declined more for men than for women seems to be the result of a
tendency for the N-to-E transition rate to decline more for men. This
fact could suggest that the differential participation rate response is
a labor demand, rather than a labor supply, phenomenon. The added
worker effect as an explanation for the sex differences in the
participation response does not get much support here, because the
N-to-U transition probability does not respond any more for women than
it does for men, at least among whites.

The racial difference in the responsiveness of the unemployment
rate during the recession is primarily the result of racial differences
in the responsiveness of the N-to-U and U-to-E transition probabilities.
Both tend to boost unemployment rates more for whites than for
nonwhites. The N-to-U difference indicates that the added and
discouraged worker effects may be important explanations here, with
whites being the added workers and nonwhites the discouraged ones. This
could simply be the result of the racial difference in the distribution
of single-parent households. However, it could also be an indication
that members of racial minorities feel that they are at a considerable
labor market disadvantage because of their race. The relatively large
decline in the N-to-E transition rate for nonwhites may very well mean
that nonwhites do suffer larger decreases in demand for their labor as
aggregate demand declines.

The major age differences in the responsiveness of unemployment and
participation rates can also be related to specific transition rates.
The unemployment rate of teenagers rose less than average as aggregate
demand fell because the U-to-E transition rate did not decline by as
much for teens as for other groups, and because the U-to-N transition
rate increased for teens (except nonwhite males) while decreasing for
other groups. The first phenomenon could indicate that reservation
wages fell more for teens than for other workers, or that the demand for
teenage labor decline less than the demand for others, while the second
phenomenon suggests that teens were more likely to become discouraged
and quit looking for work. The response of the U-to-N transition
probability also obviously contributes to age differences in the
responsiveness of the labor force participation rate. Other factors are
the age differences responses of the N-to-E and N-to-U transition rates,
especially for nonwhite males. The large N-to-E response could indicate
that a substantial portion of the participation rate decline for teens
is the result of a decrease in the demand for their labor.

The result presented here lend support to many of the hypotheses
put forth earlier regarding the sources of demographic differences in
unemployment and participation rate behavior. The male/female
difference in unemployment rate behavior is indeed probably due to
differential changes in demand, which may be attributable to the
occupational distribution of the sexes. There is no support, however,
for the hypothesis that the participation rate differences arise because
women are more likely than men to be “added workers.”
Differences between the participation responses of whites and nonwhites
and between those of teens and other workers appear to be due both to
differences in relative responses of the demand for their labor (with
the demand for labor decreasing more for racial minorities and teens),
and to difference in “supply.”

Suggestions for further research

This analysis of gross change data from the Current Population
Survey provides insights into the nature of the differential effects of
the recent recession which cannot be obtained from an analysis of
unemployment or participation rates alone. Many questions remain
unanswered, however. Foremost, of course, is, what exactly causes each
of the differential transition rate responses? If men are discouraged
more than women, hwy? That is a difficult question even with microdata.
There are also some questions relating to the methodology, including
those related to the timing of the effects of the recession and the
appropriate lag structures to use for the URATE variable in equations 1
and 2. Further, exactly what is the effect on the unemployment rate of
a 1-percent decrease in a given transition rate? Does the effect differ
by race or sex? One last question we may want to address is, how do the
effects fo the 1981-82 recession differ from those of earlier downturns?
Have there been structural changes in the relationships between
aggregate demand and transition rates which may indicate, for example,
that there is less sex or race discrimination in the labor market today,
or that there has been a profound and lasting change in women’s
attitudes toward work outside the home? Many researchers address these
issues in other contexts, but a comparison of the results presented here
with those from studies of earlier periods could lead to better

Finally, it should be noted that many cyclical changes in
employment status are not between employment, unemployment, and
nonparticipation, but rather between full-time and part-time employment.
The data used in this study do not distinguish between full- and
part-time employment. An analysis of gross flow data that make such a
distinction could be very fruitful, as could further study of gross
change data broken down by industry of employment.


I'm Tamara!

Would you like to get a custom essay? How about receiving a customized one?

Check it out