The labor supply of female household heads is distributed across a wide
variety of hours and weeks choices. This papers explores the differential
nature of the weeks per year and hours per week decisions among female heads.
A model of labor supply which separates the weeks/hours decision is presented
and several forms of this model are estimated, allowing for simultaneity in
the weeks/hours decision, as well as for the presence of either fixed costs or
minimum constraints on low levels of hours and weeks of work. The results
indicate that these two decisions are separate, although not completely unrelated
Like previous researchers, I also find strong evidence of the need to separate
the labor force participation decision from the weeks and hours of work decision.
The paper ends with a discussion of the evidence on labor rationing among
these women, either through un- or underemployment.

This paper estimates the extent to which workers with
different personal characteristics have differing probabilities of
being located in public versus private sector employment. A reduced
form two-way probit model is developed which analyzes worker choice
between the public and the private sectors, along with a three-way
probit model which breaks this down to a choice between private,
federal and state-local jobs. Significant differences in the
relationship between selection probabilities and worker
characteristics are found between these three sectors and these
differences are shown to vary in interesting ways across
occupations. These results make it possible to characterize the
type of individual who is most likely to be attracted to a job in
the private, federal or state-local sectors and provide a more
complete understanding of how workers perceive and respond to
existing sectoral employment differences.

This paper disaggregates total household income into a complete set
of components and studies the comparative cyclicality of these com-
ponent< to economic growth. Comparisons of the relative responsiveness
to GNP growth of wages, hours of work, and total labor market income of
heads and wives, and transfer income sources of households are made
across income, race, sex and age groups. This provides a picture of the
channels by which economic growth produces income change. Significant
differences in elasticities are found to exist both between different
income components and between different population groups for the same
component. The narrowing income distribution in times of high growth
occurs primarily because of large elasticities on head's labor market
income among the poor. Both wages and hours show evidence of cycli-
cality. The labor market earnings of women -- both wives and household
heads —- is far less responsive to growth. Cyclicality in transfer
income varies enormously between population groups and by type of
transfer.

This paper investigates cyclicality in real wages between 1969 and 1982,
using 14 years of data from the Panel Survey of Income Dynamics. First, it
investigates the extent to which movements in and out of the labor market
created apparent wage cyclicality. Second, it investigates whether cyclical
movements of workers between heterogeneous wage sectors within the labor market
created cyclicality. Little evidence of the first effect is found. The second
effect is much more important, and cyclicality clearly occurs in the movement
of workers between different labor market sectors. However, sector selection
is not correlated with wage determination. Thus, individual wage change
estimates of cyclicality need to control for sector location, but need not
account for sector selection. The third conclusion of the paper is that
cyclicality is present in real wages even within sectors over this time period,
and is the result of both cyclicality in overall wage levels (cyclicality in
the constant term in wage equations), as well as in the coefficients associated
with particular worker characteristics.

This paper presents new evidence on the reasons for the recent decline in
the fraction of unemployed workers who receive unemployment insurance benefits.
Using samples of unemployed workers from the March Current Population Survey,
we estimate the fraction of unemployed workers who are potentially eligible for
benefits in each year and compare this to the fraction who actually receive
unemployment compensation. Perhaps surprisingly, we find that the decline in
the fraction of insured unemployment is due to a decline in the takeup rate for
benefits. Our estimates indicate that takeup rates declined abruptly between
l98O and 1982, leading to a 6 percentage point decline in the fraction of the
unemployed who receive benefits.
We go on to analyse the determinants of the takeup rate for unemployment
benefits, using both aggregated state-level data and micro-data from the Panel
Study of Income Dynamics. Changes in the regional distribution of unemployment
account for roughly one-half of the decline in average takeup rates. The
remainder of the change is largely unexplained.