This paper presents results from an experimental evaluation of an earnings supplement program offered
to long-term welfare recipients in two Canadian provinces. The program -- known as the Self-Sufficiency
Project - provides a supplement equal to one-half of the difference between an earnings target ($2,500 or
$3083 per month, Canadian dollars, depending on the province) and the individual's actual earnings. The
supplement is similar to a negative income tax with two important differences: (1) eligibility is limited to
long-term welfare recipients who find a full-time job (30 hours per week or more); and (2) the supplement
payment is based on individual earnings rather than family income. The evaluation is based on a
randomized design that will follow 6,000 individuals for five years. Early findings for a first cohort of
2,000 individuals observed over 18 months of program eligibility suggest that the financial incentives of
the Self-Sufficiency Program significantly increase labor market attachment and significantly reduce
welfare participation.

It has become standard practice to use local linear regressions in regression discontinuity designs.
This paper highlights that the same theoretical arguments used to justify local linear regression suggest
that alternative local polynomials could be preferred. We show in simulations that the local linear estimator
is often dominated by alternative polynomial specifications. Additionally, we provide guidance on the
selection of the polynomial order. The Monte Carlo evidence shows that the order-selection procedure
(which is also readily adapted to fuzzy regression discontinuity and regression kink designs) performs
well, particularly with large sample sizes typically found in empirical applications.

Between 1960 and 1980 the gap in earnings between black and white
males narrowed by 15 percent. A detailed analysis of 1960, 1970, and 1980
Census data indicates that increases in the relative return to education
were largely responsible for black workers’ relative earnings gains. One
explanation for these higher returns is that they reflect the market
valuation of higher-quality schooling available to later cohorts of black
students. To investigate the role of school quality in the convergence of
black and white earnings, we have assembled data on three aspects of school
quality -- pupil/teacher ratios, annual teacher pay, and term length -- for
black and white schools in l8 segregated states from 1915 to 1966. The
school quality data are then linked to estimated rates of return to
education for men from different cohorts and states. Improvements in the
relative quality of black schools explain roughly 20 percent of the
narrowing of the black-white earnings gap in this period.

Racial segregation is often blamed for some of the achievement gap between blacks and whites. We study
the effects of school and neighborhood segregation on the relative SAT scores of black students across
different metropolitan areas, using large microdata samples for the 1998-2001 test cohorts. Our models
include detailed controls for the family background of individual test-takers, school-level controls for
selective participation in the test, and city-level controls for racial composition, income, and region. We
find robust evidence that the black-white test score gap is higher in more segregated cities. Holding
constant family background and other factors, a shift from a fully segregated to a completely integrated
city closes about one-quarter of the raw black-white gap in SAT scores. Specifications that distinguish
between school and neighborhood segregation suggest that neighborhood segregation has a consistently
negative impact but that school segregation has no independent effect (though we cannot reject equality of
the two effects). We find similar results using Census-based data on schooling outcomes for youth in
different cities. Data on enrollment in honors courses suggest that within-school segregation increases
when schools are more highly integrated, potentially offsetting the benefits of school desegregation and
accounting for our findings.

During the 1980s a substantial gap emerged between unemployment rates in Canada and the
United States. In this paper, we use microdata from labor force surveys at the beginning and
end of the decade to examine the sources of the emergent gap. As in earlier work, we find that
most of the relative rise in unemployment in Canada is attributable to an increase in the relative
"labor force attachment" of Canadians, rather than to any shortfall in relative employment.
Indeed, relative employment rates increased in Canada over the 1980s for younger workers and
for adult women. The relative rise in labor force attachment of Canadians is manifested by a
sharp increase in the propensity of non-workers to report themselves as unemployed (i.e. looking
for work) rather than out-of-the-labor force. This change is especially pronounced for individuals
who work just enough to qualify for unemployment insurance (UI) in Canada. Moreover, two-
thirds of the relative increase in weeks of unemployment among non-workers is associated with
the divergent trends in UI recipiency in the two countries. Both findings point to the availability
of UI benefits as an important determinant of the labor force attachment of nonworkers.

This paper estimates the effects of school quality -- measured by the
pupil-teacher ratio, the average term length, and the relative pay of
teachers -- on the rate of return to education for men born between 1920 and
1949. Using earnings data from the 1980 Census, we find that men who were
educated in states with higher quality schools have a higher return to
additional years of schooling, holding constant their current state of
residence, their state of birth, the average return to education in the
region where they currently reside, and other factors. A decrease in the
pupil-teacher ratio from 30 to 25, for example, is associated with a 0.4
percentage point increase in the rate of return to education. The estimated
relationship between the return to education and measures of school quality
is similar for blacks and whites. Since improvements in school quality for
black students were mainly driven by political and judicial pressures, we
argue that the evidence for blacks reinforces a causal interpretation of the
link between school quality and earnings. We also find that returns to
schooling are higher for students educated in states with a higher fraction
of female teachers, and in states with higher average teacher education.
Holding constant school quality measures, however, we find no evidence that
parental income or education affects state-level rates of return.

This paper considers the importance of minimum hours thresholds
for the interpretation of individual labor supply data. An analysis
of quarterly labor supply outcomes for prime-age males in the Survey
of Income and Program Participation suggests that such thresholds are
an important aspect of weekly hours choices. A simple contracting
model is presented that incorporates mobility costs and a non-
convexity in the relation between weekly hours and effective labor
input. This non-convexity gives rise to a two-part employment
schedule. In periods of low demand, some individuals are temporarily
laid off, while others work a minimum threshold level of hours. In
periods of higher demand all available workers are employed at hours
in excess of the threshold level. The model provides a simple
interpretation for the role of demand-side variables in explaining
annual labor supply outcomes. It can also explain the weak
correlations between annual hours and average hourly earnings that
have emerged in earlier studies. Under suitable assumptions on
preferences the intertemporal labor supply elasticity can be
recovered from the relationship between earnings and hours per week.
Estimation results for the SIPP panel yield elasticity estimates that
are similar to those in the literature based on annual data. If the
model is correct, however, annual labor supply is considerably more
sensitive to changes in productivity than these estimates suggest.