requirement to an optimal tax-transfer scheme is welfare-improving. Of course, these
conditions are expressed in terms of the costs and benefits of workfare. One cost of
workfare is the direct utility losses of those who must do the required work. But workfare
has behavioral effects, too. By increasing the ceteris paribus cost of remaining out of the
labor force, workfare induces workforce participation. To a first order, this is a matter
of indifference to marginal participants. However, increasing the size of the workforce
affects the public budget. If marginal participants pay a net tax, workfare eases the
public budget constraint and generates gains that can be weighed against the direct
utility loss. However, if marginal participants receive a net wage subsidy, as would be
in the case if the optimal tax-transfer system features an earned income tax credit, then
the behavioral effect of workfare actually worsens the public budget. Thus, the presence
of an earned income tax credit weakens the case for workfare when labor supply choices
are along the extensive margin.
Most of the existing literature on workfare as part of an optimal tax-transfer mix
follows Besley and Coate (1992, 1995) by modeling labor supply with variable hours of
work. In this framework, workfare helps to dissuade highly productive workers from
claiming benefits because it is the highly productive that have the highest opportunity
cost of time. As Brett (1998) notes, this channel is not available when claimants are out-
of-work, because the opportunity cost of voluntarily unemployed labor is the marginal
rate of substitution between labor and consumption, which does not depend on labor
market productivity.
Cuff (2000) introduces preference heterogeneity in a framework with intensive labor
supply responses. In some variants of her model, it is desirable to give public benefits to
individuals with a low preference for leisure. For the obvious reason, workfare helps to
screen out would-be claimants with a higher preference for leisure. Cuff also notes that
the same desire to screen out the “lazy” provide the government with a reason to provide
marginal wage subsides.1 Thus, along the intensive margin, workfare and wage subsidies
1 Due to the countervailing force of incentive effects operating along the skill dimension, the optimal