UBI decreases labor supply

We begin with labor supply, as this effect dominates discussions of the economics of means-tested transfer programs. Traditional welfare programs, with low phaseout points P and high tax rates T, unambiguously lead to reductions in labor supply through negative income and substitution effects. By contrast, the EITC, which has effectively replaced traditional cash welfare as the main income assistance program for families with children, has no transfer for nonworkers (G = 0) and a high phase-in rate (S), so it creates strong incentives to enter work. (For those with positive earnings, the EITC creates both a negative income effect and, in the phase-out range, a negative substitution effect, so it is expected to reduce working hours.) Existing program structures, the shift from AFDC to TANF, and the EITC reflect a general trend in recent decades in the United States toward programs that attempt to minimize labor supply disincentives. These take two forms. First, historically means-tested programs in the United States have used “tagging” (Akerlof 1978), limiting eligibility to those in exogenously defined groups who have low potential to work (or expectation to work). Second, current policies are increasingly designed to avoid punitive tax rates (lower T) and increase earnings disregards (higher P). This can also, as in the case of the EITC, include programs that use a positive S and no income floor (G = 0) to create incentives toward increased labor supply (Nichols & Rothstein 2016). UBI proposals move policy in the opposite direction, and in general they can be expected to reduce labor supply relative either to a no-transfer hypothetical baseline or to the status quo. First, the canonical UBI generates a pure income effect that would reduce work on the extensive and intensive margins. Second, many UBI proposals impose phaseouts, and the added P > 0 and T > 0 lead to a further work disincentive through negative substitution effects.19 Third, the high G in a UBI relative to existing cash welfare programs likely leads to larger labor supply reductions (though the higher P and lower T would arguably work in the opposite direction). Fourth, the absence of tagging means that vastly more people are exposed to these work disincentives than in our current patchwork system.