Linkages among welfare, food assistance programmes and labour supply: evidence from the survey of programme dynamics

Abstract

At the centre of the debate on welfare reform is whether incentives-based systems help poor families move out of poverty into jobs. Recent data from the first Survey of Programme Dynamics longitudinal and the 1998 experimental data files allow evaluation of the interaction among the Food Stamp Programme (FSP), Temporary Assistance for Needy Families (TANF) programme and labour force participation under different programme design and economic conditions. A model that incorporates jointly determined programmes participation and work decisions is applied to a sample of working age, lower income and asset households. Participation in TANF increases the probability of FSP participation and decreases the probability of being in the work force; working decreases the probability of receiving food stamps. Work, TANF, and FSP participation are related, and TANF and FSP participation rates among poor households who are potentially eligible for TANF are sensitive to changes in programme parameters; as expected, labour force participation is affected by the general economic conditions.

Acknowledgements

The authors thank Wally Huffman, Maureen Kilkenny, Brent Kreider, Peter Orazem and seminar participants at Iowa State University for helpful comments. This research was supported in part by a small competitive grant from the Northwestern University/University of Chicago Joint Center for Poverty Research.

Notes

Hagstrom (1996) uses data from the 1984 Survey of Income and Programme Participation (SIPP) to model the effect of food stamp participation on the family labour supply of married couples and finds that the FSP has a weak effect on the labour supply of married couples. Decreasing the FSP benefit by 25% would reduce the proportion of husbands and wives who choose not to work by less than 1%; increasing the food stamp benefit by 25% would raise the FSP participation by 7%. Keane and Moffitt (1998) estimate that a reduction of the AFDC tax rate has very little effect on labour supply and it increases the participation rate in both AFDC and food stamps. In contrast, a study by Hoynes (1996) on the effects of cash transfers on labour supply and welfare participation in two-parent families finds that labour supply and welfare participation are highly responsive to changes in the benefit structure under the Aid to Families with Dependent Children–Unemployed Parent (AFDC-UP) programme.

With changes in eligibility rules under PRWORA, the Earned Income Tax Credit (EITC) has played an increasingly important role recently in determining the eligibility and benefit levels for TANF. In essence, the income test for TANF requires that net family income not exceed a maximum benefit level that varies by family size and state of residence. Throughout the 1990s, the EITC benefits have increased, and now offer larger incentives or disincentives to work (depending on where a household falls within the range of credit). All but three states (Alabama, Connecticut, and North Dakota) disregard EITC lump-sum payments in the calculation of net income; and all but two states (Connecticut, North Dakota) disregard EITC advance payments. Because data on EITC use are not available in the SPD data, the EITC is not included in the empirical model.

Recipients are required to work, and most can collect aid for no more than five years over a lifetime. TANF recipients must secure a job after two continuous years on assistance. The legislation required states to have at least 25% of single-parent-headed households and 75% of two-parent households engaged in work activities by 1997. Single parents receiving TANF benefits were required to work at least 20 hours per week by 1997 and 30 hours per week by 2000; two-parent families were required to work 35 hours per week with the stipulation that parents can share the work hours.

States increase the severity of the sanction based on the amount of time or the number of times the individual is non-compliant. States may not use TANF funds to provide assistance to a family that includes an adult who has received assistance for more than five years, and the state may set a time limit of less than five years. The state is allowed to exempt up to 20% of its caseload due to no job availability or high unemployment rate, age of parent, disability or illness, etc. For detailed discussion of the exemptions from time limits please see Gallagher et al., 1998, p. IV-1.

The model is a so-called pure system where the simultaneity is in all latent indexes. The pure system was chosen and not a mixed system where some of the latent indexes depend on actual values of one or more outcomes. One of the reasons is that for labour participation only a binary variable is used indicating participation or not. The information on labour hours may not be accurately reported in the data.

First, the reduced form or estimate π _l , π _t , π _f are estimated by maximum-likelihood applied to each equation and form the instruments

,

and

. Second,

,

, and

are replaced on the right-hand side of the structural equations by the corresponding

,

,

and these instruments treated as fixed regressors and the resulting equations as single equation models.

1997 may be too close to PRWORA for observing the equilibrium effects of programme changes. Gleason et al. (2001) estimate that one-fourth of the decline in the FSP caseload between 1994 and 1999 was due to a one-time effect of PRWORA. The authors attribute this effect to changes in perceptions after welfare reform and not to any PRWORA policy. Changes in stigma were severe right after welfare reform. By examining 1997 the analysis may capture too much of the post-PRWORA peak in stigma.

The coefficient with respect to the unemployment rate variable was not statistically significant so the results should be interpreted with caution.