Expanding the Discourse on Antipoverty Policy: Reconsidering a Negative Income Tax

Abstract

This article proposes that advocates for the poor consider the replacement of the current means-tested safety net in the United States with a Negative Income Tax (NIT), a guaranteed income program that lifts families’ incomes above a minimum threshold. The article highlights gaps in service provision that leave millions in poverty, explains how a NIT could help fill those gaps, and compares current expenditures on major means-tested programs to estimated expenditures necessary for a NIT. Finally, it addresses the financial and political concerns that are likely to arise in the event that a NIT proposal gains traction among policy makers.

KEYWORDS:

• poverty alleviation
• public policy
• economic well-being
• social welfare policy

Notes

1. We include the Earned Income Tax Credit (EITC), the Supplemental Nutrition Assistance Program (SNAP), values for free and reduced-price School Lunch, and public housing and housing subsidy values. These estimates were generated using the Census Bureau’s CPS table creator, http://www.census.gov/cps/data/cpstablecreator.html.

2. Earned income also includes nonwage income such as interest, capital gains, and so on.

3. The five experiments tested seven different income thresholds and nine different marginal tax rates. As Widerquist (2005) noted, the decision to utilize such variation meant that the number of observations for each configuration was small compared to the total number of participants. As a consequence, the statistical reliability of the results was reduced.

4. Block and Manza (1997) also argued that any work disincentive effects for those currently in the labor force have to be weighed against the positive incentives for those who are currently provided transfer payments under categorical programs in which strict income limits mean that a recipient’s earnings above a certain level are effectively taxed at a rate of 100%.

5. As Block and Manza (1997) noted, the effects on consumer and labor demand should be stronger in poor communities.

6. We estimate NIT expenditures for 2004 rather than 2007 because the sample size in the final waves of the 2004 SIPP panel was greatly reduced for budgetary reasons. Thus, using calendar year 2004 allows us to base our estimates on a far greater sample.

7. We take the 12.5% as roughly in line with previous estimates of the work disincentive, averaged across adults (Robins, 1985; Burtless, 1986; Keely, 1981).

8. The weight for each family in the sample provides an estimate of the number of families in the civilian non-institutionalized population in the United States represented by that particular family. Because selection probabilities, response rates, and coverage rates vary across subpopulations, different families represent different proportions of the population. Using weights compensates for this differential representation and allows us to produce estimates for the entire civilian non-institutionalized population. Respondents are assigned calendar year weights if they “are present in January and interviewed (or imputed) for every month of the year that they were ‘in scope’ for the survey” (Westat & Mathematica, 2001, p. 9-8). Thus, respondents that drop out of the sample due to nonresponse are not assigned calendar year weights. However, respondents who leave the sample during the calendar year because they are no longer in scope (due to death or moving outside of the United States) are assigned a calendar year weight.

9. Earned income includes additional sources of income beyond earnings from labor, such as investment revenue, etc.