1. Refund Anticipation Loans and Social Welfare: Race and Income Effects Richard J. Smith, MSW smithrichardj@berkeley.edu

2. What is a Refund Anticipation Loan (RAL)? AKA Rapid Refund. A financial institution loans your refund as soon as your tax preparer calculates it. Refund goes to bank. Taxpayer must pay back loan if the refund is in error. In 1996, US Representative Joseph Kennedy called fees of $30-$100, if translated into effective annual percentage interest rates (APR) "would make a loan shark blush.”

3. Framing the Issue Rational Choice: The customer is simply paying extra for quick money needed today. Advocacy: The customer doesn’t have complete information about charges, alternatives, or actual time of direct deposit of refund. Institutional: The state must fairly and accurately collect taxes, provide equitable redistribution and prevent fraud.

4. The Players Tax preparers (H&R Block, Jackson Hewitt and many more in the Free File Alliance) Consumer Advocates (Association for Community Organizing and Reform Now (ACORN), National Consumer Law Center (NCLC), Children’s Defense Fund Social Work (Sheridan, Abramowitz, Karger, Brooks, Fischer et. el.) Internal Revenue Service (IRS)

5. Social Policy and the Tax System Everyone is still on welfare (Abromowitz, 2001)  HOUSING: Mortgage income deduction vs. Housing Choice Vouchers  HEALTH CARE: Medical Savings Accounts vs. Medicaid  POVERTY: Earned Income Tax Credit (EITC) vs. Temporary Assistance for Needy Families (TANF)

6. Advocacy and Social Welfare Policy The Fringe Economy (Karger, 2005)  Check cashing stores  Pawn shops  Used car financing  Payday loans  Subprime mortgages  Refund Anticipation Loans Our Money Place brought together a credit union, churches and a Check Cashing store.

7. Tax Policy and Social Welfare Connecting the Dots: EITC to Individual Development Accounts (Sherraden, 1991). The Volunteer Income Tax Assistance (VITA) Center offers Free Tax Prep approved by IRS. Communities partner with VITA and local financial institutions to market IDA participants. City of Santa Ana Outreach Daisy Wheel.

8. NCLC Research (2005) Effective APR 97.4% to 2000%. Half of RAL users poor families paid $1.14 billion in fees. December 2005, a NCLC telephone poll (N=2000) determined that while only 17% of white respondents bought a RAL, 21% of Latinos and 28% of African-Americans reported taking a RAL.

9. Research on RAL Think Tanks and Advocates Who buys RALs? Brookings found that filers in low income, high immigrant Zip codes were more likely to see a paid tax preparer, but were less likely to ask for a RAL (Berube, 2005). Brookings that found that the percent of EITC claiming taxpayers who take an RAL increases 1.6% for every 10% increase in the percent African Americans in a Zip code (cited in C. C. Wu & Fox, 2005).

10. Academic Research and Program Evaluation Stearns et. al. (2006) found that 97% of low income subjects retracted their decision to buy an RAL after hearing the interest rate. Brooks et. al. (2006) conducted a mixed method program evaluation of ACORN's VITA outreach in Hispanic and African American low income neighborhoods. They found in a random digit dialing phone survey (N=1,063) that canvassed areas had a greater proportion of persons filing for free at VITA sites than those in the comparison group. Fisher et. al. (2007) documented tactics waged nationally by the Association of Community Organizations for Reform Now (ACORN) that were effective in getting H&R Block to reduce RAL fees.

11. Data and Method RAL data (1999-2004) from the IRS available at Brookings’ website: http://www.brookings.edu/urban/eitc/. http://www.brookings.edu/urban/eitc/ Dependent variable: Percent RAL (Number of RAL taken in a zip code by EITC divided by the number of EITC returns eligible for a refund. Independent Variables: Census Summary File Three 2000 Zip Estimation Data for Race, Ethnicity, Education controlling for the average amount of the EITC refund in each zip code. Average estimate of race association of area Generalized Estimating Equation Unstructured correlation structure among panels 2000 - 2004 with robust sandwich estimator (Zips = 31676). Next, I estimated a regression model for California Zip codes (N=1668). Spatial autocorrelation and trends.

15. Smoothed Median Trend at Zip Level

17. Spatial Autocorrelation (Moran’s I = 0.5004, P <.000)

18. Generalized Estimating Equation 1999-2004

19. Variance Components  Zip codes and year had a strong interclass correlation of 87% (test-retest reliability).  There is more variability within states than between states in years.  Most of the variance is at the zip level--probably individual if we had the data.

20. One for One Race Effects In California, zip codes with a percent greater than baseline of  African-Americans,  persons who identify as having more than one race,  Hispanics, and  Low Income families receiving EITC. Conversely, zip codes with high white and educated populations were also those with low RAL utilization.

21. % RAL/# EITC Returns in California %White as Baseline

22. California %Native, Other or Multi as Baseline

23. Consumer groups recommend banning the use of RALs for the EITC stricter regulation free electronic filing directly with the IRS and deferred payment of tax preparation services banning the use of RALs for the EITC (Wu, C. C. et. al., 2007)

24. Outreach Slides Brookings has an interactive EITC database organizers can 1) increase utilization of the EITC by setting goals by zip, and 2) identify areas where taxpayers can be directed towards IRS VITA sites or free filing in general. http://www.brookings.edu/metro/EITC- Homepage.aspx

25. Brookings EITC Series

26. Interactive Downloadable Tax Data

27. Outreach management Tools ZipcodeCityStateereturn04enew04eeic04eeicam04ectc04ectcam04 94609BerkeleyCA1831825,20742,714 94701BerkeleyCA3003018,49841,938 94702BerkeleyCA9011459011,202,655178120,770 94703BerkeleyCA1,0852001,0851,407,321196118,809 94704BerkeleyCA42770427332,8064526,628 94705BerkeleyCA31742317234,3642717,364 94706BerkeleyCA4564560,18384,546 94707BerkeleyCA14114141109,713147,726 94708BerkeleyCA131813169,01185,712 94709BerkeleyCA33049330202,5832514,487 94710BerkeleyCA37759377568,5518650,678 94712BerkeleyCA3503528,14163,420 94720BerkeleyCA3503542,149104,716

30. IRS VITA Brochures

31. Conclusion Information and public outreach can help low income families make asset building decisions. Social Workers, government, advocates and even private corporations can partner to provide effective services to low income families.