The Earned Income Tax Credit and Labor Market Participation of Families on Welfare V. Joseph Hotz, UCLA & NBER Charles H. Mullin, Bates & White John Karl Scholz, Wisconsin & NBER What is the Federal EITC? It is a refundable tax credit directed primarily at low-income working families. There is a small credit available to childless taxpayers. There are three ranges to the credit: the phase-in (or subsidy); flat; ; and phase-out (or clawback) ranges. EITC is a large program: $31.5 billion in FY2000 (larger than the combined federal spending on food stamps and TANF). Dollars of EITC 0 2000 4000 6000 EITC Schedule, 2001 0 5000 10000 15000 20000 25000 30000 Dollars of Earned Income Coincident Trends: Are They Related? Expansion of the Earned Income Tax Credit Between 1990 and 1999 Real EITC spending increased from $9.6 billion to $31.9 billion (in 1999 dollars) 2+ Children 1 Child No Children 1
Millions $35,000 $30,000 $25,000 $20,000 $15,000 $10,000 $5,000 $0 Spending on Cash and Near-Cash Means Tested Transfers, 1999 Dollars 1970 1972 1974 1976 1978 1980 1982 1984 Year AFDC/TANF EITC SSI Food Stamps 1986 1988 1990 1992 1994 1996 1998 Coincident Trends (cont.) Employment Rates of Single Women with Children Between 3/1990 and 3/2000 Employment rates of single women with children rose from 55.2% to 73.9%. Rates for female-headed households on AFDC/TANF in previous year also went up during 1990s (see graph) 70 65 60 55 50 45 40 35 30 Percentage of Low-Income Female Heads Employed by Year, California Welfare Sample 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 Coincident Trends (cont.) Standard labor-leisure leisure model: Expansions of (wage) subsidy like EITC should generate increases in employment of low-wage wage workers. Question: : Did EITC play in substantial role in increases in employment of single women with children? 2
Previous Work on Relationship between EITC & Employment Fairly Large Number of Papers on this Issue Dickert,, Houser, Scholz (1995, TPE); Keane & Moffitt (1998, IER). Eissa & Liebman (1996, QJE); Ellwood (2000, NTJ); Meyer & Rosenbaum (2000 NTJ; 2001 QJE); Grogger (2003, ReStat) Hoynes and Eissa (2001, wp) ) examine EITC employment effects for 2-parent families. All find Positive, Large EITC effects on employment. Employment elasticities with respect to net income of 0.69 to 1.16. 16. (See Hotz & Scholz, 2003 for full survey of these results.) All but first two papers use Diff-in in-diff approach. Use episodic expansions in EITC and compare changes between groups who were eligible and not eligible for EITC (e.g., single mothers vs. single women). Potential Concerns about Inferences drawn from Previous EITC Employment Studies A. Use of national episodic expansions of EITC to explain national trends vulnerable to possibility that other things changed. Secular Changes in Welfare programs (AFDC/TANF, Food Stamps, Child Care subsidies) Aggregate labor market conditions could be driving changes in employment rates. $35,000 Spending on Cash and Near-Cash Means Tested Transfers, 1999 Dollars $60 Annual Real Earnings per Worker in Service Sector in California, 1992-2000 $30,000 $25,000 $50 Millions $20,000 $15,000 $10,000 $5,000 $0 1970 1972 1974 1976 1978 1980 1982 1984 Year AFDC/TANF EITC SSI Food Stamps 1986 1988 1990 1992 1994 1996 1998 1000s of $1999 $40 $30 $20 1992 1993 1994 1995 1996 1997 1998 1999 2000 Bay Area Counties Central Valley Counties Central & Southern Farm Counties Los Angeles County Northern and Mountain Counties Southern Calif. Counties, Other than LA All Counties 3
0.80 0.75 0.70 0.65 0.60 0.55 0.50 Annual Employment to Population Ratios in California, 1992-2000 1992 1993 1994 1995 1996 1997 1998 1999 2000 Bay Area Counties Central Valley Counties Central & Southern Farm Counties Los Angeles County Northern and Mountain Counties Southern Calif. Counties, Other than LA All Counties Potential Concerns about Inferences (cont.) B. Use of Second Diff in Diff-in in-diff strategy requires composition of comparison group doesn t change over time. Previous studies use Repeated Cross- Sectional data typically from CPS in Diff- in-diff analyses. Population we analyze single mothers on welfare in California during 1990s Sizeable changes in racial composition, family structure and other characteristics. Fraction of Cases that are White Fraction of Cases that are Black Fraction of Cases that are Hispanic Fraction of Cases that are Asian Number of Kids in Hshld. Fraction of Hsehlds with 2+ Kids Fraction of Children Less than Age 6 Year 1991 0.46 0.23 0.21 0.09 1.92 0.55 0.44 1992 0.47 0.23 0.20 0.08 1.93 0.54 0.47 1993 0.45 0.22 0.23 0.09 1.91 0.53 0.49 1994 0.43 0.21 0.28 0.07 1.89 0.54 0.47 1995 0.42 0.20 0.30 0.08 1.85 0.53 0.46 1996 0.41 0.20 0.30 0.07 1.85 0.52 0.48 1997 0.39 0.21 0.31 0.07 1.84 0.51 0.48 1998 0.37 0.22 0.33 0.07 1.84 0.51 0.48 1999 0.36 0.24 0.33 0.06 1.86 0.51 0.47 2000 0.30 0.26 0.35 0.07 1.87 0.51 0.49 % Chge., 1991-2000 -35.3% 17.2% 69.1% -18.1% -2.9% -6.6% 11.5% Potential Concerns about Inferences (cont.) C. If EITC expansion truly caused increases in employment rates of single mothers, should see similar systematic changes in rates of EITC take-up, i.e., claiming EITC on tax returns Analogous to studies of effects of welfare & other social programs on employment Look for changes in program participation to corroborate program effects on employment. Systematic examination of relationship btwn.. EITC expansions & differences in EITC take-up rates btwn. treatment & comparison groups has not been done. 4
Contributions of this Paper 1. We use data from a single state (California) to mitigate influence of secular changes in social policies & local labor market conditions. Over period we examine, low-income populations subject to limited set of policy changes. Better able to control for changes in state policy,, some of which vary at county level. Also control for detailed set of county-level measures of labor market conditions to capture local conditions more accurately. Contributions of Paper (cont.) 2. We exploit longitudinal data on households and focus on temporal within household changes to control for potential composition bias problem in Diff-in in-diff estimation strategy. Use longitudinal data on households in estimation. Contributions of Paper (cont.) 3. Use different Diff-in in-diff identification strategy than in most previous work. Compare differential behavior of families with 2+ children vs. 1-child 1 families before & after EITC expansion in 1990s. EITC expansion in 1994 substantially increased generosity of EITC for 2+ children vs. 1-child 1 households. Table 1: Earned Income Tax Credit Parameters, 1987-2000 (in nominal dollars) Diff. in Max Phase- Credit: In Rate Phase-In Max 2+ - 1 Phase-Out Phase-Out Year (%) Range Credit Child Rate (%) Range 1987 14.0 0-6,080 851 10.0 6,920 15,432 1988 14.0 0-6,240 874 10.0 9,840 18,576 1989 14.0 0-6,500 910 10.0 10,240 19,340 1990 14.0 0-6,810 953 10.0 10,730 20,264 1991 a 16.7 0-7,140 1,192 11.93 11,250 21,250 17.3 1,235 43 12.36 11,250 21,250 1992 a 17.6 0-7,520 1,324 12.57 11,840 22,370 18.4 1,384 60 13.14 11,840 22,370 1993 a 18.5 0-7,750 1,434 13.21 12,200 23,050 19.5 1,511 77 13.93 12,200 23,050 1994 23.6 0-7,750 2,038 15.98 11,000 23,755 30.0 0-8,245 2,528 490 17.68 11,000 25,296 7.65 0-4,000 306 7.65 5,000 9,000 1995 34.0 0-6,160 2,094 15.98 11,290 24,396 36.0 0-8,640 3,110 1,016 20.22 11,290 26,673 7.65 0-4,100 314 7.65 5,130 9,230 1996 34.0 0-6,330 2,152 15.98 11,610 25,078 40.0 0-8,890 3,556 1,404 21.06 11,610 28,495 7.65 0-4,220 323 7.65 5,280 9,500 1997 34.0 0-6,500 2,210 15.98 11,930 25,750 40.0 0-9,140 3,656 1,446 21.06 11,930 29,290 7.65 0-4,340 332 7.65 5,430 9,770 1998 34.0 0-6,680 2,271 15.98 12,260 26,473 40.0 0-9,390 3,756 1,485 21.06 12,260 30,095 7.65 0-4,460 341 7.65 5,570 10,030 1999 34.0 0-6,800 2,312 15.98 12,460 26,928 40.0 0-9,540 3,816 1,504 21.06 12,460 30,580 7.65 0-4,530 347 7.65 5,670 10,200 2000 34.0 0-6,920 2,353 15.98 12,690 27,413 40.0 0-9,720 3,888 1,535 21.06 12,690 31,152 7.65 0-4,610 353 7.65 5,770 10,380 5
Contributions of Paper (cont.) 3. We use a different Diff-in in-diff identification strategy (cont.) We systematically assess the validity of implication of this identification strategy EITC policy treated all households with 2 or more children the same (i.e., same credit). So, we should expect to see no difference in outcomes of interest (e.g., employment) between 2+ and 3+ child households. Contributions of Paper (cont.) 4. Focus on effects of EITC on employment for important population. Female-Headed households on welfare sometime during 1990s. Look at behavior of these households both on & off of welfare. Estimating effects of EITC for this population is particularly relevant from public policy perspective. Contributions of Paper (cont.) 5. Most Novel Feature of Paper: Examine differential effects (2+ vs. 1-Kid 1 households) of EITC on incidence of claiming EITC. Exploit access to data on federal tax returns for households in sample over 1990s. If our Diff-in in-diff identification strategy is isolating EITC effects on employment, should see differential rates of EITC claiming by 2+ vs. 1-Kid 1 households before & after expansion. Our Data Combine Several Administrative Sources Monthly AFDC/TANF case records. Demographic information and benefit receipt. Prior information starting in 1987 on benefits come from Medicaid data. Quarterly data from UI system. Measure employment starting in 1986 Federal tax return information. Data from CA Franchise Tax Board beginning in 1990. 6
Our Data Combine Several Administrative Sources County-level (local) local labor market data County-level policy data (from county welfare & training program implementation) Sample exclusions: Child-only cases Cases with more than 2 adults in household We focus most of analyses on AFDC-FG FG cases. Sampling Start random stratified sample of all assistance units on Welfare in California between 1987 & 2000. Drawn by Rand for another evaluation. Sample includes ~ 50% of all cases. Define a sampling date 4 th quarter of a household s spells on welfare. We determine number & ages of children when household on welfare. All cases, on and off welfare, are treated symmetrically. Avoid overweighting long-term welfare recipients in our sample. Sampling (cont.) Utilize two samples of households in our analyses Cross sectional: Employment in year following sampling date. This sample mimics repeated cross-section section data used in previous studies. Longitudinal: Employment in periods -3, -2, -1, and 0 (sampling). Longitudinal data on households allow control for household-specific specific fixed effects, so focus on within household changes to identify EITC effects. Table 2a: Employment Rates (in Percentages) by Family Size, 1991 2000, Cross-Sectional Sample Diff-in-Diff (199x - Cases with Cases with Difference 1991-93 Year All Cases One Child 2+ Children (2+ - One) Average) 1991 36.35 39.43 33.82-5.61 [27,568] [37,999] 1992 34.82 38.35 31.76-6.59 [24,433] [32,354] 1993 35.40 38.89 32.3-6.59 [23,098] [29,532] 1994 40.20 43.65 37.12-6.53-0.31 [24,005] [30,402] (0.66) 1995 44.89 48.44 41.64-6.8-0.58 [25,500] [31,164] (0.66) 1996 48.51 51.03 45.87-5.16 0.79 [25,457] [30,256] (0.67) 1997 55.20 57.73 52.78-4.95 1.27* [24,794] [28,650] (0.68) 1998 61.34 62.32 60.4-1.92 4.31*** [22,479] [26,174] (0.69) 1999 63.60 65.01 62.22-2.79 3.43*** [19,066] [21,973] (0.74) 2000 65.41 66.03 64.84-1.19 5.03*** [19,731] [22,490] (0.74) 7
Table 2b: EITC Claiming (in Percentages) by Family Size, 1991-2000, Cross-Sectional Sample Diff-in-Diff (199x - Cases with Cases with Difference 1991-93 Year All Cases One Child 2+ Children (2+ - One) Average) 1991 24.13 25.48 23.02-2.46 [27,568] [37,999] 1992 22.76 23.80 21.85-1.95 [24,433] [32,354] 1993 22.88 23.90 21.97-1.93 [23,098] [29,532] 1994 27.84 28.87 26.91-1.96 0.17 [24,005] [30,402] (0.59) 1995 32.95 34.11 31.89-2.22-0.09 [25,500] [31,164] (0.61) 1996 35.60 35.83 35.34-0.49 1.69*** [25,457] [30,256] (0.63) 1997 42.34 42.80 41.91-0.89 1.24* [24,794] [28,650] (0.66) 1998 47.46 46.72 48.18 1.46 3.60*** [22,479] [26,174] (0.69) 1999 51.28 50.18 52.35 2.17 4.31*** [19,066] [21,973] (0.75) 2000 51.94 49.80 53.99 4.19 6.32*** [19,731] [22,490] (0.75) Econometric Specifications (Diff-in-Diff) 2000 2000 9 17 Y = φ + αyear _ s + β[2 + Kids Year_ s ] + δkids_ j + ψ KidsAge_ k ict i S t s ict t j ict k itc s= 1992 s= 1994 j= 1 k= 1 55 + κx + λw + ηl + θcounty_ m + ε ic ct ct m c ict m= 1 where Y ict th 1, if at least 1 adult in i household in county c is employed in year t Emp ict = (1) 0, otherwise th 1, if i household in county cfiles tax return and claims EITC in year t ClaimEITC ict = 0, otherwise 2+Kids ict is 2+ kids indicator variable; Kids_j ict is j children in household indicator variable; KidsAge_k ict = # of children in the household age k; X ic time-invariant demographics of household; W ct county-level measures of California s welfare caseload and policies; L ct time-varying measures of county-level labor market conditions; County_m ict indicator variable for residing in county m Econometric Specifications (Diff-in-Diff) (Cont.) In some regressions for testing we also include an (exactly) 2 children indicator variable: 0, if Kidsit < 2 or Kidsit 3 + it + it = 1, if Kidsit = 2 [ 2 Kids 3 Kids ] Covariates Demographic characteristics Number of kids & number of kids by age. Local labor markets Year dummies, employment share by sector, avg. income by sector Welfare rules Proportion of population in GAIN program Time-invariant invariant covariates (in OLS Cross-Sectional Models). Race/ethnicity, county dummies, gender, age, timing of entry onto welfare. 8
Empirical Strategy for Understanding the EITC's Effect Our strategy for assessing validity of estimated EITC s effect on employment: 1. Does employment of families with 2+ children increase relative to 1-Kid 1 families? (They should.) 2. Does employment of 2+Kid families differ from effects for 3+Kid families? (They should not) 3. Do temporal patterns of EITC claiming mirror employment patterns in #1 and #2? (They should) 4. Do see any differences in employment btwn.. 2+Kid and 1-Kid households not filing tax returns?? (We should not!) Similar strategy for #1 - #3 should apply to Claiming EITC. Table 3: Estimates of EITC Effects on Household Employment OLS, Cross- Sectional Sample Household Fixed Effects, Panel Data Variable 2+ Kids in 1994 0.0001-0.0035 (0.0065) (0.0041) 2+ Kids in 1995-0.0063 0.0072 (0.0065) (0.0052) 2+ Kids in 1996 0.0040 0.0123** (0.0066) (0.0061) 2+ Kids in 1997 0.0092 0.0261*** (0.0067) (0.0071) 2+ Kids in 1998 0.0382*** 0.0324*** (0.0070) (0.0083) 2+ Kids in 1999 0.0278*** 0.0341*** (0.0076) (0.0100) 2+ Kids in 2000 0.0425*** 0.0295** (0.0075) (0.0123) No. of Observations 527,125 1,637,855 P-Value for Test of 2+ Kids in 1994-2000 = 0 0.0000 0.0004 Table 4: Estimates of EITC Effects on Whether Household Claimed the EITC on Tax Return OLS, Cross- Sectional Sample Household Fixed Effects, Panel Data Variable 2+ Kids in 1994 0.0041 0.0006 (0.0059) (0.0038) 2+ Kids in 1995-0.0032 0.0065 (0.0061) (0.0049) 2+ Kids in 1996 0.0103* 0.0184*** (0.0062) (0.0058) 2+ Kids in 1997 0.0050 0.0190*** (0.0066) (0.0068) 2+ Kids in 1998 0.0249*** 0.0170** (0.0070) (0.0081) 2+ Kids in 1999 0.0283*** 0.0233** (0.0077) (0.0098) 2+ Kids in 2000 0.0448*** 0.0194 (0.0076) (0.0122) No. of Observations 527,125 1,637,855 P-Value for Test of 2+ Kids in 1994-2000 = 0 0.0000 0.0353 Table 5a: Assessing the Validity of Strategy for Identifying EITC Effects on Household Employment [Family Fixed Effects Estimation on Panel Data] Households that Did Not File Tax Variable All Households Return (1) (2) (3) 2+ Kids in 1994-0.0035-0.0016-0.0020 (0.0041) (0.0051) (0.0065) 2+ Kids in 1995 0.0072 0.0043 0.0028 (0.0052) (0.0067) (0.0086) 2+ Kids in 1996 0.0123** 0.0114-0.0016 (0.0061) (0.0082) (0.0107) 2+ Kids in 1997 0.0261*** 0.0289*** 0.0113 (0.0071) (0.0098) (0.0131) 2+ Kids in 1998 0.0324*** 0.0331*** 0.0048 (0.0083) (0.0116) (0.0161) 2+ Kids in 1999 0.0341*** 0.0368*** 0.0037 (0.0100) (0.0137) (0.0200) 2+ Kids in 2000 0.0295** 0.0291* -0.0009 (0.0123) (0.0168) (0.0257) [2+ Kids 3+ Kids] in 1994-0.0030-0.0009 (0.0051) (0.0062) [2+ Kids 3+ Kids] in 1995 0.0055-0.0010 (0.0063) (0.0080) [2+ Kids 3+ Kids] in 1996 0.0021-0.0014 (0.0074) (0.0096) [2+ Kids 3+ Kids] in 1997-0.0042-0.0012 (0.0086) (0.0115) [2+ Kids 3+ Kids] in 1998-0.0004 0.0005 (0.0100) (0.0141) [2+ Kids 3+ Kids] in 1999-0.0039-0.0071 (0.0119) (0.0181) [2+ Kids 3+ Kids] in 2000 0.0015 0.0077 (0.0149) (0.0244) No. of Observations P-Value for Test of 2+ Kids in 1994-2000 = 0 0.0004 0.0112 0.8423 P-Value for Test of [2+ Kids 3+ Kids] in 1994-2000 = 0 0.7389 0.9989 9
Table 5b: Assessing the Validity of Strategy for Identifying EITC Effects on Whether Household Claimed the EITC on Tax Return [Family Fixed Effects Estimation on Panel Data] Variable (1) (2) 2+ Kids in 1994 0.0006 0.0057 (0.0038) (0.0048) 2+ Kids in 1995 0.0065 0.0131** (0.0049) (0.0064) 2+ Kids in 1996 0.0184*** 0.0233*** (0.0058) (0.0078) 2+ Kids in 1997 0.0190*** 0.0279*** (0.0068) (0.0094) 2+ Kids in 1998 0.0170** 0.0170 (0.0081) (0.0112) 2+ Kids in 1999 0.0233** 0.0344** (0.0098) (0.0134) 2+ Kids in 2000 0.0194 0.0255 (0.0122) (0.0165) [2+ Kids 3+ Kids] in 1994-0.0078* (0.0047) [2+ Kids 3+ Kids] in 1995-0.0100* (0.0059) [2+ Kids 3+ Kids] in 1996-0.0067 (0.0070) [2+ Kids 3+ Kids] in 1997-0.0131 (0.0082) [2+ Kids 3+ Kids] in 1998 0.0033 (0.0097) [2+ Kids 3+ Kids] in 1999-0.0158 (0.0119) [2+ Kids 3+ Kids] in 2000-0.0063 (0.0149) P-Value for Test of 2+ Kids in 1994-2000 = 0 0.0353 0.0174 P-Value for Test of [2+ Kids 3+ Kids] in 1994-2000 = 0 0.0669 Sensitivity Analyses Change base year(s) ) used to measure before- expansion outcomes. [Table 5c] Change thresholds of UI earnings used to define whether household employed. [Table 5d] Control for measure of county expenditures on childcare for welfare recipients to regressions. No qualitative difference in estimated EITC effects on employment or EITC claiming for above variants of our basic results. Sensitivity Analyses (cont.) Examine EITC effects on household employment and EITC Claiming for two- parent households on welfare (AFDC-UP cases). Do not find any evidence of positive employment effects for AFDC-UP households. Perhaps employment barriers are larger for the subset of UP households with no workers. Conclusions We find robust effects of the differential expansion of EITC btwn. 2+Kid vs. 1-Kid 1 households on employment rates & rates of EITC claiming. Our identification strategy for identifying EITC effects is validated several different ways. 10
Conclusions (cont.) EITC increased employment for families with 2+ kids by as much as 3.4 percentage points relative to 1-Kid 1 families. Illustration: 11.8% of 31 percentage point employment increase for families with two or more children between 1991 and 2000. Smaller than Meyer & Rosenbaum (2001) or Grogger (2003) But, another way to look at this is that 77 percent of the differential gain in employment for families with two or more kids. Conclusions (cont.) Implied Elasticity of Employment w.r.t. household disposable income: Average EITC differential for families with two or more children was $439 in 1998. Average disposable income (including transfers) was around $10,000. The EITC increased disposable income around 4.4 percent. Employment rates in 1998 were around 60 percent. The EITC increased the relative employment of families with two or more children 3.2 percentage points, or 5.6 percent. The implied employment elasticity with respect to disposable income is 1.3, which is at upper end of estimates in EITC-Effects Effects-on-Employment Literature. Figure 1: Quarterly Earnings Reported to AFDC Administrators and to the UI System 5000 Earnigs Reported to AFDC 2500 45 Degree Mean Q2 Q3 0 0 2500 5000 Earnings Recorded in Unemployment Insurance Records 11
Table 4: Earnings, Transfers, Disposable Income Earn Trans DispIn Welf Year 1993 1997 1993 1997 1993 1997 1993 1997 Cont $1,553 4,730 $10,363 5,585 $11,989 10,636 98.9% 61.4 Exper $1,606 5,226 $10,160 5,090 $11,836 10,696 98.6% 59.8 Diff $53 496-203 -495-153 60-0.3-1.6 % Cg 3.4 10.5-2.0-8.9-1.3 0.6-0.3-2.6 P-Val 0.316 0.022 0.055 0.001 0.141 0.404 0.135 0.106 12