What is the Federal EITC? What is the Federal EITC? The Earned - - PDF document

SLIDE 1

1

The Earned Income Tax Credit The Earned Income Tax Credit and Labor Market Participation of and Labor Market Participation of Families on Welfare Families on Welfare

• V. Joseph Hotz, UCLA & NBER
• V. Joseph Hotz, UCLA & NBER

Charles H. Mullin, Bates & White Charles H. Mullin, Bates & White John Karl Scholz, Wisconsin & NBER John Karl Scholz, Wisconsin & NBER

What is the Federal EITC? What is the Federal EITC?

• It is a

It is a refundable refundable tax credit directed tax credit directed primarily at low primarily at low-

• income working families.

income working families.

• There is a small credit available to childless

There is a small credit available to childless taxpayers. taxpayers.

• There are three ranges to the credit: the

There are three ranges to the credit: the phase phase-

• in

in (or subsidy); (or subsidy); flat flat; and ; and phase phase-

• out
• ut

(or clawback) ranges. (or clawback) ranges.

• EITC is a large program: $31.5 billion in

EITC is a large program: $31.5 billion in FY2000 (larger than the combined federal FY2000 (larger than the combined federal spending on food stamps and TANF). spending on food stamps and TANF).

EI TC Schedule, 2001

2 0 0 0 4 0 0 0 6 0 0 0 5000 10000 15000 20000 25000 30000 Dollars of Earned Income D o l l a r s o f E I T C 2+ Children 1 Child No Children

Spending on Cash and Near-Cash Means Tested Transfers, 1999 Dollars

$0 $5,000 $10,000 $15,000 $20,000 $25,000 $30,000 $35,000 1 9 7 1 9 7 2 1 9 7 4 1 9 7 6 1 9 7 8 1 9 8 1 9 8 2 1 9 8 4 1 9 8 6 1 9 8 8 1 9 9 1 9 9 2 1 9 9 4 1 9 9 6 1 9 9 8 Year M illions AFDC/ TANF EI TC SSI Food Stamps

SLIDE 2

2

Coincident Trends: Are They Coincident Trends: Are They Related? Related?

• Between 1990 and 1999

Between 1990 and 1999

• Real EITC spending increased from $9.6 billion to

Real EITC spending increased from $9.6 billion to $31.9 billion (in 1999 dollars). $31.9 billion (in 1999 dollars).

• Between 3/1990 and 3/2000

Between 3/1990 and 3/2000

• Employment rates of single women with children rose

Employment rates of single women with children rose to 73.9% from 55.2% . to 73.9% from 55.2% .

• Welfare caseloads fell to 2.3 million from 4.1 million

Welfare caseloads fell to 2.3 million from 4.1 million (though caseloads rose until 1994, peaking at 5.0 (though caseloads rose until 1994, peaking at 5.0 million families). million families).

• Standard static labor supply model:

Standard static labor supply model:

Expansions of a wage subsidy like EITC should

increase employment of low-skilled workers.

Previous Work on Relationship Previous Work on Relationship between EITC & Employment between EITC & Employment

• Large number of published papers on this issue

Large number of published papers on this issue

• Dickert, Houser, Scholz (1995, TPE); Keane & Moffitt (1998, IER)

Dickert, Houser, Scholz (1995, TPE); Keane & Moffitt (1998, IER). .

• Eissa & Liebman (1996, QJE); Ellwood (2000, NTJ); Meyer &

Eissa & Liebman (1996, QJE); Ellwood (2000, NTJ); Meyer & Rosenbaum (2000 NTJ; 2001 QJE); Grogger (2003, ReStat) Rosenbaum (2000 NTJ; 2001 QJE); Grogger (2003, ReStat)

• All find

All find positive, large EITC effects on employment. EITC effects on employment.

• Employment elasticities with respect to net income of 0.69 to 1.

Employment elasticities with respect to net income of 0.69 to 1.16. 16.

• (See Hotz & Scholz, 2003 for survey of these results.)

(See Hotz & Scholz, 2003 for survey of these results.)

• Hoynes and Eissa (2004, JPubE) find negative EITC employment eff

Hoynes and Eissa (2004, JPubE) find negative EITC employment effects ects for secondary workers in 2 for secondary workers in 2-

• parent families.

parent families.

• All but first two papers use “Diff

All but first two papers use “Diff-

• in

in-

• Diff” approach.

Diff” approach.

• Use episodic “expansions” in EITC and compare changes between

Use episodic “expansions” in EITC and compare changes between groups who were “eligible” and “not eligible” for EITC (e.g., si groups who were “eligible” and “not eligible” for EITC (e.g., single ngle mothers vs. single women). mothers vs. single women).

Given All the Previous Work, Why Given All the Previous Work, Why Another EITC Employment Study? Another EITC Employment Study?

• Other trends

Other trends

• Hard to account for with repeated, cross

Hard to account for with repeated, cross-

• sectional data

sectional data

• Welfare programs (AFDC/TANF, Food Stamps, Child Care

Welfare programs (AFDC/TANF, Food Stamps, Child Care subsidies) subsidies)

• Aggregate labor market conditions could be driving changes in

Aggregate labor market conditions could be driving changes in employment rates. employment rates.

• Use of “Second Diff” in “Diff

Use of “Second Diff” in “Diff-

• in

in-

• Diff” strategy requires

Diff” strategy requires composition of “comparison group” doesn’t change composition of “comparison group” doesn’t change

• ver time.
• ver time.
• Previous studies use

Previous studies use Repeated Cross Repeated Cross-

• Sectional

Sectional data data – – typically typically from CPS from CPS – – in Diff in Diff-

• in

in-

• Diff analyses.

Diff analyses.

• Population we analyze

Population we analyze – – single mothers on welfare in single mothers on welfare in California during 1990s California during 1990s – –

• Sizeable changes in racial composition, family structure and oth

Sizeable changes in racial composition, family structure and other er characteristics. characteristics.

• We have new data and a new approach that (we

We have new data and a new approach that (we argue) provides more reliable evidence on the EITC’s argue) provides more reliable evidence on the EITC’s employment effects. employment effects.

Y Ye ea ar r F Fr ra ac ct ti io

• n

n

• f

f C Ca as se es s t th ha at t a ar re e W Wh hi it te e F Fr ra ac ct ti io

• n

n

• f

f C Ca as se es s t th ha at t a ar re e B Bl la ac ck k F Fr ra ac ct ti io

• n

n

• f

f C Ca as se es s t th ha at t a ar re e H Hi is sp pa an ni ic c F Fr ra ac ct ti io

• n

n

• f

f C Ca as se es s t th ha at t a ar re e A As si ia an n N Nu um mb be er r

• f

f K Ki id ds s i in n H Hs sh hl ld d. . F Fr ra ac ct ti io

• n

n

• f

f H Hs se eh hl ld ds s w wi it th h 2 2+ + K Ki id ds s F Fr ra ac ct ti io

• n

n

• f

f C Ch hi il ld dr re en n L Le es ss s t th ha an n A Ag ge e 6 6 1 19 99 91 1 0. .4 46 6 0. .2 23 3 0. .2 21 1 0. .0 09 9 1 1. .9 92 2 0. .5 55 5 0. .4 44 4 1 19 99 92 2 0. .4 47 7 0. .2 23 3 0. .2 20 0. .0 08 8 1 1. .9 93 3 0. .5 54 4 0. .4 47 7 1 19 99 93 3 0. .4 45 5 0. .2 22 2 0. .2 23 3 0. .0 09 9 1 1. .9 91 1 0. .5 53 3 0. .4 49 9 1 19 99 94 4 0. .4 43 3 0. .2 21 1 0. .2 28 8 0. .0 07 7 1 1. .8 89 9 0. .5 54 4 0. .4 47 7 1 19 99 95 5 0. .4 42 2 0. .2 20 0. .3 30 0. .0 08 8 1 1. .8 85 5 0. .5 53 3 0. .4 46 6 1 19 99 96 6 0. .4 41 1 0. .2 20 0. .3 30 0. .0 07 7 1 1. .8 85 5 0. .5 52 2 0. .4 48 8 1 19 99 97 7 0. .3 39 9 0. .2 21 1 0. .3 31 1 0. .0 07 7 1 1. .8 84 4 0. .5 51 1 0. .4 48 8 1 19 99 98 8 0. .3 37 7 0. .2 22 2 0. .3 33 3 0. .0 07 7 1 1. .8 84 4 0. .5 51 1 0. .4 48 8 1 19 99 99 9 0. .3 36 6 0. .2 24 4 0. .3 33 3 0. .0 06 6 1 1. .8 86 6 0. .5 51 1 0. .4 47 7 2 20 00 00 0. .3 30 0. .2 26 6 0. .3 35 5 0. .0 07 7 1 1. .8 87 7 0. .5 51 1 0. .4 49 9 % % C Ch hg ge e. ., , 1 19 99 91 1-

• 2

20 00 00

• 3

35 5. .3 3% % 1 17 7. .2 2% % 6 69 9. .1 1% %

• 1

18 8. .1 1% %

• 2

2. .9 9% %

• 6

6. .6 6% % 1 11 1. .5 5% %

SLIDE 3

3

Contributions of this Paper Contributions of this Paper

• We use data from a single state (California) to

We use data from a single state (California) to mitigate influence of secular changes in social mitigate influence of secular changes in social policies and local labor market conditions. policies and local labor market conditions.

• Over period we examine, low

Over period we examine, low-

• income populations

income populations subject to limited set of policy changes. subject to limited set of policy changes.

• Better able to control for changes in

Better able to control for changes in state policy state policy, , some of which vary at county level. some of which vary at county level.

• Also control for detailed set of county

Also control for detailed set of county-

• level measures

level measures

• f
• f labor market conditions

labor market conditions to capture local conditions to capture local conditions more accurately. more accurately.

Contributions of This Paper Contributions of This Paper (continued) (continued)

• We are the first to use longitudinal data

We are the first to use longitudinal data

• n households when looking at the EITC’s
• n households when looking at the EITC’s

employment effects. This has two employment effects. This has two benefits. benefits.

• We can account for unobserved, household

We can account for unobserved, household-

• specific time

specific time-

• invariant factors affecting

invariant factors affecting employment. employment.

• We can also avoid bias due to unobserved

We can also avoid bias due to unobserved changes in the “treatment” and “comparison” changes in the “treatment” and “comparison” groups that might arise in repeated cross groups that might arise in repeated cross-

• sectional studies.

sectional studies.

Contributions: A 4 Contributions: A 4-

• Part Test of the

Part Test of the EITC’s Employment Effects EITC’s Employment Effects

• Test # 1:

Test # 1: We use different “Diff We use different “Diff-

• in

in-

• Diff”

Diff” identification strategy than in most previous identification strategy than in most previous work. work.

• Compare differential behavior of families with 2+

Compare differential behavior of families with 2+ children vs. 1 children vs. 1-

• child families before & after EITC

child families before & after EITC expansion in 1990s. expansion in 1990s.

• EITC expansion in 1994 substantially increased

EITC expansion in 1994 substantially increased generosity of EITC for 2+ children vs. 1 generosity of EITC for 2+ children vs. 1-

• child

child households. households.

• The groups we compare are likely more similar than women

The groups we compare are likely more similar than women with and without children. with and without children.

13,090 13,090 -

• 28,281

28,281 13,090 13,090 -

• 32,121

32,121 5,950 5,950 -

• 10,708

10,708 15.98 15.98 21.06 21.06 7.65 7.65 2,428 2,428 4,008 4,008 364 364 0-

• 7,140

7,140 0-

• 10,020

10,020 0-

• 4,760

4,760 34.0 34.01

1

40.0 40.02

2

7.65 7.653

3

2001 2001 11,610 11,610 -

• 25,078

25,078 11,610 11,610 -

• 28,495

28,495 5,280 5,280 -

• 9,500

9,500 15.98 15.98 21.06 21.06 7.65 7.65 2,152 2,152 3,556 3,556 323 323 0-

• 6,330

6,330 0-

• 8,890

8,890 0-

• 4,220

4,220 34.0 34.01

1

40.0 40.02

2

7.65 7.653

3

1996 1996 11,290 11,290 -

• 24,396

24,396 11,290 11,290 -

• 26,673

26,673 5,130 5,130 -

• 9,230

9,230 15.98 15.98 20.22 20.22 7.65 7.65 2,094 2,094 3,110 3,110 314 314 0-

• 6,160

6,160 0-

• 8,640

8,640 0-

• 4,100

4,100 34.0 34.01

1

36.0 36.02

2

7.65 7.653

3

1995 1995 11,000 11,000 -

• 23,755

23,755 11,000 11,000 -

• 25,296

25,296 5,000 5,000 -

• 9,000

9,000 15.98 15.98 17.68 17.68 7.65 7.65 2,038 2,038 2,528 2,528 306 306 0-

• 7,750

7,750 0-

• 8,245

8,245 0-

• 4,000

4,000 23.6 23.61

1

30.0 30.02

2

7.65 7.653

3

1994 1994 12,200 12,200 -

• 23,050

23,050 12,200 12,200 – – 23,050 23,050 13.21 13.21 13.93 13.93 1,434 1,434 1,511 1,511 0-

• 7,750

7,750 18.5 18.51

1

19.5 19.52

2

1993 1993a

a

11,250 11,250 -

• 21,250

21,250 11,250 11,250 – – 21,250 21,250 11.93 11.93 12.36 12.36 1,192 1,192 1,235 1,235 0-

• 7,140

7,140 16.7 16.71

1

17.3 17.32

2

1991 1991a

a

10,730 10,730 – – 20,264 20,264 10.0 10.0 953 953 0-

• 6,810

6,810 14.0 14.0 1990 1990 6,920 6,920 – – 15,432 15,432 10.0 10.0 851 851 0-

• 6,080

6,080 14.0 14.0 1987 1987 6,500 6,500 – – 11,000 11,000 12.22 12.22 550 550 0-

• 5,000

5,000 11.0 11.0 1985 1985-

• 86

86 $4,000 $4,000 -

• $8,000

$8,000 10.0 10.0 $400 $400 $0 $0-

• $4,000

$4,000 10.0 10.0 1975 1975-

• 78

78 Phase Phase-

• Out

Out Range Range Phase Phase-

• Out

Out Rate (%) Rate (%) Max Max Credit Credit Phase Phase-

• In

In Range Range Phase Phase-

• In

In Rate (%) Rate (%) Year Year Table 1: Earned Income Tax Credit Parameters, 1979 Table 1: Earned Income Tax Credit Parameters, 1979-

• 2001 (in nominal dollars)

2001 (in nominal dollars)

SLIDE 4

4

Our 4 Our 4-

• Part Test, cont.

Part Test, cont.

• The most novel aspect of our work is having

The most novel aspect of our work is having (highly restricted) access to the federal tax (highly restricted) access to the federal tax returns filed by the California households. returns filed by the California households.

• Test 2:

Test 2: We examine differential patterns (2+ vs. We examine differential patterns (2+ vs. 1 1-

• Kid households) of EITC claiming.

Kid households) of EITC claiming.

• If our “Diff

If our “Diff-

• in

in-

• Diff” identification strategy is isolating

Diff” identification strategy is isolating EITC effects on employment, we should see EITC effects on employment, we should see differential rates of EITC claiming by 2+ vs. 1 differential rates of EITC claiming by 2+ vs. 1-

• Kid

Kid households before and after expansion. households before and after expansion.

• Analogous to studies of effects of welfare & other social

Analogous to studies of effects of welfare & other social programs on employment where researchers look for programs on employment where researchers look for changes in program participation to corroborate estimates of changes in program participation to corroborate estimates of the program on employment. the program on employment.

Our 4 Our 4-

• Part Test, cont.

Part Test, cont.

• The strong economy or other factors could,

The strong economy or other factors could, possibly, result in differential employment possibly, result in differential employment effects for families of different sizes. effects for families of different sizes.

• Test 3: we should expect to see no difference in

Test 3: we should expect to see no difference in employment between families with exactly 2 and employment between families with exactly 2 and 3 or more children. 3 or more children.

• The EITC treats all households with two or more

The EITC treats all households with two or more children equally. Hence, if the EITC is causing the children equally. Hence, if the EITC is causing the employment changes we observe, they should apply employment changes we observe, they should apply to families with one child relative to families with two to families with one child relative to families with two

• r more, but not to families with 2 children relative to
• r more, but not to families with 2 children relative to

families with 3 or more. families with 3 or more.

Our 4 Our 4-

• Part Test, cont.

Part Test, cont.

• Test 4: If the EITC is causing the

Test 4: If the EITC is causing the employment changes we observe, we employment changes we observe, we should not see the these patterns for should not see the these patterns for those who do not file tax returns. those who do not file tax returns.

• The point of tests 3 and 4 is that if it is

The point of tests 3 and 4 is that if it is something other than the EITC accounting for something other than the EITC accounting for employment changes in the 1990s… employment changes in the 1990s…

• Why would it only affect those who file tax returns

Why would it only affect those who file tax returns (test 4), and (test 4), and

• Not affect families with 3 or more children relative

Not affect families with 3 or more children relative to families with exactly 2 children (test 3). to families with exactly 2 children (test 3).

Our Data Combine Several Our Data Combine Several Administrative Sources Administrative Sources

• Monthly AFDC/TANF case records.

Monthly AFDC/TANF case records.

• Demographic information and benefit receipt.

Demographic information and benefit receipt.

• Prior information starting in 1986 on benefits come from

Prior information starting in 1986 on benefits come from Medicaid data. Medicaid data.

• Quarterly data from UI system.

Quarterly data from UI system.

• Measure employment starting in 1991 through 2000.

Measure employment starting in 1991 through 2000.

• Federal tax return information.

Federal tax return information.

• Data from CA Franchise Tax Board beginning in 1991 to 2000.

Data from CA Franchise Tax Board beginning in 1991 to 2000.

• We supplement the core AFDC, UI and Tax data with

We supplement the core AFDC, UI and Tax data with

• County

County-

• level (local) local labor market data

level (local) local labor market data

• County

County-

• level policy data (from county welfare & training

level policy data (from county welfare & training program implementation) program implementation)

SLIDE 5

5

Sampling Sampling

• Start random stratified sample of all assistance units on

Start random stratified sample of all assistance units on Welfare in California between 1987 & 2000. Welfare in California between 1987 & 2000.

• Drawn by Rand for another evaluation.

Drawn by Rand for another evaluation.

• Sample includes ~ 50% of all cases.

Sample includes ~ 50% of all cases.

• Sample exclusions:

Sample exclusions:

• Child

Child-

• only cases (around 6 percent of cases)
• nly cases (around 6 percent of cases)
• Cases with more than 2 adults in household (around 8.4 percent o

Cases with more than 2 adults in household (around 8.4 percent of f cases) cases)

• Results in 626,700 useable cases. We focus mostly on AFDC

Results in 626,700 useable cases. We focus mostly on AFDC-

• FG.

FG.

• Define a “sampling date”

Define a “sampling date” – – 4 4th

th quarter of a household’s

quarter of a household’s spells on welfare. spells on welfare.

• We determine

We determine number number & & ages ages of

• f children

children when the household is when the household is

• n welfare.
• n welfare.
• This information is crucial for the analysis and we do not obser

This information is crucial for the analysis and we do not observe it ve it when the household is not receiving welfare. when the household is not receiving welfare.

• By defining the sample according to the randomly selected

By defining the sample according to the randomly selected sampling date sampling date

• All cases, on and off welfare, are treated symmetrically.

All cases, on and off welfare, are treated symmetrically.

• Avoid “overweighting” long

Avoid “overweighting” long-

• term welfare recipients in our sample.

term welfare recipients in our sample.

Sampling (cont.) Sampling (cont.)

• We utilize two samples of households in

We utilize two samples of households in

• ur analyses
• ur analyses
• Cross sectional:

Cross sectional:

• Employment in year following sampling date.

Employment in year following sampling date.

• This sample mimics repeated cross

This sample mimics repeated cross-

• section data

section data used in previous studies. used in previous studies.

• Longitudinal:

Longitudinal:

• Employment in periods

Employment in periods -

• 3,

3, -

• 2,

2, -

• 1, and 0

1, and 0 (sampling). (sampling).

• Going backwards, we are sure to correctly identify birth

Going backwards, we are sure to correctly identify birth

• timing. We can’t guarantee this going forward.
• timing. We can’t guarantee this going forward.
• Longitudinal data on households allow control for

Longitudinal data on households allow control for household household-

• specific fixed effects, so focus on

specific fixed effects, so focus on within within household changes to identify EITC effects. household changes to identify EITC effects.

Year One Child 2+ Children Difference Diff-in-Diff (2+ - One) (199x - 1991-93 average) 1991 39.43 33.82

• 5.61

[27568] [37999] 1992 38.35 31.76

• 6.59

[24433] [32354] 1993 38.89 32.3

• 6.59

[23098] [29532] 1994 43.65 37.12

• 6.53
• 0.31

[24005] [30402] (0.66) 1995 48.44 41.64

• 6.8
• 0.58

[25500] [31164] (0.66) 1996 51.03 45.87

• 5.16

0.79 [25457] [30256] (0.67) 1997 57.73 52.78

• 4.95

1.27* [24794] [28650] (0.68) 1998 62.32 60.4

• 1.92

4.31*** [22479] [26174] (0.69) 1999 65.01 62.22

• 2.79

3.43*** [19066] [21973] (0.74) 2000 66.03 64.84

• 1.19

5.03*** [19731] [22490] (0.74) Table 2a: Employment Rates (in Percentages) by Family Size, 1991 - 2000, Cross-Sectional Sample Year One Child 2+ Children Difference Diff-in-Diff (2+ - One) (199x - 1991-93 average) 1991 25.48 23.02

• 2.46

[27568] [37999] 1992 23.8 21.85

• 1.95

[24433] [32354] 1993 23.9 21.97

• 1.93

[23098] [29532] 1994 28.87 26.91

• 1.96

0.17 [24005] [30402] (0.59) 1995 34.11 31.89

• 2.22
• 0.09

[25500] [31164] (0.61) 1996 35.83 35.34

• 0.49

1.69*** [25457] [30256] (0.63) 1997 42.8 41.91

• 0.89

1.24* [24794] [28650] (0.66) 1998 46.72 48.18 1.46 3.60*** [22479] [26174] (0.69) 1999 50.18 52.35 2.17 4.31*** [19066] [21973] (0.75) 2000 49.8 53.99 4.19 6.32*** [19731] [22490] (0.75) Table 2b: EITC Claiming (in Percentages) by Family Size, 1991 - 2000, Cross-Sectional Sample

SLIDE 6

6

Econometric Specification (similar to earlier D-I-D work)

[ ] ( )

2000 2000 1992 1994 2000 1994 1 9 1 1

( 2 ) 2 3 1 ... 9 1 ... 17

ict i t it t it it t t t it it it t it it it it ict it ict

Y Year Year Kids Year Kids Kids Kid Kid KidsAge KidsAge L X φ α β γ δ δ ψ ψ η κ ε

= = =

= + + × + ∑ ∑ + × + − + ∑ + + + + + + + + + 1, if household in county is 'employed' in year 0, otherwise.

th ict ict

i c t Y Emp ⎧ = = ⎨ ⎩

1, if household in county claimed EITC in year 0, otherwise.

th ict ict

i c t Y ClaimEITC ⎧ = = ⎨ ⎩

1, if household as 2 or more children 2 0, otherwise

th it

i Kids ⎧ + = ⎨ ⎩

[ ]

0 if 2 or 3 2 3 1 if 2

it it it it it

Kids Kids Kids Kids Kids < ≥ ⎧ + − + = ⎨ = ⎩

Lict = vector of labor market indicators for county c in year t Xit = vector of demographic characteristics of ith household

Other Covariates Other Covariates

• Demographic characteristics

Demographic characteristics

• Number of kids and number of kids by age.

Number of kids and number of kids by age.

• In cross

In cross-

• sectional models (time

sectional models (time-

• invariant

invariant characteristics) characteristics)

• Race/ethnicity, county dummies, gender, age,

Race/ethnicity, county dummies, gender, age, timing of entry onto welfare. timing of entry onto welfare.

• Local labor markets

Local labor markets

• Year dummies, employment share by sector,

Year dummies, employment share by sector,

• avg. income by sector
• avg. income by sector
• Welfare rules

Welfare rules

• Proportion of population in GAIN program

Proportion of population in GAIN program

Recall: Our Empirical Strategy for Recall: Our Empirical Strategy for Understanding the EITC's Effect Understanding the EITC's Effect

• Our strategy for assessing validity of estimated

Our strategy for assessing validity of estimated EITC’s effect on employment: EITC’s effect on employment:

1. 1.

Does employment of families with 2+ children Does employment of families with 2+ children increase relative to 1 increase relative to 1-

• Kid families? (It should.)

Kid families? (It should.)

2. 2.

Do temporal patterns of EITC claiming mirror Do temporal patterns of EITC claiming mirror employment patterns in # 1 (and # 3)? (They employment patterns in # 1 (and # 3)? (They should) should)

3. 3.

Does employment of 2 child families differ from Does employment of 2 child families differ from effects for 3+ Kid families? (It should not) effects for 3+ Kid families? (It should not)

4. 4.

Do we see any differences in employment Do we see any differences in employment between 2+ Kid and 1 between 2+ Kid and 1-

• Kid households

Kid households not filing not filing tax returns tax returns? (We should not!) ? (We should not!)

T Ta ab bl le e 3 3: : E Es st ti im ma at te es s

• f

f E EI I T TC C E Ef ff fe ec ct ts s

• n

n H Ho

• u

us se eh ho

• l

ld d E Em mp pl lo

• y

ym me en nt t V Va ar ri ia ab bl le e O OL LS S, , C Cr ro

• s

ss s-

• S

Se ec ct ti io

• n

na al l S Sa am mp pl le e H Ho

• u

us se eh ho

• l

ld d F Fi ix xe ed d E Ef ff fe ec ct ts s, , P Pa an ne el l D Da at ta a 2 2+ + K Ki id ds s i in n 1 19 99 94 4 0. .0 00 00 01 1

0. .0 00 03 35 5 ( (0 0. .0 00 06 65 5) ) ( (0 0. .0 00 04 41 1) ) 2 2+ + K Ki id ds s i in n 1 19 99 95 5

0. .0 00 06 63 3 0. .0 00 07 72 2 ( (0 0. .0 00 06 65 5) ) ( (0 0. .0 00 05 52 2) ) 2 2+ + K Ki id ds s i in n 1 19 99 96 6 0. .0 00 04 40 0. .0 01 12 23 3* * * * ( (0 0. .0 00 06 66 6) ) ( (0 0. .0 00 06 61 1) ) 2 2+ + K Ki id ds s i in n 1 19 99 97 7 0. .0 00 09 92 2 0. .0 02 26 61 1* * * * * * ( (0 0. .0 00 06 67 7) ) ( (0 0. .0 00 07 71 1) ) 2 2+ + K Ki id ds s i in n 1 19 99 98 8 0. .0 03 38 82 2* * * * * * 0. .0 03 32 24 4* * * * * * ( (0 0. .0 00 07 70 0) ) ( ( 0 0. .0 00 08 83 3) ) 2 2+ + K Ki id ds s i in n 1 19 99 99 9 0. .0 02 27 78 8* * * * * * 0. .0 03 34 41 1* * * * * * ( (0 0. .0 00 07 76 6) ) ( (0 0. .0 01 10 00 0) ) 2 2+ + K Ki id ds s i in n 2 20 00 00 0. .0 04 42 25 5* * * * * * 0. .0 02 29 95 5* * * * ( (0 0. .0 00 07 75 5) ) ( (0 0. .0 01 12 23 3) ) N No

• .

.

• f

f O Ob bs se er rv va at ti io

• n

ns s 5 52 27 7, ,1 12 25 5 1 1, ,6 63 37 7, ,8 85 55 5 P P-

• V

Va al lu ue e f fo

• r

r T Te es st t

• f

f 2 2+ + K Ki id ds s i in n 1 19 99 94 4-

• 2

20 00 00 = = 0. .0 00 00 00 0. .0 00 00 04 4

SLIDE 7

7

T Ta ab bl le e 4 4: : E Es st ti im ma at te es s

• f

f E EI I T TC C E Ef ff fe ec ct ts s

• n

n W Wh he et th he er r H Ho

• u

us se eh ho

• l

ld d C Cl la ai im me ed d t th he e E EI I T TC C

• n

n T Ta ax x R Re et tu ur rn n V Va ar ri ia ab bl le e O OL LS S, , C Cr ro

• s

ss s-

• S

Se ec ct ti io

• n

na al l S Sa am mp pl le e H Ho

• u

us se eh ho

• l

ld d F Fi ix xe ed d E Ef ff fe ec ct ts s, , P Pa an ne el l D Da at ta a 2 2+ + K Ki id ds s i in n 1 19 99 94 4 0. .0 00 04 41 1 0. .0 00 00 06 6 ( ( 0 0. .0 00 05 59 9) ) ( ( 0 0. .0 00 03 38 8) ) 2 2+ + K Ki id ds s i in n 1 19 99 95 5

0. .0 00 03 32 2 0. .0 00 06 65 5 ( ( 0 0. .0 00 06 61 1) ) ( ( 0 0. .0 00 04 49 9) ) 2 2+ + K Ki id ds s i in n 1 19 99 96 6 0. .0 01 10 03 3* * 0. .0 01 18 84 4* * * * * * ( ( 0 0. .0 00 06 62 2) ) ( ( 0 0. .0 00 05 58 8) ) 2 2+ + K Ki id ds s i in n 1 19 99 97 7 0. .0 00 05 50 0. .0 01 19 90 0* * * * * * ( ( 0 0. .0 00 06 66 6) ) ( ( 0 0. .0 00 06 68 8) ) 2 2+ + K Ki id ds s i in n 1 19 99 98 8 0. .0 02 24 49 9* * * * * * 0. .0 01 17 70 0* * * * ( ( 0 0. .0 00 07 70 0) ) ( ( 0 0. .0 00 08 81 1) ) 2 2+ + K Ki id ds s i in n 1 19 99 99 9 0. .0 02 28 83 3* * * * * * 0. .0 02 23 33 3* * * * ( ( 0 0. .0 00 07 77 7) ) ( ( 0 0. .0 00 09 98 8) ) 2 2+ + K Ki id ds s i in n 2 20 00 00 0. .0 04 44 48 8* * * * * * 0. .0 01 19 94 4 ( ( 0 0. .0 00 07 76 6) ) ( ( 0 0. .0 01 12 22 2) ) N No

• .

.

• f

f O Ob bs se er rv va at ti io

• n

ns s 5 52 27 7, ,1 12 25 5 1 1, ,6 63 37 7, ,8 85 55 5 P P-

• V

Va al lu ue e f fo

• r

r T Te es st t

• f

f 2 2+ + K Ki id ds s i in n 1 19 99 94 4-

• 2

20 00 00 = = 0. .0 00 00 00 0. .0 03 35 53 3

T Ta ab bl le e 5 5A A: : A As ss se es ss si in ng g t th he e V Va al li id di it t y y

• f

f S St t r ra at t e eg gy y f fo

• r

r I I d de en nt t i if fy yi in ng g E EI I T TC C E Ef ff fe ec ct t s s

• n

n H Ho

• u

us se eh ho

• l

ld d E Em mp pl lo

• y

ym me en nt t [ [ F Fa am mi il ly y F Fi ix xe ed d E Ef ff fe ec ct t s s E Es st t i im ma at ti io

• n

n

• n

n P Pa an ne el l D Da at t a a] ] V Va ar ri ia ab bl le e A Al ll l H Ho

• u

us se eh ho

• l

ld ds s H Ho

• u

us se eh ho

• l

ld ds s t t h ha at t D Di id d N No

• t

t F Fi il le e T Ta ax x R Re et t u ur rn n ( ( 1 1) ) ( ( 2 2) ) ( ( 3 3) ) 2 2+ + K Ki id ds s i in n 1 19 99 94 4

0. .0 00 03 35 5

0. .0 00 01 16 6

0. .0 00 02 20 ( ( 0 0. .0 00 04 41 1) ) ( ( 0 0. .0 00 05 51 1) ) ( ( 0 0. .0 00 06 65 5) ) 2 2+ + K Ki id ds s i in n 1 19 99 95 5 0. .0 00 07 72 2 0. .0 00 04 43 3 0. .0 00 02 28 8 ( ( 0 0. .0 00 05 52 2) ) ( ( 0 0. .0 00 06 67 7) ) ( ( 0 0. .0 00 08 86 6) ) 2 2+ + K Ki id ds s i in n 1 19 99 96 6 0. .0 01 12 23 3* * * * 0. .0 01 11 14 4

0. .0 00 01 16 6 ( ( 0 0. .0 00 06 61 1) ) ( ( 0 0. .0 00 08 82 2) ) ( ( 0 0. .0 01 10 07 7) ) 2 2+ + K Ki id ds s i in n 1 19 99 97 7 0. .0 02 26 61 1* * * * * * 0. .0 02 28 89 9* * * * * * 0. .0 01 11 13 3 ( ( 0 0. .0 00 07 71 1) ) ( ( 0 0. .0 00 09 98 8) ) ( ( 0 0. .0 01 13 31 1) ) 2 2+ + K Ki id ds s i in n 1 19 99 98 8 0. .0 03 32 24 4* * * * * * 0. .0 03 33 31 1* * * * * * 0. .0 00 04 48 8 ( ( 0 0. .0 00 08 83 3) ) ( ( 0 0. .0 01 11 16 6) ) ( ( 0 0. .0 01 16 61 1) ) 2 2+ + K Ki id ds s i in n 1 19 99 99 9 0. .0 03 34 41 1* * * * * * 0. .0 03 36 68 8* * * * * * 0. .0 00 03 37 7 ( ( 0 0. .0 01 10 00 0) ) ( ( 0 0. .0 01 13 37 7) ) ( ( 0 0. .0 02 20 00 0) ) 2 2+ + K Ki id ds s i in n 2 20 00 00 0. .0 02 29 95 5* * * * 0. .0 02 29 91 1* *

0. .0 00 00 09 9 ( ( 0 0. .0 01 12 23 3) ) ( ( 0 0. .0 01 16 68 8) ) ( ( 0 0. .0 02 25 57 7) ) [ [ 2 2+ + K Ki id ds s – – 3 3+ + K Ki id ds s] ] i in n 1 19 99 94 4

0. .0 00 03 30

0. .0 00 00 09 9 ( ( 0 0. .0 00 05 51 1) ) ( ( 0 0. .0 00 06 62 2) ) [ [ 2 2+ + K Ki id ds s – – 3 3+ + K Ki id ds s] ] i in n 1 19 99 95 5 0. .0 00 05 55 5

0. .0 00 01 10 ( ( 0 0. .0 00 06 63 3) ) ( ( 0 0. .0 00 08 80 0) ) [ [ 2 2+ + K Ki id ds s – – 3 3+ + K Ki id ds s] ] i in n 1 19 99 96 6 0. .0 00 02 21 1

0. .0 00 01 14 4 ( ( 0 0. .0 00 07 74 4) ) ( ( 0 0. .0 00 09 96 6) ) [ [ 2 2+ + K Ki id ds s – – 3 3+ + K Ki id ds s] ] i in n 1 19 99 97 7

0. .0 00 04 42 2

0. .0 00 01 12 2 ( ( 0 0. .0 00 08 86 6) ) ( ( 0 0. .0 01 11 15 5) ) [ [ 2 2+ + K Ki id ds s – – 3 3+ + K Ki id ds s] ] i in n 1 19 99 98 8

0. .0 00 00 04 4 0. .0 00 00 05 5 ( ( 0 0. .0 01 10 00 0) ) ( ( 0 0. .0 01 14 41 1) ) [ [ 2 2+ + K Ki id ds s – – 3 3+ + K Ki id ds s] ] i in n 1 19 99 99 9

0. .0 00 03 39 9

0. .0 00 07 71 1 ( ( 0 0. .0 01 11 19 9) ) ( ( 0 0. .0 01 18 81 1) ) [ [ 2 2+ + K Ki id ds s – – 3 3+ + K Ki id ds s] ] i in n 2 20 00 00 0. .0 00 01 15 5 0. .0 00 07 77 7 ( ( 0 0. .0 01 14 49 9) ) ( ( 0 0. .0 02 24 44 4) ) N No

• .

.

• f

f O Ob bs se er rv va at t i io

• n

ns s P P-

• V

Va al lu ue e f fo

• r

r T Te es st t

• f

f 2 2+ + K Ki id ds s i in n 1 19 99 94 4-

• 2

20 00 00 = = 0. .0 00 00 04 4 0. .0 01 11 12 2 0. .8 84 42 23 3 P P-

• V

Va al lu ue e f fo

• r

r T Te es st t

• f

f [ [ 2 2+ + K Ki id ds s – – 3 3+ + K Ki id ds s] ] i in n 1 19 99 94 4-

• 2

20 00 00 = = 0. .7 73 38 89 9 0. .9 99 98 89 9 T Ta ab bl le e 5 5B B: : A As ss se es ss si in ng g t t h he e V Va al li id di it ty y

• f

f S St t r ra at t e eg gy y f fo

• r

r I I d de en nt t i if fy yi in ng g E EI I T TC C E Ef ff fe ec ct t s s

• n

n W Wh he et t h he er r H Ho

• u

us se eh ho

• l

ld d C Cl la ai im me ed d t th he e E EI I T TC C

• n

n T Ta ax x R Re et t u ur rn n [ [ F Fa am mi il ly y F Fi ix xe ed d E Ef ff fe ec ct t s s E Es st t i im ma at t i io

• n

n

• n

n P Pa an ne el l D Da at t a a] ] V Va ar ri ia ab bl le e ( ( 1 1) ) ( ( 2 2) ) 2 2+ + K Ki id ds s i in n 1 19 99 94 4 0. .0 00 00 06 6 0. .0 00 05 57 7 ( ( 0 0. .0 00 03 38 8) ) ( ( 0 0. .0 00 04 48 8) ) 2 2+ + K Ki id ds s i in n 1 19 99 95 5 0. .0 00 06 65 5 0. .0 01 13 31 1* * * * ( ( 0 0. .0 00 04 49 9) ) ( ( 0 0. .0 00 06 64 4) ) 2 2+ + K Ki id ds s i in n 1 19 99 96 6 0. .0 01 18 84 4* * * * * * 0. .0 02 23 33 3* * * * * * ( ( 0 0. .0 00 05 58 8) ) ( ( 0 0. .0 00 07 78 8) ) 2 2+ + K Ki id ds s i in n 1 19 99 97 7 0. .0 01 19 90 0* * * * * * 0. .0 02 27 79 9* * * * * * ( ( 0 0. .0 00 06 68 8) ) ( ( 0 0. .0 00 09 94 4) ) 2 2+ + K Ki id ds s i in n 1 19 99 98 8 0. .0 01 17 70 0* * * * 0. .0 01 17 70 ( ( 0 0. .0 00 08 81 1) ) ( ( 0 0. .0 01 11 12 2) ) 2 2+ + K Ki id ds s i in n 1 19 99 99 9 0. .0 02 23 33 3* * * * 0. .0 03 34 44 4* * * * ( ( 0 0. .0 00 09 98 8) ) ( ( 0 0. .0 01 13 34 4) ) 2 2+ + K Ki id ds s i in n 2 20 00 00 0. .0 01 19 94 4 0. .0 02 25 55 5 ( ( 0 0. .0 01 12 22 2) ) ( ( 0 0. .0 01 16 65 5) ) [ [ 2 2+ + K Ki id ds s – – 3 3+ + K Ki id ds s] ] i in n 1 19 99 94 4

0. .0 00 07 78 8* * ( ( 0 0. .0 00 04 47 7) ) [ [ 2 2+ + K Ki id ds s – – 3 3+ + K Ki id ds s] ] i in n 1 19 99 95 5

0. .0 01 10 00 0* * ( ( 0 0. .0 00 05 59 9) ) [ [ 2 2+ + K Ki id ds s – – 3 3+ + K Ki id ds s] ] i in n 1 19 99 96 6

0. .0 00 06 67 7 ( ( 0 0. .0 00 07 70 0) ) [ [ 2 2+ + K Ki id ds s – – 3 3+ + K Ki id ds s] ] i in n 1 19 99 97 7

0. .0 01 13 31 1 ( ( 0 0. .0 00 08 82 2) ) [ [ 2 2+ + K Ki id ds s – – 3 3+ + K Ki id ds s] ] i in n 1 19 99 98 8 0. .0 00 03 33 3 ( ( 0 0. .0 00 09 97 7) ) [ [ 2 2+ + K Ki id ds s – – 3 3+ + K Ki id ds s] ] i in n 1 19 99 99 9

0. .0 01 15 58 8 ( ( 0 0. .0 01 11 19 9) ) [ [ 2 2+ + K Ki id ds s – – 3 3+ + K Ki id ds s] ] i in n 2 20 00 00

0. .0 00 06 63 3 ( ( 0 0. .0 01 14 49 9) ) P P-

• V

Va al lu ue e f fo

• r

r T Te es st t

• f

f 2 2+ + K Ki id ds s i in n 1 19 99 94 4-

• 2

20 00 00 = = 0. .0 03 35 53 3 0. .0 01 17 74 4 P P-

• V

Va al lu ue e f fo

• r

r T Te es st t

• f

f [ [ 2 2+ + K Ki id ds s – – 3 3+ + K Ki id ds s] ] i in n 1 19 99 94 4-

• 2

20 00 00 = = 0. .0 06 66 69 9

Sensitivity Analyses Sensitivity Analyses

• Change base year(s) used to measure

Change base year(s) used to measure before before-

• expansion

expansion outcomes. [Table 5C]

• utcomes. [Table 5C]
• Cannot do much about prior trends, but there is no

Cannot do much about prior trends, but there is no evidence of trends in 1991 evidence of trends in 1991-

• 1993.

1993.

• Change thresholds of UI earnings used to define

Change thresholds of UI earnings used to define whether household employed. [Table 5D] whether household employed. [Table 5D]

• Effects get somewhat weaker as earnings thresholds

Effects get somewhat weaker as earnings thresholds increase. increase.

• Control for measure of county expenditures on

Control for measure of county expenditures on childcare for welfare recipients to regressions. childcare for welfare recipients to regressions.

• County childcare expenditures have no effect on the

County childcare expenditures have no effect on the employment results described here. employment results described here.

SLIDE 8

8

Sensitivity Analyses (cont.) Sensitivity Analyses (cont.)

• Examine EITC effects on household

Examine EITC effects on household employment and EITC Claiming for two employment and EITC Claiming for two-

• parent households on welfare (AFDC

parent households on welfare (AFDC-

• UP

UP cases). cases).

• Do not find any evidence of positive

Do not find any evidence of positive employment effects for AFDC employment effects for AFDC-

• UP households.

UP households.

• Perhaps employment barriers are larger for

Perhaps employment barriers are larger for the subset of UP households with no workers. the subset of UP households with no workers.

• Recall, our definition of employment is whether

Recall, our definition of employment is whether anyone in the household had UI earnings. anyone in the household had UI earnings.

• The EITC might still reduce employment of “secondary”

The EITC might still reduce employment of “secondary” workers. workers.

Conclusions Conclusions

• Employment rates for families with 2 or more

Employment rates for families with 2 or more children increase relative to employment rates children increase relative to employment rates for one for one-

• child families as the EITC gets more

child families as the EITC gets more generous for the former group. generous for the former group.

• Similar patterns hold for claiming the EITC.

Similar patterns hold for claiming the EITC.

• We do not see similar “EITC

We do not see similar “EITC-

• like employment” effects

like employment” effects for families with 2 children relative to families with 3 for families with 2 children relative to families with 3

• r more children.
• r more children.
• We also do not see “EITC

We also do not see “EITC-

• like employment” effects

like employment” effects for families who do not file tax returns. for families who do not file tax returns.

• The work applies to a more disadvantaged

The work applies to a more disadvantaged group (mothers on welfare) than previous group (mothers on welfare) than previous research (that typically looks at single mothers). research (that typically looks at single mothers).

Conclusions (cont.) Conclusions (cont.)

• EITC increased employment for families

EITC increased employment for families with 2+ children by as much as 3.4 with 2+ children by as much as 3.4 percentage points relative to 1 percentage points relative to 1-

• child

child families. families.

• This is 11.8 percent of a 28.8 percentage

This is 11.8 percent of a 28.8 percentage point employment increase for families with point employment increase for families with children between 1991 and 2000. children between 1991 and 2000.

• But the EITC can account for 77 percent of

But the EITC can account for 77 percent of the the differential differential gain in employment for gain in employment for families with two or more children relative to families with two or more children relative to

• ne
• ne-
• child families.

child families.

Conclusions (cont.) Conclusions (cont.)

• How do our results compare with those in

How do our results compare with those in previous studies? previous studies?

• Average EITC differential for families with two or

Average EITC differential for families with two or more children was $480 in 1998. more children was $480 in 1998.

• Average disposable income (including transfers) was

Average disposable income (including transfers) was around $10,000. around $10,000.

• The EITC increased disposable income around 4.8 percent.

The EITC increased disposable income around 4.8 percent.

• Employment rates in 1998 were around 61 percent.

Employment rates in 1998 were around 61 percent.

• The EITC increased the relative employment of families with

The EITC increased the relative employment of families with two or more children 3.2 percentage points, or 5.2 percent. two or more children 3.2 percentage points, or 5.2 percent.

• The implied employment elasticity with respect to

The implied employment elasticity with respect to disposable income is 1.1, which is at upper end of disposable income is 1.1, which is at upper end of estimates in the literature examining the EITC’s estimates in the literature examining the EITC’s effects on employment. effects on employment.

SLIDE 9

9

Figure 1: Quarterly Earnings Reported to AFDC Administrators and to the UI System

2500 5000 2500 5000

Earnings Recorded in Unemployment Insurance Records Earnigs Reported to AFDC 45 Degree Mean Q2 Q3

P P-

• Val

Val % Cg % Cg Diff Diff Exper Exper Cont Cont Year Year 0.106 0.106

• 2.6

2.6

• 1.6

1.6 59.8 59.8 61.4 61.4 1997 1997 0.135 0.135

• 0.3

0.3

• 0.3

0.3 98.6% 98.6% 98.9% 98.9% 1993 1993 Welf Welf 0.404 0.404 0.6 0.6 60 60 10,696 10,696 10,636 10,636 1997 1997 0.141 0.141

• 1.3

1.3

• 153

153 $11,836 $11,836 $11,989 $11,989 1993 1993 DispIn DispIn 0.001 0.001

• 8.9

8.9

• 495

495 5,090 5,090 5,585 5,585 1997 1997 0.055 0.055

• 2.0

2.0

• 203

203 $10,160 $10,160 $10,363 $10,363 1993 1993 Trans Trans 0.022 0.022 10.5 10.5 496 496 5,226 5,226 4,730 4,730 1997 1997 0.316 0.316 3.4 3.4 $53 $53 $1,606 $1,606 $1,553 $1,553 1993 1993 Earn Earn Table 4: Earnings, Transfers, Disposable Income Table 4: Earnings, Transfers, Disposable Income

Annual Real Earnings per Worker in Service Sector in California, 1992-2000

$20 $30 $40 $50 $60 1992 1993 1994 1995 1996 1997 1998 1999 2000 1000s of $1999 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

Annual Employment to Population Ratios in California, 1992-2000

0.50 0.55 0.60 0.65 0.70 0.75 0.80 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