Texas NCP Choices: Workforce Development for Non-Custodial ... · NCP age (years) 34.1 33.6 NCP male 96.5% 96.5% NCP number of active CS cases 1.6 1.6 Time since first observed NCP

Texas NCP Choices: Workforce Development for Non Custodial ParentsNon-Custodial Parents

Daniel Schroeder

N ti l A i ti f W lf R h dNational Association for Welfare Research and Statistics, 49th Annual Workshop

Albany, NYJuly 13th, 2009

Overview

Backgroundg

NCP Choices Program

Comparison Group Design, Selection

Results

Summary / Implicationsy p

Next Steps

Background: Previous ProgramsSeveral previous programs have attempted to engage low-income non-custodial parents (NCPs) in workforcelow income non custodial parents (NCPs) in workforce development services.

R lt h b i l S ti l tResults have been equivocal: Sometimes employment and earnings are increased, and sometimes child support payments are increased

Generally, the problem with these programs is related to either implementation or to recruitment, enrollment,to either implementation or to recruitment, enrollment, and retention of participants

Programs with voluntary participation are not typicallyPrograms with voluntary participation are not typically very successful

NCP Choices Pilot ProgramNCP Choices was implemented as a pilot starting in 2005 in four sites2005 in four sites

Serves low-income non-custodial parents (NCPs) of t f TANF i i t h d li tcurrent or former TANF recipients who are delinquent

in their child support payments

NCPs are brought before a child support judge and given a choice of either:

paying their child support,paying their child support,participating in NCP Choices, orgoing to jail

Critical Program Features

Consequences for non-participation

Co-location of workforce services in court

Case-management

Communication among partners

CHOICES Services offered:CHOICES Services offered:Job referralsSupport servicesShort-term trainingShort term trainingSubsidized employment/work experienceGED, ESL classesRetention assistance and career advancement planning p g

NCP Choices Pilot Sites

DallasDallas countyFort Worth

LubbockLubbock county

WacoMcLennan county

Tarrant county

Beaumont

El PasoEl Paso county

BeaumontOrange, Jefferson countie

HoustonHoustonHarris county

San AntonioBexar county

Texas CityGalveston, Brazoria counties

6

Original Pilot site

Expansion site

Brownsville Cameron countyEdinburg, McAllen, Weslaco

Hidalgo county

Comparison Group DesignAn experimental design, with random assignment of NCPs to be ordered into the program or not wouldNCPs to be ordered into the program or not, would have been ideal

Si d i t i f ibl iSince random assignment was infeasible, a quasi-experimental comparison group design was used instead

For each NCP in the program, a multivariate “nearest-neighbor” was selected for inclusion in comparisonneighbor was selected for inclusion in comparison group

Outcomes for NCP Choices clients and comparisonOutcomes for NCP Choices clients and comparison group then computed for estimating impact

Before comparison group selection (1 of 2)

NCP Choices All NCPs

All NCPs in target counties N=751 N=168 851All NCPs in target counties N 751 N 168,851

NCP age (years) 34.1 37.5 **

NCP male 96.5% 99.0% **

NCP number of active CS cases 1 6 1 2 **NCP number of active CS cases 1.6 1.2

Time since first observed NCP earnings (quarters) 20.5 18.7 **

NCP employed at program entry 40.1% 49.2% **

Percent of time NCP employed over 4 years prior to program 40.6% 50.8% **Percent of time NCP employed over 4 years prior to program 40.6% 50.8%

NCP average quarterly earnings over 4 years prior to program $1,912 $3,919 **

NCP experienced earnings dip of at least 20% within prior 2 years 50.5% 24.8% **

Time since CS first collected on this case (months) 33.3 32.9 **Time since CS first collected on this case (months) 33.3 32.9

Time since CS last collected on this case (months) 8.1 4.3 **

Percent of time CS collection was made in prior year 21.4% 50.8% **

Any CS collection made via federal offset in prior year 14.4% 13.7% **y CS co ect o ade a ede a o set p o yea % 3 %

Before comparison group selection (2 of 2)

NCP Choices All NCPs

All NCPs in target counties (continued) N=751 N=168,851Percent of time TANF received by associated CP(s) in prior year 17.5% 4.7% **Any Choices participation in year prior to program 1.7% .7% **Any ES participation in year prior to program 4.5% 1.8% **Any FSE&T participation in year prior to program 1.7% .4% **Any WIA participation in year prior to program 2.8% .8% **Any Project RIO participation in year prior to program 2 4% 6% **Any Project RIO participation in year prior to program 2.4% .6% **Capias issued for arrest of NCP in 6 months prior to program 21.7% 1.6% **Approximate arrears balance at program entry $33,256 $11,887 **Total current support obligation (PP1) at program entry $510 $332 **Total current support obligation (PP1) at program entry $510 $332Total arrears obligation (PP2) at program entry $123 $74 **

Those experiencing an earnings dip N=379 N=70,562Time since earnings dip occurred (quarters) 4.6 4.8 **Percent of earnings which earnings dip represents 85.3% 76.8% **

Comparison Group SelectionPool of potential comparison group was restricted to NCPs in same counties as the demonstrationNCPs in same counties as the demonstration

Exact matches were required on: county, gender, h th i di f 20%whether an earnings dip of 20% or more was

experienced, and time of entry

Weighted multivariate distance was computed on remaining variables, with greater weight given to those dimensions that more strongly differentiate the twodimensions that more strongly differentiate the two groups

One nearest neighbor chosen for each NCP ChoicesOne nearest-neighbor chosen for each NCP Choices participant, and the resulting groups follow

After comparison group selection (1 of 2)

NCP ChoicesComparison Group

All NCPs N=751 N=751

NCP age (years) 34.1 33.6

NCP male 96.5% 96.5%

NCP number of active CS cases 1.6 1.6

Time since first observed NCP earnings (quarters) 20.5 21.2

NCP employed at program entry 40.1% 39.8%

Percent of time NCP employed over 4 years prior to program 40.6% 42.3%

NCP average quarterly earnings over 4 years prior to program $1,912 $1,947

NCP experienced earnings dip of at least 20% within prior 2 years 50.5% 50.5%

Time since CS first collected on this case (months) 33.3 33.8

Time since CS last collected on this case (months) 8.1 7.9

Percent of time CS collection was made in prior year 21.4% 21.5%

Any CS collection made via federal offset in prior year 14.4% 12.1%

After comparison group selection (2 of 2)

NCP ChoicesComparison Group

All NCPs N=751 N=751Percent of time TANF received by associated CP(s) in prior year 17.5% 14.6%Any Choices participation in year prior to program 1.7% 2.4%Any ES participation in year prior to program 4 5% 2 7%Any ES participation in year prior to program 4.5% 2.7%Any FSE&T participation in year prior to program 1.7% .7%Any WIA participation in year prior to program 2.8% 2.5%Any Project RIO participation in year prior to program 2.4% .8% *Any Project RIO participation in year prior to program 2.4% .8%Capias issued for arrest of NCP in 6 months prior to program 21.7% 21.7%Approximate arrears balance at program entry $33,256 $28,562Total current support obligation (PP1) at program entry $510 $452Total arrears obligation (PP2) at program entry $123 $111

Those experiencing an earnings dip N=379 N=379Time since earnings dip occurred (quarters) 4.6 4.7Percent of earnings which earnings dip represents 85.3% 84.9%

Results: Workforce Development

Any Workforce Development, 1st year

82.5%90.0%

60.0%70.0%80.0%

7 6%20 0%30.0%40.0%50.0%

7.6%

0.0%10.0%20.0%

NCP Choices Comparison Group

Over 82% participated in workforce development (a 10-foldincrease; p<.01)

Results: Capias (Warrant)Capias (warrant) issued for arrest of NCP

19.6%

20.0%

12.7%15.0%

5.0%

10.0%

0.0%NCP Choices Comparison Group

A 54% increase in rate of capias issuance (6.9 pct points, p<.01), but this is not the best measure of jailing rate…

Results: Child Support CollectionsPercent of time any child support collected

45.5%42.8%

%

50.0%

30.5% 32.4%

30.0%

40.0%

10.0%

20.0%

0.0%NCP Choices Comparison Group

Year 1 Years 2-3

A 49% increase in frequency of collections in year 1 (p<.01)

Results: Child Support Collections

Monthly average child support collections

$176$161

$130$160$180

$122 $130

$$100$120$140$

$20$40$60$80

$0$20

NCP Choices Comparison Group

Year 1 Years 2-3

A 44% increase in amount of collections, year 1 ($54 per month,p<.01)

Year 1 Years 2 3

Results: Child Support Consistency

Consistent payment of CS, 3 out of 3 months

30.4%35.0%27.2%

19.5%23.4%

20.0%25.0%30.0%

35.0%

5.0%10.0%

15.0%

20.0%

0.0%

5.0%

NCP Choices Comparison Group

Year 1 Years 2-3

40% increase in consistency of collections, year 1 (8 pct points,p<.01)

Year 1 Years 2 3

Results: Employment

Percent of time NCP employed

48.4%43.8% 41 1%50.0% 41.1%

37.0%

30.0%

40.0%

50.0%

10.0%

20.0%

0.0%NCP Choices Comparison Group

Year 1 Years 2-3

18% increase in employment rate in year 1 (7.3 pct points, p<.01)

Year 1 Years 2 3

Results: Earnings of the Employed

Avg quarterly earnings, among employed NCPs

$2 838

$3,435 $3,303 $3,298$3,500 $2,838

$2,000$2,500$3,000$3,500

$500$1,000$1,500$2,000

$0$500

NCP Choices Comparison Group

Year 1 Years 2-3

Initial decrease in earnings of employed NCPs disappearsIn later years

Year 1 Years 2 3

Results: Unemployment ClaimsNCP files unemployment claim within 1 year of

program entry

6.6%7 0%

program entry

3 0%5.0%

6.0%

7.0%

3.0%

2.0%

3.0%

4.0%

0.0%

1.0%

NCP Choices Comparison Group

A 55% decrease in unemployment insurance claims filed (3.6pct points, p<.01)

Results: Welfare (TANF) Receipt

Percent of time CP(s) receiving TANF

8 0%

9.6%10.0% 8.0%

6.7%7.7%

6.0%

8.0%

10.0%

2.0%

4.0%

0.0%NCP Choices Comparison Group

Year 1 Years 2-3

An 17% decrease in TANF receipt among associated custodialparents in year 1 (1.6 pct points, p<.01)

Year 1 Years 2 3

Summary and ImplicationsNCP Choices met most of its program goals:

NCP Choices clients far more likely to participate in workforce development

Increased frequency, amount, and consistency of child support payments

Increased employment rates; reduced earnings levels of the employed, but effect goes away in later yearsthe employed, but effect goes away in later years

Reduced rate of filing unemployment claims

Reduced TANF receipt among custodial parents

Next Steps

Presently working on final report that expands the evaluation to include six additional sites that started in 2007

Ultimately, given the success so far, the goal is to expand the program statewide

Future evaluation will include a modification of the program to target low-income NCPs in p g gestablishment cases, in hopes of serving them before they get behind on their CS payments

Contact

Daniel Schroeder (512) [email protected]@

Latest report available:http://www.utexas.edu/research/cshr/pubs/pubs.htm

Updated impacts to be posted in Sep 2009p p p p