Personal Reemployment Accounts: Simulations for Planning Implementation Christopher J. O’Leary and Randall W. Eberts W.E. Upjohn Institute for Employment Research 300 South Westnedge Avenue Kalamazoo, MI

1. Introduction  Back to Work Incentive Act of 2003 (HR 444)  $3.6 billion nationwide budget for 2 years  Budget allocations based on share of unemployment  Federal to state ($85.3 million to Georgia)  State to Service Delivery Regions (SDR)  Assume service use patterns in Georgia under WIA  Zero, one, and two-week duration impacts imputed

2. Simulated PRAs for Georgia under HR 444  Top WPRS profiled (neither union hiring hall, nor standby)  Eligible for at least 20 weeks of UI benefits  PRA uniform offer of up to $3,000  Can purchase intensive, supportive and training services  Cash bonus (60%) for reemployment within 13 weeks  Cash bonus (40%) after 6 months continuous reemployment  No second bonus if services purchased after first bonus  UI exhaustees may draw weekly support payments from PRA  Top 30% of profiling distribution 5 quarters of WIA: 46,855

3. Services Chosen Table 4. Service Participation Rates for Top 30 Percent of Profiling Scores First 13 weeksAfter 13 weeks DescriptionParticipantsRateParticipantsRate Service Coordination Customer Service Plan In-Depth Assessment Counseling Expanded Workshop Ref to Support Services Training , , , , , , Total Intensive/Training17, ,

4. Prices for Services Table 5. Prices for Intensive, Supportive, and Training Services Georgia UI Claims 7/1/2000 to 9/30/2001 Total Services Cost Factors Services Prices Intensive Services Service coordination Customer service plan In-depth assessment Counseling Expanded workshop 1,051 36, ,661 1, $356 $712 Supportive Services3,6233.0$1,068 Training Services7,0244.0$1,424 Note: Customer service plan is the numeraire (cost factor = 1).

5. PRA Participant Groups Employed Period 1 – UI less than 13 weeks Employed Period 2 – (Quarterly earnings 1, 2000, 1 ) Yes = 1 (0.366) No = 0 (0.634) Yes = 1 (0.402) E11 (0.267) E10 (0.135) No = 0 (0.598) E01 (0.099) E00 (0.499)

6. Characteristics of PRA Participant Groups UI and Demographic Summary by Bonus Qualification Group for the Top 30 Percent of the Profiling Distribution Group Proportion of sample UI weeks drawnMean WBAAge > 45 Lower income E11 E10 E01 E Total

7. Statewide Simulation Results Group Proportion of sample Bonus CostsServices Spending Extended compensation Spending for group First period Second period First period Second period E11S00 E11S10 E11S01 E11S E11S00 E11S10 E11S01 E11S E11S00 E11S10 E11S01 E11S E11S00 E11S10 E11S01 E11S Total

7. Statewide Simulation Results—Continued Table 11. Average cost per PRA offer and total number of potential PRA offers over two years given the Georgia PRA budget PRA ScenarioAverage CostNumber of Offers Bonus, services, exhaustee pay 1-Week Impact 2-Week Impact $2,475 $2,515 $2,551 34,473 33,924 33,446 Bonus and purchase services 1-Week Impact 2-Week Impact $1,452 $1,491 $1,528 58,760 57,223 55,838 Bonus only with free services 1-Week Impact 2-Week Impact $1,040 $1,086 $1,131 82,038 78,564 75,438

8. Within State Budget Allocation Alternatives SDRRegion Share of Georgia Budget Percent difference UnemployedUI claimants Northern Georgia Atlanta Northern Georgia Balance of Georgia Coastal Georgia

9. Summary of Results  With Georgia grant of $85.32 million, if every PRA recipient spent the entire $3,000 grant, then:  28,440 offers over two years  With bonuses, purchases of services, and extended UI type payments assuming a 100% acceptance of PRA offers, then:  $3,000 offered to top 30%, 34,473 offers over two years with no response  One and two week duration responses would permit 33,924 and 33,446 offers  Excluding payment of remaining PRA funds to UI benefit exhaustees  70% more offers could be made (58,760)  Under a pure bonus PRA (free services and no exhaustee payments)  138% more offers could be made (82,038)  Sensitivity of results to prices for services (Appendix spreadsheet)  Doubling prices reduces offers possible by only 20%  Halving prices increases offers possible by about 20%

10. Extensions  Could customers pay for their preferred bundle of services under PRA?  With bonuses less than 1% have a budget shortfall  Without bonuses (services only) about 0.5% have a budget shortfall  Setting the bonus as 10 times the WBA with a minimum of $1,500  Permitted more bonus offers  May have smaller distortions for low wage workers  Allocations for PRAs within states to SDR  A uniform WPRS score cutoff cannot be applied state-wide  Unemployment rate and insured unemployment shares differ  Entry and Displacement Effects  Should be reduced by targeting offers using WPRS  Wage effects of early reemployment  No effects in bonus experiments requiring 4 months reemployment  An increased risk given immediate payment of 60% upon reemployment