Financial Performance Incentives for United States Government Programs: WIA, TANF, and SNAP

1
Financial Performance Incentives for United
States Government Programs WIA, TANF, and SNAP
Stephen A. WandnerSenior EconomistU.S. Department of Labor Michael WisemanResearch Professor George Washington University
2
The question

• What can the European Social Fund learn from the
  American experience with operation of training
  programs under the Workforce Investment Act?
• This question has many facets
• WIA attempts to increase effectiveness and
  efficiency through performance bonuses
• This paper (in progress!) expands the perspective
  to include the TANF and SNAP High Performance
  Bonuses

3
The Workforce Investment Act

• Federal-state program, operated by the states
  through local one-stop career centers
• Supported by formula-based grants to states
• Beyond the formula allocations, High Performance
  Bonuses are paid states on the basis of measures
  of achievement
• Bonuses are small and, over time, have gotten
  smaller

4
WIA Procedures

• Performance targets are set through negotiations
  between the (6) federal government regional
  offices and the states
• Final targets depend on negotiation techniques
  and skills of regions and states
• Performance targets cover many program outcomes
• States have some control over measurement
  implementation

5
Issues

• Regional DOL office strategies, capabilities, and
  enthusiasm differ
• States vary in analytical capabilities for
  response to regional targeting proposals.
• Measures adopted offer numerous opportunities for
  gaming by states
• Results have been uneven, regionally and by state
  
• Connection of HPB accomplishment to actual
  achievement challenged by research

6
Lessons (?)

• Negotiation may be useful, but attention must be
  paid to leveling playing field
• Should start with control for variation in
  characteristics of clients served, economic
  environment
• Addressing problems of motivation essential
• Bottom line No evidence of significant positive
  effects
• Watch developments with reauthorization

7
Temporary Assistance for Needy Families (TANF)
High Performance Bonus

• TANF famously replaced Aid to Families with
  Dependent Children beginning in 2007
• Enabling legislation called for high
  performance
• Employment targetsjob entry, job retention,
  earnings gainhave face validity
• Goals multiplied over time, as did winners
• Like WIA, stakes were small
• Program died, unmourned, in 2005

8
Procedures

• Began with data available to states, but it was
  clear procedures were not uniform and states
  lacked access to some data
• Introduced a new resource, the National
  Directory of New Hires
• Ultimately major measures wholly computed at
  federal level, with substantial lag between
  reference performance year and award
• Federal computation not always intelligible,
  reliable

9
Issues

• What to measure
• How to measure
• Control for context
• Strategic response
• Missing feedback

10
Lessons (?)

• Give thought to the objectives
• Be cautious about statistical inference
• Plan for improvement
• Institutional development may be an important
  by-product of performance assessment effort

11
Supplemental Nutrition Assistance (Food Stamp)
Program (SNAP)

• SNAP is a national negative income tax operated
  outside of the tax system. Does not purport to
  be adequate for minimum subsistence
• Delivered by electronic benefits transfer (EBT)
  and collected when recipients purchase food
• Arguably the nations most important means-tested
  benefit
• Plays a significant role in economic stimulus

12
Incentives

• SNAP is operated by states
• Benefits are wholly federally funded
  administration costs are split between states and
  federal government
• Incentive problems addressed by sample-based
  quality control system
• States liable for cost of errors, but attempts
  made to reduce emphasis on penalty and shift to
  rewards
• Result (2002) was --

13
High Performance Bonus

• Based on QC audit, other sources
• Four bonus categories
• Payment accuracy
• Negative error rate
• Application processing timeliness
• Program access
• Only 48 million (total state administrative
  costs were about 3 billion in FY 2007)
• Awards delivered by September of following year

14
(No Transcript)
15
Issues

• Assessing sample-based penalties
• Program access measures
• Change versus levels
• Technical assistance
• Net effect

16
Lessons (?)

• Link to ground-level operations
• Audit the outcomes
• Take care with statistics
• Link to better practice
• But dont get carried away
• Task is relatively simple Deliver a
  well-defined benefit to a target population each
  month
• Outcome immediate
• Broad political support

17
Conclusions

• Take in the museums
• Keep watch
• Get back to the Open Method
• Start on the ground
• Reward process