What Have We Learned From the Illinois Reemployment Bonus Experiment

1
What Have We Learned From the Illinois
Reemployment Bonus Experiment?

• Bruce D. Meyer, Northwestern University and
  National Bureau of Economic Research

2
Background

• For a brief period of time in Illinois, randomly
  assigned Unemployment Insurance recipients
  received 500 if they returned to work within 11
  weeks.
• This experiment has interesting policy
  implications, especially considering that the
  Illinois UI office actually saved money.
• The bonus experiment provides a rare case of
  observing exogenous changes in monetary
  incentives on job finding rates.

3
Illinois Experiment and Results

• Three groups control group, Claimant Experiment,
  and Employer Experiment.
• To qualify for Claimant Experiment, one must find
  a job less than 11 weeks after filing for UI,
  hold that job for at least 4 months, and work at
  least 30 hrs/wk.
• Note in the following slide how the mean number
  of weeks of compensated unemployment are lower
  for the Claimant Experiment.

4
(No Transcript)
5
Search Theory Predictions

• Escape rate during the period of bonus
  eligibility will be higher for the experimental
  group than the control group.
• Hazard rate will rise for the experimental group
  just before the end of bonus eligibility.
• The bonus will have a larger impact on low-income
  individuals since the bonus amount is fixed at
  500.

6
Graphically, one sees a budget set discontinuity
that may lead to a rising hazard rate at 10 weeks.
7
Empirical Results Claimant Experiment Hazard
Clearly Lies Above for 1st 10 Weeks.
8
Another Viewpoint Notice No Real Difference Over
the 11 to 24 Week Interval.
9
Main Empirical Results
10
Interpretations of Results

• The five typical experimental variables all are
  significant and have the expected signs.
• Nonetheless, the variables Claimant experiment
  (CE) and Spike are not significant!
• CEindicating a strong response to the 500
  incentive.

11
Since the income interaction variables are not
significant, Meyer looks at the following
comparison.
12
Policy Implications

• Currently unemployed who do not file could
  respond to the increased incentives to file for
  UI benefits.
• Currently not unemployed but changing jobs could
  start work slightly later and file for UI
  benefits.
• Firm layoffs could increase since workers receive
  increased compensation for short UI spells.

13
Conclusions

• Only one of the original search theories has
  strong support in the data.
• Meyer concludes that this experiment does not
  provide enough evidence that governments should
  make the program permanent.
• From a cost benefit analysis standpoint, the
  program succeeds since the treatment group earns
  several hundred dollars more in the year
  following the initial claim for UI.

14
Critiques

• Meyer does not provide sufficient reasons for why
  the data contradicts the theory that low income
  individuals are more likely to partake in the
  bonus.
• Meyer does not advocate making the bonus a
  permanent program, but he admits that the
  temporary program saved government revenue,
  induced the treatment group to return to work
  sooner, and did not lead to the treatment group
  taking advantage of the government program. If
  it was a successful temporary program, why should
  it not be permanent?