Randomized Evaluation in Legal Assistance: What Difference Does Representation (Offer and Actual Use) Make?

By: D. James Greiner and Cassandra Wolos Pattanayak. Published by: Yale Law Journal. Published in June 2012.

Link to article

Link to PDF

Yale Law Journal, Vol. 121

The researchers ask: What difference does representation (offer and actual use) make in unemployment insurance appeals?

This was a randomized study that analyzed outcomes in unemployment insurance appeals. It compared the results people appealing unfavorable unemployment decisions to Administrative Law Judges had when they received an offer of representation from the Harvard Legal Aid Bureau (HLAB) with those that received no such offer. It analyzed the unemployment appeal cases of 207 persons who sought the assistance of the Harvard Legal Aid Bureau between the Summer of 2008 and the Spring of 2010. In total, 103 persons received offers of representation and 104 did not receive an offer.


Did an offer of HLAB representation increase the probability that the claimant would prevail? There was no evidence of that an offer of HLAB representation increased the probability that a claimant would prevail. Did an offer of HLAB representation delay adjudication? The short answer is yes. On average … cases receiving an HLAB offer took 53.1 days from HLAB interview to ALJ decision, while those not receiving the offer took 37.3 days. The sixteen-day difference is statistically significant.

Did an offer of HLAB representation impose costs on the unemployment system? Due to the delay in adjudication noted above, the answer is a tentative yes.

When we shift focus from an offer of HLAB representation to the actual use of representation (from HLAB or some other source), do we still find a delay and no statistically significant effect on the win rate? It turns out that the question is challenging to answer. The reason is that, as noted above, only the offer of, not actual use of, representation was randomized, and the randomization is what produces treated and control groups identical up to random variation. Some claimants (under 10%) randomized to receive an HLAB offer of representation did not in fact receive representation, ordinarily because HLAB was unable to contact them before their ALJ hearings. Some claimants (between 39% and 49%) randomized not to receive an HLAB offer obtained representation from other service providers, such as private attorneys, Greater Boston Legal Services, the Volunteer Lawyers Project, or the clinical program at Northeastern Law School.

Were there subsets of cases, subsets identifiable at intake, as to which an offer of Harvard Legal Aid Bureau assistance did increase the probability of a win? Nothing in these results supports the conclusion that an HLAB offer actually did benefit these subsets of claimants.

Responses to this study

Jeffrey Selbin, Jeanne Charn, Anthony Alfieri & Stephen Wizner,Service Delivery, Resource Allocation, and Access to Justice: Greiner and Pattanayak and the Research Imperative,30 July 2012

Steven Eppler-Epstein, Passion, Caution, and Evolution: The Legal Aid Movement and Empirical Studies of Legal Assistance, 126 Harv. L. Rev. F. 102 (2013).

Categories: Consumer/finance, Employment, Employment, Legal Aid Attorneys, Legal Aid Practitioners, Policymakers and Funders, Researchers and Academics, State-Specific, Unbundling

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: