Thursday, May 31, 2007

Refugee Roulette

Today’s NY Times presents in brief the findings of Refugee Roulette, an article forthcoming in the Stanford Law Review that I co-authored with Philip Schrag and Andrew Schoenholtz (full study available here). Our empirical study found disturbing disparities in decision-making in all four levels of the asylum process in the United States. Not only were there dramatic differences in grant rates to asylum seekers from the same country between offices and courts in different parts of the country, but we found vast disparities in grant rates within offices and courts. In other words, judges sitting on the same court during the same time frame granted asylum cases from the same country at wildly varying rates -- for example, taking only Albanian asylum claims in the New York Immigration Court, one judge granted 5% of these cases, while another granted 96%. There were similar disparities in grant rates between and within asylum offices, and great variability in the federal courts of appeals, with the 7th Circuit remanding 36% of the asylum decisions before it and the 4th Circuit remanding less than 2% of its asylum cases in the time period studied. It appears from our study that the outcome of asylum cases may depend to a great extent on the personality, background, and prior experience of the adjudicator, rather than on the merits of the claim. To test this hunch, we correlated immigration judges' biographical information with grant rates. Perhaps the most interesting finding was that the gender of the judge appeared to be an important factor, with male judges granting asylum at a rate of 37.3% and female judges at 53.8%, a 44% higher rate. While we present various theories of gender and judging in our paper, I'm interested in hearing your thoughts on why this might be the case!


by the beach at sunset said...

Important topic, excellent paper; long enough that I may have missed some points, so excuses if this comment is covered in the paper.

My question is did the study try to control for asylum claim issues more associated with women: rape, forced sterilization, and even genital mutilation in finding that grant rates are higher for women judges. It would be easy to hypothesize that as a group women judges would bring a heightened sensitivity and ability to create a court environment allowing for the development of this type of testimony, even with the specter of cross examination. Do you know if women immigration judges grant asylum claims by women at a higher level than men judges do?

Secondly, was any hypothesis examined to try account for the difference between findings that at the asylum office level the male officers were granting at a statistically equivalent rate vs. immigration court. If the gender of the asylum adjudicator is important variable, why was it an insignificant variable at the asylum office level?

Fairy said...

I got a grant from the federal government for $12,000 in financial aid, see how you can get one also at