Blackwell on Judicial Identity & Legal Outcomes

Michael Blackwell (London School of Economics & Political Science (LSE)) has posted Measuring the Length of the Chancellor's Foot: Quantifying How Legal Outcomes Depend on the Judges Hearing the Case and Whether Such Variation Can Be Explained by Characteristics of the Judges on SSRN. Here is the abstract:

    This paper considers the extent that judicial decisions are dependent on which judge(s) hears the case and whether any variation in decision making can be attributable to factors associated with the judge(s). This paper addresses these questions, using multi-level modelling, by a statistical analysis of the 1,308 appeals from the Immigration Appeal Tribunal and Employment Appeal Tribunal heard by the Court of Appeal (Civil Division) of England and Wales between 2001 and 2009 assembled by the author for this purpose.

And from the paper:

    These results suggest that there can be considerable variation between litigation outcomes dependent on which judges hear the case. In the typical case discussed – involving an appeal by an employee or immigrant heard in late 2005 with a judgement of about 3,000 words – it was shown, in section 4, how a change of judge giving the lead judgement can be association with a doubling of the expected probability of a liberal outcome. When judges sit on panels section 5 showed that about two-thirds of the variability attributable to panel composition is explained the judge who delivers the lead opinion. By using a measure of judicial variation based on appeals to the House of Lords, section 6 demonstrated that such variation cannot be explained by away by differences in the type of cases heard by different judges – so the differences in outcomes appear at least partly due to, and not merely associated with, different judges hearing the cases.

    It was shown in section 7 that such variation is not explained by factors that the existing literature on judicial decision making suggests as responsible for judicial prejudice. Rather it appears attributable to the idiosyncratic preferences of particular judges.