Is Bigger Always Better? On Optimal Panel Size with Evidence from the Supreme Court of Canada

I have recently posted a new paper on SSRN entitled, Is Bigger Always Better? On Optimal Panel Size, with Evidence from the Supreme Court of Canada. It is work coauthored with Andrew Green and Ed Iacobucci.  In the paper we model how panel size should be determined by High Courts that provide for this kind of flexibility (and most do).  Our model supports the following predictions: (a) all else the same, more important appeals should be heard by larger panels; (b) all else the same, very easy and very difficult appeals should be heard by smaller panels; and (c) all else the same, appeals of intermediate difficulty should be heard by the largest panels.  We find that the predictions of our model are borne out by an analysis of the panel assignments at the Supreme Court of Canada from 1984-2005.