Different baselines, different outcomes: Reassessing the representative promise of within-party sortition

Cowley, Webb and Bale’s critique of within-party sortition argues that lotteries among party members cannot produce a socially representative UK Parliament because party memberships are unrepresentative. Drawing on Party Members Project data, they show that random selection would produce only mixed improvements in descriptive representation. While their empirical contribution is significant, this article argues that the broader sceptical conclusion rests on a narrow modelling baseline. Cowley et al. effectively test only one institutional configuration—unstratified random selection from static party memberships—while extrapolating broader claims about sortition as a reform. This article develops three lines of critique. First, their simulations treat party membership as exogenous and unaffected by institutional change, ignoring evidence that candidate selection rules shape recruitment and participation. Second, their analysis evaluates only a single design variant, despite evidence that modern lottery-based institutions employ stratification and corrective mechanisms. Third, their evaluation relies on implicit ‘mirror of society’ standards without specifying formal metrics for aggregating representational trade-offs. Drawing on scholarship on sortition, deliberative democracy, and party organisation, the article proposes a dynamic and design-sensitive framework for evaluating within-party lotteries. Once alternative institutional baselines are considered, the representational potential of within-party sortition appears considerably less constrained than Cowley et al.’s analysis suggests.