Election Forecasting: The Evolution of Political Economy Models in the UK

• In Event: Comparative Election Forecasting

Abstract

In this paper we assess how Harold Clarke’s contributions to election studies in the UK – specifically the valence and political economy approaches – have informed the growing body of political economy forecasting models for British elections. While Clarke’s models were almost invariably explanatory, he occasionally turned to prediction of outcomes. He was a devoted empiricist, with richly estimated models. These models could easily be converted into substantive forecasting models and this is something Clarke sometimes did (Whiteley et al 2011).

In this paper, we focus on the British case and trace how Clarke’s work fits within an election forecasting frame, and has aided, directly and indirectly, the shape of election forecasting models in Britain. We then chronologically review the political economy regression-based forecasting studies beginning way back with Anthony Mughan's first piece (1987) to the most recent forecasts, and we compare these to the accuracy of vote intention polls. This works to answer the question: what works better in the UK, models (namely political-economy models) versus polls (namely vote intention polls).

Works cited:

Mughan, A. (1987). General election forecasting in Britain: a comparison of three simple models. Electoral Studies, 6(3), 195-207.

Whiteley, P., Sanders, D., Stewart, M., & Clarke, H. (2011). Aggregate level forecasting of the 2010 general election in Britain: The Seats-Votes model. Electoral Studies, 30(2), 278-283.

Authors

• Mary Stegmaier, University of Missouri

• Steven Jokinsky, University of Kentucky

• Michael S. Lewis-Beck, University of Iowa