This paper explores the industry cost of equity capital for the UK. We replicate the <link rid="b10">Fama and French (1997)</link> US analysis for UK industries, but additionally investigate the industry cost of equity capital obtained from a conditional CAPM, the <link rid="b3">Cahart (1997)</link> four factor model, and the <link rid="b1">Al-Horani, Pope and Stark (2003)</link> R&D model. In line with the Fama-French US results, the out of sample performance of all the models is disappointing Whilst the FF3F model has a somewhat higher explanatory power than the CAPM in terms of explaining past returns, the SMB and HML factor slopes show considerable variability through time. However, all our models of the cost of equity capital in the UK outperform a simple 'beta one' model, a result that has implications for the regulatory process. There is also some evidence to suggest that a conditional CAPM may be of interest to regulators. The new R&D model of Al-Horani et al. clearly has potential, in that over the limited period for which data is available it yields return errors not dissimilar to those found under the FF3F model, but exhibits slope coefficients on the fourth R&D factor that seem to be relatively stable. Copyright (c) 2009 The Authors Journal compilation (c) 2009 Blackwell Publishing Ltd.
