The Time-Varying Risk Return Tradeoff in the Long-Run: UK Evidence

The risk-return tradeoff implied by time-invariant conditional CAPM and ICAPM is rather weak with the two century history of UK data from 1836 to 2010, contrary to the findings of Lundblad (2007). I develop a nonlinear ICAPM with multivariate GARCH-M based on Harvey et al. (1992) to allow for the time-varying risk-return tradeoff and hedging coefficients. I find that the risk return relation is largely positive over the time. More importantly, I show that the seemingly negative risk-return relation could be entirely spurious because it is not statistically different from zeros with the 95% confidence bounds. I conclude that the time-varying risk-return tradeoff is the main reason for the weak relation.