The existing empirical literature fails to agree on the nature of the intertemporal relation between risk and return. This paper attempts to resolve the issue by estimating a conditional two-factor model motivated by Merton's intertemporal capital asset pricing model. When long-term government bond returns are included as a second factor, the partial relation between the market risk premium and conditional market variance is found to be positive and significant. The paper also helps explain the convoluted empirical relation between the market risk premium, conditional market variance, and the nominal risk-free rate previously reported in the literature.