We show that the risk-return trade-off parameter is inconsistently estimated when observed or estimated conditional variances measure risk. The inconsistency arises from misspecified, unbalanced, and endogenous return regressions. We propose a 2SLS estimator that produces consistent estimates without observing the variance premium. Using this method, we find a positive risk-return trade-off and long-run return predictability. Our approach outperforms commonly used risk-return estima- tion methods, and reveals a significant link between the VP and economic uncertainty.