The equity risk premium is generally considered to be a reward that investors earn on top of the prevailing risk-free return, implying that, all else equal, total expected stock returns should increase with the level of the risk-free return. We examine whether this notion is true using long-term historical data. Our statistical tests strongly reject the hypothesis that a higher risk-free return implies higher total average stock returns. Instead, expected stock returns appear to be unrelated (or perhaps even inversely related) to the level of the risk-free return. Thus, the equity risk premium tends to be higher when the risk-free return is low and vice versa. This result appears to stem from the operating performance of firms. Our findings challenge the conventional wisdom about expected stock returns and have important implications for asset allocation decisions, in particular when risk-free rates are at extreme levels.