Discrete events, particularly transitions between jobs, are thought to be one of the largest sources of wage dispersion. These switches often induce changes in wages that can be quite large, highly persistent, and largely uninsurable. Labor market transitions can be idiosyncratic tail events, potentially having a disproportionate impact on welfare without aecting aggregate quantities. The nature of these events is highly cyclical; transitions are more likely to be favorable if they occur during expansions relative to recessions. As such, agents may require a premium for investing in assets which underperform when labor market event risk is high, a feature absent from leading asset pricing models. This paper explores the potential importance of this channel in aecting asset prices. We provide empirical evidence on the plausibility of event risk in explaining the shape of the idiosyncratic distribution of income growth rates as well as the relationship between event risk and aggregate variables. Next, we formalize its role within a general ane, jump-diusion asset pricing framework with heterogeneous agents and incomplete markets, making our results immediately applicable to a wide class of existing models for aggregate dynamics. We propose a model with a single state variable{capturing the probability of large, idiosyncratic decreases in consumption{ that quantitatively matches the level and dynamics of the equity premium, even with i.i.d. aggregate consumption growth. Consistent with the model’s predictions, initial claims for unemployment, suitably normalized, is a highly robust predictor of returns, outperforming a number of conventional predictors, including the dividend yield.