Chasing ghosts: the elusive ambiguity premium in U.S. equities

This study investigates whether ambiguity about model uncertainty is priced in the cross-section of U.S. equities. Specifically, it tests whether ambiguity functions as a distinct risk factor or instead acts primarily as a systemic state variable. A weekly composite ambiguity factor (CAF) is constructed using principal component analysis on implied volatility, policy uncertainty, and news sentiment. In the post-2014 sample, the ambiguity proxies exhibit structural decoupling, as the first component is driven by volatility and sentiment, while policy uncertainty contributes little at weekly frequency. Using balanced panel of 448 constituents from the S&P 500 for November 2014-May 2025, the firm-level CAF exposures are estimated, and pricing is evaluated via portfolio sorts and Fama–MacBeth regressions. The estimated price of CAF risk is economically small and statistically indistinguishable from zero. The portfolio sorts do not exhibit the monotonic return gradient implied by linear factor pricing. In contrast, regime-switching evidence at the aggregate level indicates that CAF is strongly associated with market valuation states and its impact is larger in high-volatility regimes. Overall, the results reconcile strong time-series relevance of ambiguity with weak cross-sectional pricing. Future research should therefore pivot away from linear factor models and toward frameworks that capture the regime-dependent aggregate transmission of ambiguity.