Abstract

I identify a "Staleness Trap" in market anomalies: while signals are highly persistent, their associated return opportunities decay much faster than the underlying signals. Consequently, even when using the freshest data, standard portfolio methodologies mechanically accumulate stale positions whose economically relevant premiums have largely dissipated. I develop a measure of Signal Age and show that these stale continuers absorb one-third of portfolio capital while generating negligible returns, whereas fresh entrants drive roughly half of the aggregate premium. In contrast, stale continuers in accounting anomalies continue to yield stable returns, revealing a fundamental difference in the lifecycle of profitability across market and accounting anomalies. Taken together, these patterns are most consistent with fresh entrants in market anomalies commanding greater risk compensation when they are harder to absorb at signal activation.

Abstract

A stock that is exposed to an anomaly is likely to be traded by investors in a certain direction. However, a stock can be included in long positions and short positions for different sets of anomalies at the same time. Using 153 anomalies from Jensen, Kelly, and Pedersen (2023), We study the relationship between exposure to anomalies and market quality. We find that exposure to anomalies widens spreads, increases volatility, and increases adverse selection. The relationships are strengthened when exposure to long and short positions is balanced. The results overall suggest that exposure to anomalies harms market quality.

Presented at: CICF 2026

Abstract

Li et al. (2023) show that intraday risk factor exposure leads to predictable returns. In this paper, we focus on the unexplained price movements from the factor-based intraday model. We document an economically large and statistically significant return reversal based on the previous period’s residual return. This residual reversal strategy, which buys stocks with negative residuals and sells stocks with positive residuals, earns an annualized return of 162.3%. The strategy captures the returns to liquidity provision to the transitory component of stock returns.

Abstract

Industrial banks (IB) are state-chartered, FDIC-insured institutions that commercial or financial parents may own, and they occupy a distinctive niche in the U.S. banking system. We document that IBs differ from other FDIC-insured banks along three core margins: (i) ownership and supervision (CALMA/PCA commitments that contractually obligate parents to act as sources of strength), (ii) business models (specialized, often nationally distributed lending with heavy use of brokered deposits), and (iii) performance and resilience. From 2000 to 2025, we demonstrate that IBs—especially those with commercial parents—exhibit persistently higher ROA/ROE and stronger capitalization than other banks. Critically, this exists even during the Global Financial Crisis and the COVID-19 shock. IB funding relies disproportionately on (insured) brokered deposits and higher capitalization. As a result, IB failures are rare and small in aggregate relative to the broader system. IBs have niche business models that lead them to decrease market concentration and enable them to increase product innovation. Their market segments focus on specialized and underserved segments, often using information advantages from their parent company. We discuss the policy tradeoffs of layering additional regulation onto a framework in which FDIC/state oversight already mirrors that of comparable banks. We find little evidence that restrictions on commercial ownership would deliver stability benefits, but that they could increase compliance and innovation costs.

Abstract

The negative relation between idiosyncratic volatility and stock returns presents a puzzle, and many papers try to explain this phenomenon. Since the idiosyncratic volatility is unobservable, it is commonly estimated by the standard deviation of residuals relative to a specific asset pricing model. Thus, the presence of a missing pricing factor in the model might distort the estimation of the idiosyncratic volatility, and it can mislead the empirical analysis. We propose a methodology to test whether the idiosyncratic volatility puzzle is driven by the effect of a missing pricing factor. Using the Fama-French (2015) five-factor model to estimate the idiosyncratic volatility, we find that a missing pricing factor does not drive the negative relation between the idiosyncratic volatility and stock returns. This result suggests that the idiosyncratic volatility does affect the risk premium of its stock.