Forthcoming Issue
Volume 24, No. 2, Second Quarter 2026
-
Article
Adapting to AI: Rethinking Labor Income and Retirement Design in a Changing Economy
In this paper, we discuss how artificial intelligence (AI) could impact lifetime income through the channel of occupational tasks and wages. We first derive a life-cycle model of consumption, then incorporate AI as a shock to income and longevity. We find that AI-derived impacts on labor income and longevity can influence wealth accumulation and consumption over an individual’s lifetime. We test a key input to this model in which occupations that experienced a net drop in tasks demanded of workers suffer declines in total wages. We then use a large language model (LLM) to estimate how susceptible occupations can be to task reduction from automation. Our results suggest that most occupations will be affected by AI-driven automation but with a wide dispersion across jobs and industries. We conclude with a discussion on why AI-driven changes in the labor market should prompt new approaches to retirement and explore how financial planning can help workers adjust to an AI-enhanced economy.
-
Article
A Tail of Five Skews
We show that a highly statistically significant total skewness risk premium is embedded in the cross-section of returns within a large universe of multi-asset futures and forwards, within a broad set of 215 long/short style factors formed on that universe, and within the cross-sectional equity factor zoo. The skewness risk premium is most robust when it is measured in a relatively new, intuitive way that minimizes the impact of outliers while still capturing information in the tails, which we demonstrate by evaluating five candidate methods across a battery of empirical tests. We show there is compensation for bearing skewness risk in both the long run and ex ante on a point-in-time basis available to investors.
-
Survey & Crossover
Option Return Anomalies
This paper surveys empirical option return anomalies in both unhedged and delta-hedged strategies. We document persistent deviations from classical models, including the sharply negative returns on out-of-the-money puts, the negative performance of zero-beta straddles, and the large losses of delta-hedged portfolios documented in many studies. These patterns remain difficult to reconcile because most option-pricing models provide valuation formulas but not the formulas for finite-horizon expected returns or higher-order return moments. Equivalent Expectation Measures (EEMs) and Multiverse EEMs (MEEMs) supply closed-form expressions of these moments, enabling cleaner comparisons between model-implied and realized returns. We outline how these tools can be used to address longstanding anomalies and highlight future research opportunities for explaining cross-sectional predictability, volatility risk premia, and finite-horizon option returns.