Causal Factor Analysis is a Necessary Condition for Investment Efficiency

Abstract

This article reveals the dire consequences of factor model misspecification in the context of portfolio optimization. We show that causal factor modeling is a necessary condition for investment efficiency, and that the prevailing (associational) factor modeling paradigm leads to investment inefficiency. Furthermore, we show that the biases introduced by factor model misspecification can be so large as producing portfolios where investors buy what they should sell and vice versa, even when investors have perfect knowledge of the mean and covariances. Our findings challenge the scientific soundness of the portfolio construction paradigms currently used by the asset management industry. To overcome these pitfalls, academics and practitioners should rebuild the financial economics literature on the more scientifically rigorous grounds of causal factor investing.

Keywords: Causal inference, causal discovery, confounder, collider, factor investing, portfolio construction, efficient frontier, investment losses