Key Takeaways
- • Hudson Bay Capital's Sander Gerber developed the "Gerber Statistic," a new measure of co-movement between securities that challenges the traditional covariance framework used in Modern Portfolio Theory since Harry Markowitz's foundational work.
- • Research published in the February 2022 Journal of Portfolio Management demonstrates that portfolios constructed using the Gerber Statistic outperform those using historical covariance across three key metrics: cumulative return, average geometric return, and Sharpe ratio.
- • The Gerber Statistic filters out noisy data in correlation analysis, enabling investors to identify genuine relationships between securities and construct portfolios that are actually as diversified as intended, rather than falsely assuming diversification based on flawed covariance measures.
- • Markowitz, the father of Modern Portfolio Theory, collaborated with Gerber to validate the new statistic rather than defend the existing framework, demonstrating institutional acceptance of improved portfolio construction methodologies.
Summary
Sander Gerber, portfolio manager at Hudson Bay Capital, set out to address a fundamental limitation he identified in traditional portfolio construction: the inadequacy of historical covariance as a measure of how securities move relative to one another. While building a portfolio management system, Gerber developed an alternative metric—the Gerber Statistic—designed to provide clearer insight into actual portfolio diversification by measuring co-movement between assets.
The impetus for this work stemmed from Gerber's observations that historical covariance, the standard measure employed by practitioners of Modern Portfolio Theory, failed to capture the true diversification characteristics of his portfolios. His background as an equity options market maker on the floor of the American Stock Exchange informed his perspective, giving him what he described as a "guttural" understanding of how markets actually function beyond theoretical models.
Rather than facing resistance from academia, Gerber found an unexpected ally in Harry Markowitz himself, the architect of Modern Portfolio Theory. The two collaborated on research validating the new approach, with findings published in the February 2022 issue of the Journal of Portfolio Management. The research demonstrates measurable performance advantages: portfolios constructed using the Gerber Statistic outperformed those relying on historical covariance across cumulative return, average geometric return, and Sharpe ratio metrics.
The Gerber Statistic operates by establishing thresholds that filter noise from correlation data, distinguishing genuine relationships between securities from statistical artifacts that may suggest spurious connections. This refinement addresses a critical flaw in traditional covariance analysis: investors may construct portfolios that appear diversified based on historical correlation measures while unknowingly concentrating risk. By more accurately identifying whether securities move in tandem, opposition, or independence, the statistic enables more sophisticated risk management and genuine portfolio diversification.
The impetus for this work stemmed from Gerber's observations that historical covariance, the standard measure employed by practitioners of Modern Portfolio Theory, failed to capture the true diversification characteristics of his portfolios. His background as an equity options market maker on the floor of the American Stock Exchange informed his perspective, giving him what he described as a "guttural" understanding of how markets actually function beyond theoretical models.
Rather than facing resistance from academia, Gerber found an unexpected ally in Harry Markowitz himself, the architect of Modern Portfolio Theory. The two collaborated on research validating the new approach, with findings published in the February 2022 issue of the Journal of Portfolio Management. The research demonstrates measurable performance advantages: portfolios constructed using the Gerber Statistic outperformed those relying on historical covariance across cumulative return, average geometric return, and Sharpe ratio metrics.
The Gerber Statistic operates by establishing thresholds that filter noise from correlation data, distinguishing genuine relationships between securities from statistical artifacts that may suggest spurious connections. This refinement addresses a critical flaw in traditional covariance analysis: investors may construct portfolios that appear diversified based on historical correlation measures while unknowingly concentrating risk. By more accurately identifying whether securities move in tandem, opposition, or independence, the statistic enables more sophisticated risk management and genuine portfolio diversification.
Go deeper than the headline
Research every firm in this story, track the ones that matter and get the daily briefing — free for 7 days.