Key Takeaways
- • Equity market correlations have nearly tripled from 5.7% (1962-1997) to 15.9% (2021-April 2026), with correlation itself now 90% correlated to market volatility, meaning diversification benefits collapse precisely when investors need them most.
- • The diversifiable share of equity risk has fallen from 94% in early periods to significantly lower levels in recent years, with peak correlation reaching 36.8% in April 2025—more than six times pre-1998 averages—demonstrating that traditional Modern Portfolio Theory frameworks are inadequate for current market conditions.
- • Daniel Gelernter, Director of AI and Machine Learning at R.G. Niederhoffer Capital Management, presented evidence that most family offices, allocators, and CIOs maintain "Diversification In Name Only" (DINO) portfolios that appear diversified on paper but behave as single concentrated bets during market stress.
- • Modern Portfolio Theory, fundamentally unchanged since Markowitz's 1952 framework, no longer reflects market mechanics that have transformed over seven decades, requiring portfolio construction and risk management approaches to be reconsidered for contemporary volatility environments.
Summary
An Opalesque webinar featuring Daniel Gelernter, Director of AI and Machine Learning at R.G. Niederhoffer Capital Management and Chief Investment Officer of the Niederhoffer Family Office, presented institutional investors from 34 countries with updated empirical evidence that conventional diversification strategies have become largely ineffective in modern markets. Gelernter's analysis challenges the foundational assumptions underlying diversification frameworks still widely employed by family offices, allocators, and institutional investors.
The core argument centers on what Gelernter terms "Diversification In Name Only" (DINO)—portfolios that maintain apparent diversification across asset classes and holdings but demonstrate highly correlated behavior during periods of market stress. This phenomenon reflects a fundamental disconnect between the theoretical models guiding portfolio construction and the empirical realities of contemporary markets. Modern Portfolio Theory, which has remained essentially unchanged since Harry Markowitz's seminal 1952 publication, was developed for market conditions that no longer exist.
Recent correlation data reveals the magnitude of this shift. Average pairwise daily-return correlations across U.S. equities have increased substantially: from 5.7% during the 1962-1997 period to 13.4% from 1998-2020, and further to 15.9% between 2021 and April 2026. More critically, these correlations are not static across market conditions. Gelernter's research demonstrates that equity correlations themselves are approximately 90% correlated with market volatility, meaning correlations spike dramatically during market downturns—precisely the periods when portfolio diversification is most essential. The April 2025 data point illustrates this risk acutely: monthly correlation reached 36.8%, more than six times the pre-1998 average and occurring during peak market stress when diversification benefits are most needed.
The practical consequence is measurable: the diversifiable share of equity risk, measured as one minus average pairwise correlation, has contracted significantly from historical levels of 94%. This erosion of diversification efficacy suggests that institutional investors relying on traditional correlation-based portfolio construction face material concentration risk during volatile periods, despite holdings that appear diversified by conventional metrics. The presentation underscores the need for portfolio managers to reassess risk frameworks and stress-testing methodologies in light of empirical evidence that market structure and correlation dynamics have fundamentally shifted from the assumptions embedded in decades-old portfolio theory.
The core argument centers on what Gelernter terms "Diversification In Name Only" (DINO)—portfolios that maintain apparent diversification across asset classes and holdings but demonstrate highly correlated behavior during periods of market stress. This phenomenon reflects a fundamental disconnect between the theoretical models guiding portfolio construction and the empirical realities of contemporary markets. Modern Portfolio Theory, which has remained essentially unchanged since Harry Markowitz's seminal 1952 publication, was developed for market conditions that no longer exist.
Recent correlation data reveals the magnitude of this shift. Average pairwise daily-return correlations across U.S. equities have increased substantially: from 5.7% during the 1962-1997 period to 13.4% from 1998-2020, and further to 15.9% between 2021 and April 2026. More critically, these correlations are not static across market conditions. Gelernter's research demonstrates that equity correlations themselves are approximately 90% correlated with market volatility, meaning correlations spike dramatically during market downturns—precisely the periods when portfolio diversification is most essential. The April 2025 data point illustrates this risk acutely: monthly correlation reached 36.8%, more than six times the pre-1998 average and occurring during peak market stress when diversification benefits are most needed.
The practical consequence is measurable: the diversifiable share of equity risk, measured as one minus average pairwise correlation, has contracted significantly from historical levels of 94%. This erosion of diversification efficacy suggests that institutional investors relying on traditional correlation-based portfolio construction face material concentration risk during volatile periods, despite holdings that appear diversified by conventional metrics. The presentation underscores the need for portfolio managers to reassess risk frameworks and stress-testing methodologies in light of empirical evidence that market structure and correlation dynamics have fundamentally shifted from the assumptions embedded in decades-old portfolio theory.
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.