Harin de Silva on Portfolio Management (Podcast)

Episode Summary

Executive Summary: In this Masters in Business podcast, Barry Ritholtz interviews Harindra Da Silva, a pioneer in low-volatility and factor-based investing who leads Wells Fargo's Analytic Investors Group managing over $20 billion. Da Silva shares his journey from engineering in Sri Lanka to quantitative finance, explaining key concepts like the low-volatility anomaly, factor momentum, and the fundamental law of active management. He discusses how factors evolve, the importance of understanding hit rates rather than expecting constant outperformance, and offers insights on navigating the post-COVID market. The conversation covers specific strategies like their Global Dividend Opportunity Fund and Low Volatility U.S. Equity Fund, plus Da Silva's views on ESG as a risk screen, alternative data, and the crucial role of investor sentiment and environment in decision-making.

Main Topics: Low-Volatility Anomaly and Factor Investing (Priority: 5/5): Da Silva explains how low-beta stocks, contrary to traditional CAPM, offer better risk-adjusted returns than the market. He discusses factor momentum, persistence, and mean reversion—emphasizing that even the best factors outperform only 6-7 out of 10 times. Factors in the Current Market Cycle (Priority: 4/5): Da Silva analyzes which factors are working now (price-to-sales, estimate revisions, small cap) and which to avoid (trailing earnings yield, ROE/ROA due to pandemic-distorted data). He identifies early-cycle characteristics in today's factor behavior. Fundamental Law of Active Management (Priority: 4/5): Da Silva explains the formula he co-developed linking investment success to breadth, information coefficient, and transfer coefficient—quantifying how much outperformance a portfolio can achieve. ESG as a Risk Screen, Not Alpha Source (Priority: 3/5): Da Silva argues ESG factors, especially governance, are crucial for predicting tail risk and future volatility, but not for return forecasting. Poorly governed companies have fat-tailed returns invisible to standard risk models. Investor Sentiment and Decision-Making Environment (Priority: 3/5): Da Silva emphasizes how emotions, environment, and attention fatigue affect investment decisions. He recommends deliberate, unhurried decision processes and awareness of one's mental state when managing portfolios. Quantitative vs. Qualitative: The Misunderstood Nature of Factor Returns (Priority: 4/5): Da Silva challenges the expectation that factors will always deliver positive returns, highlighting recency bias, manager cycles, and the importance of understanding that factors have momentum and can underperform for years.

Key Arguments: Low-volatility anomaly: Portfolios of low-beta stocks provide equity risk premium with ~70% of market volatility, yielding superior Sharpe ratios. Factor persistence: Factors that worked in the last 1-2 years tend to continue working, but after 3 years mean reversion emerges—crucial for portfolio construction. Even best factors outperform only 60-70% of the time; three-year underperformance periods are normal and poorly communicated to investors. Current cycle favors forward-looking data: price-to-sales, analyst estimate revisions, high asset turnover; avoid trailing earnings and other pandemic-distorted accounting metrics. ESG factors (especially governance) are powerful volatility/risk predictors, not return predictors. They identify stocks with 15-30 year tail risk events. Investor environment matters: nature exposure improves decision-making (20% higher test scores); attention restoration is critical for portfolio managers. Alternative data is useful but must match investment horizon: satellite imagery suits short-term models, carbon footprint data suits long-term horizon strategies.

Data Points: Assets under management: $20 billion+ - Da Silva's Analytic Investors Group at Wells Fargo manages over $20 billion across quantitative strategies. Factor outperform rate: 60-70% - Even the best factors only outperform 6 or 7 out of 10 times, meaning 3-4 out of 10 years they underperform. Low vol fund target volatility: 70% - Da Silva's Low Volatility U.S. Equity Fund targets a volatility level of about 70% of the equity market. Factor persistence window: 1-3 years - Factors show very strong persistence over one year, less strong over 2-3 years, then start mean reverting after 3 years. Data costs: Millions of dollars annually - Da Silva notes their team's data costs are in the millions and continue rising despite lower overall costs for some data types. Bad governance tail risk frequency: Every 15-30 years - Companies with poor governance face significant negative return events roughly every 15-30 years, invisible to standard risk models. Attention restoration benefit: 20% higher test scores - Students walking through an arboretum before an exam scored 20% higher than those walking through urban environments.

Pivotal Quotes: "Even the best factors will outperform six or seven out of ten times. So you can think of that in one of the things. Meaning it's going to miss three or four out of ten times. So they underperform three to four out of ten times." — Harindra Da Silva: Da Silva explains the realistic hit rate of factor investing, countering the misconception that factors should consistently deliver positive returns. "If you're doing well, you're a genius, but if you're doing poorly, your style is out of favor. That's the euphemism. It's an asymmetrical bet." — Harindra Da Silva: Da Silva critiques how managers attribute success to skill and underperformance to temporary style headwinds, highlighting the importance of factor momentum. "I think the hardest thing in starting in the business right now is figuring out whether you're actually interested in investing or you're interested in what you think is investing." — Harindra Da Silva: Da Silva advises young professionals to distinguish between genuine interest in long-term research and the allure of short-term trading edges.

Implications: For investors: Factor-based strategies require patience—expect 3-4 years of underperformance even for good factors. ESG is best used for risk screening, not alpha generation. In current cycle, favor forward-looking metrics over trailing data. For portfolio managers: Account for factor momentum and mean reversion; manage decision environment to combat attention fatigue. For the industry: The rise of alternative data demands matching horizon to strategy, and carbon/sustainability factors will grow in importance.