Episode Summary
Executive Summary: Kent Daniel traces his path from physics to finance and explains how his research challenged the CAPM and Fama-French framework by showing that characteristics, not just covariances, drive returns. The conversation ties behavioral finance, intangible information, short-selling constraints, and changing market efficiency to practical investing and the Goldman quant experience.
Main Topics: Kent Daniel’s background and path into finance (Priority: 4/5): Daniel describes growing up in Southern California, studying physics at Caltech, working in aerospace, then moving into an MBA and PhD at UCLA, where exposure to finance research redirected his career. Testing predictability in market returns (Priority: 5/5): He explains his dissertation on time-varying expected market returns and the importance of test power in detecting predictability across bear and bull markets. Behavioral vs. rational explanations in asset pricing (Priority: 5/5): The discussion revisits the 1990s debate over whether anomalies reflected risk premia or investor mistakes, especially in value stocks and market-level predictability. Critique of the Fama-French three-factor model (Priority: 5/5): Daniel summarizes his influential characteristics-versus-covariance paper, arguing that stock characteristics like book-to-market matter more than factor covariation in explaining returns. Behavioral finance, momentum, and underreaction (Priority: 5/5): He revisits his classic paper linking overconfidence and self-attribution to momentum, long-term reversal, and delayed price reactions to corporate announcements. Intangible information and mispricing (Priority: 5/5): Daniel explains that value reversals are driven by the part of returns not explained by accounting fundamentals, suggesting markets misprocess intangible information. Limits to arbitrage, short selling, and market efficiency (Priority: 5/5): He discusses how high borrow costs and short-sale frictions can sustain mispricing, and how this affects both market efficiency and the persistence of overpriced securities.
Key Arguments: Predictability in market returns exists, but detecting it requires powerful tests; weak tests can miss real effects. The market return varies with the business cycle: expected returns tend to be high in recessions and low in booms. The value effect is not fully explained by rational risk; behavioral explanations remain compelling, especially when survey expectations are pessimistic at recession troughs. The Daniel-Titman characteristics-vs-covariance test suggests that stock characteristics, not Fama-French factor loadings, drive expected returns. Momentum, underreaction, and long-horizon reversal can be unified by overconfidence and self-attribution bias. A large share of return variation over five years is explained by accounting fundamentals, but the residual “intangible” component is where mean reversion is concentrated. High borrow costs create a direct and persistent limit to arbitrage, helping overpriced stocks remain overpriced. Short selling is economically useful because it supports price discovery, but it is constrained by lending frictions and regulation. Indexing in broad, low-information funds is not a major threat to efficiency, but leveraged and exotic ETFs may pose real risk to uninformed investors. Market efficiency should be understood as nuanced and friction-filled, not as a binary state.
Data Points: Caltech selectivity: Most selective college in the United States - Used to emphasize the academic rigor of Daniel’s undergraduate education Goldman QIS leadership timeline: 2004–2010 - Daniel worked in Goldman Sachs Asset Management’s Quantitative Investment Strategies group during this period Managing director promotion: 2005 - He became managing director and head of equity research at QIS Co-CIO appointment: 2009 - Daniel became co-chief investment officer at QIS Dissertation structure: 3 chapters / 3 papers - Daniel’s PhD dissertation consisted of three papers on return predictability Regression explanatory power: about 60% - Over a five-year horizon, fundamental measures explained roughly 60% of cross-sectional return variation Residual return component: about 40% - The remaining return variation was attributed to intangible information not captured by accounting data Median borrow cost for U.S. stocks: 25 basis points (0.25% per year) - Typical short-borrow fee for most large stocks Historic borrow cost extreme in DiVoglia’s paper: 70% - Highest borrow cost observed in the referenced 2002 study over its sample period Current borrow cost extremes: 100%+ for many stocks; up to 1,000% annualized - Daniel describes contemporary hard-to-borrow stocks as dramatically more expensive to short Hard-to-borrow portfolio return: -80% alpha per year - Portfolio of stocks with very high borrow costs lost about 80% annually over 2010–2025 Sample period for short-borrow study: 2010–2025 - Time frame for the team’s new short-selling and borrow-cost analysis Quant crisis timing: August 2007 - The one-week period in which quant strategies collapsed dramatically Goldman factor count during quant crisis: 26 factors - Daniel says the group was using a multifactor optimized portfolio at the time Borrow cost persistence: high and persistent - Daniel emphasizes that hard-to-borrow mispricing persists for long periods
Pivotal Quotes: "if you couldn't explain the ideas to people who didn't have a strong background in physics, that probably meant that you yourself didn't understand it" — Kent Daniel: Reflecting on lessons learned from Richard Feynman and the value of clear explanation "What seemed to matter was the characteristics and not the covariances." — Kent Daniel: Summarizing the main result of the characteristics-versus-covariance paper "I think we've come to now is we need a more nuanced understanding of market efficiency." — Kent Daniel: His view on how the literature and practice should interpret market efficiency today
Implications: Investors should expect mispricing to persist where frictions, short-sale constraints, and intangible signals matter. The future of asset pricing likely depends on combining behavioral insight, better fundamentals, and realistic limits to arbitrage.
About Value Investing with Legends
Value investing is more than an investment strategy — it's a fundamental way of thinking about finance. Value investing was developed in the 1920s at Columbia Business School by professors Benjamin Graham and David Dodd, MS '21. The authors of the classic text, Security Analysis, Graham and Dodd were the very pioneers of their field and their security analysis principles provided the first rational basis for investment decisions. Despite the vast and volatile changes in the economy and securities markets during the last several decades, value investing has proven to be the most successful money management strategy ever developed. Value investors' success over the second half of the twentieth century proved not only the validity of the value approach, but its preeminence over even the most widely taught and practiced modern investment theory, which was developed in the 1950s and '60s and remains dominant even today. Our mission today is to promote the study and practice of Graham & Dodd's original investing principles and to improve investing with world-class education, research, and practitioner-academic dialogue. In this podcast you will hear from some of the world's greatest investors, their views on the investment management industry, how they developed their investment process and how they see the field changing over time.