Why Cliff Asness Believes Markets Are Getting Dumber

Episode Summary

Executive Summary: Cliff Asness argues markets are useful but imperfect, and he believes they’ve become more prone to extreme inefficiency due to bubbles, social media-driven mob behavior, and prolonged low rates. He defends a balanced blend of theory and data, sees ML/AI as improving many small inefficiencies, and is broadly skeptical of Bitcoin, while presenting ESG shorting as a logical extension of values-based investing.

Main Topics: Markets are imperfect and may be less stable than before (Priority: 5/5): Asness rejects perfect-efficiency straw men and argues markets have shown repeated episodes of extreme mispricing, especially in the dot-com era and during COVID-era trading. He frames markets as generally useful but vulnerable to bubbles and crowd behavior. Social media and mob psychology as market distortions (Priority: 5/5): He links social media to coordinated, less independent opinion formation, arguing it can turn the wisdom of crowds into a dangerous mob and amplify both market and political distortions. How quants and active managers affect efficiency (Priority: 4/5): The conversation explores whether strategies like value, factor, and momentum improve or worsen efficiency. Asness says value strategies are clearly pro-efficiency, while momentum can either correct underreaction or exploit/strengthen feedback loops. Data, theory, and machine learning in finance (Priority: 5/5): Asness describes his process as roughly half theory and half data, but says the mix has shifted toward data and ML because modern tools can detect subtle, nonlinear patterns humans miss. Bitcoin, crypto, and skepticism about monetary value (Priority: 4/5): He is mostly aligned with Gene Fama that Bitcoin is a bubble, though he allows that a future use case could change that. He views crypto largely as speculative and notes AQR trades some crypto only through trend-following models. ESG, shorting, and cost of capital (Priority: 4/5): Asness argues that excluding controversial companies lowers their prices and raises expected returns, and that allowing shorting is an even more powerful way to raise cost of capital for firms clients want to penalize. Politics, Trump, and the Overton window (Priority: 3/5): He says Trump has rapidly shifted the range of acceptable policy debate, and that politics increasingly resembles markets through crowd behavior, polarization, and reduced independence of opinion.

Key Arguments: Markets are not and have never been perfectly efficient; the real question is how imperfect they are and whether that has worsened over time. He believes markets have become prone to rare but extreme episodes of inefficiency, citing the dot-com bubble and COVID-era speculative surges. Social media can destroy independent judgment by coordinating opinions, undermining the wisdom-of-crowds mechanism that helps markets work. Value investing and other fundamental strategies tend to make markets more efficient by buying cheap assets and avoiding expensive ones. Momentum can be either pro-efficiency or anti-efficiency depending on whether it reflects underreaction to information or feedback trading driven by trends. Asness thinks machine learning will improve identification of subtle, nonlinear, and cross-factor relationships in finance, but it is unlikely to solve major bubble psychology quickly. He is skeptical of Bitcoin because its strongest uses appear tied to speculation and illicit activity, and a capped supply alone does not create value. ESG restrictions work by lowering prices and raising expected returns for non-ESG buyers; shorting intensifies that effect and can be a logical extension of ESG. Politics and markets both involve group behavior, and both can become less rational when coordinated by media and social platforms. He favors markets over political allocation because even flawed prices are generally better than centralized decision-making.

Data Points: AQR assets under management: $128 billion - Asness was described as managing this amount across various strategies. Chicago Booth donation: $60 million - Disclosure noted a gift from Cliff Asness and John Liu to name the master in finance program. AQR founding disclosure year: 1996 - He said he had managed ESG-restricted accounts for clients since 1996. Timeline of concern: Two major extreme-inefficiency episodes in about 20 years - He identified the dot-com bubble and COVID-era market mania as the clearest examples. Academic data horizon: Data going back to the 1920s; confidence mainly from the 1950s onward - He said one can literally go back to the 20s, but he feels more confident starting in the 50s. Popularity of low-rate era: Several decades of super low interest rates - Asness suggested long-running low rates may have encouraged excessive risk-taking and speculation. AQR ESG example: Restricted lists since 1996 - He cited longstanding client-driven exclusions, including a religion-based account that prohibited several 'sin stocks'. Machine learning shift: From roughly 50-50 to more data-heavy, about 65-35 (illustrative) - He suggested the firm has moved toward data mattering more relative to story over the last five years.

Pivotal Quotes: "Our goal is to make our clients money, not to make markets more efficient." — Cliff Asness: He explained that market efficiency is a secondary benefit, not the firm’s primary objective. "I think our goal is to make our clients money, not to make markets more efficient. That is a lovely secondary thing that I believe we help with, I hope we help with." — Cliff Asness: A more extended clarification that he believes active management can improve efficiency even when that is not the explicit aim. "has there ever been a better vehicle for turning a wise, independent crowd into a coordinated, clueless, even dangerous mob than social media?" — Cliff Asness: He described social media as a major mechanism behind market and political crowd distortion.

Implications: Listeners get a sharp case for treating markets as broadly informative but vulnerable to crowd-driven distortions. The episode suggests AI/ML will refine trading edges, not eliminate bubbles, and that ESG/crypto debates ultimately hinge on incentives, not slogans.