Michael Mauboussin - Overcoming Biases for Effective Decision-Making

Episode Summary

Executive Summary: The conversation explores how value investing blends rigorous valuation, competitive strategy, psychology, and process discipline. Mike Mabusin argues that markets are shaped by individual biases, group dynamics, incentives, and institutional frictions, so investors need diversity, structured decision processes, and margin-of-safety thinking to identify mispricing and avoid outcome-driven mistakes.

Main Topics: Value investing as an integrated framework (Priority: 5/5): Mabusin explains how Graham and Rappaport shaped his thinking: focus on cash flows, competitive strategy, and expectations rather than accounting earnings alone. How biases enter markets (Priority: 5/5): The discussion distinguishes psychology, social psychology, and sociology, arguing that biases matter most when they become correlated across investors and institutions. Diversity in decision-making and markets (Priority: 5/5): The podcast emphasizes cognitive diversity over demographic diversity, citing Scott Page’s work and the idea that groups can outperform individuals when diversity and aggregation mechanisms work well. Expertise, judgment, and the expert squeeze (Priority: 4/5): Mabusin is skeptical of expertise as commonly used; he distinguishes experience from predictive expertise and argues that machines and crowds are compressing the traditional role of experts. Process over outcomes (Priority: 5/5): A major theme is outcome bias: investors should evaluate decisions by the quality of the process, not just whether a trade made money. Tools to reduce bias (Priority: 5/5): He recommends practical safeguards such as investment journals, probabilistic language, base-rate analysis, premortems, and red-team/blue-team review. Passive investing, inefficiency, and information costs (Priority: 4/5): The discussion concludes that passive investing is rational in many cases, but markets cannot become fully efficient because some inefficiency is needed to reward information gathering and price discovery.

Key Arguments: Value investing should start with cash flows, returns on capital, and competitive strategy, not accounting earnings or P/E multiples. Stock prices encode more than fundamentals; they also reflect behavioral biases, institutional frictions, and incentives, so investors must study market structure as well as psychology. Individual biases do not automatically show up in prices; mispricing is more likely when diversity collapses and beliefs become correlated. Wisdom of crowds works only under three conditions: heterogeneity, a functioning aggregation mechanism, and proper incentives. Cognitive diversity can outperform raw intelligence alone on complex problems, because the best collective answer is often better than the average member of the group. The Asch experiment shows that people conform under social pressure, and neuroscience suggests conformity can alter perception while independence triggers fear responses. Expertise is domain-specific and most reliable in stable, linear environments; in rapidly changing, nonlinear systems, judgment matters more than formal expertise. Outcome bias is dangerous in investing because good luck can mimic good process and bad luck can punish good process; decisions should be logged and reviewed. Base-rate thinking and premortems help counteract overconfidence by forcing investors to compare a thesis with historical reference classes and imagine failure in advance. Machines and checklists can improve repeatable tasks and sizing decisions, but they cannot fully replace human theory-building in novel regimes or major structural change. Passive investing makes sense when information costs exceed likely alpha, but some inefficiency must remain for markets to function and reward price discovery.

Data Points: Start of Columbia Business School faculty role: 1993 - Mabusin notes he joined Columbia Business School in 1993, before behavioral finance was mainstream there. Teaching awards: 2 - He received the Dean’s Award for Teaching Excellence twice, in 2009 and 2016. Asch conformity result: about one-third - In the line-matching experiment, about a third of subjects conformed to the incorrect majority answer. Number of people in Asch setup: 8 - Seven confederates and one unsuspecting subject were used in the classic conformity experiment. Amazon revenue example: just over $100 billion - Used to illustrate the difficulty of extrapolating very high growth from a huge base. Revenue growth assumption in Amazon example: 15% for the next 5 to 10 years - Presented as an overly optimistic forecast that should be checked against base rates. Probability language survey: thousands of people - Mabusin and his son surveyed thousands of people to translate words like 'likely' and 'possible' into probabilities. Probabilistic phrasing: 20% example - He notes that phrases such as 'possible' can be interpreted very differently, sometimes around 20% or 80% depending on the listener. Margin of safety concept: steep discount to expected value - He defines margin of safety as buying far below expected value so errors do not cause permanent damage. Selected quote on value investing: “the marriage of a contrarian streak and a calculator” - Cited from Seth Klarman as a concise description of disciplined value investing.

Pivotal Quotes: "“value is about cash flows at the end of the day, not accounting numbers.”" — Michael Mabusin: Explaining the first core lesson he took from Alfred Rappaport's work. "“The collective will be more accurate than the average person within the collective.”" — Michael Mabusin: Summarizing Scott Page’s diversity prediction theorem and its implications for teams and markets. "“go down to Dwayne Reed and buy a notebook for $2 and write down your decisions.”" — Daniel Kahneman: Advice to keep an investment journal so investors can audit their reasoning and learn from outcomes.

Implications: Listeners should focus on disciplined process, cognitive diversity, and base-rate thinking rather than chasing narratives or outcomes. For the industry, the future likely blends human judgment, machine tools, and passive strategies, while preserving enough market inefficiency to reward genuine insight.