Reading 87

Traditional finance assumes individuals are perfectly rational and process all relevant information objectively, producing efficient markets. Research by Kahneman and Tversky (1970s onward) showed instead that human decision-making contains systematic biases that lead to predictably irrational decisions. Better understanding of these biases by clients and professionals should, over time, push prices and returns closer to what traditional theory expects.

Behavioral finance argues that certain biases are widespread and predictable, not random.

• Cognitive errors — driven by faulty reasoning: failure to apply statistical analysis, information-processing mistakes, illogical reasoning, or memory errors. Can be reduced through awareness, training, and more information.
• Emotional biases — driven by feelings, impulses, or intuition rather than conscious thought. Difficult to overcome and often must be accommodated rather than corrected.

Cognitive errors fall into two groups:

1. Belief perseverance — irrational reluctance to change prior conclusions.
2. Processing errors — flawed analysis of new information.

Cognitive dissonance is the stress felt when a current belief conflicts with new information. Individuals reduce dissonance by either updating their belief or (more commonly) discounting the new information — producing bias in favor of currently held beliefs.

1. Conservatism bias — investors form an initial view rationally but then fail to update it as new information arrives. Prior probabilities are overweighted; new data is under-incorporated, especially when it is complex.

Implication: investors hold positions too long; they avoid the mental effort of revising forecasts.

2. Confirmation bias — selectively seeking information that supports prior beliefs and avoiding/diminishing conflicting evidence. New car buyers read positive reviews and skip recall stories. May lead to:

• Considering positive but ignoring negative information
• Designing decision processes that confirm a preferred conclusion
• Overconfidence in present beliefs

Mitigation: deliberately seek out contrary views and analyses.

3. Representativeness bias — using surface characteristics to classify an investment, then assuming it shares the rest of the category's traits. Two sub-forms:

• Base-rate neglect — ignoring the base rate of a characteristic in the population (e.g., judging a "shy man" as a librarian, ignoring that male salespeople vastly outnumber male librarians).
• Sample-size neglect — drawing strong conclusions from small samples (e.g., treating a 3-year hot streak as proof of skill).

4. Illusion of control bias — believing you can influence outcomes you cannot. Often paired with illusion of knowledge, self-attribution, and overconfidence. Investors may overweight stocks of their employer or familiar companies, producing under-diversified portfolios.

5. Hindsight bias — selective memory: remembering correct predictions, forgetting wrong ones, and treating past events as having been more predictable than they were ("I-knew-it-all-along"). Symptoms:

• Distorting earlier predictions when looking back
• Viewing past events as inevitable
• Believing past outcomes were foreseeable

Implication: overconfidence in forecasting ability; abandoning sound techniques that happened to underperform in favor of weak ones that worked by chance.

6. Anchoring and adjustment — basing expectations on a prior reference number and adjusting insufficiently from it. Examples: valuing a security relative to its current price; estimating EPS relative to a previous figure. New data should be assessed objectively, not as deviations from an anchor.

7. Mental accounting — treating money differently depending on which "account" or source it came from, instead of applying a portfolio view. Examples: investing a bonus in a speculative biotech because it is "found money"; segregating an inheritance into low-risk bonds out of sentiment. A common form is treating income differently from capital appreciation, leading to suboptimal income/growth mixes.

8. Framing bias — how a question is presented changes the answer. Tversky & Kahneman's 600-lives example:

Investment implication: over-emphasizing short-term volatility (a "loss" frame) produces overly conservative portfolios. Risk-tolerance questionnaires must be carefully worded.

9. Availability bias — judging probability by how easily examples come to mind. Recent and vivid events feel more likely; distant or unmemorable ones feel less likely.

Implications: selecting investments by name recognition or recent advertising; limiting the universe to familiar firms (poor diversification); over-reacting to recent or media-heavy events.

Six biases generally arise from emotion rather than reasoning:

1. Loss-aversion bias — the pain of a loss is felt more strongly than the pleasure of an equivalent gain (Kahneman & Tversky 1979). Faced with gains, people prefer certainty; faced with losses, people accept risk to avoid them.

Both scenarios produce identical outcome distributions ($15 sure vs. $10/$20 coin flip). The flip in preference is driven by the reference point.

Implications: selling winners too early ("disposition effect") and holding losers; taking excessive risk to recover; framing-dependent treatment of the same position as a "win" or a "loss".

2. Overconfidence bias — overestimating one's own ability or reasoning. Two flavors:

• Prediction overconfidence — underestimating uncertainty / standard deviation of forecasts.
• Certainty overconfidence — overstating the probability of being right.

Reinforced by self-attribution: credit successes to oneself (self-enhancing), blame failures on circumstances (self-protecting). Although partly cognitive, it is treated as emotional because it is driven by the desire to feel good and is hard to correct. Implications: underestimating risk, overestimating return, insufficient diversification.

3. Self-control bias — favoring short-term satisfaction over long-term goals (a.k.a. hyperbolic discounting). Implications: insufficient retirement savings; later forced to take excess risk to catch up; over-allocation to income-producing assets to fund near-term consumption. Mitigation: set an investment plan and savings budget, review regularly.

4. Status quo bias — sticking with current allocations because change is uncomfortable. Studies show automatic enrollment (with opt-out) sharply raises retirement-plan participation vs. opt-in (Thaler & Sunstein 2008, Nudge). Implications: portfolios drift away from appropriate risk; better alternatives are ignored.

5. Endowment bias — valuing an asset more once you own it. Common with inherited holdings. Diagnostic question: "Would you buy this with new money today?" Studies show minimum-sale prices (e.g., $25) systematically exceed maximum-purchase prices (e.g., $23) for the very same asset. Implications: failure to sell positions that no longer fit; comfort-driven concentration in familiar names.

6. Regret-aversion bias — doing nothing because action might be wrong. People weigh errors of commission (acting and being wrong) more heavily than errors of omission (failing to act when the action would have been right). Herding is a form of regret aversion — going with the crowd reduces personal blame. Implications: excessive portfolio conservatism, since risky assets sometimes underperform and the regret is sharp.

Many "anomalies" once cited as evidence of inefficiency have been re-explained by small samples, time-period bias, or risk-model misspecification. Behavioral finance offers complementary explanations.

Bubbles and crashes. Behavioral finance does not provide a full explanation, but several biases are visible during bubble episodes:

• Overconfidence → overtrading, underestimation of risk, lack of diversification.
• Self-attribution + hindsight → persistently good outcomes feed further overconfidence (investors give themselves credit for choosing winners in a bull market).
• Confirmation → ignoring or reframing warning signs; treating early declines as buying opportunities.
• Anchoring → believing recent highs are "fair" prices even after the decline starts.
• Regret aversion → even skeptics stay invested, fearing the regret of missing further upside.

Note: arbitraging a bubble is hard — short-selling is restricted and risky in a rising market, and leaving the market too early can be just as career-damaging as staying too long.

Value vs. growth anomaly. Value stocks (low P/B, low P/E, high dividend yield) have historically outperformed growth stocks. Fama & French (1992) showed that adding firm-size and book-to-market risk factors largely eliminates the anomaly — i.e., value-stock returns can be interpreted as compensation for additional risk. A behavioral counter-explanation is the halo effect: a version of representativeness in which a company's good qualities (fast growth, rising price) get extended into "this is a good stock", overvaluing growth stocks.

Home bias. Investors over-allocate to domestic firms (or to firms in their region of a country) versus what global diversification would suggest. Drivers: belief in better local information access, or simply emotional preference for "closer-to-home" companies. A related effect: investors may underestimate risk or overestimate return for firms whose products or marketing they are heavily exposed to.