Anchoring bias in investing: why the first number you see distorts every decision after

The first number an investor encounters—a purchase price, an analyst target, a 52-week high—tends to stick. Even when that number is irrelevant to the current decision, it shapes subsequent judgements in ways investors rarely notice. Tversky & Kahneman (1974) identified this tendency as anchoring: the cognitive habit of adjusting estimates insufficiently away from an initial reference point. In investing, it consistently distorts sell decisions, valuation judgements, and assessments of what constitutes a fair price.

What anchoring bias is

Anchoring is a psychological mechanism described by Tversky & Kahneman in their landmark paper Judgment Under Uncertainty: Heuristics and Biases (1974). When forming a numerical estimate, the mind starts from an initial value—the anchor—and adjusts from it. The adjustment is typically insufficient: the final estimate remains too close to the anchor, even when the anchor is arbitrary or clearly irrelevant.

The brain uses anchors because processing information from scratch is cognitively expensive. A reference point reduces the computational load. The cost is systematic bias: decisions are shaped by whatever number happened to come first, not by a fresh assessment of available evidence.

How it manifests in investing

Anchoring appears in investment decisions in several recurring forms.

Anchoring to purchase price. An investor who bought a stock at £80 and watches it fall to £50 often refuses to sell until it "gets back to £80"—even if the fundamentals no longer support that valuation. The purchase price is functioning as an anchor, not as information. The relevant question is whether the asset is worth holding at £50 today, which has no necessary connection to what was paid for it.

Anchoring to analyst price targets. Published price targets are treated as authoritative fair-value estimates long after the analysis underpinning them has become stale. When a target is revised, investors update insufficiently—the original figure continues to influence their estimate of what the asset is worth.

Anchoring to 52-week highs and lows. Investors frequently treat arbitrary historical price levels as signals of fair value—believing a stock is cheap because it trades near its 52-week low, or expensive because it trades near its 52-week high. These levels reflect past prices, not current fundamentals.

Anchoring is distinct from the disposition effect—the tendency to sell winners too early and hold losers too long—but both biases often reinforce each other when the purchase price is used as the anchor.

The cost

Anchoring to purchase price causes investors to hold losing positions well beyond the point at which a fundamentals-based assessment would support selling. The cost is not only the continued loss but also the opportunity cost of capital that could be redeployed. Odean (1998), in a study of retail brokerage accounts, found that investors who held losers rather than selling and reinvesting bore an opportunity cost of approximately 3.4 percentage points per year relative to those who acted on current prices rather than historical anchors. The anchor does not protect the investor from further loss—it only makes it harder to act on information suggesting the loss should be cut.

What helps

Systematic investment strategies evaluate assets on the basis of current price signals and factor scores, not historical reference prices. A momentum strategy, for example, ranks assets by recent price performance—a current observation—and rebalances accordingly. The purchase price of any given position plays no role in the sell decision. Removing the historical anchor from the decision process removes the primary surface on which anchoring bias operates.

Anchoring bias in pfolio

pfolio's systematic portfolios evaluate assets exclusively on current price signals and factor scores. The entry price of any position plays no role in any calculation: there is no mechanism through which the original purchase price can function as an anchor in the buy or sell decision. A momentum strategy exits when the signal reverses; a volatility-scaling model adjusts when current volatility changes. Neither depends on what was paid for the position. This removes the primary reference point through which anchoring bias operates in investment decisions. See how we build portfolios for a full description of the signal logic.

Related articles

• Disposition effect
• Loss aversion
• Confirmation bias