Anchoring and Adjustment

Anchoring and adjustment is a mental shortcut, or heuristic, in which people make an estimate by starting from an initial value, called the anchor, and adjusting away from it to reach their answer. The adjustment is usually too small, so the final judgement stays biased towards the anchor, even when the anchor is irrelevant or random. The effect appears in pricing, negotiation, forecasting and investing.

The idea was described by the psychologists Amos Tversky and Daniel Kahneman in a 1974 paper on judgement under uncertainty, and it became a cornerstone of behavioural economics. It sits alongside other biases such as the framing effect and loss aversion, and it is one reason economists question the assumption that people process information like fully rational calculators, a theme of bounded rationality.

When a market trader says "these shoes were $200, today only $90", the first number sticks in your mind. Even if the shoes were never really worth $200, $90 now feels like a bargain. Your brain started from the anchor and moved down, but not as far as it would have if you had judged the shoes on their own.

Investors do the same with prices they remember. A share that once traded at $100 can feel "cheap" at $60, even if nothing about the company supports $100 any more. The old number becomes a mental hook that is hard to shake off.

In their 1974 paper in the journal Science, Tversky and Kahneman described an experiment in which participants watched a wheel of fortune, rigged to stop at either 10 or 65, and were then asked to estimate the percentage of African countries in the United Nations. Those who saw 10 gave a median estimate of 25%; those who saw 65 gave 45%. The random number clearly affected answers. In another study, people asked to estimate the product of 8×7×6×5×4×3×2×1 within a few seconds gave much higher answers than those shown the same numbers in ascending order, because the first few multiplications served as an anchor. Later research found anchoring among experts too, including judges, estate agents and financial analysts.

• Past prices. Investors anchor on the price they paid, an all-time high or the 52-week high and low, and may hold losing positions waiting to "get back to even".
• Analyst forecasts. Analysts tend to revise earnings estimates slowly from previous figures, which may help explain why prices keep drifting after earnings surprises.
• Negotiation and deals. In mergers, the first price offered strongly shapes the final agreed price.
• Initial public offerings. The price range in an IPO prospectus anchors expectations of value.
• Credit cards. Research suggests that printing the minimum payment on a statement leads some borrowers to pay less than they otherwise would.
• Valuation. Using a round-number price target or a peer's multiple as a starting point can bias a valuation.

Crypto markets offer many anchors. Traders often compare a token's price with its previous peak and describe a 90% fall as a "discount", even when the project has changed. Round numbers such as $100,000 for Bitcoin act as psychological anchors where buy and sell orders cluster. Token launches set anchors through presale prices and the headline fully diluted valuation. Price predictions shared online can anchor beginners' expectations about what is "normal", fuelling fear of missing out or unrealistic targets. Because crypto assets often lack clear measures of fundamental value, anchors may play an even larger role in price discovery than in traditional markets.

Researchers offer two main explanations. The original adjustment account says people start at the anchor and move step by step until they reach a plausible value, stopping too early because they are uncertain. The selective accessibility account says the anchor makes information consistent with it easier to recall, so people build their estimate around it. Both may operate together. Anchoring is stronger when people are uncertain, tired or rushed, and when the anchor seems relevant, but it can persist even when people are warned about it or offered rewards for accuracy.

Common techniques for reducing the effect include building an estimate from first principles before looking at prices or forecasts, considering reasons why the anchor might be wrong, using several independent reference points, and writing down a valuation before seeing others' views. Investment committees sometimes gather individual estimates before discussion to avoid anchoring on the first speaker. Pre-set rules, such as reviewing every position as if buying it today, help investors detach from purchase prices.

Some replication studies have found smaller anchoring effects than early experiments suggested, especially when anchors are clearly irrelevant or when people have strong knowledge of the subject. Using an anchor is also not always irrational: past prices and prior estimates often contain useful information, and adjusting gradually can be sensible when new data are noisy. The challenge is distinguishing helpful reference points from arbitrary ones.