In one of the most famous experiments in cognitive psychology, Tversky and Kahneman (1974) spun a wheel of fortune in front of participants. The wheel was secretly rigged to land on either 10 or 65. Participants were then asked: "What percentage of African countries are members of the United Nations?" The wheel number had nothing to do with the question — everyone knew it was random. And yet it mattered enormously. Participants who saw 65 guessed a median of 45%. Participants who saw 10 guessed a median of 25%. A completely arbitrary number had dragged their estimates in opposite directions.

This is the anchoring effect: the first piece of numerical information we encounter becomes a reference point — an anchor — that biases all subsequent estimates, even when we consciously know the anchor is irrelevant. The mind adjusts away from the anchor, but never far enough.

Median estimates by anchor group — Tversky & Kahneman (1974)

Anchoring is not a laboratory curiosity. Real estate agents shown a high listing price for a house give higher appraisals — even experienced professionals with market data in hand. A 2006 study by Ariely, Loewenstein and Prelec had MBA students write down the last two digits of their social security number before bidding on wine, chocolate, and computer equipment. Students with high two-digit endings bid 60–120% more than those with low endings.

Salary negotiation is one of the highest-stakes arenas for anchoring. Research consistently shows that whoever names a number first — employer or candidate — sets the anchor that pulls the final offer toward it. A 2011 study in Organizational Behavior and Human Decision Processes found that candidates who made the first offer received significantly higher salaries than those who waited, purely because they owned the anchor.

Fun fact: Retailers exploit anchoring constantly. The "was £199, now £99" label does not just communicate a discount — the crossed-out price becomes an anchor that makes £99 feel reasonable, even if the item was never sold for £199 in the first place. Several countries now require proof that a "was" price was genuinely charged for a minimum period before a sale can be advertised.

The anchoring effect is one of the few cognitive biases that does not diminish with expertise. Judges, doctors, auditors, and financial analysts all show anchoring effects of similar magnitude to undergraduates. Awareness of the bias helps slightly — but not enough to eliminate it. The practical defence is procedural: generate your own estimate before seeing any external number, commit it to paper, and only then expose yourself to the anchor. The pre-commitment moves the reference point to your own reasoning rather than the first figure you encountered.