The Original Finding
In 1974, Amos Tversky and Daniel Kahneman published their landmark paper "Judgment Under Uncertainty: Heuristics and Biases" in Science. One of the phenomena they documented was anchoring: the tendency for people to rely disproportionately on the first piece of numerical information they receive when making subsequent estimates or decisions. In their original experiment, participants watched a wheel of fortune spin and land on a number between 0 and 100. They were then asked to estimate the percentage of African countries in the United Nations. Despite the wheel being demonstrably random and entirely unrelated to the question, participants who saw the wheel land on 65 guessed significantly higher than those who saw it land on 10.
The implication was direct: an arbitrary number, even one that everyone present knew to be arbitrary, exerted measurable pull on subsequent judgment. The anchor did not need to be credible to be effective. It needed only to exist.
Insufficient Adjustment: Why Smart People Do Not Escape It
The cognitive mechanism underneath anchoring is what Kahneman later termed insufficient adjustment. When a person receives an anchor, they begin from that point and adjust toward what they believe is a more accurate estimate. The problem is that the adjustment consistently falls short. The mental energy required to move away from the anchor is real, and most people stop adjusting before they have fully corrected for the anchor's distortion. This is not a failure of intelligence. It is a feature of how the brain processes numerical reference points. More information, more time, and more deliberate reasoning reduce the effect at the margins but do not eliminate it.
Research by Gregory Northcraft and Margaret Neale, published in 1987 in Organizational Behavior and Human Decision Processes, demonstrated the effect among real estate professionals. Experienced agents shown an inflated listing price for a property consistently appraised that property higher than agents shown a deflated listing price, even when both groups had access to identical information about the property's actual market comparables. When debriefed, the agents denied that the listing price had influenced them. Their final numbers showed otherwise.
"The anchor does not need to be credible to work. It needs only to arrive first."
Where It Operates
Salary Negotiation
The research on salary anchoring is consistent and practical. A 2011 study by Galinsky and Mussweiler in the Journal of Personality and Social Psychology showed that job candidates who named a specific salary first obtained significantly better outcomes than those who waited for the employer to open. Candidates who anchored high also fared better than those who anchored at market rate, because the final offer adjusted from the anchor rather than from a neutral midpoint. The common advice to let the employer speak first reflects a misunderstanding of the mechanism. In most negotiation contexts, the person who sets the anchor exerts more influence than the person who responds to it.
Legal Settlements and Civil Damages
Plaintiff attorneys routinely use anchoring to shift jury expectations on damages awards. When a plaintiff's attorney requests $10 million in a personal injury case, the amount functions as an anchor even if the eventual award is a fraction of that figure. Research by Gretchen Chapman and Brian Bornstein published in 1996 in Applied Cognitive Psychology showed that mock jurors awarded higher damages when exposed to a large plaintiff demand than when exposed to a small one, regardless of the underlying facts. Defense attorneys counter with their own anchoring: low damage figures introduced early, expert testimony emphasizing specific low numbers, and explicit challenges to the plaintiff's figure designed to re-center the jury's reference point before deliberation begins.
Retail Pricing
The crossed-out original price displayed next to a sale price is an anchor. The consumer is not evaluating whether the sale price is a good price. The consumer is evaluating the sale price relative to the anchor, the original price, which the retailer controls entirely. Williams-Sonoma famously used this when it introduced a bread maker at $429 alongside its existing $279 model. Sales of the $279 unit increased substantially, because the $429 anchor made the lower price feel like a deal even though nothing about the $279 bread maker had changed. The $429 unit did not need to sell. Its function was to reframe what the $279 unit cost.
Strategic Deployment: Extreme First Offers
The most sophisticated practitioners of anchoring do not aim for a reasonable opening position. They aim for the most extreme number they can justify entering without destroying the negotiation entirely. Adam Galinsky's research at Columbia Business School on "first offer advantage" shows that extreme anchors shift final outcomes more than moderate ones, provided they remain within the threshold of plausibility. An anchor that is dismissed as absurd loses its pull. An anchor that is extreme but defensible forces the other party to spend their cognitive and negotiating energy on the adjustment process rather than on establishing their own reference point.
This is why skilled labor negotiators, M&A advisors, and commercial real estate brokers open at positions that would appear unreasonable to an outside observer. The goal is not to settle at the anchor. The goal is to own the center of gravity of the negotiation before the other side has had a chance to establish one.
Anchoring Signals
- The other party states a number, price, or valuation before you have established your own position
- A crossed-out or "original" price is visible alongside a discounted figure
- A salary range is presented before you have named your number
- An appraisal, estimate, or assessment is shared with you before you form your own view
- Your counteroffer feels like it is responding to their number rather than asserting your own
- The discussion of range begins from a point that you did not choose and have not challenged
- You find yourself adjusting from a figure someone else provided rather than arriving at your own independently
The Counter: Independent Generation Before Exposure
The only reliable defense against anchoring is to form your own reference figure before receiving the other party's opening. If you are about to enter a salary negotiation, calculate the number you intend to name before the conversation begins. If you are evaluating a property, estimate its value from comparables before reading the listing price. Once an anchor is received, awareness of the effect reduces it modestly but does not eliminate it. The cognitive contamination has already occurred. The practical leverage is in the preparation, not the recognition.
When anchoring is being used against you and you have not prepared a prior figure, the most effective response is to explicitly reject the anchor before engaging with any adjustment. Name it as an anchor, state that you are not treating it as a reference point, and introduce your own figure. This does not neutralize the effect entirely, but it prevents the negotiation from taking the anchor's gravitational field as its operating premise.