Anchoring: How the First Number Shapes Every Number After

What this means

Ask one group of people whether a used car is worth more or less than $18,000, then ask them to name a fair price. Ask a second group the same questions with $11,000 substituted. The second group will name a lower figure. The car is identical. The only difference is the number that entered the conversation first.

This is anchoringThe tendency to rely heavily on an initial piece of information when making subsequent judgments, so that estimates are pulled toward that starting value.. An anchor is any figure presented before a judgment is made, and it exerts a pull on that judgment even when the person knows the anchor is arbitrary, and even when they consciously try to discount it. The effect has been demonstrated with numbers chosen by spinning a wheel in front of the participants, which is about as clearly uninformative as a number can be.

The mechanism is usually described as insufficient adjustment. People treat the anchor as a starting point and move away from it toward what seems right, but they stop too early. The final estimate ends up closer to the anchor than an independent judgment would have been. Notice that this is a claim about the direction of a systematic bias, not a claim that people simply believe whatever they are told.

Anchoring interacts with a second condition that makes it far more powerful in practice: uncertainty about the true value. If you know precisely what a good is worth, an anchor has little room to work. The used car market is fertile ground for anchoring exactly because almost no buyer knows what a specific used car with a specific history is genuinely worth, and the seller knows considerably more. Where valuation is confident, anchoring weakens; where valuation is vague, it dominates.

This explains the opening move in almost every negotiation. A seller who would accept $12,000 posts $16,500. The posted price is not a mistake, a lie, or an expectation. It is a deliberately placed anchor, and it does two things at once. It shifts the buyer's sense of the reasonable range upward, so that a "negotiated" price of $14,000 registers as a win rather than an overpayment. And it establishes the seller's bargaining rangeThe set of prices a negotiator treats as plausible outcomes, which the opening offer helps define. before the buyer has stated one.

A related effect compounds it. Once the buyer has moved from $16,500 to $14,000, the reduction itself feels like value received, independent of whether $14,000 is a good price. That framing — evaluating the outcome by movement from the anchor rather than by the merits of the final number — is what makes high posted prices worth the risk of scaring some buyers away.

The defense is simple to state and hard to execute: determine your own valuation before encountering the seller's number. Research comparable sales, set a maximum, and write it down. An anchor cannot pull a judgment you have already made on independent evidence. A buyer who arrives knowing the model typically sells in a certain range treats $16,500 as information about the seller, not about the car.

Why it matters

You will negotiate salaries, rents, and prices your entire adult life, and in most of those negotiations someone will name a number first. Understanding that the first number is a tactical choice rather than a statement of value changes how you respond to it. It also tells you when to move first: if you have the better information about value, opening sets the anchor in your favor; if you have the worse information, letting the other side open gives you data at the cost of ceding the anchor.

Anchoring is also embedded in ordinary retail, where no negotiation happens at all. A crossed-out "original price" beside a sale price is an anchor. A premium option placed beside the one the seller actually wants to sell makes the middle option look moderate. A suggested donation amount printed on a form raises average donations. In none of these cases is anyone lying, and in all of them the first number is doing work.

Try it

1. Run the anchoring experiment. Split the class in half without telling anyone why. Give both halves the same short description of an item whose value is genuinely uncertain — a used bicycle, a signed poster, a phone two generations old. Ask group A whether it is worth more or less than a high figure you choose, and group B whether it is worth more or less than a low one. Then ask both groups to write down a fair price.
2. Collect and compare the two sets of estimates. Compute each group's average and range, and plot them. Present the results before revealing the manipulation.
3. Reveal the design and discuss. Ask students who were anchored high whether they believe the anchor affected them. Most will say no. Record that response, because the gap between the measured effect and the perceived effect is a central finding here.
4. Answer the dealer question in writing. Explain why a car dealer posts a price above what the dealer will accept. Address both functions of the anchor: shifting the buyer's sense of the plausible range, and making any concession feel like value delivered.
5. Answer the haggling question. Explain why, when bargaining over any good, the seller opens above the seller's true minimum. Then work the mirror image: explain why the buyer opens below the buyer's true maximum, and what happens when both sides anchor aggressively at once.
6. Consider the cost of an aggressive anchor. Anchoring high is not free — an extreme opening can end a negotiation before it starts or signal bad faith. Describe the trade-off a seller faces in choosing how high to open, and identify what the ideal anchor depends on.
7. Build the defense. Choose a real item you might actually buy. Before looking at any seller's listing, research comparable sales and write down your maximum price with the evidence supporting it. Then look at three listings and record whether your number moved. If it did, identify what moved it.
8. Audit anchoring in retail. Photograph or note five examples of an anchor used in a store or online where no haggling occurs: crossed-out prices, "compare at" figures, suggested donation amounts, tiered pricing with an expensive top option, quantity anchors like "limit 4 per customer." For each, explain what judgment the anchor is shaping.
9. Take a position in one paragraph. A posted price above the acceptable price is not a false statement, since the seller will genuinely sell at that price if someone pays it. Argue whether the practice is deceptive, and identify what would have to be true for it to cross into deception.

Check yourself