Four groups hear the same account of a young woman in constant pain, expected to die within months. Only the damages request changes: $100, $20,000, $5 million or $1 billion.

The $100 group awards an average of $990.

That result is grotesque because the request is obviously silly. Nobody should need an advanced degree to realise that terminal cancer is worth more than a second-hand television. Yet the number still affected the judgement.

Anchoring is often demonstrated with spinning wheels and trivia, which makes it sound like a parlour trick for people bad at guessing African countries. Damages expose the uglier possibility. A figure can influence the scale before anyone has established that it belongs in the calculation.

The commercial versions are quieter. A supplier opens at £80,000 and £60,000 starts to feel restrained. A house advertised far above comparable sales makes the eventual reduction look like generosity. A project estimate spoken by the most senior person becomes the figure everyone else mysteriously reaches ten minutes later.

Even rejecting the opening number requires attention to it. “That is absurd” begins a negotiation from the absurd figure unless somebody brings a separate estimate. The room may argue fiercely while staying on a scale chosen by the person who spoke first.

Independent estimates help, but experts arrive with favourite numbers too. A benchmark based on comparable cases deserves more respect than a demand plucked from ambition; both occupy memory.

The jurors in the $100 condition had the description of the pain in front of them. They also had $100.

Behavioural principles

Behavioural ideas at play in this post

Short, plain-English explanations of the principles behind this post, with links to related books and examples in the archive.