“The anchoring effect: the first number you see, even if you know it to be arbitrary, will have a disproportionate influence on your estimate. People make estimates by starting with an initial value and adjusting it. Adjustments are typically too small.”

Daniel Kahneman Thinking, Fast and Slow (2011); originally Kahneman & Tversky, Judgment Under Uncertainty (1974)

Commentary

Anchoring demonstrates that our estimates are irrational—not in the sense of being illogical, but in being unduly influenced by irrelevant information. Show someone a random number, then ask for an estimate; the random number influences the estimate.

This has practical implications: salary negotiations (the first offer anchors), real estate pricing, negotiations generally. It's also theoretically significant: it shows that System 1 (automatic processing) dominates, and adjustments by System 2 are insufficient. Awareness of the bias doesn't eliminate it.