Coherent Arbitrariness
The phenomenon where arbitrary initial anchors create internally consistent but externally unfounded preferences and valuations.
Also known as: Arbitrary coherence
Category: Cognitive Biases
Tags: psychology, cognitive-biases, decision-making, behavioral-economics, pricing
Explanation
Coherent arbitrariness, a concept introduced by Dan Ariely, George Loewenstein, and Drazen Prelec, describes how people's preferences and willingness to pay can be shaped by arbitrary initial anchors, yet remain internally consistent once established. The initial anchor may be completely random, but subsequent decisions maintain a logical relationship to it.
In their landmark experiment, participants were asked to write down the last two digits of their Social Security number, then bid on various products (wine, chocolate, electronics). Those with higher Social Security numbers consistently bid more — the arbitrary number anchored their valuations. Crucially, while the absolute amounts were arbitrary, the relative ordering was coherent: participants still bid more for premium wine than cheap wine.
This reveals something profound about how preferences form: our sense of what something is 'worth' is far more malleable than we assume. The first price we encounter for a category of product becomes our reference point, and all subsequent valuations are adjustments from it.
Implications for everyday decisions: the first salary you accept anchors your career earnings trajectory, the first rent you pay anchors your housing budget expectations, and the first price you see for a product category sets your value framework for that category.
For knowledge workers and decision-makers, coherent arbitrariness is a warning: just because your preferences feel rational and internally consistent doesn't mean they're based on objective value. Being aware of this effect helps question initial anchors and evaluate options on their actual merits rather than relative to arbitrary starting points.
Related Concepts
← Back to all concepts