A major institution announces unexpectedly poor results. Within hours, theories circulate: there must have been hidden accounting manoeuvres, coordinated sabotage, a secret agreement between competitors, or suppressed internal data that changed everything.
The simpler explanation - that the results reflect a genuinely difficult operating environment, compounded by a series of individually minor decisions that cumulated over time - is also available. But it doesn't spread as readily.
The simpler explanation lacks scale. A big, unexpected outcome demands a cause with commensurate size and significance. Proportionality bias is the cognitive pressure that produces that demand.
What Proportionality Bias Means
Proportionality bias is the tendency to assume that large, significant events must have large, significant causes - and correspondingly, to find explanations more plausible when their scale matches the scale of the event being explained.
The bias is not a universal commitment to wrong answers. Large events often do have large causes. The problem is the prior: before any evidence is assessed, proportionality bias tilts the evaluation toward the proportional explanation. A mundane cause for a dramatic outcome feels insufficient, incomplete, and slightly implausible - regardless of the evidence.
How It Works: The Mechanism
The mechanism draws on the representativeness heuristic: the tendency to judge the probability of an explanation by how much it resembles what a "good explanation" of a given type should look like.
For large events, a "good explanation" looks large. It feels proportional to the outcome it is explaining. When a genuinely major, visible, consequential event is explained by something small - a single act of carelessness, a minor miscalculation, bad luck at a critical moment - the explanation fails the intuitive representativeness test. It doesn't look like an adequate cause, regardless of whether it actually is one.
The cognitive discomfort created by disproportionate explanations creates a pull toward elaboration. The explanation expands to include additional actors, additional intentions, additional mechanisms - not because evidence demands it, but because the expanded explanation restores the sense of proportionality that the simple explanation violated.
This is one reason conspiracy theories about large events tend to be more elaborate rather than simpler. The elaboration is partly a response to the proportionality demand: a significant enough cause must be involved, so the explanation must be enriched until it feels adequate.
Why This Matters
In understanding complex events, proportionality bias makes simple, compound, or probabilistic explanations feel unsatisfying compared to narratives involving intentional large-scale causes. Historical events attributed to individual human error, systemic failure, or cascading small causes often acquire over time a version of events that restores proportionality - a more powerful, more deliberate cause is added to the explanation.
In everyday judgement, proportionality bias makes unlikely, complex explanations for personal setbacks more credible than they warrant. A job application that failed, a relationship that ended, a project that went wrong - each of these can generate elaborate causal narratives if the proportionality demand is applied. The failed application becomes evidence of systematic discrimination; the ended relationship becomes a coordinated pattern of manipulation; the failed project becomes sabotage.
In media and public discourse, proportionality bias drives the appetite for narratives that locate elaborate, intentional causes behind significant events. Explanations involving incompetence, bad luck, or compound small failures are genuinely less compelling, even when they are more accurate. Proportionality bias is part of what makes misinformation narratives attractive - they deliver on the demand for scale.
The Common Misunderstanding
Proportionality bias does not mean that large events never have large causes. They sometimes do. The bias is not an error about the world; it is an error about the assessment process. It involves applying a prior - cause-scale must match effect-scale - before the evidence has been assessed, and allowing that prior to inflate the perceived plausibility of elaborate explanations over parsimonious ones.
The corrective is not to assume simple causes. It is to assess causes on their evidence, not on their scale.