Correlation-Causation Fallacy
Also known as: Cum Hoc Ergo Propter Hoc, Post Hoc Ergo Propter Hoc, Correlation Is Not Causation
Formulated by David Hume (1739)
Definition
The error of concluding that because two variables move together, one must cause the other. A statistical correlation between two series (for example, income concentration and GDP growth) says nothing on its own about which one, if either, drives the other: both could be caused by a third factor, the causal arrow could run the opposite way, or the relationship could be coincidental. Establishing causation requires identifying an actual mechanism, not just a coefficient. The philosophical root of the problem goes back to David Hume, who argued that we never directly observe causation, only a constant conjunction of events, and that inferring a necessary causal link from repeated co-occurrence is a habit of the mind rather than a logical proof. In economic debate this fallacy is common: a regression showing that a policy or a group's income share moves alongside some macroeconomic outcome is routinely presented as proof that the policy or the group produced that outcome, skipping the harder question of what mechanism, prices, capital allocation, entrepreneurial decisions, would actually transmit the effect.