The idea of a “natural” experiment is misleading. The so-called natural experiment typically depends on some prior political intervention—in other words, a project or an experiment of some sort, which arranges the sociotechnical world in a way that offers further opportunities for experimentation. This intervention must be beyond the direct control of the investigator. But that does not insulate the second experiment from the effects of the first.

Philip Mirowski & Dieter Plehwe (eds.), The Road from Mont Pèlerin, loc. 1977