About Natural Experiments

Natural experiments are characterized by the presence of a “naturally” occurring event (e.g., new law, natural disaster, economic crisis) that intervenes in an environment (e.g., Dunning, 2008) and assigns subjects (e.g., people, companies) into a treatment group and a control group. Researchers can then compare the treatment group with the control group with regard to an outcome variable to provide evidence for a (causal) relationship between the treatment and the outcome variable (Dunning, 2012).

Natural experiments have received considerable attention in economics and political science to establish causal relationships that are otherwise difficult to identify. For instance, researchers used natural experiments to analyze the effect of institutions on economic growth (e.g., Acemoglu, Cantoni, Johnson, & Robinson, 2011) and the effect of the incumbency advantage in elections (e.g., Lopes da Fonseca, 2017). In management and organization studies, we can observe some first studies that use natural experiments to identify causal relationships. For instance, researchers analyzed the effect of corporate social responsibility on firm performance (Flammer, 2015) and the legitimation of inequality (Haack & Sieweke, 2018). Yet, despite the potential of natural experiments to provide further insights into causal relationships, they are used rather sparsely. On this website, you will therefore find information about natural experiments that should help you to apply them successfully in your research.