Despite a heated debate on the perceived increasing complexity of financial regulation, a comprehensive framework to study regulatory complexity is lacking. We propose one inspired by the analysis of algorithmic complexity in computer science. We use this framework to distinguish different dimensions of complexity, classify existing complexity measures, develop new ones, compute them on two examples—Basel I and the Dodd-Frank Act—and validate them using novel experiments that involve the computation of risk-weighted assets under various rules. Our framework offers a quantitative approach to the policy trade-off between the precision and the complexity of regulation. The toolkit we develop is freely available and allows researchers to measure the complexity of any normative text as well as test alternative measures of complexity.