Allocative and redistributive rules in the public sector are often less contingent on available information than normative theory would suggest. In this paper we offer a political economy explanation. Under different rules, even if the observable outcomes of policies remain the same, the informational content which can be extracted by these observations is different. Simpler rules are more transparent because they allow citizens to gain more information on politicians. Since there are limits to what voters can observe, this may be a relevant insight into the functioning of the political system. We present several institutional examples supporting our argument and discuss a simple model which supports our theory. By using the same model, we also carefully discuss the links and the differences with several other strands of literature, such as the macroeconomic trade-off between rules and discretionality and the principal-supervisor-agent model in the regulatory context. Finally, we offer suggestions for further developments and possible empirical tests of the theory dicussed here.