We investigate if a more populous or fragmented executive or legislative increases public spending, an effect known as “the Law of 1/n”. We test this for the supra-local governments of Belgium. On the basis of our dataset – including rather few “cross-sections” as well as rather few...