For more than 150 years, many economists have assumed public subsidies on higher education to have a regressive distributional effect. The German debate on this issue is kept alive by many empirical studies. Most of them confirm the thesis of a \"perverse distribution of income\" (Milton Friedman). Nevertheless, a methodological analysis of these studies unveils some problems. As the following paper tries to show, the empirical studies on the distributional effect of public higher education funding are not able to confirm the prima-facia plausibility of the thesis mentioned above. Therefore, an alternative approach to ascertaining the incidence in the long run of public subsidies on higher education seems to be necessary.