EN
This paper analyses the interaction between reporting earnings and pension benefits, using a very simple, elementary model. Workers can be classified in three groups: 1. well-paid, reporting in full, 2. well-paid, reporting to a minimal extent (free-riders) and 3. poorly paid, reporting in full. In the basic version, it is assumed that free-riders save a significant proportion of their hidden earnings for their old age. Three pension systems are compared: 1. contributory, 2. contributory combined with a universal basic pension, and 3. contributory with means testing. The major result of the paper is that if free-riders are distinguishable and can be excluded, a means-tested system is welfare-superior to the basic system. The robustness of the observations can be checked by changing the assumptions of the basic model methodically. In a future model, the report and the saving are to be derived from individual optimization, and the system parameters will be set by maximizing the social-welfare function.