When the State Mirrors the Family : The Design of Pension Systems

Family is a primal institution. Family organization (e.g. inheritance rules) can be transmitted over time, or even embedded into newly born economic institutions. We argue that when the family is substituted in one of its economic role by another, perhaps new, institution, the economic organization that was prevailing within the family is likely to be adopted by the new-born institution. To study this transmission mechanism from family culture (or organization) to economic institutions, we concentrate on the impact of the family structure on the design of the most widely spread welfare state program in the world: the public pension system. We first build a simple OLG model which predicts that, when pensions systems are introduced, in society with weak family ties they act as a safety net, while in societies with strong ties pensions replicate the tight link between generations and tend to provide benefits related to earnings. We turn to a historical perspective and we find that medieval family structures (based on Emmanuel Todd's classification) have influenced the design of pension systems since their introduction, shaping the fundamental characteristics that are still entailed in the current systems, and that differentiate them across countries. In particular, in societies dominated by absolute nuclear families, i.e. weak family ties (f.i. Anglo-Saxon countries), we observe the emergence of a pension system which acts as a flat safety net entailing the largest within-cohort redistribution than societies dominated by any other type of family. This link between the type of families and the design of pension systems is robust to including several other variables, which may constitute alternative explanations of the introduction of the pension systems, such as legal origin, religion, urbanization and democratization of the country at the time of the introduction, and the current GDP, share of elderly in the population, electoral rules and forms of government. Interestingly, historical family types matter for explaining the design of the pension systems, which represents a persistent feature, but not their size, which have largely changed over time