Intergenerational Altruismt Rcardian Equivalence and the Relevance of Distributional Policy

In recent years there has been a growing controversy concerning the effective role of distributional policy in economies in which agents hold altruistic preferences. We show that in order to establish cross sectional Distributional Neutrality households require information on consanguinity links prior to making their decisions on consumption and bequests. If such information is not available due to the veil of time, then the Distributional Neutrality property disappears. However Ricardian Equivalence under stated conditions, remains and is therefore a property independent of DN. Summary: In recent years there has been a growing controversy concerning the effective role of distributional policy in economies in which agents hold altruistic preferences. In an overlapping generations model with altruistic behaviour, Bernheim and Bagwell(1987) have purported to show the irrelevance not only of fiscal and monetary policies which switch resources between generations (Ricardian Equivalence (RE)); but also of policies which switch resources between members of the same generation (cross sectional Distributional Neutrality (DN)); and argue that DN is a necessary consequence of RE. In the present paper we show that in order to establish DN households require information on consanguinity links prior to making their decisions on consumption and bequests. If such information is not available due to the veil of time, then the DN property disappears. However Ricardian Equivalence, under stated conditions, remains and is therefore a property independent of DN. We show that for RE, households must know only that they will be related to a given number of households in the future, whereas for DN they also require precise details of the characteristics of these households. Thus the BernheimBagwell arguments concerning the irrelevance of all redistributional policies arising out of intra-family linkages appear somewhat misplaced, and the theoretical case for Ricardian Equivalence is rather more robust than previously supposed. However, we show that for growth models in which intra-family linkages are explicitly introduced rather different results are generated than are found in the more usual formulation of the overlapping generations model, e.g. A Nash equilibrium with undercapitalization in steady growth. In this respect it becomes more difficult to accept results which flow from models which postulate families with asexual reproduction; i.e. the assumption found most often in the overlapping generations literature. An earlier version of this paper was presented at the Econometric Society European Meeting, University of Cambridge, 26 September 1991. WEB VERSION: July 30th 1996