The significance of the market portfolio: theory and evidence

Abstract

The market portfolio (world portfolio) is in one sense a least important portfolio to provide to investors; there is always a better portfolio for social planners to make available to them. In a J-agent one-period stochastic endowment economy, where preferences are quadratic, the market portfolio is never spanned by the optimal markets a social planner would create. With identical preferences, the market portfolio is orthogonal to all J — 1 portfolios which achieve a first best solution. These conclusions rely on the assumption that the social planner has perfect information about agents’ utilities. We also show that as the contract designer’s information about agents’ utilities becomes more imperfect, the optimal contracts approach contracts that weight individual endowments in proportion to elements of eigenvectors of the variance matrix of endowments. If there is a substantial market component to endowments then a social planner, for reasons of robustness and simplicity, may conclude that creating a contract to allow trading the market portfolio would be a significant innovation. To gauge the empirical relevance of the world portfolio, we estimate the optimal contracts, using Maddison [1995] data on per capita income, for Canada, France, Germany, Italy, UK and the USA. We find that when the contract designer has no information about utilities, the first two contracts nearly span the world portfolio.

We thank Peter Bosaerts, Subir Bose, John Geanakoplos, Chiaki Hara, Jesus Santos, Paul Willen, Eric van Wincoop, participants at the University of Chicago Finance Workshop, the Economic Theory Conference in Turkey (1997), Summer Economnetrics Society Meetings in Pasadena (1997) and participants at the NBER, conference on Asset Pricing (1997) for helpful comments. The authors are responsible for any remaining errors.