Portfolio allocation and borrowing constraints

Abstract

We explore the empirical relationship between borrowing constraints and household financial portfolio allocation. To motivate our analysis we develop a mean-variance model of portfolio allocation with three tradable asset classes defined by increasing risk, and establish a link between borrowing restrictions and financial portfolio allocation at the household level. Under non-restrictive assumptions the proportion of wealth allocated to the medium-risk asset is ambiguous. We also demonstrate that in the presence of both correlated background risk and borrowing constraints the domain of the non-binding risk-return space will be a function of background risk. We then analyse the US Survey of Consumer Finances with a view to empirically exploring the predictions of our theoretical framework. The distribution of medium-risk assets in US households is remarkably similar to that for high-risk assets, and suggests the presence of a more general ‘risk puzzle’, which our proxies for borrowing constraints partially explain. Our findings indicate that such constraints are inversely related to the proportion of financial wealth allocated to both high-risk and medium-risk assets, but are positively related to low-risk asset holdings. In light of our findings, further work aimed at accounting for the allocation of medium-risk assets in US households is considered expedient. it would also be unable to handle boundary observations of 0 or 1 shares; and would likely embody heteroskedasticity in u ij . We have also explored the use of a multi-nominal fractional response model, see for example, Becker In this setting, the inherent risk ordering of asset classes is not accounted for in the estimation strategy, instead the multi-nominal probit model is used as the foundation of the estimation strategy. We obtain similar results to those presented when we adopt this alternative modelling strategy.