Back to Basis with Investment Consultants

The thesis includes two papers that investigate return predictability across asset classes and agency issues associated with investment consultants.

The first paper provides an explanation to the pervasive pattern of return predictability across asset classes discussed in Cochrane (2011). The paper shows analytically that the basis between spot and futures contracts contains information about future returns of securities across the asset classes of commodities, equity indices, fixed income and foreign exchange. The bases in commodities are positively correlated with a leading indicator of the business cycle whereas the bases in the financial assets are negatively related to the short-term rate. The return predictability of the basis can be captured with a simple multi-asset long-short strategy which produces an out-of-sample sharpe ratio of 0.5 and an alpha of 2.5%-4.5% per annum with respect to commonly used asset pricing models. Specifically, the analysis includes five Fama-French Factors, a bond index and futures risk premia of multi-asset momentum, value, time-series momentum and four asset-specific carry factors. The strategy performance is
counter-cyclical and robust to transaction costs. 
The second paper reconciles a previously unexplained discrepancy between the pervasive tendency of institutional investors to rely on investment consultants and an apparent inability of the consultants to improve investment decisions and strong preference for large funds. The paper introduces a model of decentralized investment management, in which the CIO acts as the principal in the best interest of pension fund beneficiaries and hires an investment consultant who serves as an agent responsible for portfolio management of an asset class relative to an external benchmark. I show that within a standard information asymmetry setting misalignment of incentives can lead to over-conservative investments and negative agency costs that are passed to investment consultants through lower wages in equilibrium. The agency issue is magnified if the CIO is ambiguity-averse. The predictions of the model are consistent with empirical evidence for the lack of value in consultants’ recommendations and preference for large funds. The problem is exacerbated further with additional agency costs incurred by the fund beneficiaries if the CIO derives a private benefit from reducing her career risk by delegating responsibility to investment consultant.