The Efficiency of Decentralized Investment Management Systems

1981 
The primary purpose of this paper is to demonstrate that decentralized investment management systems may not always be efficient. Specifically, within the context of a particular portfolio choice paradigm it is shown that a given decentralized investment management system is (weakly) efficient if and only if the joint probability distribution of asset rates of return satisfy certain covariance restrictions. If these restrictions do not obtain then the asset portfolios generated by this decentralized structure will generally be inferior to those which would be generated by a completely centralized structure. This paper also discusses how the managers of departments within an efficient decentralized structure should behave so as to generate portfolios which are optimal from the point of view of the institution as a whole. Generally, departmental managers should behave as if they have less risk aversion than the institution as a whole. In fact, a given manager should be more risk averse the greater the value of his portfolio. Finally, we note that the efficiency concept employed in this paper is equivalent to the proposition that certain assets admit consistent simple sum aggregation. It is shown that this implies that the efficient decentralization of investment decisions permits the institution to economize on the information which must be passed to higher level departments.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    2
    References
    1
    Citations
    NaN
    KQI
    []