Introducing Objective Benchmark-Based Attribution in Private Equity

Abstract

Private equity asset owners (“LP”s) seeking to reduce downside risk and increase upside probability would logically benefit from indexing prospective asset managers (“GP”s) by their skill. However, theoretical deficiencies and a lack of rigorous market calibration prevent the metrics and techniques commonly used in private equity from isolating GP skill. In this paper we introduce a new conceptual framework for a repeatable decomposition of private equity returns that disambiguates the quantification of GP skill. Modern proxy benchmarks are a key component of the framework for their definition of systemic returns specific to the target asset. They satisfy Andrew Lo’s (2016) fundamental properties of an index (systematic, transparent, and investable), and the CFA Institute’s “SAMURAI” criteria for a valid benchmark. However, we propose that the integrity of the decomposition requires the benchmark’s similarity (to target) and its stability be systematically derived, measured quantities. We discuss these two new properties in conjunction with the technology that enables the construction of modern proxy benchmarks and their active management over time. With systemic returns thus defined, excess returns against the modern proxy benchmark are attributed to dynamic elements under the control of the GP, which we define as manager alpha. Systemic returns in excess of a broad/policy benchmark are deemed static elements. Static elements measure the portion of returns attributable to size and sector selection, which a GP tends to ‘specialize’ in and which are known to the LP prior to investment. While both static and dynamic elements contribute active returns to the investment, it is the dynamic elements – alpha – that should merit attention (and high fees) from LPs.