Does portfolio emulation outperform its target funds?

Abstract

An emulation fund is designed to reduce trading activity, thereby lowering costs, for a multi-manager fund. It does this by delaying, and potentially combining, trading decisions from each employed fund manager to eliminate offsetting trades (e.g. one manager may buy a stock for her fund while another manager sells the same stock at approximately the same time for his fund). While lowering transaction costs is a key benefit of an emulation strategy, there has been little research that compares the reduction in transaction costs with the opportunity costs of delaying trade. Using reported equity trades for a large Australian pension fund we simulate the consequences of an emulation strategy. We find that simulated emulation trades underperform those trades made by the employed (or target) fund over our sample period. That is, the opportunity cost of delayed trading significantly outweighs transaction cost reductions. Overall, we do not find strong evidence to support emulation from a cost-benefit perspective before management fees and taxes. JEL classification: G23

Nonlinear hedge fund index clones?

“…An important caveat to this economic analysis is that despite the improvement in out-of-sample tests demonstrated above, the benefits of our nonlinear technology may be eroded after trading costs are included (Chen et al, 2013). This is because recreating nonlinear hedge fund exposures via nonlinear clones may be prohibitively expensive due to the more frequent rebalancing requirements vis-à-vis linear clones.…”

A model of emulation funds

“…While it is possible to simulate emulation funds ex‐post based on historical trade flow (Chen et al ., ), there are a number of limitations with this approach. The primary concern of this paper is to address the ‘black box’ characteristic of simulation‐based analysis, which relies solely on historical data.…”

“…Chen et al . () provide a case analysis of emulation funds. The authors simulate an emulation fund using trade level data and show that a hypothetical emulation fund on average underperforms its target fund.…”

“…We build on the Chen et al . () study in a number of ways. Our model enables us to develop performance expectations of an emulation fund, which can then be compared to realised performance.…”