Evidence of Style Gaming and the Moral Hazard Created in US Real Estate Mutual Funds

The moral hazard problem has been examined in the context of REITs (Edelstein, et al., 2010), securitized real estate (Fan, et al., 2006) and mortgage back securities (Keys, et al. 2009 and 2010; Kau, et al. 2012). Most past studies investigate moral hazard through managerial incentive, with little explicit empirical evidence on how moral hazard is created by information asymmetry between agent and principal. This study presents empirical evidence of a moral hazard existing amongst US Real Estate sector mutual funds. Using basis assets (Brown and Goetzmann, 1997; Ahn, et al., 2009) to determine undisclosed style, an average of 33% of RE funds are found to be misclassified; i.e. they change style without disclosing the change to investors. We take this, prima facie, as evidence of style gaming, which results in moral hazard for RE mutual funds. RE style ‘gamers' earn on average, a 3.06% p.a. premium above non-misclassified RE funds (non-gamers). This validates that misclassification by RE funds may not be innocent but rather driven by a strong economic incentive for fund managers to ‘game' their style. The success at style gaming by some RE fund managers is an adverse incentive for others to do the same as it generates improved ranking relative to the notional RE peer group, greater fund inflow and ultimately higher fee income. Such behavior causes a moral hazard where the misrepresentation of style dislocates the risk-return expectations of an investor from the actual investment behavior of the fund. We track the style shifts of RE gamers to determine how they earn such high returns. We find that the moral hazard of style gaming is real, widespread, volatile and severe when it occurs. Holdings of RE fund gamers and non-gamers are compared to examine whether capacity constraint is an additional factor motivating style gaming. The study is the first to recommend that Real Estate moral hazard could be addressed by explicitly reducing the information asymmetry between agent and principal through the use of basis assets to track style independently of fund manager disclosure