Mutual Fund Performance Attribution: 1994-2005
Gary E. Porter,L. Brooks
2012-10-01
Abstract:AbstractWe examine the performance of nearly 2000 actively managed equity funds using attribution analysis, a process commonly used by mutual funds to isolate management contributions from the sector allocation and stock selection components of fund returns. For the period 1994 through 2005 we show that, on average, potential gains from sector allocation were offset by poor stock selection. However, equity managers excelled at sector allocation and stock selection during the bear market of 2000-2002. Over the 12-year sample period, returns from actively managed funds dominated returns from passive portfolios with less risk, primarily because of their performance during the bear market.© 2012 Academy of Financial Services. All rights reserved.JEL classification: G11; G2Keywords: Managed mutual funds; Attribution performance1. IntroductionManagers of actively managed funds require sound and reliable performance measurement to evaluate and guide their investment decisions. Performance evaluations reveal how managers are performing relative to their competition, how they should be rewarded, and whether changes in strategy are required. Sound and reliable evaluation also provides individual investors with information necessary to guide their financial planning.Performance typically uses variations on the Sharpe (1966) index and alphas derived from either the single-index market model or multifactor models. However, performance attribution or attribution analysis, which is widely used in the fund industry for compensation purposes, differs from widely studied performance models in that it uses a fund-specific benchmark, or bogey, to evaluate fund performance. Based on the fund's bogey, the equity fund can precisely determine how their managers' sector allocation and security selection decisions each contribute to the fund's total return.1 While fund managers are commonly judged on their performance relative to an index such as the SP while factor derived alphas are not. Why? Because attribution analysis differs from factor-based models by providing estimates of both sector allocation and stock selection components of a fund's return regardless of the fund's objective or investment strategy. Accurately tracking both components enables active fund managers with a variety of investment objectives to effectively monitor them, and to make appropriate changes as information arrives. Importantly, if data services provided performance data based on attribution analysis, investors could more effectively monitor management's potential by observing their ability to generate wealth from each attribute.Recent research challenges the assumption that betas derived from historical data represent investor risk. Fama and French (2007) report that their size (SMB) and value (HML) factors contain excess positive returns resulting from unanticipated higher growth in U.S. business productivity over the last century. This growth represents exploitable opportunities in sector allocation and stock selection for fund managers who are able to anticipate forthcoming productivity gains. Additionally, Moskowitz and Grinblatt (1999) conclude that astute investors can capture the Carhart (1997) momentum factor premiums (PYR) of Jegadeesh and Titman (1993). The same explanation applies for the single-factor market model, which incorporates the net impact of the three remaining Carhart factors, plus any unobserved priced factors representing productivity gains or losses because of changes in expected cash flows and/or risk. …
Economics,Business