Accurately measuring gold mutual fund performance

Abstract

Since the price of gold began climbing dramatically over a decade ago, gold-related industries have received a great deal of attention from investors. Moreover, investing in gold mutual funds has become a promising alternative to investing in gold directly because of the inherent difficulties involved in purchasing the commodity. However, gold’s insensitivity to market factors requires a novel benchmark to adjust the returns of gold funds to risk. In the absence of this risk correction, gold mutual fund managers appear to perform systematically better than equity fund managers, although incorporating the new benchmark shows that this appearance is deceptive.

Keywords:

• performance measure
• gold funds
• mutual funds

JEL Classification:

• G11
• G12
• G23

Notes

¹ For example, Baur and McDermott (2010) and McCown and Zimmerman (2006) analyse gold as a safe-haven asset.

² Morningstar defines gold mutual funds as funds that pursue capital appreciation by investing primarily in the equity securities of companies engaged in the mining, distribution or processing of precious metals. However, gold mutual funds may also have a proportion of their portfolios invested directly in bullion.

³ Our study utilizes all gold mutual funds in the American market for the analysed time period with the exception of two funds that are excluded from the sample because the number of available observations for each of these funds is lower than 20.

⁴ The Weisenberger, Lipper and Strategic Insights codes that were utilized were GPM (gold and precious metals), AU (gold-oriented funds that invest primarily in shares of gold mines, gold-oriented mining finance houses, gold coins or bullion) and GLD (equity gold), respectively.

⁵ Gruber (1996) or Moreno and Rodriguez (2009), among others.

⁶ This random sample of equity mutual funds contains 100 randomly selected funds of different sizes. We classify equity mutual funds into quintiles by total net-asset value and randomly choose 20 funds from each quintile.

⁷ In addition to computing the average values of the coefficients for all mutual funds, which are presented in Table 1, we repeat the analysis using a unique equally weighted portfolio of all gold mutual funds and reach identical conclusions.

⁸ We repeat all analyses using the conditional models proposed by Ferson and Schadt (1996) and reach the identical conclusion that a gold index should be included in models to appropriately evaluate the performance of gold mutual funds.