Mutual Fund Performance: Evidence from South Africa Ömer Faruk Tan Research Assistant, MEF University Faculty of Economics, Administrative and Social Sciences, Department of Economics, Turkey | e-mail: omerfaruk.tan@mef.edu.tr / omerfaruktan34@gmail.com Volume 5 No 2 (2015) | ISSN 2158-8708 (online) | DOI 10.5195/emaj.2015.83 | http://emaj.pitt.edu Abstract This paper aims to evaluate the performance of South African equity funds between January 2009 and November 2014. This study period overlaps with the period of quantitative easing during which developing economies in financial markets have been influenced severely. Thanks to the increase in the money supply directed towards the capital markets, a relief was experienced in related markets following the crisis period. During this 5-year 10-month period, in which the relevant quantitative easing continued, Johannesburg Stock Exchange (JSE) yielded approximately 16% compounded on average, per year. In this study, South African equity funds are examined in order to compare these funds’ performance within this period. Within this scope, 10 South African equity funds are selected. In order to measure these funds’ performances, the Sharpe ratio (1966), Treynor ratio (1965), Jensen’s alpha (1968) methods are used. Jensen’s alpha is also used in identifying selectivity skills of fund managers. Furthermore, the Treynor & Mazuy (1966) and Henriksson & Merton (1981) regression analysis methods are applied to ascertain the market timing ability of fund managers. Furthermore, Treynor&Mazuy (1966) regression analysis method is applied for market timing ability of fund managers. Keywords: Mutual Fund, South Africa, Performance Evaluation, Equity Funds New articles in this journal are licensed under a Creative Commons Attribution 3.0 United States License. This journal is published by the University Library System of the University of Pittsburgh as part of its D-Scribe Digital Publishing Program, and is cosponsored by the University of Pittsburgh Press. mailto:omerfaruk.tan@mef.edu.tr mailto:omerfaruktan34@gmail.com http://www.library.pitt.edu/ http://www.pitt.edu/ http://www.library.pitt.edu/articles/digpubtype/index.html http://www.upress.pitt.edu/upressIndex.aspx http://creativecommons.org/licenses/by/3.0/us/ Volume 5 No 2 (2015) | ISSN 2158-8708 (online) | DOI 10.5195/emaj.2015.83 | http://emaj.pitt.edu Omer Faruk Tan Emerging Markets Journal | P a g e |49 Mutual Fund Performance: Evidence from South Africa Ömer Faruk Tan 1. Introduction Mutual fund performance has always been one of the most researched areas of finance studies. Using diverse technical measurement methods, these types of studies analyze fund performances of various markets from different perspectives. Especially, following the period of liberalization of the financial markets, mutual funds have gained much more significance in the eyes of investors, resulting in numerous studies that have been carried out on performance evaluations. Mutual funds bring investors who share a common goal together. According to Deepak (2011), investors invest the money they collect into capital market instruments such as shares, debentures and other investment securities. The total income acquired from investments and the capital appreciation is equally shared among unit holders by taking into account the units owned by them. As a consequence, mutual funds are a suitable investment for the common man, as they provide the opportunity to invest various professionally managed securities at a relatively low cost. The main objective of investing in a mutual fund scheme is to diversify risk. The mutual funds invest in diversified portfolio and the fund managers take different levels of risk so as to achieve the scheme’s objectives. Hence, while evaluating and comparing the schemes, the returns should be measured by taking into account the risks involved in achieving the returns. (Rao, 2006). The global crises emerged in America in 2008 and later spread to other countries, affecting especially the economies of Europe and America and their financial markets a great deal. The American and European economies went into recession and some significant financial investment banks collapsed, such as Lehman Brothers. Also, in Europe, banking crises occurred in various countries led by Portugal, Ireland, Spain, Greece, and Italy. This situation, in the eyes of investors, made America and Europe lose their reputation of being the “safe port” and making investors turn towards other stock markets for investment purposes. To ease the recession, the FED applied a policy of quantitative easing. Between December 2008 and October 2014, the FED bought huge quantities of government bonds and bills from the markets to enhance the money supply for the sake of encouraging the revival of the economy. Quantitative easing policy started in December 2008 and finished in October 2014. Quantitative easing policy separates four terms QE1 (December 2008- June 2010), QE2 (November 2010- June 2011), QE3 (September 2012- October 2014) and finally QE4 (January 2013- October 2014). (Useconomy). During the period, huge amount of money influx from developed countries to developing countries experienced. Hence, in this paper, it is tried to analyze fund performances of South African equity funds between 09 January 2009 - 31 0ctober 2014 in the era of quantitative easing. South Africa is considered as one of the emerging markets and over the study period of 5 years - 10 months, Johannesburg Stock Exchange (JSE) grew by 15.9% compounded annually on average. Johannesburg Stock Exchange performed better than major developed European markets. In the sample period, developed market indices DAX, FTSE 100, CAC 40 yielded 12.1%, 6.8% and 4.1%, respectively. 2. Literature Review Beginning from the 1960s, there have been several studies carried out on mutual fund performance. Treynor (1965), Sharpe (1966) and Jensen (1968) are among those who measure fund performance related to risk and return measurements. Sharpe (1966) measured 34 open-ended mutual funds between 1954-1963 using the Sharpe ratio and Treynor ratio. As the result of the study, it has been found out that while 11 funds out of 34 show a better performance than the index, 23 funds underperform their benchmarks. Jensen (1968) examined 115 mutual funds - which were active between 1945-1964 – by using an alpha indicator that he generated. His alpha indicator shows the selectivity skills of fund managers. Based on his results, funds could not outperform the market performance, revealing that mutual fund managers, in general, did not have selective ability. Malkiel (1995) used the Jensen method to calculate the performance of American funds between the years 1972 and 1990. He revealed that mutual funds could not show positive excess return. Detzler (1999) searched 19 global bond funds by using monthly returns between the years 1985 and 1995. In the study, a multiple regression analysis was used and it was found out that funds could not show better performance than indexes. Dahlquist, Engström and Söderlind (2000) evaluated 201 Swedish mutual funds – including only domestic funds - from the period between 1993 and 1997. They found that regular equity funds seemed to over perform while bond and money market funds performed less. Furthermore, actively managed funds demonstrated better performance than passively managed funds. Mutual Fund Performance: Evidence from South Africa Page |50| Emerging Markets Journal Volume 5 No 2 (2015) | ISSN 2158-8708 (online) | DOI 10.5195/emaj.2015.83 | http://emaj.pitt.edu With the aim of detecting the market timing ability of the fund managers, Treynor and Mazuy (1966) established the quadratic regression analysis method. They applied this method to 57 open-end mutual funds (25 growth funds and 32 balanced funds). They revealed only a single fund as having statistically significant market timing ability. Henriksson and Merton (1981) and Henriksson (1984) developed both parametric and nonparametric statistical models to the test market timing ability of portfolios. Having been introduced by Henriksson and Merton (1981), the parametric and non-parametric tests in question were applied by Henriksson (1984) to evaluate the market timing ability of 116 open-end funds between 1968 and 1980 in the U.S. market. The results revealed that there wasn’t any support for market timing ability. Moreover, Henriksson found an inverse relationship between selection ability and market timing ability. Chang and Lewellen (1984) tested the market timing ability of 67 U.S. funds covering the period from 1971 to 1979 by using the Henriksson & Merton (1981) method. It was found that there were weak indications of fund manager market timing ability. Gallo and Swanson (1996) tested 37 U.S. mutual funds by using the Treynor & Mazuy model for market timing, yet found no evidence of market timing of funds. Christensen (2005) evaluated 47 Danish funds between January 1996 and June 2003. He found that fund managers did not have selectivity skills in general and, in terms of timing ability, the results were also negative, due to the fact that only two funds had significant timing ability. Gilbertson and Vermaak (1982) evaluated seven South African mutual funds over the period 1974 to 1981. According to results, in general, returns of funds were lower than market indexes. Only one fund - Guardbank – showed significantly outperformed than indexes. Manjezi (2008) investigated 15 South African funds during the period between 2001 and 2006. According to his results, the index showed a better performance than funds. In addition, only one fund displayed both selective and market timing ability during the study period. Mbiola (2013) examines 64 South African domestic general equity unit trusts over the period from 1992 to December 2011. According to his result, funds could not show strong evidence of superior performance than market. 3. Methodology and Data 3.1. Methodology In this study, it is tried to evaluate both funds and funds managers’ performance of South African equity funds. A total of 10 equity funds performances’ are analyzed. In order to evaluate fund performance, Sharpe (1966), Treynor (1965) and Jensen’s alpha (1968) ratios are computed. Jensen’s alpha method also shows the selectivity skills of fund managers. In order to test mutual fund managers’ market timing ability, the Treynor & Mazuy (1966) and Henriksson & Merton (1981) methods are applied. 3.1.1. Treynor Ratio According to Kouris, Adam, & Botsaris (2011) the Treynor ratio is the first risk-adjusted performance measure of mutual funds that was put forward by Treynor in 1965. It is calculated as the ratio of the excess return of the mutual fund divided by its beta (systematic risk) and is defined as: Ti = (Rp-Rf) / P (1) where Ti = Treynor’s performance index Rp = portfolio’s actual return during a specified time period Rf = risk-free rate of return during the same period P = beta of the portfolio According to Reilly (1992), whenever Rp > Rf and p > 0, a larger T value means a better portfolio for all investors regardless of their individual risk preferences. In two cases, a negative T value may result: when Rp < Rf or when p < 0. If T is negative because Rp < Rf, then we deduce that the portfolio performance is very poor, whereas if the negativity of T comes from a negative beta, the fund’s performance is excellent. 3.1.2. Sharpe Ratio According to Noulas &Lazaridis (2005), the Sharpe technique was developed in 1966 and is fairly similar to the Treynor technique, but the Sharpe technique uses the total risk of the portfolio rather than systematic risk. This technique computes the risk premium earned per unit of the total risk. The Sharpe value can be calculated as follows: Sp =(Rp – Rf /) p (2) where Sp = Sharpe Ratio Rp = the average rate of return for a fund Rf = the average risk-free return p = the standard deviation of the fund. The Sharpe ratio (Sp) evaluates the performance of its level of total risk. A higher value of this ratio indicates Volume 5 No 2 (2015) | ISSN 2158-8708 (online) | DOI 10.5195/emaj.2015.83 | http://emaj.pitt.edu Omer Faruk Tan Emerging Markets Journal | P a g e |51 that the fund delivers a higher performance by using standard deviation ( p). (Duggimpudi, Abdou, & Zaki, 2010, p. 79). 3.1.3. Jensen’s Alpha As Jensen (1968) explained, “a portfolio manager’s predictive ability – that is, his ability to earn returns through the successful forecast of security prices that are higher than those which we could presume given the level of his riskiness of his portfolio” (p. 389). Jensen’s model can be written as: Rpt – Rft = p + p (Rmt – Rft) + ept (3) p = the excess return on the portfolio after adjusting for the market Rpt = the return on the portfolio p at time t Rft = the return on a riskless asset at time t Rmt = the return on the market portfolio at time t p = the sensitivity of the excess return on the portfolio t with the excess return on the market. The sign of the alpha displays whether the portfolio manager are superior to the market after adjusting for risk. A positive alpha denotes better performance relative to the market, and a negative alpha designates poorer performance. (Mayo, 2011). 3.1.4. Treynor&Mazuy Regression Analysis Investment managers may well beat the market, if they are able to adjust the composition of their portfolios in time when the general stock market is going up or down. That is, if fund managers believe the market is going to drop, they alter the composition of the portfolios they manage from more to less volatile securities. If they think the market is going to climb, they shift in the opposite direction. (Treynor&Mazuy, 1966). Mutual fund managers may hold a higher proportion of the market portfolio if they are qualified to predict future market conditions and envisage the stock market as a bull market. On the other hand, mutual fund managers may hold a lower proportion of the market portfolio if they expect the market to underperform in the future. Treynor and Mazuy (1966) developed the following model to evaluate market-timing performance: (4) where i is the timing-adjusted alpha, which represents the timing-adjusted selective ability of mutual fund managers. The quadratic term in equation (4) is the market timing factor and the coefficient of the market timing factor, , represents mutual fund managers’ market timing ability. If is positive, mutual fund managers have superior market timing ability i.e., the investment portfolios of mutual funds are adjusted actively to well-anticipated changes in market conditions. A negative implies that mutual fund managers do not exhibit market timing ability. (Chen et al., 2013). 3.1.5. Henriksson&Merton Regression Analysis Another return-based approach for estimating performance is the option approach developed by Merton and Henriksson. The regression used is similar to the Treynor & Mazuy regression. In contrast to the linear beta adjustment of the Treynor and Mazuy framework, the portfolio beta in the Henriksson and Merton study is assumed to switch between two betas. A large value if the market is expected to do well, i.e., when Rm>Rf up market and a small value otherwise i.e., when RmRft (up market), D is equal to 1 and when Rmt Rft Rit-Rft = i + i (Rmt – Rft) + i1 + Rmt