Paper Review: Using downside risk in evaluating the performance of Malaysian mutual funds

This article gives a critical review of the Reza Baghdadabad's journal article titled "Using downside risk in evaluating the performance of Malaysian mutual funds".

Review written by: Iliyas Ismail

The article discusses on the various measurements for mutual funds in Malaysia and attempts to provide an alternative measuring tool for the more common practice of using systematic risk and conventional standard deviation. The alternative posed by the author is using the downside risk and semi-standard measurement instead, and this is due to his argument that it is a better calculation tool to meet the asymmetric market condition of mutual fund, rather than the common model, which is considered to be better applied for portfolio returns that are symmetry, which is not the majority of mutual fund returns.

The article is aimed at investors and practitioners in Malaysia to assist them in making decisions for their stock and mutual fund choices, and the author hopes that the intended readership would be able to utilize the findings in the decisions.

It found that the proposed method is more suited to calculate the performance of mutual funds in Malaysia, more specifically in evaluating risk. It presents a new framework of using downside beta, defined as the uncertainty of investors in the event of a loss, rather than standard beta which calculates the systematic risk of a security or portfolio in contrast to the whole market. The author challenges the notion that conventional means of risk calculation namely using systematic risk and variance should be the common norm. Apart from that, the usage of systematic risk and variance is characterized as not able to “make a distinction between the good and bad returns, rather than the average return meaning that those returns are equally considered undesirable”. This is considered a good reason why the paper seeks to propose a new method for calculation that differs from the conventional practice.

To be clear, this is not a new attempt nor a novel opinion. The author’s critique of standard deviation and risk analysis seems incessant, but it is supported by a wide body of literature. Nawrocki (2000) put forth a brief history of downside risk and stated that even Markowitz thought that downside risk is a better tool for asymmetric returns but he stayed with variance instead of semi-variance optimization model due to the latter being computationally simpler. Sortino and Satchell (2001) also presented arguments in favour of semi-variance for downside risk instead of conventional ones. Indeed, numerous authors have suggested that the downside risk to be taken more seriously since the 90s (Balzer, 1994, Merican, 1994, Sortino and Forsey 1996).  

Chong et. Al (2013) also propose that downside beta be used to address the asymmetric return from investments. Abbas et al (2013) recommends downside risk measure compared to Markowitz’s mean-variance, especially when the returns on asset is skewed, Parello (2007) applied downside risk to hedge funds and stated that the ability to distinguish between good and bad return (returns greater or lower than the investor’s expected return) is an advantage point compared to CAPM, for instance. What the author seeks to do is using this argument and implement them on Malaysian mutual fund market.

The author begins with a short review of the background development of eight measures of investment result from different funds and gave his reasonings for choosing them. Among the reasonings include direct comparison of the risk adjusted returns of each fund notwithstanding their correlations with a benchmark, or the consideration of total market risk.

The chosen eight measures were applied to the chosen mutual funds for this study, and the exercise is to seek to answer four questions, namely, whether they are able to evaluate the performance of the mutual funds, in what way can the funds be ranked separately using these measures, how one measure (the modified leverage factor) can raise the returns of a mutual fund with low risk and how mutual funds can be sorted according to ranking by comparing them with a benchmark.

As mentioned, the author derives his formula by using, as stated, the eight existing formulas for measuring risk in investment portfolio or return stock, and he outlines them as in their standard form and also later in the paper provides a modified form of the equation. The equations are those of Sharpe, Treynor, Jensen’s alpha, M2, Information ratio, MSR, SPI and leverage factor.

The main point of the paper is to analyse if the semi variance and downside beta components are better measures of risk in the investment markets by using these formulas, as a further contribution for downside risk. The question of whether using downside risk is better, is handled quite well by the author. After listing the various equations, he offered a modified version of each of the equations that has not yet been given any modifications and attempted to implement it. These equations according to him are M2, IR, MSR, SPI and leverage factor in downside risk factor. Some of these factors are already modified versions, namely Modigliani and Modigliani’s M2 is a derived from Sharpe’s measurement.

The analysis of the author did not include much discussion on the reason as to why variance and normal beta is still continuously being used to assess investment returns, and this aspect is something that should have been touched upon. If there are many literatures supporting the authors viewpoint, why is it that the author still feels the need to present this argument against the common practice as though it is something new? This is something that is left unanswered in the discussion. Indeed, certain scholars have argued that there is not much difference between variance and downside risk (Grootvelda and Hallerbachb, 1999).

The author’s brief description of the fact that the five measurements chosen to be modified in this study, may leave the intended reader wondering as to why these modified measurements have been largely ignored up till then. The author could have provided reasons as to why it is important to provide these modified measures specifically and whether it is crucial to meeting its objectives. Furthermore, to provide the measurements as it is, inadvertently misguides the average reader to assume that lesser than these eight measurements would result in a heavily disadvantaged position, whereas it might very well be not the case, considering measurements such as SPI is not widely used, based on a search on the literature.

Despite my inability to grasp the entire discussion on the modified equations, I was greatly interested in the author’s intended innovation this approach, namely the five measures.

Nonetheless, as it stands, the central focus of the title is well supported by its empirical findings, the study result using the modified measures are compared with the traditional measures, by finding its correlation coefficient, and the modified measures were also analysed for its economic significance. Using the measurements, the 91 mutual funds chosen for this study is ranked based on its returns, which meets one of the main objectives of the study. The findings of the study show that some of the modified measures have high correlation with the traditional measures and can be used as a substitute and also offers a better performance based upon the lower abnormal returns in the findings.

One of the findings of the study with is rather unclear and not given enough justice, namely, the conclusion that conventional measure does not have a crucial influence on the evaluation of funds. He discusses this finding under his analysis on the correlation coefficient between the modified and traditional form but provides little discussion on why this is the case. It might have been better if the author clarifies further on how the conventional measure can have insignificant influence compared to modified measure, considering that both have a high correlation to each other, and can be replaced by one another.

Overall, this article offers an analytical look at mutual fund evaluation and the issue of assessing risk and returns and offers itself as a contributor in the discussion. Despite some criticism, the research offers interesting insights into the further discussions on downside risk and a starting point for investors assessing mutual funds in the country using the method given. It is a well-argued paper and met its objectives given.

Iliyas Ismail is a financial analyst

Bibliography:

Balzer, Leslie A. "Measuring Investment Risk: A Review," Journal of Investing, 1994, volume 3(3), 47-58

Chong J., Jin, Y. and Philips G.M. (2013) The Entrepreneur’s Cost of Capital: Incorporating Downside Risk in the Buildup Method. MacroRisk Analytics Working Paper Series

Grootveld, H. and Hallerbach, W. (1999). Variance vs downside risk: Is there really that much difference?. European Journal of Operational Research 114 (2): 304

Merriken, Harry E. (1994). "Analytical Approaches To Limit Downside Risk: Semivariance And The Need For Liquidity," Journal of Investing, volume 3(3), 65-72.

Nawrocki, David. (2000). A Brief History of Downside Risk Measures. The Journal of Investing. 8 (3)

Perello, J. (2007). Downside Risk analysis applied to the Hedge Funds universe. Physica A: Statistical Mechanics and its Applications. 383 (2). 480-496.

Sortino, F. & Satchell, S (2001) Managing downside risk in financial markets London: Butterworth Heinemann

Sortino, F. A. and Forsey H. J. (1996) "On The Use And Misuse Of Downside Risk" Journal of Portfolio Management, volume 22 (2), 35-42.

Tahir, M., Abbas, Q., Sargana, S.M., Ayub, U. and Saeed, S.K. (2013), “An investigation ofbeta and downside beta based CAPM-case study of Karachi Stock Exchange”,AmericanJournal of Scientific Research, Vol. 85, pp. 118-135