How to arrange mutual funds in terms of their returns?

Last Updated on December 29, 2021 at 5:49 pm

We received this interesting question from a reader who wishes to be anonymous: “There are too many categories of mutual funds. Is there a way by which we can arrange them in terms of their returns?” In this article, we discuss a simple way to do this. We say that equity, debt, gold, real estate are different asset classes. What makes them different?

Is it the returns that they offer? If an equity mutual fund has produced an 18% return over some period of time, we can always find some other point in history where debt (eg gilt funs) or gold has produced the same return for the same period. The same is true for negative returns too. Using returns directly to classify asset classes or different mutual funds will not be consistent. We do use returns for classification but indirectly.

Instead, we can classify mutual fund or asset classes by how much returns fluctuate. Or in effect how much the NAV moves up or down. There are two ways to do this. We shall see how with an example.  We shall consider a three-year window for this study, from 26th April 2018 to 26th April 2021 and compute the average of monthly returns of different indices. The indices represent different mutual fund categories.

The standard deviation of monthly returns is a measure of how much the monthly returns deviate from the average monthly return. We have used this measure several times in the past to classify mutual funds. See for example: When to choose what mutual fund? And The key to successful mutual fund investing. In this article, we shall directly use returns to appreciate how much the returns fluctuate and classify them.

The average monthly return of the CRISIL Liquid Fund Index over the three year period mentioned above is 0.49%. We can convert this into an annualized return: (1+0.49%)^12 -1 = 6%. The annualized return over this three year period (using only start date NAV and end date NAV) is 5.95%.  The difference between the two calculations (6% – 5.95%) is a measure of volatility.

Why? In a fixed deposit, each monthly return would be the same. So both return calculation would give you the same result. So with zero difference to represent no volatility, we get an estimate of the volatility of a liquid fund index: 0.05% (approximately). Any departure from zero represents fluctuating returns.

To compute the standard deviation, we compute the difference between each monthly return and the average monthly return (there are 36 such data points for our duration). Take the square of this difference and compute the average. Then take the square root of that average. This will always be positive. Higher the standard deviation, the higher the volatility in price. This is only one type of risk.

Even this can be misleading. If you look only at the last three years, gold could have smoothly gone up or crashed or moved nowhere, while equity could have done the opposite. So a rolling standard deviation analysis (like return) is necessary for better consistency. The data presented below should be only be considered as a flavour of what to expect only represents the volatility for the period considered. The standard deviation of some indices can swing as wildly as their returns!

Some segments are always more volatile. For example, small cap indices are typically always more volatile than large cap indices and so on. So interpreting this data requires some perspective.

How does using standard deviation help? Higher the value, the higher the volatility in NAV, the higher the fluctuations in returns. Higher fluctuations in returns only mean that and do not mean the possibility of higher returns! Higher the spread in returns, the higher the uncertainty in what return we would get! See: Do not expect returns from mutual fund SIPs! Do this instead!

We present standard deviation and return difference data (bare value) for 161 indices.  We recommend readers see how the standard deviation increases as we move from debt funds to gold to international equity, hybrid funds, diversified equity, thematic indices and other commodities.

Several interesting inferences (over the time windows studied) can be made. A few are:

• Gold is three times more volatile than gilt funds
• The aggressive hybrid index is only a bit less volatile than equity indices
• Merval (Argentine index) and Brent Oil are the most volatile (pole apart from the overnight index!)
• Midcap, smallcap and thematic indices occupy the bottom of the table.

Note: The difference between the return from the arithmetic mean and annualized return (geometric mean) should always be positive since AM >= GM. Thanks to Siva in AIFW for pointing this out.