The majority of actively managed equity schemes for 1-year, 3-year and 5-year periods ending June 2012, have underperformed their benchmarks over the last five years, says a study. But does it make sense to do a fixed period analysis of a product that is not fixed income in nature? Here are three reasons why the study is flawed
S&P Indices Versus Active Funds (SPIVA) scorecard, produced by S&P Dow Jones Indices in partnership with CRISIL, highlights that the majority of actively managed Indian equity mutual funds have underperformed their respective benchmark indices over the last five years’ period ending June 2012. According to the research report, majority of large-cap equity schemes failed to beat the S&P CNX Nifty benchmark index for large-cap companies, 53.33% underperformed their benchmark over the last five years, 57.14% over the last three years and 52.63% over the last year.
Over 53.10% of diversified schemes outperformed the benchmark S&P CNX 500 in the one year period ending June 2012. This number increased to 61.6% in the three year period but again dropped to 49.5% in the five year period.
This has becomes headlines everywhere leading confirming to fence-sitters and believers of bank fixed deposits that equity mutual funds are best avoided.
However, this kind of analysis and conclusion carries a few fundamental flaws. Nobody has suggested that equities have to make money no matter when they are bought.
One, it is pertinent to note that the returns taken are for just a fixed period from June 2007 to June 2012. So which investor would have underperformed by buying mutual funds? Exactly those who had bought in June 2007. If their investment was made in June 2006, the results would have been different. Take June 2003 and it would have been different still. Returns vary, depending on the start and end date. This is natural because equity funds are not fixed income products. It is obvious that to come to any conclusion, one will have to take into account different periods. This is simply done by calculating what is called rolling returns.
According to our analysis over the last five rolling periods with a quarterly frequency, the average returns of 63% of the schemes were better than that of the Nifty index. The large-cap oriented equity schemes averaged a return of 7.59% compared to the Nifty which returned 6.62%. The headline should have read “A majority of schemes beat the market index”.
Two, no investor jumps into a scheme with all his money at one go. He is consistently advised to invest systematically.
Three, no investor invests in all the schemes. He is advised to invest in the better ones. There is no shortage of good schemes to choose from. In the rolling periods taken for our analysis, the top 10 schemes returned an average of 13.13%. While one can always argue that we know which of the schemes are better only with the benefit of hindsight, the fact is choosing good quality and index-beating schemes in India has been surprisingly easy compared to the US market.
Inside story of the National Stock Exchange’s amazing success, leading to hubris, regulatory capture and algo scam
Fiercely independent and pro-consumer information on personal finance.
1-year online access to the magazine articles published during the subscription period.
Access is given for all articles published during the week (starting Monday) your subscription starts. For example, if you subscribe on Wednesday, you will have access to articles uploaded from Monday of that week.
This means access to other articles (outside the subscription period) are not included.
Articles outside the subscription period can be bought separately for a small price per article.
Fiercely independent and pro-consumer information on personal finance.
30-day online access to the magazine articles published during the subscription period.
Access is given for all articles published during the week (starting Monday) your subscription starts. For example, if you subscribe on Wednesday, you will have access to articles uploaded from Monday of that week.
This means access to other articles (outside the subscription period) are not included.
Articles outside the subscription period can be bought separately for a small price per article.
Fiercely independent and pro-consumer information on personal finance.
Complete access to Moneylife archives since inception ( till the date of your subscription )
1)Although the figures are factually correct, people draw wrong conclusions from these figures by extrapolating them into the future. But, what works most of times is reversion to mean, although the mean may shift slightly upwards to adjust for inflation & normal growth.
2) Even though some schemes, fund houses, fund managers have consistently outperformed the benchmark, it would be a mistake to presume that they would continue to do so.
3) Because data is numerically expressed and mathematical or statistical formulas can be applied it still cannot be a pointer to the future performance. We simply do not know. (same is with ratings too) For most, volatility is a measure of risk . . . but risk is the probability of being right and the consequences that you may suffer if you are wrong.
A mutual fund investor will do well to disregard such ratings and focus on the basics; long term investing, asset allocation, diversification & periodic re-balancing. If he saves enough and invests for the long haul he will do well. Investing in mutual funds is never a race, between funds.
Even if the fund manager's poll is conducted they (the fund managers themselves) may not be able to predict which of their schemes will outperform the rest & over what period of time.
The database is purchased from moody's rating agency - mutual funds india and Madam Megha vora published how old funds are bad.
I that time also pointed out that Mastershare generated 18% cagr since 1986..but till on date u have not confirmed..
How reliable your research is when u claim that mastershare generated only 7% cagr since launch..
You have also based your article on rating agency data.. Whereas lessed aware news paper & other media has taken crisil data reseach and published..whats the difference?
The database is purchased from moody's rating agency - mutual funds india and Madam Megha vora published how old funds are bad.
I that time also pointed out that Mastershare generated 18% cagr since 1986..but till on date u have not confirmed..
How reliable your research is when u claim that mastershare generated only 7% cagr since launch..
You have also based your article on rating agency data.. Whereas lessed aware news paper & other media has taken crisil data reseach and published..whats the difference?
I have been planning to tell Dhirendra Kumar and his Mutual Fund analysis to change their valuation methodology.
1. They don't take rolling returns.
2. They take 1-year, 3-year and 5-year returns. However, these include the latest year's returns thrice and the last 3 years' returns twice. That is called Double Counting.
If you did badly in the last year, you did badly in all the 3 numbers. That is why funds that betted on infra and cap goods in the last year have languished at the bottom. Recency Bias is not only wrong, it is downright dangerous, if you believe in 'reversal to the mean'. If the market bounces back in the next year, the funds that will outperform are exactly the ones at the bottom of the list that Dhirendra Kumar posts. And the reason is the mathematical lacunae in calculating the outperformers.
3. Mathematically, they should take only 1-year rolling returns for the full period of existence and take Geometric mean of those returns. That is because gains are compounding in nature and multiply over a period of time.
In fact, to be really rigorous, they should calculate 1 - Geometric Mean of (1 + %Gain) for all one year rolling-periods the fund has been existing.
After all, an average (which is also the arithmatic mean) is a lousy way to treat gains. If I gain 60% one year and gain -60% next year, the average is not 0% (If you put Rs 100, at the end of the first year, you have Rs 160. And at the end of the second year, you have Rs 160 x (1-60%) = Rs 160 x 40% = Rs 64 in hand. You are actually down 36% in two years or a loss of 20% per year (100 to 80 and 80 to 64).
The only way to get this picture cleared is if you take the geometric mean of (1+60%) and (1-60%)
= square root of 1.6 x 0.4
= square root of 0.64
= 0.8
= (1-20%)
Or on an 'average', 20% loss a year.
Now, it will be interesting to see which funds outperform after this new criterion for evaluation. :-)
If we want to calculate the real rate of return of an investment over a period of time, then the final value of the investments is PRINCIPAL*(1 + RATE_OF_RETURN/100)^(NO_OF_YEARS). This is derived from the formula of compound interest. First year you earn PRINCIPAL * (1 + RATE/ 100), second year you earn PRINCIPAL * (1 + RATE / 100)^2 and likewise. The NO_OF_YEARS parameters explains the reason why we have to take the geometric mean here. So instead of multiplying the returns on a yearly basis and then doing the geometric mean, one may just do ((FINAL_VALUE / PRINCIPAL)^(1/NUMBER_OF_YEARS) -1).
An example might make more sense. In the above case, an investment of 100 is 64 after 2 years. So the real return is square root of (64/100) -1.
Actually this formula can also be applied to SIP. Only thing is that we have to thinking of it a bit like a recurring account where we have a fixed rate of return. Some websites have their own formula for dollar cost averaging which relies on the average cost - which I think is not the right way to calculate the returns on SIP.
In order to apply the above formula for an SIP, you will have to think of this like a series of investments growing gradually over time.
The final formula is bit length but it is bit like this
P*(1+R/100/12)^(NUMBER_OF_MONTHS) + P*(1+R/100/12)^(NUMBER_OF_MONTHS -1) + P* (1 + R/100/12)^(NUMBER_OF_MONTHS-2) and so on for the entire period. What you are doing here is for every monthly investment you are calculating the return and then adding the returns for all the months. Also note that the rate of return is divided by 12 to cater to monthly investments. If you observe this sequence, it is a geometric progression and it can be shortened easily.
While doing all these calculations, I have assumed a constant return on investment - which is of course not the case in equities. My intention was here to calculate what return I would have earned had I invested the same amount in a fixed income product which would have given me the same return.
Welcome any suggestions or comments.
1. Coming to theoretical nitty-gritty, you actually want geometrical mean of DAILY gains over the entire period of the fund's existence (adjusted for holidays in between, i.e. Friday to Monday gains are to be taken as over 3 days and so adjusted to CDGR, Compound Daily Growth Rate).
Thus, even the geometric mean of rolling 1 year periods is not right (as it gives less weightage to the boundary values of performance, i.e. the first 11 months and the last 11 months of the fund's performance).
2. For the sake of simplicity, you can take monthly gains over the entire period and take their geometric mean to get the monthly returns.
3. I used the rolling 1-year returns because people understand returns in terms of annual gains.
4. The starting point of the fund can bias the comparison. The funds that started at 2003 will do better than the funds that started in 2000 or Jan 2008, even after taking averages the way mentioned earlier. I have no answer to that challenge.
5. A great addition would be to plot monthly gains above or below benchmark gains over a period of time. That would tell how the fund performs overall. Is the outperformance due to one big 'matka', like Kwality Dairy or is it a regular small outperformance, adding into big gains. Of course, the risk is that some funds don't adhere to their mandate (some infra funds invest in Apollo Hospitals & Lupin calling them 'Healthcare Infrastructure' stocks and in ICICI Bank calling it 'financial infrastructure' stock).
6. One could compare performance over a business cycle. But that would exclude many of the new funds.
7. Probably, nothing to substitute proper detailed checkup and your judgement.
I think we don't need to do a daily or monthly or yearly comaprison to get the actual numbers. You can just take two points in time and still get the same value as rate of return whether you use monthly gain or daily gain. Perhaps my comment posted above might shed some light.
It would be interesting to see "top rated" funds with your methedology.