The Tale of Two Studies | Value Research S&P has just concluded that a majority of active funds don't outdo benchmarks. That's bang opposite our findings last month. Here's how it came about
Top of Mind

The Tale of Two Studies

S&P has just concluded that a majority of active funds don't outdo benchmarks. That's bang opposite our findings last month. Here's how it came about

Fund performance numbers, like beauty, seem to lie in the eyes of the beholder. Our cover story last month highlighted how an overwhelming majority of actively managed equity funds in India have beaten their benchmarks.

Surprisingly, S&P Dow Jones Indices has recently published a study showing that a 'majority of actively managed equity funds' in India have underperformed benchmarks!

So what's the true position? How can two studies, based on the same fund data, come up with such starkly different conclusions? Well, it turns out they can, based on the universe and benchmarks they considered.

Only large-caps?

One, there's the difference in fund classification. S&P Dow Jones classifies equity funds into three broad categories of Large-cap, ELSS and Mid/small-cap funds. Taking stock of their returns for the last 5 years, it then finds that a majority of large-cap funds (57.94 per cent), underperformed the index.

Value Research classifies equity funds into large-cap, large/midcap, multicap and mid/smallcap funds. Taking stock of the first two categories, it found that 74 per cent of the funds had beaten their benchmarks over the last five years.

Now, it is not clear how S&P defines its 'large-cap' category. In India, there are hardly any pure large-cap funds. Most funds in the large-cap category do take on sizeable mid-cap stock exposures (upto 40 per cent in some cases). Value Research considers funds with over 80 per cent large-cap exposure as large-cap funds and those with over 60 per cent large-cap exposure as large/mid-cap funds. It was in these two categories combined, featuring 96 funds, that our study found 74 per cent outperforming the indices. S&P's finding that only 59 per cent did could be because of a smaller fund universe or a different classification. We don't know which.

If one sets aside large-cap funds, S&P's study does not really show active funds 'under- performing' benchmarks. In fact, 69.44 per cent of ELSS funds and 64.15 per cent of mid/small cap funds considered by S&P outperformed the indices. This suggests that in these categories at least, active investing did fare better than indexing.

Therefore, S&P's claim that a majority of active equity funds underperformed has a qualifier- Only in the large-cap category!

Benchmarks matter

The benchmarks used by the two studies are very different too. Because Indian equity funds are not purists in sticking to one style or market-cap, it is hard to find a common index that can serve as a good yardstick for each category.

Value Research solved this problem by pitting each equity fund's returns against its own chosen benchmark. On this basis, we found that 74 per cent of large-cap funds, 95 per cent of the mid and smallcap funds and 91 per cent of multicap equity funds outperformed their chosen benchmarks over 5 years. After all, when you buy an active fund, the promise it is making is that it will beat its chosen benchmark.

But S&P's study measures large-cap funds against the S&P BSE 100, mid-cap funds against S&P BSE Midcap and ELSS funds against S&P BSE 200. This seems to have resulted in its more adverse findings. That adds another qualifier to the statement. It should probably read- A majority of active funds failed to beat our benchmarks.


Three, while the Value Research study considers only open end funds that were in existence when we did this analysis, S&P's considers all funds that were in existence at the start of the 5,3 or 1 year period. It reckons that taking only current funds leads to a survivorship bias. Because poorly performing funds get merged or close down over the years, in the long term, only the best funds survive. Therefore, if you take stock of 5 year performance of the funds existing today, their numbers will look unusually good.

While this is a valid point, there are counters to it. In India, instances of funds closing down and returning investor money due to poor performance are few and far between. Fund houses saddled with chronic underperformers often merge these funds into their better performers to improve their record. And most ownership changes in India have nothing to do with performance. Some really good funds have failed to 'survive' over the past five years simply because their global sponsor decided that India was not a focus market. Such funds haven't vanished but have been taken over and relabelled or merged into the acquiring house. Unitholders in these funds would probably still be investors in the acquiring fund, which would figure in any analysis done today. This makes it hard to conclude that active fund numbers look better due to survivorship bias.

Finally, even the two studies threw up more similar findings, they've drawn vastly different conclusions. For instance, S&P's study found that most categories of equity fund did well against the index in the 1 year period. According to it, only 23.8 per cent of large cap funds, 5.4 per cent of ELSS funds and 10.8 per cent of mid/small-cap funds trailed the index in the one year to December 2014. These numbers match to a large extent with Value Research findings.

But S&P uses this finding to state that actively managed funds may outperform the benchmark over 'shorter time periods', but find it difficult to maintain the performance over longer periods. Based on our numbers, we in fact inclined to think the opposite. That is, active funds may struggle to outperform markets during shorter time frames but over 5 or 10 year periods they convincingly trounce their benchmarks.

Based on the facts and figures, you can draw your own conclusions. In any case, to investors, the whole episode is a good illustration of why you must not get carried away by statistics. They tell you only part of the story.

And even that part depends on the point of view the researcher wants to portray.

Other Categories