Follow The Leaders
By Devangshu Datta with Narayan Krishnamurthy and Sameer bharadwaj December 28, 2006
There's scarcely any necessity to caption the pictures appearing on this page and on the next few pages. These gentlemen are all members of the India Inc A Team. Their pictures, biographies and sound bytes appear everywhere - across news channels and in billionaires' lists. And, also in the databases of large institutional investors.
The latter is hardly surprising. Collectively they run the biggest and most successful of India's companies and that is where the bulk of mutual fund corpuses are invested. Fund managers are cautious individuals; before they start hunting for unusual stories andextraordinary gains, they park around 56% of their corpus in the usual suspects - that is, the giant stocks that comprise the population of mainstream indices such as the Nifty and Sensex.
Market values in most stock markets exhibit what is known as the Pareto Principle (sometimes called the "80:20 rule" where 20% of the total population performs 80% of the useful activity). A few big stocks generate the lions' share of market capitalisation. There are over 5,000 stocks listed on Indian exchanges. But the top 50 stocks (which make up the population of the Nifty) together generate over 50% of total market capitalisation. At the other end of the scale, there are over 4,000 stocks, which together generate less than 10% of the total market cap.
The top 10 picks of Indian fund managers (and this is closely mirrored by foreign institutional investors as well) consist of giant stocks with an average market capitalisation of Rs 87,000 crore.
The fund managers are making what is known in gaming terminology as a "safety-play". As a bloc, a portfolio of Sensex/Nifty stocks will offer returns that are fairly close to that of the indices themselves. By parking over 50% of their cash within his population, the fund managers ensure that their returns will not drop below a certain acceptable level in the worst circumstances. Even if a big loss is reflected in a dropping NAV, it will be mirrored by similar losses across the market.
It's a win-win gambit since India is very much a growth market and even the largest businesses are still delivering double-digit annual returns. In fact, well over 65% of fund flows are into growth stocks - regardless of size. Very few of the top 20 large cap holdings of mutual funds have disappointed their shareholders in the bull run of the past three years.
In the past five years, the Nifty has delivered an absolute return of 274% (about 30% per annum) and the stocks that comprise the top 20 mutual fund holdings, have returned an average of 42%. In the past three years, the Nifty has returned 35% per annum while the top 20 has returned 56%. In the past year, the top 20 has returned an average of 59% while the Nifty has risen 49%.
As you can see, the returns are better than the Nifty although the top 20 is a more volatile set. That's natural with a smaller universe of stocks and an unweighted average. A comparatively small stock that under/out performs will affect thetop 20 more than it would affect the Nifty, which is weighted according to overall market capitalisation.'
Of course, there are extreme outperformers such as Grasim (+104%) and Zee Telefilms (136%). And there are underperformers - notably HPCL (-13%). But this set as a whole has delivered pretty satisfactory returns that are in line with, and better than, the broad market. That means fund managers have not been under pressure to think out of the box when picking their core holdings.
You can pretty much stick a pin into the top 20 list and find a safe stock. Can you do better than that? John Maynard Keynes once compared successful stock-selection to picking the winner at a beauty contest. It isn't a question of picking the girl you consider the most beautiful; it's a question of picking the girl the judges will consider the most beautiful (incidentally Keynes was married to a beautiful Russian ballerina).
Similarly, when you pick a stock, you get decent returns only if other people also think the stock is investment-worthy. This entire study is an attempt to win such a beauty contest - we are tracking stocks that other people, namely fund managers, consider attractive. According to our logic, the higher the degree of consensus on a stock being investment-worthy, the more likely is the stock to yield safe returns.
That's why we have given an estimate not only of total fund holdings but of the total number of funds invested in a given stock. This is to ensure that there is a high degree of consensus about the stock. Otherwise, there is a chance that one or two funds making massive bets could skew the picture.
In a beauty contest, judges normally use a weighted system where they rank the contestants 1, 2 and 3 with points awarded for each placement. Well, we haven't done that but we can offer a snap shot of the number of companies, which have committed to holding the top 10 stocks as their top, second or third holding (see box Popularity Contest). As is obvious from the table, Reliance and Infosys are clear winners in terms of widespread popularity.
Instead of simply mirroring stocks that the funds' love, we have also outlined a method of stockpicking, which is mechanical but it should keep an investor out of trouble and, in the long term, yield results that equal the benchmark market indices at the least. We suspect that it could produce returns that beat the funds.
Here's how the logic works. All these stocks are covered in obsessive fashion by the in-house research departments of the institutions that own them as well as by other investment advisory houses and brokerages. Each of the top 20 has been the subject of multiple research reports. It's easy to derive both consensus estimates on financial projections and advice (buy/hold/sell) as well as the occasional contrarian view.
We accessed multiple sources with a concentration on ICICI Direct to derive our consensus estimates. Our sources include Merrill Lynch, First Global, JP Morgan Securities Equities, Macquarie Research Equities, UBS Equities, Motilal Oswal Securities, Kotak Securities, Prabhudas Liladher, Brics Securities and Edelweiss Capital among others.
We have used those estimates and a specific valuation tool to derive ballpark price-target projections. First, we have multiplied the 2005-6 price earnings ratio (PE) with the estimated 2006-7 earnings per share (EPS) to derive a price target for April 2007.
The extent of possible nearterm capital appreciation can be gauged from this, although there are bound to be both pleasant and unpleasant surprises in store for anybody who thinks that target is set in stone. The full methodology is explained in the box (see Arriving at Target Price)
What's interesting is that the price-targets we derived from a consensus of analyst estimates, suggest that there is a lot of room for appreciation for these stocks from even the exalted levels of Nifty 3962.
But let's start with a statutory warning. Any statistical system will have a few weaknesses. In stock analysis involving PE ratios, the problems occur in the case of a stock, which has a very low but positive earnings per share. This leads to situations when the PE ratio is unnaturally high. If the stock is a turnaround, the PE ratio returns to within normal limits as the EPS climbs. But there is a period when it throws up absurd results.
Zee Tele is a complete outlier on this consensus target system because its 2005-6 PE is absurdly inflated at 463. Zee is in a fiercely competitive market and it won't make the numbers our targeting system throws up.
Reliance Communications is a newcomer and, while we believe that the 2006-7 EPS estimates of Rs 13 are reasonable, the 2005-6 PE of 123 is very likely to ease down close to competitor Bharti Airtel's 42 PE. If it does, the April 2007 target will drop to about Rs 525-550 and yield a return of about 17-20%, which is nice enough but far less startling than the current target of Rs 1,623.
Jaiprakash Associates is another relative newcomer and the target of Rs 1,637 is built around reasonable consensus PE expectations but a high 2005-6 PE ratio of 47 plus suggests that it may not meet that demanding target.
But even if we ignore the targets which seem to be aberrations, that leaves us with long-established businesses. Among others, TCS (implied appreciation of 210%),Grasim (67%) , Siemens (88%), Crompton Greaves (60%), Bharti (35%), ONGC (32%) Infosys (24%) etc, are all looking like good investments on this system.
Each of these companies has had a fairly predictable PE ratio over a long period of time and most have consensus EPS growth estimates and management guidance that appears quite credible. Also, while low returns are implied for Tata Motors (3.5%), Maruti (6.3%) and Reliance (2.6%), there are no implications of negative returns in the picture. Perhaps the entire analyst community consists of manic optimists?
To try and offer another filter on the valuation front, we have also used the popular price-earnings to growth ratio or PEG to get a fix on current valuations adjusted for potential earnings growth.
A PEG ratio is mechanical but it offers a rule-of-thumb valuation. In practice, a low PEG stock is likely to be safer and to offer more capital appreciation than a high PEG stock. Again the methodology is explained in detail in the box (see PEG and its Implications).
As we have said above, every statistical tools will be prone to certain aberrations. The PEG throws up strange numbers if growth drops to a low rate or goes negative. In these cases, the PEG can rise to absurd heights and any projections made on the PEG's basis is very error-prone. We will come across this situation several times in the following analysis.
There are few surprises to be found inside the ambit of the top large-cap holdings of mutual funds. Most of these companies have 100- odd funds nested amongst their shareholder base and, of course, the FIIs are also heavily invested in these highly liquid stocks.
However, there are a few new inclusions in the list. Most largecaps attain that distinction only after years of struggling to grow new businesses. Infosys for instance, took seven years from its listing to its inclusion in the heavyweight indices (incidentally, Infy has just been included in the Nasdaq 100 - the first Indian company to make it to the premier global technology index). Crompton Greaves took decades to climb the ranks from mid-cap to large-cap.
But some companies take the "elevator to stardom, rather than walking up the stairs" as Vinod Kambli famously described his mate Sachin Tendulkar's ascent. Reliance Communications, for instance, has only been listed since May 2006. Prior to that, in the days when the Ambani siblings still spoke as one, the company was an unlisted subsidiary of the undivided group's flagship, Reliance Industries Limited.
Once it was spun off, R Comm promptly forced its way into the list of elite companies by dint of sheer size, setting a record for the fastest inclusion into the major indices. Jaiprakash Associates is another relatively new concern that has entered the ambit of the largecaps and become a name to be reckoned with.
In terms of sector allocations, the top 20's sector biases reflect overall allocation biases (see story India's Most Wanted Sectors). The set is overweight in technology services - the list includes IT majors such as Infosys, TCS and satyam as well as telecom service providers such as R Comm and Bharti Airtel. The other perennial favourites are FMCG (ITC and HLL), and engineering (Bhel, Siemens, L&T).
According to our PEG calculations, most of the top 20 are either close to fair-value or underpriced in that the PEG is below 1 or close to unity. This is surprising until you realise that earnings growth was truly excellent in 2005-6 and it has been pretty good even in the first half of 2006-7.
Maruti, Bhel, ONGC, Satyam and TCS are amongst the cheapest stocks in terms of low PEGs. None of these are really cheap in terms of PE ratio in itself. However, all have displayed earnings growth at rates that exceed the PE and that translates into a low PEG, which is a buy.
A few stocks are however, definitely over-valued. The PEG is a growth-oriented tool. While it often signals a "buy" on apparently highpriced stocks that are growing atbetter than their PE ratios, it always signals a sell on any stock that has even a minor slowdown in momentum.
For example, L&T, Grasim and ITC are "sells" due to negative and low 2005-6 EPS growth of -2%, -2.5% and +1.9% respectively. If you're feeling optimistic about stocks such as these and believe that the dip was temporary, we suggest that you apply the same PEG formula with estimated 2006-7 EPS. Remember however that projections are rarely met with exactitude so, keep a margin for error.
Incidentally, if you are going to trade the large-cap universe, you need to be certain that you don't fall for any one out of several market myths. Take the quiz (see box Fund Familiarity Quiz) and check if you are carrying baggage in terms of false beliefs.
Don't be intimidated by high share prices either - the price of a single share is almost irrelevant because the minimum market lot is just one share. Large-caps are no more expensive than mid-caps or small-caps for that matter. Just one out of the total population of 20 large-caps, 10 mid-caps and 10 small-caps, which we focused on, is priced at less than three digits. The key is to ensure that you don't buy at high valuations, not the nominal price of the share.
The other thing is that mechanical trading systems are all very well when it comes to creating initial filters. By tracking fundtrading patterns, you establish a first filter, which offers some comfortin that you know that institutions with vast resources have already vetted the stock and put their money on the line.
By using a second filter in the form of PEG, you attempt to pick stocks, which are on the cheap side in terms of valuations relative to their growth prospects. This is an aggressive strategy.
We have added a third filter in projecting forward to calculate a likely price for April 2007, if the same PEs are maintained and consensus EPS estimates are also met. By comparing that April target price with the current price, we have an estimate of which stocks could offer excess short-term returns.
All this is wonderful. But you should not blindly buy stocks for the long-term on the basis of black-box mechanical systems. Stock-picking is an art, not a science. Once you've used these filters to narrow down your potential universe of choices, you must dig deeper. Ultimately, a share represents a small share of a business. Try to understand the business and be comfortable with its future direction. If you can't do this, don't take a long-term position in individual stocks. Leave it to the professionals and buy equity mutual funds instead.