My latest at Yahoo on the absence of Nifty EPS Growth:

I’ve been complaining about the lack of Nifty Earnings Per Share (EPS) Growth for at least two years now. Amazingly, three Septembers ago, in 2007, the Nifty 50 companies provided an EPS of about Rs. 220, and the number today is about 238. What this tells us is that the strongest Indian companies have grown their earnings (per share) at just 8% in three years, which is less than 3% a year. This sounds peculiar for a country that has shown 6% GDP growth or higher for the last three years – shouldn’t our top companies be growing faster?

Earnings Per Share for an index like Nifty are difficult to understand. I can understand EPS for a company, you say. If a company has 100 shares, it makes 1,000 rupees in net profit; I can calculate the EPS as Rs. 10 per share. And the Price to Earnings ratio, or the P/E, is simply the market price of the company share divided by its EPS. If Reliance earns 4000 crores and has 80 crore shares, it’s EPS is 50. At a price of 1,000, it’s P/E is 1,000/50 , or 20.

Simple enough. But when it comes to an index, which has a number of companies in it, what is the P/E or EPS? This is where it gets slightly complex. Indian indices do not give each of their constituent companies equal importance. Most weigh companies according to market capitalization – that is, number of free-float shares multiplied by the per-share price. Why “free-floating”? Basically stock owned by promoters is not usually traded, so only non-promoter stock is considered, and multiplied by the market price of the stock to get the “Free Float Market Cap”. (Some indices consider promoter stock as well, but let me not confuse you) Then, each company in the index is ranked by its free float market cap, and its weight is calculated accordingly. This data is revealed on a daily basis. Recent data shows that Reliance Industries is nearly 10% of the index, followed by Infosys, ICICI Bank, Larsen and Toubro, ITC and HDFC. Together, the top six comprise 40% of the Nifty by weight.

Now if we take the last four quarter earnings (net profits) of all the Nifty 50 companies, weigh them by the individual company’s Nifty weight, they will add up to a figure called the Weighted Earnings of the Index. Given that we know the free float of every Index stock, and its share price, we can multiply those with the stock’s weight and get the index’s “Weighted Market Capitalization”. The weighted market cap divided by the weighted earnings should be the Price to Earnings ratio of the Index or the “Index P/E”. And then, the index value divided by the Index P/E is the Index Earnings Per Share.

Luckily we don’t have to calculate this by hand every day. The NSE and BSE provide daily index P/E levels for all their indices. We can use that to calculate the Index EPS. For a Nifty value of 5500, if the Index P/E is 20, the Index EPS is about 275.

Currently, Nifty is around 5,500 and the Nifty P/E is 23, so the index EPS is at 238; from last year the growth has been an abysmal 6%. Some say this is due to the weak performance of the top few companies which are heavily weighted in the index. Not quite. Reliance grew its profits by over 25% in the last quarter compared to the June 2009 quarter (and its EPS as well). ICICIBank, HDFC and ITC grew their bottomlines more than 20%. Infosys and L&T were subdued, but they couldn’t have accounted for all of the fall – the remaining 44 companies that account for about 60% of the Nifty have underperformed severely.

And it’s not just the Nifty; the CNX 100 – a broader index consisting of the top 100 stocks, and the S&P 500, which is the broadest index available, have both seen their EPS growth (This year’s EPS compared to last year) falling dramatically – at 8% and 10.5% respectively, from highs of 18% earlier this year. Sensex EPS growth is similarly subdued; for three years, analysts have been predicting a Sensex EPS of 1,000 – and today it’s still at 840, an unimpressive 8% growth over last year.

The growth of a company’s EPS is useful in valuing companies. If a company grows earnings 20%, but in order to do that, has to raise money and dilute capital by say 1/4th, the true EPS will actually drop. (by 4%, for the mathematically inclined) And that is useless to the investor, because by the dilution he gains lesser and lesser from his ownership. The price you pay for a company might be looked at as a multiple of the EPS growth you expect – so if you expect a company with an EPS of 10 to grow by 20% year on year, you might want to pay Rs. 200 for the stock, which is a “P/E multiple” of 20.

Indices are expected to perform similarly – and the Sensex and Nifty did indeed deliver stellar returns in terms of EPS growth prior to 2007, up as much as 50% year-over-year at one point. But since then the growth has been absent – to some extent one might argue that fear of a recession played a role; but we have definitely recovered in the last year, haven’t we? Yet, most of the growth in the index has been an expansion in the P/E ratio, not earnings. Meaning, investors are willing to bet that the future is much brighter, that our companies will shrug off the dirty past and get better. For three years, that bet hasn’t worked in terms of earnings; as the Nifty and Sensex stay at two year highs, will it win from now on?

In our favour – most companies aren’t required to report consolidated earnings which means the profits made in subsidiaries might not be reported. The Nifty P/E calculations must then take only “standalone” earnings; so the real Nifty EPS might be higher if companies like Tata Motors or Tata Steel make much larger profits in their subsidiaries.

Also, the Nifty P/E is heavily skewed towards just a few sectors: Oil and Gas, IT and Financial Services. India’s growth may be more secular, and as other companies gain market capitalization, we should see a better representation. And last, focusing on the historical EPS may not be worthwhile; after all the market should discount the future, and the P/E has traditionally stayed very high during turnarounds. Yet, if this was true, the markets should have predicted the last three miserable years of low EPS growth – other than for a few months in 2008-09, the P/E of the Nifty has remained about 15 for the last three years.

Looking through rose coloured glasses is useful sometimes – optimism builds confidence – but it’s not prudent to ignore warning signs. Not everything can be explained away with those four beautiful but dangerous words – “This time it’s different”.

>Q : Lets say EPS of a company is 100. If stock splitted by twice, there are twice number of shares,so, EPS = 100/2 = 50. So it looks as if EPS has goes down substantially, but in effect nothing happened to company earnings, it is same.

So assuming stock splits happened in some index companies (eg airtel, reliance happened i guess across these 3 years), does this index EPS reflect a accurate picture of actual earnings of companies across years ?

Shouldnt one look at total earnings of companies across years and compare to see how much earnings has changed ? .

(ie total earnings of all companies was 100cr, now it is 120 crs, so earnings increased by 20%)

[ if some issue with my Q, pardon my ignorance, i only have limited understandings of finance. I think there is some flaw in my argument, when stock splits, for index calculation, possibly some appropriate weight is added to it to make total index value same as before a stock splitted ]

>Splits and bonuses are taken into account. For a single stock EPS is "adjusted" – like Reliance EPS for Q1 10 was 14, but Q1 09 was 23, and reliance had a 1:1 bonus last year. So adjusted Q1 09 EPS is 11.5, from which the current 14 is a 25% growth.

With index EPS it doesn't ever depend on the number of shares, only on free float market cap and weighted earnings (not weighted EPS). the total earnings of the company and the market cap remain the same even if you split 100 times, so individual company splits or bonuses don't matter for index EPS calculations

>I guess NIFTY will hit 6000, even break all time highs by diwali, the correction is due and september series expiry might me around 5500 odd level so be cautious at higher levels

http://marketsinandout.blogspot.com/