earnings ratio or commonly referred to as the P/E ratio. Though commonly
used, it is also misunderstood for various reasons. Here is an attempt to
simplify this valuation metric.
*How is P/E calculated?*
It is calculated by dividing market price of a stock by EPS (earnings per
share). EPS in turn is calculated by dividing the net profit of the company
by the number of shares outstanding.
*Having calculated the P/E, what does it stand for? *
Lets assume a stock is trading at Rs 100 and its EPS is Rs 20. The P/E
multiple is 5 (100 upon 20). Assuming that the company's EPS is likely to be
Rs 20 each year, it will take 5 years for the investor to realize Rs 100. Of
course, the assumption here is that the company's EPS is not growing at all.
Now taking the example of commonly traded stocks like Infosys and Tisco.
While the former trades at a P/E multiple of 25 times, the latter trades at
7 times. Why is it so? It is believed that the stock price of a company
tracks its long-term earnings growth potential. In an economy, some
companies (or sectors) are likely to grow at a faster (like say software or
pharma) rate. So, the P/E multiple of companies from these sectors are
likely to be higher and vice versa. Depending upon growth expectations, the
P/E multiple could vary.
There is one crucial factor here i.e. expectations. Though Infosys may be
trading at 25 times earnings, if EPS is expected to grow by 25% per annum,
the investor could realize the money in four years.
*P/E Is it a discount or a multiple? *
There are two ways of quoting P/E valuations:
Tisco is currently trading at Rs 350 *discounting* its earnings by 5.5
Tisco is currently trading at Rs 350 at a P/E *multiple* of 5.5 times
Which is right? The answer to this lies in the formula for calculating P/E
P/E is Market price divided by EPS. If we were to reverse the formula,
Market price = P/E *multiplied* by EPS. Stock prices reflect future earnings
potential and not past performance. Discounting the current price with
historical EPS is not a right way to analyse companies.
Take a hypothetical case. If Tisco's EPS for the next year is expected at Rs
50 and the growth in EPS is around 15%, the market price is calculated by
multiplying Rs 50 with 15 times i.e. Rs 750. When determining the stock
price, one does not discount earnings but multiply earnings.
*What is the 'right' P/E multiple for a stock? *
The answer to this question is not easy. In the previous example, we have
assigned a P/E multiple of 15 times because EPS is expected to grow by 15%
in the immediate year. Is this the right way? Not necessarily. Here, it is
important to understand industry characteristics of the company.
For a commodity stock like Tisco, EPS tends to grow at a faster rate when
steel prices are recovering or are at the peak and the EPS is likely to
decline at a faster rate during downturns. To qualify this statement, if we
look at EPS growth of Tisco from 1994 to 2004, the compounded growth in
earnings is 17%. However, the CAGR growth in the last three years was 193%
(the recovery phase). So, if one believes that steel demand is likely to
trace long-term economic growth and that 15% growth is unsustainable, the
P/E multiple should be ideally much lower than 15 times. Similarly, the
long-term growth prospects for software companies could be much higher than
commodities. So, the P/E multiple for software stocks could be at a premium.
Determining the P/E multiple for a stock/sector also depends on:
*Historical performance* Why does Infosys trade at a higher P/E
multiple compared to Satyam? By historical performance, we mean, focus of
the management (without unrelated diversifications), ability to outperform
competitors in downturn/upturns and promise vs performance. This can be
gauged if one looks at the last three to five year annual reports of a
*The sector characteristics* Margin profile, whether it is asset
intensive and intensity of competition. Less asset intensive sectors (say,
FMCG) are considered defensive and therefore, could trade a premium to the
And more importantly, *expectations*. Take the case of textile stocks.
Expectations of significant growth opportunities post the 2005 quote regime
phase out has resulted in upgradation of P/E multiple of the textile sector.
*When is P/E not useful? *
*Economic cycles *- In FY02, Tisco was trading at a P/E multiple of
20.5 times its FY02 earnings. Was it expensive? Based on FY05 expected
earnings, Tisco is trading at a P/E multiple of 5 times its earnings (at Rs
250). Is it cheap? If one ignored Tisco in FY02 on the basis that it was
'expensive' on the P/E multiple in FY02, the opportunity loss is as much as
350%. Businesses operate in cycles. During downturn, EPS will be low but P/E
will be inflated and vice versa. At the same time, during expansionary
phase, corporates invest in capacities. In this case, high depreciation
costs suppress earnings. P/E, in this context, may mislead investors.
*Not actively tracked* There are number of companies in the Indian
stock market that are not actively tracked by investors, analyst and
institutions. For example, Infosys' average price was Rs 2 in FY94 and the
P/E multiple was 17 times. At times, P/E multiple may be lower because some
sectors/stocks are not in the limelight.
*Expectations* On the downside, some stocks may be trading at a
significant premium because earnings expectations are higher. High P/E also
does not mean a good stock to buy. What if the expectations are unrealistic?
One needs to exercise caution to this extent.
*Means little as a standalone number* P/E, as a standalone number,
means little. Besides P/E, it is also important to look at margins, return
on net worth, cash generating ability and consistency in performance over
the years to assign a value to a stock.
*Market sentiment * During bear phases or when interest in stocks is
low, valuations could be depressed. Since equities are considered less
attractive during these periods, valuations are likely to be below
historical average or below earnings growth prospects.
*When is P/E useful? *
*A powerful metric* Unlike metrics like discounted cash flow method and so
on, P/E is relatively a simple and at the same time, a powerful metric from
a retail investor perspective. Though the factors behind determining the
'right' P/E multiple are important, a historical perspective of a stock's
P/E could make this exercise less complex.
To conclude, valuation of stocks involves subjectivity. A person X may
assign a higher P/E multiple to the stock as compared to a person Y
depending on the risk profile and growth expectations. In the end, it all
boils down to how the company is likely to perform.
It is not that stock market is always right when it comes to valuing a
stock! As Mr. Benjamin Graham puts it "in the short term, the market is a
'voting' machine whereon countless individuals register choices that are
product partly of reason and partly of emotion. However, in the long-term,
the market is a 'weighing' machine on which the value of each issue
(business) is recorded by an exact and impersonal mechanism". Watch the