Tuesday, August 21, 2007

Which EPS to use? (for calculating PER)

We are back to my favourite topic on PER (Lao Jiao value investors are yawning right?). But I think I need to clarify one issue on PER (which stands for Price Earnings Ratio) which I KIVed for some time. For the un-initiated on PER, pls refer to this post.

Ok in short, PER is simply stock price divided by its earnings per share and it measures the cheapness of a stock. The stock price does not tell you anything about whether the stock is cheap or not!

Now to determine Price is easy, this is the all impt Price that you get from TVs, Yahoo, Your broker's system. Singtel is $3.30, SMRT is $1.75, even grandmas know this.

But EPS? Where to find this? And which year's EPS should we use?

For the sake of newbies, we go back to Finance 101. EPS is actually the net profit for the company for the year, divided by its no. of outstanding shares. These no.s are usually inside the company's annual report.

However, things published in annual reports are dated, ie. We can only get last yr's EPS. In the stock market, nobody likes to look at the past. The market is always forward looking. So we need to know next yr's EPS.

This next yr's EPS is usually an average of all the analysts' estimates which are usually not available for most folks but are easily accessible from financial service providers like Bloomberg, Reuters and Thomson One.

As convention, the PER that is usually quoted is the 1-yr forward PER (ie. Next yr's PER using analysts' estimates of 1-yr forward EPS). But for growth stocks, ie. the company is growing its profits really fast, then the 1-yr forward PER is usually too ex. Then you need to look at 3-yr forward or even 5-yr forward PER so that it gets reasonably cheap.

But it's always very dangerous to use such futuristic PER bcos the probability of error will be very very big. You may think that you are buying a 10x 5-yr forward PER stock but if the company fails to grow in its 3rd and 4th year, then Ha Base, 10x become 30x PER and the stock plunge 60%!

As for value investors (or rather any prudent investors), we should be determining what is the long-term sustainable EPS and hence what is the PER of the stock today.

There is no magic formula here to help you predict the long-term sustainable EPS. For the professional investment analysts, they talk the company's management, study their markets, do some research and analysis and try to come up with a sustainable EPS, and still usually get them wrong. So for the simple folks, what's the chance that you can get it right? Not much higher than winning Toto.

Nevertheless, that doesn't mean that you shouldn't try though. Bcos when you can get an estimate for this EPS, and apply a margin of safety, if the stock is still reasonably cheap after that, then probably it's safe to buy. But with investment, nothing is for certain, so you may still be wrong. Well at least, you learn from your mistakes.


  1. Sometimes earnings on the income statement may not actually represents economic earnings. True earnings which I would like to highlight is more accurate by looking it in the form of earnings that truly belongs to the owners.

    Owners earnings = a) Operating earnings plus b) depreciation expense and other non-cash charges and then substract c) the average amount of capitalized expenses for plants and equipments that a company needs to maintain its long term competition position and unit volume.

    For most biz, c) usually exceeds b), so cash flow analysis usually overstates economic reality. When b) & c) are different, cash flow analysis and owners earnings analysis also differs too.

  2. That is quite true, it is definitely better to use some form of revised earnings to gauge to the real economic reality.

    Using EPS offers a quick and dirty way to gauge the co's profitability and should only be used as a first cut when doing analysis on co.s

  3. The problem with the income statement is sometimes an ignorance or some other times, a blessing by the one who signs off it as correct.

    One of Abraham Lincoln favorite riddles goes like this: "How many legs does a dog have if you call his tail a leg? "The answer: "Four, because calling a tail a leg does not make it a leg." It behooves managers to remember that Abe's right even if an auditor is willing to certify that the tail is a leg.