Saturday, October 14, 2006

Intrinsic Value - Part 2

This post is updated in 2013.

The intrinsic value is determined by the future cashflows that an asset earns. In the case of stocks, it is earnings and is partially transferred to investors via dividends. In the case of property, it's the future cashflows coming from the rental income. In bonds, it is interest income. In businesses, it is real cash generated. When an asset has no cashflow, we can safely say that it has no value. This means that in today’s context, there are many things actually being masqueraded as investments such as wine, art, watches and some may even consider gold.

The ways and means to calculate intrinsic value is called Valuation. In the purest form, it is by discounting all the future cashflows back to the present and that determines the intrinsic value. This methodology is known as discounted cashflow or DCF. This is very high level stuff and for those who are really, really interested, please continue to read my 1,500 word thesis under the Value Investing Page.

Luckily, for the rest of us, I have updated this simple methodology to determine a stock's intrinsic value right here, in this post. For an earlier tongue-in-cheek discussion, please refer to my previous post on intrinsic value. In this post, I shall attempt to calculate the intrinsic value of some stock using PER and EPS. Now please understand this is really cutting really big and round corners.

Anyway, the idea behind is actually quite simple. Basically you must first have an idea of what is the true PER of the stock, and what is its sustainable EPS, say 5 to 10 years in the future. PER stands for price earnings ratio and it represents a multiple that we should pay when buying something based on its earnings power. EPS, which stands for earnings per share, is one simple way to measure that earnings power.

A simple analogy is property, Singaporeans' favourite topic. The equivalent of EPS would be the rental income and the equivalent of the multiple is the inverse of the rental yield that we are prepared to pay. Say Sky Habitat (my favourite property in Bishan) has an annual rental of $48,000 (or $4,000 per month) and the right rental yield should be 4%, or inversely a "mulitple" of 1/0.04 = 25x. This means that the price of this Sky Habitat condo should be 48k x 25 which gives us $1,200,000. Today Sky Habitat condo sells for $1,500,000 to $2,000,000. What does this says about its price vs intrinsic value? :)

In the case of stocks, we can use the same methodology by multiplying the PER with EPS and get the intrinsic value. For those still blur, don't worry, here's the formula:

PER = Share price / Earnings per share (EPS)

if we swap the formula around,

Intrinsic value = true PER x sustainable EPS

Simple right? But do take note of the words in italics: true and sustainable.

So, how do we know what is the true PER of the stock? The true PER isn't next year's PER as it is commonly used in today's market. It is a reflection of the really correct multiple that we should use. This has to be determined by looking at the historical multiple of the stock in question, or by comparing with peers, or by thinking in terms of the growth, stability of the business etc. Usually, the multiple ranges from 12-18x. Personally, I would be very reluctant to pay for anything more than 18x PER. There is a lot more in-depth discussion on this and it can really intellectual simulating and interesting. I am not kidding. Interested parties please continue to read my 1,500 word thesis under the Value Investing Page.

Next we need to determine its sustainable EPS. Again this is not the current EPS, or next year's EPS. It has to be a sustainable EPS over a decent time period, say, 5 to 10 years. In fact, for most value investors, they would be looking for sustainable EPS over much longer periods, like 10 or even 20 years. Of course it is almost impossible to be able to determine with accuracy over such time frame. Which is why Buffett has a tendency to invest in certain kinds of businesses which allows for some predictability over long time frames.

Ok let's use a real stock to illustrate this intrinsic value thing. I would like to use Singtel, our largest stock that has the longest history.

Singtel has managed an average EPS of 22c to 28c over the last 10 years. So assuming that it can maintain this, we can say that it's sustainable EPS is probably 25c. Now we need to determine its true PER. Again looking at its own history, it has traded around 12x to 17x.  Globally telcos have also traded around 8x to 20x PE. So what's the right number to use? There isn't really much science here, I would use 15x.

So Singtel's intrinsic value should be 0.25 x 15 = $3.75. Today it is trading at $3.60. So it's a buy?

Well, the answer is no, because our calculations may be wrong. What if the true PER is 14x? Or what if the potential EPS is only 22c?. If you work if this new set of no.s, (0.22 x 14 = $3.08) Singtel is now 15% overvalued!

Intrinsic value is not just a single number, it has a range of probable values. It changes when we use different assumptions and nobody ever gets it right. And I really mean nobody, not even the Gahmen and definitely not the Korean pop group who declared "Nobody, Nobody but You".

Ok jokes aside. Since we can never be sure we got the intrinsiv value, that's where we need the margin of safety comes in. For those not-so-familiar, pls read the post on margin of safety.

However, if Singtel ever trades at $2.00. Then it is definitely a buy, because no matter how you tweak the two no.s (PER or EPS), you probably find it very hard to get Singtel's intrinsic value to go below 2 bucks. In fact, Singtel has only traded at 2 bucks for a brief period over the last 10 years, the recent bout during the Global Financial Crisis in 2008 which is like 5 years ago. In other words, it is damn bloody difficult to find great companies trading below its probable range of acceptable intrinsic values for long. This is the challenge and this is value investing.

Singtel's stock price from 2004 to 2013

See also Definition: Value investing
and Expectations vs Reality


  1. Sounds easy! But lots of guess work isn't it?! And isn't the results of inaccurate assumption/estimate disasterous?!

  2. Well you have to guess 2 things correctly, that's quite good compared to some other models like DCF.

    In DCF, you have to guess roughly 5-10 inputs. I hope to write something on that some day. Just the idea, though. Not sure if it is useful.

    I think the main point is that when you have an idea of how the intrinsic value is like, e.g. SIA should be around $12-$17, it gives you a starting point.

    From there, as you refine your assumptions, you will have a better grasp of its true intrinsic value.

    It is never easy. Like everything else in life.

  3. any stock ur holdin that fits those criteria now ?

  4. People's food, Singtel. Stocks that I don't have but looks interesting include memtech, food junction.

    I won't be right, so do your own homework. Nobody is right more than 5 out of 10 times. The winners like Buffett let profit runs and cut loss early.

    Also more impt than what to buy is WHY? I tend to focus on the why. And to know exactly why and to know why the stock you have bought will not make you lose your shirt is the key.

    It is never easy. Like everything else in life.

  5. nothing is easy