Thursday, December 14, 2006

Method of evaluating and selecting stocks (Part 2)

Again this is in response to Jojo, a fellow value investor, it's great to share ideas with someone who has a similar mindset towards investing, though may not be exactly the same in totality but the basic foundations are the same.
I too subscribe in total to Benjamin Graham basic notion of investing which is to find a dollar worth of business for 50 cents. This may be sniffed out in many ways, for instance, through the pure Grahamites method, or the improved method like how Warren and Charlie did.
Ok, I shall go part by part to elaborate how I select my business thus far. As to what I said earlier, I got to add that the ratios are just a trigger to consider investing in it. Well, PER and PBR are important triggers to telling you if a business is undervalued in a large part but it don't give you an idea of what the intrinsic value is.
The level of margin of safety is dependent on how confident an investor is to his valuation. Let's say if you are driving a 10-tons truck across a bridge over the Grand Canyon, you would feel you need a lot more in the way of margin of safety. But if you are driving over the Sin/Johor causeway, you need a lot less.
As for intrinsic value, I do not necessary base on that for my decision in investment. This is because if I can find a stock that is a great business and that it will survive, and have certain organic growth in the years ahead, and coupled with a valuation which is almost at its low for a long time, say, 10 years or more, I am sure the intrinsic value of its future flow of earnings at the present value today will be worth much more than the price that I pay today. Further, I got to add that I like to consider businesses which has a long history of getting good average return on shareholder equity.
Though I do not calculate the intrinsic value as always in all my purchases, I do have a method to calculate it. But the method is not simple and it is just an estimation though it is an all-important concept that offers a logical approach to evaluating the relative attractiveness of investments and businesses. To start, intrinsic value is the discounted value of the cash that can be taken out of a business during its remaining life. So what is meant here is really a long run for all the great businesses who has a great moat that has the durability to survive and grow its moat and more importantly, the durability of the moat. It is quite pointless to do an estimation of intrinsic value if the duration is 3 years, 5 years or such because you will not be able to see the effects of cash flowing in to make up for the price you pay at with such a short time frame.
And the inputs to the calculation of the intrinsic value is all important as well. And it has to be changed yearly when interest rates changes or future cash flows are revised.
And what importance intrinsic value brings to the table is it also helps to determine the fair amount of what you should pay irregardless of what is carried on the book or its book value at the point of valuation. In other words, book value by itself is meaningless as an indicator of intrinsic value.
Here's an example on how I calculate the intrinsic value for a business, I shall take this business called, USG, as an example.
In calculating intrinsic value, it is basically to compare between two options. In this case, we are to compare between investing in USG or to invest in risk-free equities like government bonds.
The process of estimating intrinsic value is firstly, to estimate the earnings that the business (USG) will receive over its lifetime or for the duration of the time frame you are willing to invest for (10, 15, 20 years or so on). Then once you manage to estimate the earnings for the duration, you must substract from that figure an estimate of what the investor would have otherwise earned if he had deployed it in some other securities like risk-free bonds. This will give you an excess earnings figure which must then be discounted at an appropriate interest rate back to the day you invested in. The dollar result equals the intrinsic economic value of the investment. This figure will then give you an idea of how much you should pay for at maximum and it brings you back to the old Aesop's axiom of "if a bird in hand is worth two in the bush."
Here's the exact intrinsic value that I calculated for USG and see if it makes much sense to you. But be cautioned that the outcome of it at the end is really dependent on two important variables, the current interest rate that I used at 5% for risk-free interest and the return of equity of the business at roughly about 26% historically.
A few months back, USG was traded at US$46.50. I used a time frame of 11 years. Then the book value of USG was about US$20, and by the year 2016, it is calcuated to be worth about $210. On the other hand, if I had opted to invest in bonds which pays 5% till 2016, the initial value of $46.50 would grow to US$75 by 2016. Obviously, at first glance, USG is a better investment but that is not the intrinsic value and it does not show you what the present value is for USG future value. So now, you got to take the difference between $210 and $75 and discount it back to today at an appropriate rate of say, 5% again. So the present value of the difference which is $135 is worth $82.90 today. So effectively, you are paying $46.50 for $82.90 of future cash flows.
I shall give another example that may make it much clearer in distinguishing the intrinsic value to the price that you should pay. I think some people like to value things too much by having the book value of the business or something as the yardstick but it isn't exactly a good way to do so.
You can gain a lot of insight from a college education. Think of education cost as its book value. And if this cost is to be accurate, it should include the earnings that were foregone by the student because he chose college over a job. Here is how it works. Firstly, you must estimate again the earnings the student will receive over his lifetime and then subtract from that figure an estimate of what he would otherwise have earned had he lacked his education. Again, the difference will give you an excess earnings figure, which must be discounted at an appropriate interest rate back to graduation day. This will then give you the intrinsic value. So the next question is if the parents of the student got his money (the cost of the education which is the book value) worth? Some students will find that the book value of their education far exceeds the intrinsic value, which means the parents or who ever paying for his education did not get his money worth. In other cases, the intrinsic value of an education will far exceed its book value, a result which proves capital is wisely deployed.
What we can glean from these 2 cases is clearly that book value is meaningless as an indicator of intrinsic value.
I am not sure if I describe clearly or I may have missed out any important point, which you may point out to me. I am extremely interested to learn more and improve on my method.
Aside from the intrinsic value, just as important, I can afford to hold businesses like these for many years or even throughout my lifetime because they have great business models which gives investor great returns if gotten at a fair price, or even better at a great price.
As for the value of USD, yes, it will definitely fall in the future because they are buying more than they are selling. So slowly, they got to sell away their assets and things like that and it will cause their currency to be worth lesser. Well, in any thing, there will be rock bottom or hit its bottom and then goes up again. So if my time frame for USG is 11 years, I really cannot tell if the exchange rate today of 1.55 will be worth more or less in 11 years but somewhere between now and then, it will probably be worth less. If I recall, in my poly days in mid 90s, SGD hit 1.42 to the Dollar, in 2002 it was at 1.82. I am no currency expert but if my time frame is long enough, I do not worry about the exchange rate of USD to SGD. Moreover, Singapore imports more from US than US imports more from SG, the only thing is SGD is pegged to a basket of currency which it is not easy to predict which direction it goes. I am pretty comfortable with the exchange rate for now which is probably at the middle between the bottom of 1.42 and high of 1.85 in recent 10 years.


Anonymous said...

Hi Bershire
Thanks for your reply. I do use NAV as a means of comparing if the company's equity has been growing and is also part of the criteria for selection of buying a company.

For intrinsic value, I look at the FCF and try to establish a growth rate (something I find it hard to achieve at times). Based on the growth rate and each year's FCF, I determine the NPV of that in today's term. A sumation of that till infinity will give be a fair value of the company's stock price. If the price is of certain margin of safety, then it is a company that I would consider.

The most difficult part for me is to determin a fair growth rate. In most companies the FCF varies throughout the years and I'm still trying to figure out the best way to establish that ups and downs. THe most common method i think people would use is the averaging effect. Let me know what you think.

Anonymous said...

Hi Berkshire
Was reading your blog again and I not sure if I have comprehend fuly but I do have the following comments:

1. How do you determine the ROE of 26%? Do you average out the companies ROE over the years?

2. By mutiplying the NAW by the ROE for the said number of years to derive the future NAV of the company, you are assuming that the company is not giving out any dividend and that all profits generated is retained for growing the company. Otherwise I think you will need to deduct the dividend amount from calculation. Another way your calculation holds true will be that you re-invest all your dividends earnings back into the company to achive the same ROE.

Do like to hear from you again.


Berkshire said...

Hi Jojo, i gotta admit it is very difficult to estimate an intrinsic value to the most precise number. All that is done is just a rough estimate. Of course, you are right that my calculation is an assumption of either the company reinvest the whole amount in it or either I reinvest whatever dividend they pay me into the same business. That's why intrinsic value is intriguing in fact and no two person can come out with the same numbers. As Warren remarked, Charlie and him does not have the same intrinsic value for the same business.

Having said that, of course it is better to know exactly how much dividend should be excluded from the earnings each year. That is really not easy to predict and maybe too complicated. For eg, Walgreens was already big enough in the 1980s with a mkt cap of US$2.3B in 1986. At that time, they pay out about US3.25cents per share which is about a distribution of 31% of earnings. Most people will assume that as a biz get bigger, the less likely a biz is able to reinvest their earnings as much as when they were at a smaller scale, right? So people will assume as they grow bigger, the more distribution in percentage they will have to return to their investors. But in reality, in Walgreens case, as the years went by, the distribution went smaller, in 2006, the distribution of dividends formed 16% of earnings.

Another eg is Coca Cola, in 1983, the distribution was over 64%, then gradually it slowly went downwards hitting a low of 34% of dividends distribution in 1997, and only in recent times, it went to 50%.

But of course there are also many great businesses whose dividends distribution as a percentage of earnings increases with time.

So it is difficult to get this part right. So what i may have given as an example may differs alot from how you derive your intrinsic value. But i suppose the basic ideas are the same whereby it is the present value of all future cash flows after deducting the earnings which would otherwise could have been earned in some alternate investments like bonds.

Frankly, if you would, i would really like to hear from you on your method of deriving intrinsic value.

For USG, 26% if i recall correctly is an average of its historic earnings excluding extraordinary charges.

Anonymous said...

Hi Berkshire
Apologies if there are any delays in replying.

Ok.. lets use an example which I have. Say SPH.
I estimate the FCF of SPH which is about $0.196/share for FY2006. My way of determining FCF is the Op profit + Dep + Amotization - Tax - Cap Expenditures. By determining the FCF of SPH for the past years, I can also estimate the compounded FCF growth which is about 13%.

Further, I estimate to infinity, the FCF for each year based on the compounded growth. I have estimated that the next 10 yearsm the growth would be 13% followed by 3% for year 11-30 and after that growth will be at 0%. This is a rather conservative estimation. With this, I'm able to estimate the NPV of FCF for each year. Hence, by adding these together, I obtain the intrinsic value of ~$4.40. Thus this would be the maximum amount that I would have paid for SPH.

Recently, I read a book on Buffettology by Mary Buffett (Warren B's daughter in-law) and this book gave a different perspective to valuing shares (Not intrinsic value). I will try to explain below:

According to the book, she estimates the compounded NAV growth of the company. In my example above, SPH, the compounded growth is about 15%. Hence we are able to estimate the NAV in say 10 years time which is about $5.4. By mutiplying this with ROE, we can determine the EPS of the company at Yr10 and hence determining the compounding rate of return of your investment. This method doesn't look at the intrinsic value but tries to value the NAV and EPS of the company. With this, we can estimated the compounded rate of return and see if the return matches your own selection criteria.

I'm still in the midst of deciphering which is a better method.

Would like to hear your comments too. Cheers and Happy new year.


Anonymous said...

BTW, the methods and example I gave doesn't constitute to whether this company is a good or bad to buy. It's soley an example to show my calculation : )


Berkshire said...

Hi Jojo, thank you for dropping by again. I will think about what you say and hopefully give you some thoughts about what I think soon. In any case, there are some food for thoughts pointed out by Warren and Charlie in recent times on newsprint business. I will post their thoughts on the blog.

The Buffettology by Mary Buffett is a very recent new print of the many books written on Buffett. However, Mary is an ex-daughter-in-law of Warren. She used to Peter Buffett's wive.

I am not sure how good this book written by Mary is. But recently, Warren disown the daughter (no blood link to the Buffetts) of Mary. Here is the link

As many will know there are many books written on Warren but none are written by the man himself. But there is one particular title which in a recent interview where he was asked which book does he thinks presented him in the most accurate manner. He answered "The essays of Warren Buffett by Lawrence Cunningham." I suppose this book should be a must-read for all serious investors who want to learn his the master himself.

Anonymous said...

Hi Berkshire,
I'm beginning to like this blog of yours and seems I'm writting quite frequently here too : )

I'll try to get this book and see what it has to say. You read it already and any different from the way you determine the intrinsic value or other pointers to note from the book? Trying to get a summarised version from you if you have read it. :)

ok... back to the comment on newspapers. I agree that the world has evolved and with internet and many other means of reaching individuals, newspapers is slowly becoming non-consumer monopoly.

In order to evaluate this type of company I think we need to study to see if the company itself is also evolving and changing with times and challenges. They must cast the net wider and go beyond newspapers printing.

For SPH, what I do see is that the company is slowing getting into internet base businesses (maybe late but let's hope they play catchup). They have recently acquired and looking into broadening multimedia business. Such company (in my opinion) is worth having a re-look. Since already have a monopolistic business in Spore (possibly shrinking in years though) but it is still a monopolistic business as there are no other newspaper print company. Spore has tried to open up this sector but has since retrack back - suggesting the market may be too small for 2 of such.
With addition of multimedia as a possible avenue of income, it may be worth considering... provided the price is right. Cheers again.


Berkshire said...

Hi Jojo,

I just bought the essay of WB last week when there was a 20% off at Kino together with a few other titles which i think are pretty good. I have not started on this book yet. I'm currently on another book, Barbarians at the gate: the rise and fall of RJR Nabisco, it is a very intriguing read.

Ok, as for my way of determining intrisic value, i think you may have seen my one of my writings somewhere last week on it. And i will pose the exact words of what Warren remarked about the foundation of intrinsic value. I personally follow exactly the same steps as what he says except maybe the inputs of the variables are different because at the end of the day, you need to determine the variables which must sound logical and meaningful to you. For example, all future earnings estimated are dependent on your estimate of the business return on equity, I may put it as 20%, someone else may think otherwise. And then after determining all this earnings, you must then discount it back at an appropriate interest rate. I may use 5%, someone else may use at the end, my value, your value or Warren value are very different but the way to doing it is the same. I feel firstly, the fundamental of deriving is most important, then slowly, you can adjust the variables through what you read and your experience.

As for your estimation on SPH I.V, I reckon you use the base of 0.196 and compound it by 13% for the first 10 years and then 3% for the subsequent 20 years and when you add all those 30 years of earnings together, you then discount the total earnings by roughly about 5% to get $4.40?

If that is the method, it is roughly the same way as how i do.

Berkshire said...

Hi Jojo,

If possible, you may leave some comments on what you think of what is written on that latest Intrinsic Value write-up that I used on Home Depot. If there is anything I miss out or you may feel there can be something that is of value to adding in, pls drop me a note.

And yeah, you are apparently one of the most active reader and commentor of my writings and I appreciate that. Besides you, i think there is one or two more but sigh, it seems not many people appreciate the value of intelligent investing.


Anonymous said...

Hi Berkshire
Sure will go read your article and leave comments if any.

I'm not sure if I have misunderstood you... thought your method of determining intrinsic value was to use the NAV multiply by the ROE in accordance to the number of years of investment and taking the difference between that of the risk free bond returns ?


Berkshire said...

Hi Jojo, personally, i think basically, using FCF which means net cash from operating activities - capital expenditure + disposal of capital assets or using NAV, are the same.

In fact, let's say that the free cah flow is 0.196, where does the 0.196 in effect comes from. By definition, it comes from net profit per share for that particular year after accounting for all capital expenses. Ok, for sure, this 0.196 is less than what the actual earnings that can be churned out by the capital assets. But it is also important to include those into account, whereby those earnings that are used for capital expenses to be included in. Not sure you get what i mean.

So effectively, if i use NAV and ROE to derive at the intrinsic value. It is a combined figure of both the FCF and those capital expenses.

As to whether you want to take the difference in earnings in an alternative investment is really entirely up to the individual. For me, it makes more sense to do so.


Excellent info