Valuation Techniques Help Please

[Buckeye]

Hello All-

 

Would anyone care to recommend a book/article/publication that would help an investor learn and practice valuation techniques?  Also, what are the boards most favored techniques?  I know that the technique may depend on the type of security, but it would be much appreciated if members would list their favorite techniques, the strengths and weaknesses of those techniques and the most applicable circumstance for that type of technique.

 

I realized that I spend most of my time (more like all of time) reading ideas and I need to start practicing valuations. 

 

Thank you for any help.  I greatly appreciate your feedback and value your opinions.

 

Buckeye

[JBird]

On valuation: http://www.berkshirehathaway.com/letters/2000.html

 

On owner earnings: http://www.berkshirehathaway.com/letters/1986.html

[LC]

My process is generally this:

Read annual reports

Try and form a rough understanding on how the business works and where it is going

Figure out the last 5-10 years of owners earnings, and what the company has done with those earnings

 

Then I actually write up a conversation between two hypothetical parties, a buyer and seller, for the entire company.

 

I'll start with something like, "I'll offer you 10 dollars for the business" to which the seller will respond, "no way, we earned 30 mil last year! How about you buy it from me for 1 billion?"

 

And then from there I just go back and forth zoning in on a reasonable range of what that transaction would look like. And that is my estimate of "intrinsic value" so to speak. At each price point, I write out what the opposite party would say. Something long the lines of, "okay well 150m is a bit of a lowball bid...we made 30m but we're in a growing market, have 20m of cash, another 50m of ppe and working capital, and no debt."

 

Then I just compare to market prices, if there's a fair enough margin of safety I make a decision from there.

[Buckeye]

Jbird-

 

Thanks! I am reading Max Olsens Compilation of Berkshire Letter and am on year 1982.  I will make a note to read those two years extra carefully.

 

LC-

 

When you say owners earnings are you referring to net cash flows?  Do you extrapolate that out into the future and the discount it back to present values?

 

 

 

[JBird]

The '86 letter will answer your question about owner earnings.

 

The '00 letter will tell you that future owner earnings discounted to the present value is the definition of intrinsic value.

[ExpectedValue]

Hello All-

 

Would anyone care to recommend a book/article/publication that would help an investor learn and practice valuation techniques?  Also, what are the boards most favored techniques?  I know that the technique may depend on the type of security, but it would be much appreciated if members would list their favorite techniques, the strengths and weaknesses of those techniques and the most applicable circumstance for that type of technique.

 

I realized that I spend most of my time (more like all of time) reading ideas and I need to start practicing valuations. 

 

Thank you for any help.  I greatly appreciate your feedback and value your opinions.

 

Buckeye

 

To me it comes down to figuring out what a company is going to earn, often there might be accounting nuances here because you'll want to strip out non-recurring expenses and other one time items to figure out the earnings power of the firm. Then it's a matter of making a call on growth and using that to discount your cash flows over time to arrive at a value for the company and see how it compares to the price. Once you've done this/understand this you can revert to something more shorthand like using a simple free cash flow multiple. These things can get tricky when you start thinking about growth rates, so it's best to be conservative and look for things with large margins of safety.

 

Understanding the business is critical. If you underestimate things like the industry or competitive dynamics you'll almost always misestimate your normalized earnings power and growth rates, without those you're going to be in a tough spot when it comes to valuation.

 

I think with things like comps or asset based valuation methods you really have to be careful. Comps to me are tricky because an industry's valuation might just be inflated because of some temporary phenomenon. When things revert, those comp valuations would compress. Asset-based valuations are tricky to me because you usually use those when things are going really really poorly.

 

The way I always think about it though is that if the company cannot achieve an adequate return on capital with their asset, why should you think anyone else can? It's dependent though on what the asset is. Specialized equipment is different than a securities book. Book value is a poor means of measuring asset value. You also have to understand that management's incentives are not aligned with your own, so even if liquidation value is there, management might do whatever they can to keep their jobs and take their salaries.

[Buckeye]

This is the article that got me started on this topic:

 

http://www.gurufocus.com/news/225351/how-to-practice-valuation

 

He uses a similar company and its multiple of sales and its multiple of EBIT as a comparison and then averages that companies multiples and the one he is researching to find the applicable multiple.

 

So, after reading this I noticed the same thing that ExpectedValue pointed out that you might just end up comparing one inflated value to another. 

 

He also seems down on the the DCF model.  If I were to use a DCF what is the discount rate I should be using?  Is this a dumb question or is this the 1,000,000 dollar question?

 

Thanks for your replies!

[JBird]

It's a great question to ask. The discount rate is your required return rate. Since you always have the option of investing at the risk-free rate, you're required return rate must at least match the risk-free rate. Above that rate, it's your call. If the risk-free rate is extremely low, you may find it appropriate to use a much higher rate.

[LC]

Jbird-

 

Thanks! I am reading Max Olsens Compilation of Berkshire Letter and am on year 1982.  I will make a note to read those two years extra carefully.

 

LC-

 

When you say owners earnings are you referring to net cash flows?  Do you extrapolate that out into the future and the discount it back to present values?

 

Owners earnings I define the way Buffett does, net income + depreciation/amortization - capex + certain non-cash charges. I don't make an actual calculation in terms of assigning a Growth rate and then picking a discount rate to do a DCF. What I do is figure out the past history of owners earnings, look at that in terms of how much capital is employed to achieve those earnings, and (here's the rub), hope to know enough about the industry to know how certain those owners earnings will be in the future. Whether they will grow (tailwinds or superior moat) or shrink (headwinds, poor operations/management, poor economics vs competitors). And I try to keep I very general. I'm not looking to get so specific. That is, I'd rather be rougly correct than precisely wrong.

 

I think if you know enough about a business and the environment in which it operates, and you have a history of how well the business has done in the last ten years, you can come up with a rough range under which that business would sell on a fair basis.

 

Coke makes about 7-10 billion/ year in owner earnings per anum. It's worth somewhere in the 150 to 225 billion range, at my guess. I dont know whether its 159b or 203b. But in a sale, today, it would fetch somewhere in that range at my best educated guess. And At a price below that range you have the opportunity for exceptional returns, that's all I'm trying to do.

 

Why 150-225b as my valuation? Well, compare the certainty of earnings to a government bond. At 150 billion, that 7b of earnings represents a 4.6pct return. At a 225b valuation, that 7b of earnings is a 3pct return. That equates to slightly higher than a t-bill, which is due to the certainty as a business which those earnings have.

[Icarus]

I'm currently reading The Investment Checklist by Michael Shearn and so far it has some good ideas.

[vinod1]

I think "Value Investing: From Graham to Buffett and beyond" is THE book for you. To me more than any other book, this helped me in how to think about valuation.

 

Vinod

[Packer16]

I have a tendency to use multiples as I think there are too many assumptions in classical DCF models to very helpful in finding undervalued securities.  In addition multiples will get you 90% of the way there and the additional 10% takes alot of time for limited additional benefit.  You can also focus on the issues that make a difference that cannot be modeled - investor psychology, security concentration, industry structure, moats/competitive advantage, other market (bond, option) signals, mgmts track record and incentives.  I also do not do this full time so I try to maximize effort vs. time.  See my AKA thread for valuation examples using comps.  I do valuation as a business so I am familiar with the techniques and the uncertainty in the assumptions.

 

Packer

[Buckeye]

Wow!  Thank you all for taking the time to reply!

 

JBird- I will be sure to read and reread those years annual reports.  I understand the theory of the discount rate it has always just seemed to me as producing a wide range of outcomes. 

 

On a side note, I wonder if there is a certain years letter that is Buffett's favorite.  We have all heard him repeat over and over that chapters 8 and 12 (I believe) are the two most important chapters in the intelligent investor, I wish some one would ask him which is his favorite "chapter" of the Berkshire letters.

 

LC - Great tip on not trying to focus so much on the exacts of future cash flows, but concentrate more on past and current exacts and be general about future prospects.

 

LanceSanity- I will read your article tonight.  Thanks!

 

Icarus - I just added that book to my wish list after reading a review from Jae at Old School Value.

 

Vinod - You just reminded me that I have the book you mentioned.  After digging it out of a pile of books, it looks like I read the first chapter or two and some how it got put down and forgot about. 

 

One of my issues is starting one book reading part of it, getting excited about and starting another book and then the next thing I know I have about 6 or 7 books I am "trying" to read.  Does anyone else do this?  Or is it just my attention deficit disorder kicking in?

 

Packer - I guess I didn't realize that the multiples are as valuable as the are.  I always assumed that was too simplistic.  But now that I think about it, after reading through many of the old Graham writings he used these multiples quite often.  I remember and think about the most common multiples and gloss over some of these that may be more important.

 

As part of this learning experience I am going to try to go back through some of my readings and make notes of the mentioned types of multiples.  After some of my further research I will list what I found and board members can weigh in on which ones they like or use the most.

 

Thanks again all!!

 

 

 

 

 

[rjstc]

I have a tendency to use multiples as I think there are too many assumptions in classical DCF models to very helpful in finding undervalued securities.  In addition multiples will get you 90% of the way there and the additional 10% takes alot of time for limited additional benefit.  You can also focus on the issues that make a difference that cannot be modeled - investor psychology, security concentration, industry structure, moats/competitive advantage, other market (bond, option) signals, mgmts track record and incentives.  I also do not do this full time so I try to maximize effort vs. time.  See my AKA thread for valuation examples using comps.  I do valuation as a business so I am familiar with the techniques and the uncertainty in the assumptions.

 

Packer

 

Sorry. AKA thread. What or where is it? Thanks

[augustabound]

Sorry. AKA thread. What or where is it? Thanks

 

I'm guessing it's the "Ask Packer" thread

 

[augustabound]

Here's an example of valuing businesses: http://webreprints.djreprints.com/3026020366910.pdf

 

Understanding the business is key. I'm not sure how to choose a discount rate. I only feel comfortable estimating intrinsic value when there are comps to look at. Usually if something is trading at low multiples, I don't think about a target price.

 

When I read about EBITDA I can't help but think of Charlie Munger calling it bullshit earnings, yet so many people seem to use it. 

Most of Geoff Gannon's articles cite EBITDA. But I like his thought process in researching the actual company.

 

If I understood the article, this guy uses an industry comp as a basis for EBITDA and cash flow.

 

Different strokes for different folks I guess. 

It's times like this that I see why people use Buffett's owner earnings number.

 

 

[Packer16]

I find EBITDA helpful when used with other cash flow measures like price to FCF/owners earnings.  It is also useful in situations where the firm is going through a one-time large capital expenditure period where FCF is negative and as a comparitive measure.  Using it as the only measure is not right but in combination with other measures it can be useful.  It is also helpful for international comparisons where depreciation rules are different and for calculating debt coverage and leverage ratios as this is what banks use for their covenants.

 

In the case of real estate, FFO (a form of EBITDA) is commonly used to value property.  If you use earnings you will be understating free cash flow because a good portion of the depreciation is not real.

 

Packer

 

[Uccmal]

I have a tendency to use multiples as I think there are too many assumptions in classical DCF models to very helpful in finding undervalued securities.  In addition multiples will get you 90% of the way there and the additional 10% takes alot of time for limited additional benefit.  You can also focus on the issues that make a difference that cannot be modeled - investor psychology, security concentration, industry structure, moats/competitive advantage, other market (bond, option) signals, mgmts track record and incentives.  I also do not do this full time so I try to maximize effort vs. time.  See my AKA thread for valuation examples using comps.  I do valuation as a business so I am familiar with the techniques and the uncertainty in the assumptions.

 

Packer

 

I concur.  I have never used DCF of any kind - too much BS.  I will estimate future net income using back of the envelope usually for when things normalize.  i.e. From 2.5 years ago: BAC should earn a minimum of 20 Billion/1 B shares = $2 x pe12=24. 

 

There are two many variables when looking at a company to make even remotely useful etimates of the future.  I prefer to deal with where we are now, and then focus on the intangibles such as Packer suggests: the business, sustainability of the business model, competition, moats etc.

 

LVLT is a perfect example of something I couldn't being myself to buy because the business model didn't account for wireless which was growing on an annualized basis at a rate that was making landlines obsolete. 

 

BAC, AIG, JPM, WFS were much easier to see.  They were insanely cheap in turn.  Their business has at least weak moats, probably strong moats if you consider that they have been deemed TBTF. 

 

Q. I ask:

1) Will the business stay in business - more relevant now than ever.

2) Is it cheap P/B, P/e

3) Can the business return to health.  goes back to 1)

4) Can I visualize easily how they make money. 

Other basic Buffett questions.

 

This value investing is simple but not easy.  Anything that gives you a false of actually knowing something such as DCF is dangerous.  My mental models basically eliminate whole industries such as mining, E&p for oil and gas, small retail  based on fickle customers (lululemon), and badly run retail (SHLD).  Anything that requires me to predict the future is dangerous imho. 

 

 

[Guest deepValue]

Figure out what the businesses' normal owner earnings are in its present state, then figure out what your return would be if you bought it at the current price (buy $1 earnings for $10 --> 10% annual return). Then just figure out if normal earnings are more likely to increase or decrease in the future, and decide if the market is offering a satisfactory return based on your broad view of the future (usually linked to the company's competitive position).

 

Coming up with a discount rate does not have to be anything scientific. You'd want to buy Coke at a 9% yield but would pass on an E&P company w/three wells in the Gulf of Mexico at a 12% yield. Just apply common sense.

[SharperDingaan]

Keep in mind that P/B, & P/E is secondary to your investment knowledge - & your industry circle of competence.

 

Re investment knowledge: B is the last reported BV - but if you are expecting a significant gain this quarter; the BV is really lower than it should be & the current P/B is higher than it should be. Passing simply because the P/B (or P/E) ratio was too high, is to miss the opportunity entirely. And if you consistently seem to miss the upswings ... this may well be the reason why.

 

Re bank industry competence: If you do not fully understand the dynamics of regulatory capital - & a banks ability to lend, you are materially handicapped. There is zero spread income if the bank cannot lend, & that ability is driven by BIS - & the central banker/global regulator/political bargaining behind it. Most would argue that current BV, or future 1 year E , is not a useful metric when the underlying  regulation is undergoing such massive change. And if you were not fully aware of that (industry competence), you were really just placing a bet on either red/black.

 

Back of the envelope works great - but it works because it forces you to think.

 

SD

 

 

 

 

[LC]

 

LC - Great tip on not trying to focus so much on the exacts of future cash flows, but concentrate more on past and current exacts and be general about future prospects.

 

 

I'd recommend reading this speech: http://turnkeyanalyst.com/wp-content/uploads/2013/02/Williams-Trying_too_Hard.pdf

 

Some of my key takeaways (and a humorous line):

 

-Confidence in a forecast (or valuation) comes from the amount of information we put into it. However the accuracy of your forecast/valuation stays the same.

 

-Respect the virtues of a simple investment thesis. Successful investments must have a simple thesis, by nature.

 

-Complication is evidence of a poor thesis

 

 

Now for the humorous line:

"An investor without a forecast is like a fish without a bicycle"

[twacowfca]

Figure out what the businesses' normal owner earnings are in its present state, then figure out what your return would be if you bought it at the current price (buy $1 earnings for $10 --> 10% annual return). Then just figure out if normal earnings are more likely to increase or decrease in the future, and decide if the market is offering a satisfactory return based on your broad view of the future (usually linked to the company's competitive position).

 

Coming up with a discount rate does not have to be anything scientific. You'd want to buy Coke at a 9% yield but would pass on an E&P company w/three wells in the Gulf of Mexico at a 12% yield. Just apply common sense.

 

Very good.  :)

[oddballstocks]

Keep in mind that P/B, & P/E is secondary to your investment knowledge - & your industry circle of competence.

 

Re investment knowledge: B is the last reported BV - but if you are expecting a significant gain this quarter; the BV is really lower than it should be & the current P/B is higher than it should be. Passing simply because the P/B (or P/E) ratio was too high, is to miss the opportunity entirely. And if you consistently seem to miss the upswings ... this may well be the reason why.

 

Re bank industry competence: If you do not fully understand the dynamics of regulatory capital - & a banks ability to lend, you are materially handicapped. There is zero spread income if the bank cannot lend, & that ability is driven by BIS - & the central banker/global regulator/political bargaining behind it. Most would argue that current BV, or future 1 year E , is not a useful metric when the underlying  regulation is undergoing such massive change. And if you were not fully aware of that (industry competence), you were really just placing a bet on either red/black.

 

Back of the envelope works great - but it works because it forces you to think.

 

SD

 

Great comment!

 

I use multiples, but they are best used in context.  Often something might appear to have a high P/E, or low book value, but upon a deeper look it becomes obvious that the multiple is wrong because of some obscured or hidden variable.  What's best is when you realize this variable, and management is also very aware of the variable, but the rest of the market isn't.

 

Banks are interesting right now, there are a lot of dynamics at work in the industry.  Yet some of these can be taken advantage of, for example there is a sweet spot of banks that are very cheap, and given the regulatory change will either be forced out of business, or will be forced into an acquisition.  It can be very profitable to buy these potentially extinct ones cheap and wait for them to be acquired, but Sharper's point is very relevant, one needs to understand the dynamics and what's at play.

[Uccmal]

Keep in mind that P/B, & P/E is secondary to your investment knowledge - & your industry circle of competence.

 

Re investment knowledge: B is the last reported BV - but if you are expecting a significant gain this quarter; the BV is really lower than it should be & the current P/B is higher than it should be. Passing simply because the P/B (or P/E) ratio was too high, is to miss the opportunity entirely. And if you consistently seem to miss the upswings ... this may well be the reason why.

 

Re bank industry competence: If you do not fully understand the dynamics of regulatory capital - & a banks ability to lend, you are materially handicapped. There is zero spread income if the bank cannot lend, & that ability is driven by BIS - & the central banker/global regulator/political bargaining behind it. Most would argue that current BV, or future 1 year E , is not a useful metric when the underlying  regulation is undergoing such massive change. And if you were not fully aware of that (industry competence), you were really just placing a bet on either red/black.

 

Back of the envelope works great - but it works because it forces you to think.

 

SD

 

SD, I am not sure I understand what you are getting at.  I used The big banks as an example.  In December 2011 BAC was around $5.00-6.00 per share.  JPM was at 30+-.  Based on a few knowings, if you will, you could get at what normalized earnings might look like.

 

These knowings:

- banking is a sticky business. 

- It is universal - Nearly every transaction an American makes will pass through the hands of the big 4 or 5 somewhere along the way.

- regulators need the industry it to be profitable or banks from Dubai, or elsewhere will take over.  No regulator wants to be on the hook for BAC or JPM failing, or falling into the hands of Prince Alwalaheed. 

- government needs banks to lend and will ensure they are profitable - most countries do this. 

- banks need investors to help shore up their regulatory capital.  Investors will only invest in enterprises that make money for them

- gov't will ensure that banks can make a spread by lowering overnight lending rates as low as zero, or less, as we have seen in the present climate.

- Once you cleared away the garbage, BAC, JPM, and all the others had very viable franchises remaining that are protected, and moated.

 

I contend that the regulatory environment is and always will be secondary to the requirement of banks to be profitable.  Sure the government sanctions them periodically but that is purely to satisfy the publics need for blood. 

 

You can go back through the various BAC threads and my theme has remained the same for over 3 years. 

 

Sd, As a Canadian you certainly know the degree to which the Cdn. Banks are protected by government. 

 

Note: All of these things above require an understanding of how banks make money and the interplay between banking and government.  Not one of them require me to do a DCF or other phony estimate of future earnings. 

 

I read ARs and Qs but the most valuable, valuation technique, aside from the very basics, I make use of is 'kicking the tires'.  My best investments have been those I have held for awhile and have gotten to know in ways other than purely financial:  Seaspan, Fairfax (last decade), Northbridge and ORH, , Mullen Group, Russell Metals, Bell Canada (after the merger collapse), BAC, JPM, Sbux, Axp, HD, Cfx, WFC, and now AIG, and Extendicare. 

 

Not one of the best was a Graham style valuation only investment.  Many of my worst, conversely, I have had the most information on such as sfk.un/fbk . 

[WarrenWatsa]

Does anyone believe that Buffett (arguably the greatest investor of all time) does more than back-of-the-envelope calculations when making investment decisions?

 

He barely knows how to use Excel and rarely touches a computer, so, think long and hard about that.

 

The complexity of the calculations he uses when making investments are limited.

 

Either he veers away from overly complicated businesses that he cannot value with his relatively simple methods or he figures out what kind of back-of-the-envelope calculation makes plenty of sense for the business he's considering.

 

If it takes a Masters or PhD caliber of person / knowledgee to arrive at your valuation, the market will definitely never look at it like you do. Most market participants can't think on those levels and will never be comfortable with your thought process, therefore.

 

I believe Buffett's purchase of Bank of America, despite the company's complexity, was based significantly on a back-of-the-envelope calculation that made a lot of sense...a la Bruce Berkowitz.

 

And, I believe Buffett is a heavy user of multiples (derived based on his experience)...Alice Schroeder has said that all Buffett wants to do is make a 10% pretax return on his investment with little risk, for example. In his younger days (when he had less capital to manage), it was probably 15-20% instead of 10%.