How do you measure value?

[MrB]

Intrinsic value is the discounted future cash flows, so knowing the discount rate used is key.

Buffett puts a high premium on certainty and since the bulk of your future value is made up of the first 3-4 years, the initial yield is key.

So say you want to make 10% then you have to discount at, at least 10% and you don't need a calculator if you simply work along the following formula. Only invest when you get at least a 10% yield in year one and can establish with a high degree of certainty that the cash flows will be higher from year 2 going forward. 

 

That is the general concept and does not assume a margin of safety.

[twacowfca]

When Warren says that he doesn't understand a business, what he means is that he doesn't understand how that business is almost guaranteed to make good money for many years, even decades, that will be available for the shareholders or wisely reinvested  with high returns on capital  employed.

 

Indeed. Sometimes that's because it's an unpredictable or bad business/industry/management/etc, and sometimes he just doesn't have the specific technical knowledge to really understand the products and competitive dynamics deeply enough to be comfortable.

 

I think a lot of the time he does understand that the permanence of the edge a company may have is inherently unknowable.  Charlie used to invest in tech companies before he hooked up with Warren.  He said that as soon as one of those companies developed something superior, someone else would come up with a new technology that leapfrogged them.

 

Bill Gates and Paul Allen had no idea how important owning a superior disc operating system would become for their business.  They planned to set IBM up with another  company to provide it, but when there was a problem, they decided to take on that job under their ownership.

[AZ_Value]

Intrinsic value is the discounted future cash flows, so knowing the discount rate used is key.

Buffett puts a high premium on certainty and since the bulk of your future value is made up of the first 3-4 years, the initial yield is key.

So say you want to make 10% then you have to discount at, at least 10% and you don't need a calculator if you simply work along the following formula. Only invest when you get at least a 10% yield in year one and can establish with a high degree of certainty that the cash flows will be higher from year 2 going forward. 

 

That is the general concept and does not assume a margin of safety.

 

I'm not sure I agree with your last statement about no margin of safety being assumed in your IV calc.

 

When you say:

Only invest when you get at least a 10% yield in year one and can establish with a high degree of certainty that the cash flows will be higher from year 2 going forward.

the corollary to that statement is that if you don't have a high degree of certainty about the cash flows and/or the yield you will be getting, you shouldn't invest! So in my opinion the margin of safety principle is inherently embedded in your statement even without adjusting any number (like requiring a higher yield).

 

[MrB]

Intrinsic value is the discounted future cash flows, so knowing the discount rate used is key.

Buffett puts a high premium on certainty and since the bulk of your future value is made up of the first 3-4 years, the initial yield is key.

So say you want to make 10% then you have to discount at, at least 10% and you don't need a calculator if you simply work along the following formula. Only invest when you get at least a 10% yield in year one and can establish with a high degree of certainty that the cash flows will be higher from year 2 going forward. 

 

That is the general concept and does not assume a margin of safety.

 

I'm not sure I agree with your last statement about no margin of safety being assumed in your IV calc.

 

When you say:

Only invest when you get at least a 10% yield in year one and can establish with a high degree of certainty that the cash flows will be higher from year 2 going forward.

the corollary to that statement is that if you don't have a high degree of certainty about the cash flows and/or the yield you will be getting, you shouldn't invest! So in my opinion the margin of safety principle is inherently embedded in your statement even without adjusting any number (like requiring a higher yield).

 

You are correct, but to clarify I was merely pointing out that I did not address MOS specifically in the discussion. I was only talking about intrinsic value.

I assume you will agree that if using the above example again and if we assume cash flows yield 10% in year one and are exactly the same from year two going forward then I will have no margin of safety?

[Guest valueInv]

Intrinsic value is the discounted future cash flows, so knowing the discount rate used is key.

Buffett puts a high premium on certainty and since the bulk of your future value is made up of the first 3-4 years, the initial yield is key.

So say you want to make 10% then you have to discount at, at least 10% and you don't need a calculator if you simply work along the following formula. Only invest when you get at least a 10% yield in year one and can establish with a high degree of certainty that the cash flows will be higher from year 2 going forward. 

 

That is the general concept and does not assume a margin of safety.

 

Thanks, this is a great description.

 

I've also heard that he pays a lot of attention to ROE. How does ROE figure into all this? For eg, for KO, ROE is in the 25%-30% range but FCF growth is in the around 6%-7%and it has a high P/B. What makes KO attractive to Buffet? 

[petec]

MrB: any business fitting your description will be a good investment, I think, but I don't think it is how Buffett thinks.  Not exclusively, anyway.  He understands that the vast bulk of the value of long lived companies is in the out years, not the first 3-4, and occasionally pays a lot more than a 10% yield.  BNSF, for example, and Coke for nearly all the time he has held it (I'm not surer what he paid initially).  He also covers this concept in his assessment of managers who use stock to buy other companies and claim that they have achieved value because the deal is 'accretive' - Buffett points out that it may be accretive in year 1, but whether it is accretive in year 10 depends on the growth rates of the two companies.  So while your framework is a very good one, it will miss a lot of great investments.

[MrB]

Intrinsic value is the discounted future cash flows, so knowing the discount rate used is key.

Buffett puts a high premium on certainty and since the bulk of your future value is made up of the first 3-4 years, the initial yield is key.

So say you want to make 10% then you have to discount at, at least 10% and you don't need a calculator if you simply work along the following formula. Only invest when you get at least a 10% yield in year one and can establish with a high degree of certainty that the cash flows will be higher from year 2 going forward. 

 

That is the general concept and does not assume a margin of safety.

 

Thanks, this is a great description.

 

I've also heard that he pays a lot of attention to ROE. How does ROE figure into all this? For eg, for KO, ROE is in the 25%-30% range but FCF growth is in the around 6%-7%and it has a high P/B. What makes KO attractive to Buffet?

 

You are always simply trying to buy a discounted $1. Remember Buffett always talks about the cigar butt (horrible, but free puff) and then goes on to saying that it was a mistake. Firstly, it is the obvious and fairly easy thing to do, but creates an ongoing problem, which is you have to consistently look for the next butt as soon as you took the last free puff on the last one or re-investment risk.

 

What is more difficult but the smarter thing to do is to look for a $1 that is generally always being bought and sold at a premium. In the public markets the company with the good franchise will generally always be trading at a premium. So let's say that Coke will generally trade on a PE of 15 v the average drinks company that trades on a PE of 10. Investors are generally prepared to pay up, because of the wide moat around the Castle (the castle is the $1). However, if you were to lose that premium you are in serious trouble, you will lose 33% of your value from the average PE contraction.

So the basic point is that you can put value on a high quality franchise, a high quality $1 so to speak. Also, being invested in a high quality $1 reduces your re-investment risk, which you have in a cigar butt $1.

 

It then begs the question of how do you identify a high quality $1? In pure cash flow terms it simply means the company that can increase the value of the $1 the most in a given year, hence return on equity. That will be a result of some kind of moat, but you still need to tie it back to the fact that you returned X on the $1 you started.

 

Finally to tie it in with your example of KO. Please double check the numbers, but as I recall Buffett paid $5.22 for $0.36 in earnings in 1988, which is an initial yield of 6.9% or PE of 14.5 and in 1994 when he bought a few more he paid $21.95 for $0.99 or an initial yield of 4.51% or PE of 22. However, by 1994 the average PE for the previous 10 years was 21, so for his investment in 1988 he got a PE multiple expansion of 51%, but note that the PE really just came back to the average, he did not pay up for the premium that the market ascribed to KO on average over the 10 years. He basically got the premium at a discount. Now the PE is obviously a multiple of the E or earnings, which compounded by 18% from 1988 - 1994

 

Another way to think about it is that the investment results will be driven by premium expansion/contraction and underlying earnings. So if I buy a stock on a PE of 5 and it grows earnings by 10% and I sell it on a PE of 5 then my investment will return 10%. However, if the same happens but I sell on a PE of 10 then I will return 120%.

 

The last thing to do is now go back and listen to the video/discussion by Buffett mentioned earlier in the thread. However, don't stop listening at Aesop. The Aesop discussion is an important part of the equation. It talks about certainty of getting the two birds and when. In some ways it all relates to ROE, because it speaks to the heart of the matter. The quality of the franchise. The moat around the $1, provides certainty and if the $1 is well protected and ALSO happens to be growing every year then when is less important. With a high ROE company you can afford to wait whereas time with a cigar butt is death by a 1,000 cuts.

Whether a cigar butt or a franchise stock it all comes back to understanding the very basics of it all, which is that investment is about buying a company for less than its intrinsic value and that intrinsic value is $1 discounted at your chosen rate.

 

[MrB]

MrB: any business fitting your description will be a good investment, I think, but I don't think it is how Buffett thinks.  Not exclusively, anyway.  He understands that the vast bulk of the value of long lived companies is in the out years, not the first 3-4, and occasionally pays a lot more than a 10% yield.  BNSF, for example, and Coke for nearly all the time he has held it (I'm not surer what he paid initially).  He also covers this concept in his assessment of managers who use stock to buy other companies and claim that they have achieved value because the deal is 'accretive' - Buffett points out that it may be accretive in year 1, but whether it is accretive in year 10 depends on the growth rates of the two companies.  So while your framework is a very good one, it will miss a lot of great investments.

 

The basic concept is still that you are buying $1 at a discount and the fact that the market will ascribe a premium to the $1 or that Buffett is willing to incorporate that into his assessment of value should not confuse the issue. You are in very tough territory when you start assuming growth and what premium the market is willing to put on it 10 years down the road, because these are issues of certainty and when you get your birds as discussed by Buffett when he talks about Aesop. A "value" investor needs to crawl before he can walk and unless you really understand the very basic concept of what "buying a company for less than its intrinsic value" means, particularly if you are using a high discount rate then it will be impossible to appreciate the risk in paying a premium for a company with a good franchise. The basic frame work does not exclude paying up for growing companies with fantastic franchises it serves to highlight the risk of paying up for a heavily discounted cash flow more than 5 years down the road. Hence, it should make one stop and think really hard about why Buffett refuses to buy Microsoft when it is sitting on a EV/NI of 8 times, but was willing to buy Coke in 1994 when it was sporting a PE of 22!

 

[twacowfca]

Intrinsic value is the discounted future cash flows, so knowing the discount rate used is key.

Buffett puts a high premium on certainty and since the bulk of your future value is made up of the first 3-4 years, the initial yield is key.

So say you want to make 10% then you have to discount at, at least 10% and you don't need a calculator if you simply work along the following formula. Only invest when you get at least a 10% yield in year one and can establish with a high degree of certainty that the cash flows will be higher from year 2 going forward. 

 

That is the general concept and does not assume a margin of safety.

 

Thanks, this is a great description.

 

I've also heard that he pays a lot of attention to ROE. How does ROE figure into all this? For eg, for KO, ROE is in the 25%-30% range but FCF growth is in the around 6%-7%and it has a high P/B. What makes KO attractive to Buffet?

 

Well over 100 years of high returns on capital and the opportunity to reinvest some of their profits at high rates of return in developing countries.  Management that returns earnings that aren't used for profitable reinvestment to shareholders as dividends or share repurchases.  Perhaps the strongest brand name in the world.  Habit forming products.  The low cost distributor in that industry.  Not impacted much by technological change.  A business that isn't very cyclical and isn't dependent on financial markets.  In short, a better business than 999 out of 1000 other businesses.  Plus, Warren likes to drink Cherry Coke.

[Martian]

If I remember correctly, from poor charlie's almanack, Charlie  is not a big fan of DCF. he said that when you mix correct data with imaginary data (discount rate) you will still get imaginary data (same as mixing turd and raisin still gives you turd)

 

Charlie says that whenever we see EBITDA we need to replace it with Bullshit since that number is meaningless :-)

[biaggio]

Good succinct post Martian.

 

I have been guilty of using DCF too much.

 

Very helpful thread.

 

Thanks everyone for sharing some very important information.

 

[Guest valueInv]

Finally to tie it in with your example of KO. Please double check the numbers, but as I recall Buffett paid $5.22 for $0.36 in earnings in 1988, which is an initial yield of 6.9% or PE of 14.5 and in 1994 when he bought a few more he paid $21.95 for $0.99 or an initial yield of 4.51% or PE of 22.

 

My numbers are for KO today. I'm trying to answer whether Buffet would by KO today if he didn't own it.

[alwaysinvert]

MrB: any business fitting your description will be a good investment, I think, but I don't think it is how Buffett thinks.  Not exclusively, anyway.  He understands that the vast bulk of the value of long lived companies is in the out years, not the first 3-4, and occasionally pays a lot more than a 10% yield.  BNSF, for example, and Coke for nearly all the time he has held it (I'm not surer what he paid initially).  He also covers this concept in his assessment of managers who use stock to buy other companies and claim that they have achieved value because the deal is 'accretive' - Buffett points out that it may be accretive in year 1, but whether it is accretive in year 10 depends on the growth rates of the two companies.  So while your framework is a very good one, it will miss a lot of great investments.

 

The basic concept is still that you are buying $1 at a discount and the fact that the market will ascribe a premium to the $1 or that Buffett is willing to incorporate that into his assessment of value should not confuse the issue. You are in very tough territory when you start assuming growth and what premium the market is willing to put on it 10 years down the road, because these are issues of certainty and when you get your birds as discussed by Buffett when he talks about Aesop. A "value" investor needs to crawl before he can walk and unless you really understand the very basic concept of what "buying a company for less than its intrinsic value" means, particularly if you are using a high discount rate then it will be impossible to appreciate the risk in paying a premium for a company with a good franchise. The basic frame work does not exclude paying up for growing companies with fantastic franchises it serves to highlight the risk of paying up for a heavily discounted cash flow more than 5 years down the road. Hence, it should make one stop and think really hard about why Buffett refuses to buy Microsoft when it is sitting on a EV/NI of 8 times, but was willing to buy Coke in 1994 when it was sporting a PE of 22!

 

To be fair, Buffett has actually said outright that Microsoft is cheap but that it's off-limits because of his close relationship with Gates. I think he would have bought it last year if it weren't for that, but that's obviously just a guess.

[MrB]

Finally to tie it in with your example of KO. Please double check the numbers, but as I recall Buffett paid $5.22 for $0.36 in earnings in 1988, which is an initial yield of 6.9% or PE of 14.5 and in 1994 when he bought a few more he paid $21.95 for $0.99 or an initial yield of 4.51% or PE of 22.

 

My numbers are for KO today. I'm trying to answer whether Buffet would by KO today if he didn't own it.

 

Please correct me if I'm wrong, but you seem to think he will not buy more KO, simply because he already owns it? Why?

[Guest valueInv]

Finally to tie it in with your example of KO. Please double check the numbers, but as I recall Buffett paid $5.22 for $0.36 in earnings in 1988, which is an initial yield of 6.9% or PE of 14.5 and in 1994 when he bought a few more he paid $21.95 for $0.99 or an initial yield of 4.51% or PE of 22.

 

My numbers are for KO today. I'm trying to answer whether Buffet would by KO today if he didn't own it.

 

Please correct me if I'm wrong, but you seem to think he will not buy more KO, simply because he already owns it? Why?

I don't think that. It just simplifies the question for our discussion.

 

 

 

[Green King]

I thought the reason he would by by more is one because he owns a lot of it he would be over weighting and the other being he can get better compounding buying businesses out right since the money can go to brk daily for reallocation. Also of cheaper things are out there but I am not vey sure about it I have not look at all of them yet.

[ExpectedValue]

Good succinct post Martian.

 

I have been guilty of using DCF too much.

 

Very helpful thread.

 

Thanks everyone for sharing some very important information.

 

I think it's important to be really really careful about making a distinction here.

 

DCF is not bad, it's a sound way to think about valuing a company.

 

The bad thing is when people relax and massage the inputs that go into the DCF to make excuses for buying inferior or expensive companies. That's where you abuse discounted cash flow analysis.

 

You want to be really conservative and assume a real discount rate and be very conservative about your growth rate to come up with a valuation. And then obtain a margin of safety on top of that.

 

 

[Rabbitisrich]

Any valuation multiple is just one side of various DCF equations.

[mcliu]

DCFs are inherently sound if you can correctly estimate the cash flow and you have an appropriate discount rate.

 

I think the problem is using the CAPM as a basis for determining the discount rate.

 

 

[WarrenWatsa]

Regarding discount rate for DCF, almost everyone uses 8-10%. Some use CAPM to justify, others just say “that’s my required return and I won’t invest if it doesn’t meet that hurdle.

[petec]

 

I think it's important to be really really careful about making a distinction here.

 

DCF is not bad, it's a sound way to think about valuing a company.

 

The bad thing is when people relax and massage the inputs that go into the DCF to make excuses for buying inferior or expensive companies. That's where you abuse discounted cash flow analysis.

 

 

 

+1

 

Actually, + about 10!

[anders]

I dont think conventional metrics such as p/e, p/b give you that much in terms of valuation. These nrs only tell you what has been put into the company, while value derives from what can be taken out.

 

The law of valuation of any asset, whether its farmland, oil in the ground, a new technology, real estate, is and will always be; the future cash produced from present until kingdom come, discounted at an appropriate discount rate.

 

Ive stopped using DCF since I believe the model is flawed. And last, you can never adjust the discount rate to protect against risk. The risk lies in the probability of losing purchasing power over the contemplated holding period.

 

My two cents,