Tuesday, March 5, 2013

Buffett Letter 2012

OK, so there are a lot of comments about the Berkshire Hathaway letter to shareholders all over the place and I really don't have too much to add, but I thought I'd post some comments for the people I tell to read these every year (and I know they don't).

So consider this just a quick takeaway sort of thing that may not be of much value to the folks who wisely took the time to read it.  And by no means is this a summary.  I'm only writing about certain things that came to mind.  I don't mention most subjects that was in the letter.

Anyway, here goes.

Good return!
First of all, Berkshire Hathaway (BRK) grew book value per share (BPS) +14.4% in 2012 compared to +16.0% for the S&P 500 index.  Buffett called this return subpar, but that's only because it underperformed the S&P 500 index (total return which includes dividends).  Aside from that, +14.4% is a very good, decent return.  BRK tends to not do well on a relative basis in strong market years, but does much better in not-so-good market years.  So we shouldn't be too worried about BRK underperforming in big up years.  In fact, Buffett says that  BRK only underperformed in 9 out of 48 years, but that in 8 of those 9 years the market was up 15% or more.

Buffett says, "We do better when the wind is in our face".

Just one year doesn't really tell you much, so let's take a quick look at the longer term.   

                                     BRK growth in book           S&P 500 total return
Since 1965                    +19.7%                                +9.4%
Last 10 years                +10.6%                                 +7.1%
Last 5 years                  +7.9%                                   +1.7%

BRK has grown an astonishing +19.7%/year since 1965 versus +9.4% for the S&P 500 index.

Please note that despite all that has happened since 1965 (Vietnam war, dollar devaluation/70's inflation, disco, black monday, internet bubble, financial crisis, gungnam style etc...) the market has returned +9.4%/year.  Can you imagine telling Buffett in 1965 all of the horrible things that will happen in the next 47 years so he should consider another line of work?

The five and ten year rates are also favorable compared to the S&P 500 index.
Interestingly, he said:
Charlie and I believe the gain in Berkshire's intrinsic value will over time likely surpass the S&P returns by a small margin.  (emphasis mine)
I've suspected that for a while now so that doesn't surprise me, but I don't remember him actually saying that this way before.  Of course, he has said many times in the past that BRK's growth will be nowhere near what it has been in the past.   Anyway, that is not really a big concern for conservative investors who own BRK.  It's a solid stock to own.

Don't Worry So Much
I know this is a recurring theme on this blog and it's beating a dead horse.  But other than "what do you think of Apple/Facebook (or whatever stock is hot or cold on that day) the most common question I tend to get when people realize I deal with the stock market is, "What's going to happen with the fiscal cliff?" or "I got out of the market because I'm worried about xxxxx".  Some say, "you better get out of the market because I heard some guy on TV say that the Dow can get down to 6000!".

So here is Buffett again on the topic:

This is in sharp contrast to many shareholder letters these days that spend so much time talking about the many problems that we have.  This is not to make light of the many real problems that we face.  It's also not prudent to be overexposed, leveraged or anything like that where if another crisis happened you might blow up.

Granted, the reason why mutual fund and hedge fund managers have a different mindset than Buffett on this issue is partly due to the flakey nature of their capital.  Buffett manages real permanent capital.  Mutual fund and hedge fund managers manage hot money.   If they act like Buffett, they will get killed when the market tanks and they have to face redemptions.  Buffett never has to worry about investor redemptions!  This is a key difference.  This is also why individual investors should act more like Buffett than the ever-so-worried mutual fund and hedge fund managers.

New Investment Managers
As we know, Buffett hired two portfolio managers to invest BRK's capital.  We don't have a whole lot of detail on their performance since they only just started, but Buffett says that both of them have outperformed the S&P 500 index by double-digit margins.  That means each of them returned more than 26% (the S&P 500 index returned 16%) in 2012.  Look through your year-end mutual fund statements; how many of your funds have a 26% return?

Of course, it's too soon to say whether these managers are any good (well, you can say they are good because they had good track records before joining BRK) but it's not a bad start!

Wells Fargo (WFC)
As of year-end, Wells Fargo is the largest stockholding at BRK.  This may come as a shock to many who don't like banks, but there it is.  WFC is a larger position than Coke, and the interesting thing is that it is big not because it has gained the most (like Coke) but because Buffett has been buying a lot of it lately.  Only IBM is larger on a cost basis.  On CNBC the other day, Buffett said he has bought WFC this year too and said it's cheaper than Coke.

I looked at WFC in this Wells Fargo is cheap post; the fact that he has still been buying supports the view that it is still a cheap stock.  (Buffett also mentioned in the letter that WFC's earnings is understated by $1.5 billion due to an amortization charge that doesn't make sense).

At the end of the letter to shareholders, Buffett explains why a dividend doesn't make sense right now for BRK, and how people can create their own dividend by selling a little bit of BRK every year.  It's a great explanation of why it would be better to do this than have BRK actually pay a cash dividend.   If you don't understand how selling 4% or so of your holdings every year is actually better than BRK paying a 4% dividend, read this section of the letter.  It is very clear.

Buffett often recommends books, especially the Graham books, but I don't remember him recommending books in his letter to shareholders.  You would be a fool not to read books he recommends (if you have any interest in Buffett and his ways, that is).

Here are the books he recommends.  I haven't even read them yet, but I put them in the store anyway.

Of course, he has to push the new book out by Carol Loomis.

Tap Dancing to Work by Carol Loomis
The Outsiders by William Thorndike Jr, a book about CEOs that were good at capital allocation and includes a chapter on Tom Murphy.
The Clash of the Cultures by Jack Bogle about investment versus speculation
Investing Between the Lines by Laura Rittenhouse

The following section gets into some valuation issues for BRK.  It's a hint to friends and relatives who are not particularly interested in financial stuff to skip the rest. 

BRK Valuation
One big parlor game on the internet is to figure out what BRK is actually worth.  You can find all sorts of models and estimates all over the place.  Buffett, in recent years, has gone out of his way to tell us that book value per share way understates the intrinsic value of BRK.  In this year's letter, he again says, "we use book value as a significantly understated proxy " of intrinsic value (emphasis is Buffett's!).

So as in last year's letter, he is telling us that BRK is worth way more than stated book value, which by the way comes to around $114,214/share (from the annual report).

I will probably make some posts about BRK's valuation in the future, but for now I just wanted to make a quick comment on the two-column method BRK valuation that Buffett seems to endorse.

Let's take a quick look at this.

Buffett's idea is to look at BRK as two separate businesses; the operating business and the insurance business.  The insurance business is where the stock and other investments are held (roughly speaking) and the operating businesses are the wholly owned (or consolidated) subsidiaries that include things like See's Candies, Nebraska Furniture Mart, Dairy Queen etc...

So what Buffett does is he looks at a single metric for each part:  For the operating business he looks at the pretax earnings per share of businesses other than insurance and investments.  For the insurance business, he looks at the per share total investments held by BRK.

The per share pre-tax profit of the operating businesses was $8,085.  Since Buffett says he likes to pay 9-10x pretax profits, I will value the operating businesses at $81,000/share (around 10x pretax profit).

Investments per share comes to $113,786/share.  So according to this method of valuation, BRK is worth $81,000 + $113,786 = $194,785/share.  By the way, this comes to 1.7x book value.  According to this methodology, BRK is indeed worth far more than stated book value.

This is a good, simple model and it's hard to argue against.  There is no question that the investments per share are worth what it is on the books for.  $1.00 of cash is worth $1.00 of cash.  $1.00 of bonds is worth $1.00 of bonds.  $1.00 of stocks is worth $1.00 of stocks.  So what's the problem?

Nothing, really.

Another Way to Look at It
I don't want to argue what is the right way or wrong way to look at these things.  Valuation is a tricky business and people will have different ways of looking at this and that's totally fine.

My own personal reservation with this valuation is that to me, investments per share may not be worth investments per share.

Why?  Let's just say, for example, that all of the investments were in bonds yielding 1% and that there is no real prospect of this changing in the near or far future.  There is no plan to distribute this or to sell the bonds and invest the proceeds in higher return assets.

In this case, if my required rate of return is 10%, then the investments per share is only worth 10% of face value to me.  A business is only worth the future cash flows it generates, right?  So if the future cash flows are low, then it's not worth much.

Some will argue that bond funds don't trade at such steep discounts.  I think that is different.  Money market funds and bond funds are valued differently; people don't expect 10% returns from a money market fund; just safety and stability of principle and some interest.  This is the same with bonds. A bond is a bond, not a business.

If all of the BRK investments per share was invested in stocks managed by Buffett, then as an equity investor, it would obviously be worth investments per share dollar for dollar.   It may even be worth more if we expect Buffett to outpeform the stock market.

If half of the assets is invested in low yielding bonds and cash, then to me it would be worth quite a bit less.

So what is the investments per share worth to me?  I will look at this in detail in a later post but for now I'll keep it really simple. 

First of all, I need to know what the expected return on the investments per share is.  If the expected return is 10%, then of course the investments per share would be worth the whole amount as that is the same as my required rate of return (or discount rate).  If the expected return is 5%, then the investments per share would be worth 50% of the amount.

Expected Return on Investments
I noticed that the expected long term rate of return on BRK's pension plan assets is 6.6%.  I also see that the asset allocation for the plan assets is roughly 50% equities and 50% fixed income and cash.  I actually don't know how the outside managed assets are invested but I assume it is not listed equities as most of what is in there is not Level 1 asset (assets with quoted prices); so I assume they are some sort of corporate and other fixed income investments.

Interestingly, BRK's investments per share is also close to 50% equities and 50% fixed income and cash.

Buffett has said many times that pension funds that have expected returns of 8-9% is pure fantasy; they will never achieve that sort of return in this environment.

Note, also, that this is the expected return on plan assets for the long term.

So you see where this is going.  With investments per share of $113,786, the return on this amount may be 6.6%.   If that is the case, then to me, the investments per share is only worth 66% of that or around $75,000/share.   (If float can grow, then of course the increase in investments per share would be higher than 6.6%, but Buffett has been hinting that float may not grow all that much in the future and may even decline (but no more than 2%/year)).

Add this number to the operating businesses value of $81,000/share and I get a total value of $156,000/share.   BRK closed today at $154,425/share so it is trading right around fair value using these assumptions.

The portion allocated to equities in the investments per share may rise, but I have established in my previous post So What is BRK Really Worth?  that the fixed income and cash portions of BRK's total investments typically don't go below the float amount (which is currently $73 billion).   This makes sense; float is an insurance liability so it is held in liquid securities.

It's fair to argue that BRK currently has way more cash than usual.  There seems to be $47 billion cash and Buffett said he would like to keep it at no less than $20 billion.  So maybe we should give full credit to the $27 billion excess cash because that can actually be deployed into high return equities or acquisitions.  So we can take that excess cash out and value it at 100% and then keep the 34% discount for the rest of the investments.  This would bump up the value of the investments to $80,000/share, so the total value of BRK would come to $161,000/share.

Either way, from this point of view, it looks like BRK to me is worth around $156,000 - $161,000/share. 

Anyway, I do have more thoughts on valuation but I'll post those later. 

Oh, and I do own BRK in my long term account and plan to keep it there a long time, but have sold out most of it in my more active account (it was a pretty big, levered (LEAPed) position so...).


  1. This comment has been removed by a blog administrator.

  2. Great commentary. I agree fully that the float is not as valuable as equity capital, because Berkshire must keep much of the float invested in low returning, safe bonds or cash. This seems to be glossy over in most analysis.

    But, I have a question that might make a good follow on post. What do you think the intrinsic return of Berkshire going forward will be at its current valuation? (Said another way, if you held Berkshire for 50 years, what % return would you expect to have earned at the end of those 50 years?)

    Thanks again for the great blog.

    Tom L.

    1. Hi,

      That's sort of an easier question than what BRK is actually worth right now. Over time, a stock will reflect the growth in intrinsic value and Buffett said that BRK will grow IV at a pace beating the S&P 500 index by a small margin. So that's what BRK holders will earn over time.

      But what about current stock price versus IV? BRK is now around 1.3x book, let's say. If fair value is 2.0x book, then over 50 years, if BRK trades at 2x book by then, it will add 0.9%/year to performance. If IV is 1.7x book, then the price getting up there over 50 years would add 0.5%/year to total return. If IV is just 1x book, then it will take away 0.5%/year from returns.

      So the longer you hold something, the less price matters. (This may be a hint why Buffett has been willing to pay up for good businesses over the years, like KO and the more recent acquisitions. If you hold it for so long, the growth in business will far exceed the gain from a price-to-IV adjustment).

      So ironically, even though I can't tell you what BRK is worth today, it's safe to say that BRK holders will beat the S&P by a small margin over time.

      Thanks for reading and commenting.

    2. Hi kk,

      Thanks for the response. I fully understand your math - in fact, it helps to illustrate the issue that I was trying (unsuccessfully) to point out. Let me try again by asking a question that looks backwards instead of forward.

      What was the intrinsic value of a share Berkshire Hathaway 50 years ago in 1963? (No need for a specific number, a range will do.)

      In theory this should be an easy question to answer today, because we know how the 50 years since 1963 have played out. However, I personally don't know how to answer the question. I could say that BRK's intrinsic value is roughly $150,000/share today and then discount back to 1963. However, then the question becomes what interest rate to use? I don't know how to answer that. If I use a 10% interest rate, I'd conclude that a share of BRK was worth $1277 in 1963, a factor of 100x greater than its actual price.

      My point is if we were living in 1963, the right question to ask was NOT: What is the intrinsic value of BRK? The right question to ask was: What intrinsic return do I expect BRK to earn going forward? Today, we know the answer to the intrinsic return question is roughly 20% annualized.

      I'm just trying to start a discussion. Thanks for your engagement.

      Tom L

    3. That's an interesting question. I guess if you knew that BRK would earn 20%/year for 50 years, and you use 10% discount rate, then you are correct.

      And you are right that the question for long term owners is more about what the growth rate in intrinsic value is going to be rather than price.

      But BRK is a big exception. How many companies grow 20%/year for 50 years?

      You could've paid 100x IV in BRK and made money (but not much more than the S&P), but that won't be the case for BRK going forward or for most other companies.

      But then this goes against what we are trained; look for things that are trading at less than they are worth etc... If you are a long term investor, why would a discount matter that much?

      These are all good questions. I think the answer depends a whole lot on each case.

      For me, I tend to like to buy cheap and get out at fair, rinse-repeat. Unless there is something that can grow for a long time like BRK.

      Thanks for the discussion.

    4. You are correct that not many companies grow at 20% per year for 50 years, but again that speaks to my point. I think the focus should be on intrinsic returns and identifying companies that could provide those kind of returns based on today's price.

      To me, the concept of intrinsic value loses meaning if we talk as though IV can grow at different rates. Clearly Buffett talks this way, so you are in good company. I think that a more accurate description of how Buffett thinks is in terms of intrinsic returns, not intrinsic values. But, you are clearly free to laugh at me - because Buffett talks in terms of intrinsic value. I'll grant you that phrasing things in terms of IV makes more sense when one is looking at a static balance sheet value, like Ben Graham did. I just think that IV loses meaning for companies that will have a lot of growth at high ROEs.

      "Why would a discount matter that much?" I'd say that discounts don't mean much. To me, intrinsic returns matter - after all, that is what I will earn over the long-term.

      I don't mean to be argumentative. Just trying to think independently - so I appreciate any response that rips my point of view apart.

      But, I'll repeat my question. What was the IV of BRK in 1963? If we say that IV was between $12 and $1200, then I'd reply that IV seems like a strange concept. That would imply that as the future unfolds, the past intrinsic value becomes more uncertain. However, the intrinsic return of BRK at 1963 prices becomes more certain as time goes on (about 20% annually).

      Tom L

    5. Hi,

      You are exactly right in what you are saying. But it's not either or. It really depends on the situation. Growth investors don't care too much about valuation because they think the company will outgrow any 'valuation' problem, just like BRK did.

      But this doesn't mean the concept of IV is wrong. For many companies, this concept is very important.

      For example, someone published an interesting paper a while back. They looked at valuation of stocks and growth forecasts. Analysts projected earnings out into the long term for many companies and the study tracked it. They also looked at the valuations of the companies at the initial period.

      One strategy was to buy the stocks with strong growth projections and the other strategy was to buy the cheap stocks.

      Guess who won? The cheap stock strategy won. Why? Because it turned out that the analyst projections were wrong. Analysts couldn't predict the future so paying high prices for companies with high expected growth rates didn't work out too well.

      However, playing low prices meant that some of those companies actually grew alot.

      So in that sense, valuation matters. When you pay low prices, good things tend to happen. Tweedy Browne and many others have published studies that show that low p/e, low p/b strategies actually work over time.

      Things like BRK are really the exceptions, I think. If you can find something like BRK in front of 50 years of 20% growth, that's amazing. But the odds of that happening is very low.

      As for the IV of BRK in 1963, I agree with the below post that you can come up with a CAPM fair value because you know the inputs. But I don't know what value there is in finding that out.

      Buffett talks alot about IV, but he also says that value and growth investing is joined at the hip. Buffett's point is that he doesn't usually want to pay up for growth. He wants to pay a reasonable price and then get the growth for free.

      Anyway, there is nothing wrong with what you are thinking.

    6. We don't disagree on any fundamental level - it is more a matter of words. I fully agree that growth needs to be incorporated into IV, as you said.

      I guess the concept of IV loses a lot of meaning to me when you or Buffett talks about IVs growing at different rates. Clearly BRK in 1963 was a superior investment (at $30/share) to a crappy company trading at half its cash value (that stayed that way). Does this mean that buying something for 2x IV can be superior to buying something for 0.5x IV? You may say yes. Buffett might even say yes. I just think that the concept of IV has lost a lot of its meaning when it is used this way.

      I have found it useful to force myself to think about what I expect the business's intrinsic return to an owner who buys at a certain price. Anyway, this is what I was trying to get at with my original question. How much will a purchaser of BRK at today's price earn over the long term?

      Well, sorry for beating a dead horse. I'll sign off now. Thank you again for the discussion and the blog. I very much enjoy reading your thoughts.

      Tom L

  3. I would say...

    - if you knew with 100% certainty that it would be worth $150,000 today, you discount back at a 50-year risk-free rate. You'd have a risk-free arbitrage if you could borrow at less then the BRK return rate, and you knew for certain what that return would be. (barring Corzine/MF Global type risk)

    - If you thought it had a beta of 1, you would discount it at the then-prevailing risk-free rate plus the equity risk premium.

    - Suppose you thought it had a huge beta, that would give it a risk premium that would take it to 20% required return. Then the value would be... exactly the then-prevailing price.

    Basically, in order to answer this question, you have to ask yourself what the a priori distribution of expected returns looked like at that time, in other words, how much actual risk was taken to achieve that return.

  4. "Despite our poor showing last year, Charlie Munger, Berkshire’s Vice Chairman and my partner, and I expect that the gain in Berkshire’s intrinsic value over the next decade will modestly exceed the gain from owning the S&P. We can’t guarantee that, of course. But we are willing to back our conviction with our own money. To repeat a fact you’ve heard before, well over 99% of my net worth resides in Berkshire. Neither my wife nor I have ever sold a share of Berkshire and — unless our checks stop clearing — we have no intention of doing so. Please note that I spoke of hoping to beat the S&P “modestly.” For Berkshire, truly large superiorities over that index are a thing of the past. They existed then because we could buy both businesses and stocks at far more attractive prices than we can now, and also because we then had a much smaller capital base, a situation that allowed us to consider a much wider range of investment opportunities than are available to us today."

    From the '99 letter

    1. Good memory (or good find)! Either way, thanks for posting.

  5. You should value things based upon MTM b/c you could just sell at the market and distribute the cash. Therefore I think its fine to take the investments at market and maybe the multiple might be low given the huge proportion of earnings coming from railroads and utilities. The MTM/break up value is one data point. The other I would use is what type of ROE can the business make (I think a teens return). So a 1.5 to 2 P/B sounds OK to me.

    1. Yes, that's true in many cases. Things should just be valued at market.

      But in some cases, where the liquidation/monetization is not in the cards, then things may be different.

      Much of the investment portfolio at BRK is not distributable as it is there to support the insurance business. Buffett couldn't just spin off or sell the investments and distribute cash (without liquidating the insurance business).

      For example, let's say there is an insurance company that breaks even on underwriting every year. Let's say they have 2.5x investment leverage, meaning they have $2.50 in investments per $1.00 in shareholders equity. And furthermore, let's say that they only invest in fixed income investments (like most other insurance companies).

      With an after tax yield these days of 2%, this insurance company can only earn 5% ROE. Is this insurance company worth book value?

      Most financials are valued using a 10% cost of capital or return hurdle. If a company can't earn 10% ROE, then it trades below book value, and if it can do better, then it trades above it.

      Buffett has said the same thing about banks recently. He said companies with ROA below 1% trades for below tangible book and if it can earn above 1% ROA, then it trades above tangible book. So his view is similar.

      In the above case, where the insurance company has 2.5x leverage and only earns 2% after tax on an all fixed income portfolio, equity investors may only value this company at 50% of book value. 50% of book value would be only worth 20% of investments.

      So in that case, the bonds the insurance company holds is worth only 20% of face value.

      Does that make any sense?

      That's the way I (and many others) look at this.

      Thanks for reading and commenting!

    2. I get what you are saying and for the most part agree. That's why I said it's one data point. The other one is the ROE. In the above example, I think that Company is probably worth at least book value given the fact pattern, but not because of the assets.

      Anyways since we are being specific to BRK in this post, the way I try to deconstruct BRK is look at it as where is the equity allocated and what is the ROE of each segment. And then figure out what will the combined ROE look like.

      Also, instead of looking at investments per share, I value the insurance ops on a BV approach. I then look at the earnings per share as well. I get a high book multiple on insurance and a minimum multiple of 10 pretax on the OpCos. The utilities and railroads should be given a higher multiple because they produce cash predictable cash flows and utilities are somewhat uncorrelated to the economy. So anyway you slice it, BRK is worth above 1.5 BV in my eyes and probably closer to 2 BV.

      Also, separate from this discussion, I notice you have discussed Investment Banks. I don't look too much at IBs but am intrigued by the LUK JEF merger. Any insights on how to look at an IBs balance sheet? I feel pretty lost, any primers you could point me to would be very helpful.

    3. Hi,

      Most of the discussion above about investments per share is about the insurance ops and not all of BRK. I wrote about that in previous posts (search tags for BRK).

      If you look at the insurance business based on BV, then we are doing the same thing.

      As for investment banks, I don't know that I can say anything in particular. I think it's hard for people to get comfortable with the financial statements of investment banks and banks in general and there is not much I can say to help.

      I worked in the business and feel like I have a sense of the culture and people at the various firms. That's really more important than what you see in the GAAP financials.

      You have to really rely on how the firm has done over time, what kind of culture they have, what kind of businesses they are in, how they have done in good times and especially bad times etc.

      My view on Goldman Sachs, for example, or JP Morgan is based on years of following them while I was in the business, hearing from people that work there or worked there, people who worked directly with Dimon etc... All of this stuff accumulated over the years and this gives me the comfort level with these companies.

      If I didn't have that background and information, I don't know that reading the financial statements in detail would make me comfortable.

      It's a good, tough question.

      I deal with this when I read about foreign companies or small companies. Sometimes the metrics are great and the valuation is great too, but if I can't get a sense of what kind of business it is from a customer's view point, or what kind of people work there or some other information like that, I generally can't really get comfortable.

      In that case, no amount of historical ROE, EBIT/EV or whatever is going to make me comfortable so...

  6. Nice Blog keep sharing more information about furniture.
    leather furniture fort worth

  7. youre not valuing the insurance business here. I know a lot of it is in the float which is invested but the goodwill of the insurance business is on the books for something like $15bn and buffett has said this significantly understates what they would pay to acquire an insurance business of similar float quality. You either need to add insurance earnings to the operating earnings or otherwise account for it...


Note: Only a member of this blog may post a comment.