David Harris

Long-term horizon
David Harris
Long-term horizon
Contributor since: 2011
Thank you, yes it is excellent and I had not seen it before. I really like the way Fundoo Professor ties it all together. I had read about the salad oil scandal before but had not understood the significance of it in this way.
You probably know already but for anyone who might not, the situation with the derivatives liabilities has changed a little since Fundoo Professor's article was written (by the sound of it, not sure when it was written).
When Berkshire originally wrote the derivatives contracts, there was no requirement to put up collateral. That still applies to the existing contracts but new rules mean Berkshire would have to post collateral for any new contracts. That is not attractive to Buffett so he intends not write any new contracts. That means the derivative liabilities will not be a revolving fund, unlike the insurance float.
On the other hand, it is quite likely that a substantial chunk of the derivatives contracts will expire with nothing to pay out and the corresponding liabilities simply become a chunk of equity. So Berkshire will get a one-off gain plus the income earned on investing other people's money in the meantime.
This is an addition to the AMV and SALT (at most about two per cent I think) in my article above. That number jumps around a bit due to some contracts written against the level of the S&P and other stock indexes. I am not sure how to deal with that so I have left it out.
Yes, I would love to find some more, sturdy, fully self-painting fences. Please let me know if you come across any. In the meantime I quite enjoy the odd bit of painting, even if it does not quite make sense!
A very interesting and thought provoking article and wow, what a storm you caused!
Buffett explained the benefit of compounding tax free and paying all the tax at the end in his 1993 letter, with his usual clarity and humour. (See for example David Fish's comment above, 16 May 10:47, point 2). The government ends up with more money that way too.
I am delighted to have Buffett, Munger, Weschler and Combs investing on my behalf but there is no other company I trust as much. So I would be happy to have a dividend in many other cases (companies in cyclical industries, for example) for reasons people set out in some of the comments above.
You are right that other investors actions determine price but that does not mean you have to be guided by it.
As Buffett is fond of quoting, "In the short-run, the market is a voting machine [reflecting market sentiment]... ...but in the long-run, the market is a weighing machine. [reflecting intrinsic business value]"
If you read the "Marketable securities" section of Warren Buffett's 1987 letter to shareholders this should become clear.
I agree with Buffett!
At the risk of a huge sidetrack, "In God we trust, all others bring data."
As far as I know, Buffett has never said that float should not be considered a liability at all:
"Neither item, of course, is equity; these are real liabilities."
That is what Warren Buffett has been saying in the Owner's Manual section of the annual report for years.
But yes, Buffett has pointed out that if there were to be any drop off in float "at some future time" it is likely to be very gradual; and yes, float keeps growing; and yes, Buffett expects it to grow again in 2013; and yes, the new commercial insurance venture may generate significant new float.
So I agree with you that the bulk of float is likely to be perpetual (but not eternal), with the investment returns it earns going to Berkshire for many years to come.
And yet... ...well pages 7 - 9 of the 2012 annual report say it better than I can.
So I feel comfortable adding back 85% of the roughly $122bn of float and deferred taxes, but that is only a guess. You may yet be right about value!
NL TInvestor,
My guess is that Berkshire will continue to prosper way in to the future with or without Buffett and Munger.
They have assembled a collection of excellent businesses and a source of low risk, low cost gearing via the float from Berkshire's insurance companies, which are themselves unusually effective. I think that will endure long in to the future.
How the market might react is another matter, but that causes me no concern at all as a long-term investor.
Thank you for flagging up Daily Journal. Size gives Charlie a huge advantage (Buffett has been quoted as saying he could average 50% p.a. with small sums), I will investigate further!
Yes you have understood correctly.
Over the past few years there has been a fairly unusual double margin of safety from Berkshire's attractive growth prospects and low price. At the current price, I think Berkshire's growth outlook still provides a margin of safety, more typical of the past 17 years.
On your question about Berkshire's effective tax rate, were you thinking about a possible boost to earnings if the tax rate fell back in future? I have not looked in to that.
That depends how big a margin of safety you want.
To me, the thick lines on the charts (AMV and SALT) indicate that Berkshire has grown robustly over the past 17 years, undaunted by all that has happened.
If you think about Berkshire's subsidiaries such as Geico, BNSF, MidAmerican Energy and the 70+ others, and you think about Berkshire's equity holdings such as American Express, Coca-Cola, IBM, Wells Fargo and the rest, how do you rate Berkshire's future prospects? If you think those businesses have above average growth potential, does that give you a big enough margin?
Short-term I have no idea what will happen but long-term, Berkshire looks good to me.
Yes, you identify some very important issues. However, the point of the SALT calculation is that even if Todd and Ted were duffers, and even if everyone were to shun the new CEO, Berkshire still looks very cheap.
If you think the existing businesses will match the US average, Berkshire is worth its SALT. It is worth more than SALT if the subsidiaries continue to beat the average. And that is all without Todd, Ted, Warren, Charlie or Jamie doing anything at all.
Warren and Charlie add value over and above that. So, your nagging questions are perhaps about how much additional value the new team can contribute. My guess is quite a bit.
Well, I think it is always easy to lose money buying stocks! And as Berkshire does not pay a dividend you only make money if you can sell it for more than you paid.
But I agree with your general idea. What it says to me is that Berkshire has continued to prosper in spite of all that has been happening. (Buffett loves a cheap market.) I strongly suspect the market will recognise that in due course and that the price will adjust accordingly, but I do not know when.
Very good point on the puts. Yes, if you take those in to account the discount is even greater.
It would be interesting to see look-through earnings over the years but they do not always publish the figures. Someone asked about that at the shareholder meeting so perhaps we will see some numbers in the next annual report... ...or perhaps you could do an article?
You might be right about electric cars and newspapers, although Warren and Charlie seem just as smart as ever to me. However, one of the great things about Berkshire is that it would not matter even if they had lost it. That is because BNSF, Geico and (almost) all the other Berkshire subsidiaries would still run without interference from Warren and Charlie, just as they do today. And they would still be great companies!
Great article Jeremy, thank you. You have cut through the horrible complexities of these companies' financial statements to shine a very clear and interesting light on how they have changed their ways.
Actually you’re not alone, there are plenty of people it doesn’t make sense to! Where you’re different is that you’ve explained your thinking in public, so full marks and thank you for doing that.
Also, you’re quite right to make whatever adjustments to a balance sheet make sense to you: you just come up with a different gauge if you do that. For example, we could plot book value against market price, without any adjustments . That would draw a line similar to the SALT line but lower down the graph. Or we could plot total assets, without any deductions for liabilities. That would give another similar line, but this time higher up the graph.
The reason I’ve made the adjustments I have is that it gives me a feel for the funds available long-term for Berkshire to invest (but it is a guess). That allows us to greatly simplify the valuation equation because lots of terms just cancel out. We can simply say that, so long as those funds (i.e. SALT as described in the article above) earn an average rate of return, they have $1 of economic value to shareholders for each $1 invested.
We could do it on your basis instead and let’s assume, for arguments sake, that we find each dollar of book value (less your deductions) produces 1.5 times an average rate of return, due to the gearing effect of float and deferred taxes. In that case we would apply a multiple of 1.5 to your adjusted book value. However, doing that requires that we start to make judgements about rates of return. And small differences in assumed rates result in big differences in calculated value, so errors tend to get magnified.
Similarly, if we do a discounted cash-flow valuation or a market price comparison, there are lots of additional judgements and opportunities for error.
Whichever approach we use, we have to make judgements anyway about how well or badly Berkshire’s businesses will do in future. But doing it on the basis of SALT, we don’t need to try and turn that in to numbers, we just need a feel for whether they will do averagely well, better or worse.
So am I saying this is the right way to value Berkshire and that other measures are wrong? Certainly not. I prefer to look at things from several different points of view and if the picture looks consistent from all angles I feel more confident that I’m on the right track.
Some further explanation of float and deferred taxes would have helped a lot of people, so here it is:
Float is insurance premiums received by GEICO, General Re etc that is reserved for paying out some time in the future against insurance claims. Because it will need to be paid out in future, the float is recorded on the balance sheet as a liability and so is deducted from assets in the calculation of book value.
However, until the float is paid out, Berkshire gets to use it. Some will be paid out quite quickly (e.g. float relating to car insurance) and some will be around for many years (e.g. float for workers insurance). So why am I saying it can be considered very long term capital? That is because, so long as Berkshire’s volume of insurance business doesn’t shrink, it will have $69 billion of float to use.
Why do I say that float has value to Berkshire shareholders? Unlike most insurers, Berkshire’s cost of float is generally nil (actually, most years they get paid for holding the float) because they are very disciplined when they take on business. So Berkshire gets the benefit from investing $69 billion using capital that doesn’t cost them anything. It’s quite a pleasant experience!
This is probably easiest to picture in relation to the shares Berkshire owns in companies like Coca Cola, Wells Fargo and American Express. They have a market value of $61 billion but cost only $34 billion to buy. If Berkshire sold those shares for the stated market value, it would have to pay taxes on the gain. The amount of tax that would be due is included on the balance sheet as a liability, and so is deducted from assets in the calculation of book value.
Berkshire does sometimes realise gains and have to pay taxes, plus the value of its share portfolio (and of the deferred tax liability) fluctuates with the market. So I’ve guessed we can consider 85% of it as interest-free very long term capital.
Does that make sense?
Yes, very good Greg, up to your usual high standards. Also like the time series data.
Great article and I really like the "look-through" earnings data. Do you have a time series of that? It would be interesting to compare with stock price over a number of years and with other valuation pointers eg SALT.
Hi Shahar,
It's a question lots of people ask and your guess is as good as mine. Mssrs Buffett and Munger claim the secret of their success is that they don't really do much, and with only 21 people at HQ against more than 250,000 at the businesses they own, I believe them.
They say those businesses run autonomously, just as they did before Berkshire bought them, so I don’t foresee the absence of Buffett and Munger changing that. Will the new guys be as good at doing not very much as Buffett and Munger are? That’s hard to imagine but I recon they’d still be pretty spectacular.
(By the way, Bill Gates has made a long term commitment to serve on the board and make sure things don't get out of shape. As well as having great personal loyalty to Warren Buffett, The Bill and Melinda Gates Foundation has a long term interest in the future health of Berkshire to the tune of tens of billions of dollars.)
Presumably there would be a sell-off when they go, but that wouldn't worry me (it might even be another of these rare buying opportunities). As I see it, Buffett and Munger have built a self-propelled, amazing compounding machine that presently looks cheap compared to what you get. So we’re getting Buffett and Munger for as long as they’re involved at no extra cost. That strikes me as good value.
Delighted to see so many smart people can disagree so widely on the value of Berkshire (even those who have read the letters). That’s what makes an opportunity!
My spread sheet keeps track of Berkshire’s “intrinsic value” quarterly since 1996 based on a simple method Charlie Munger referred to in the 1999 Wesco letter. CM’s method was book value plus one fifth of deferred taxes. Charlie mentioned that each dollar of book value was far more valuable in WB’s hands at Berkshire than at Wesco. I wasn’t sure what to make of that or why he went for one fifth. Anyway, I track book value plus one fifth of float and deferred taxes and multiply by a premium that gets smaller as Berkshire grows relative to total US GDP to give “Adjusted Munger Value” (AMV).
AMV matches pretty well with market price since 1996, showing over-valuations of 20% or more from Sep ’97 to Sep ’98; winter '03 (oops, selling then would have missed a lot of upside); and winter '07. Also showing under-valuations of 15% or more in Sep ’99 to May ’00; Sep ’08 most of the time through summer 2010; and now for the last couple of months.
Not claiming that “AMV” is right and I always felt one-fifth of float and deferred taxes was conservative (CM said as much in respect of Berkshire). I like J. Stew’s reasoning above of a liability discounted 100 years at 5%. Any further thoughts on that and on why Charlie Munger would have been so much more conservative than that when talking about Wesco – e.g. does Wesco expect to sell holdings more often than Berkshire for some reason?