This is an article I have been meaning to write for a while, and the release of Berkshire Hathaway's (BRK.B, BRK.A) 2013 results and Buffett's annual letter gave me up-to-date info and the push to write it now. What prompted me in the first place was this recent article by Larry Swedroe which made the case against using Berkshire as a substitute for index investing. I disagreed with the argument, but it was the best kind of disagreement. He raised good points. He suggested that my intuitive view had not been sufficiently examined. He made me think.
On the face of it, the best argument against using Berkshire over an index is contained in a common behavioral finance fallacy. Diversification is a defense against many things, including blowups in specific areas. Asked whether they would prefer protection from death or damage of all kinds in California or protection from earthquakes in California, many test subjects choose protection against earthquakes. Death or harm from earthquakes is, of course, a subset of death from all causes. You could apply this, in a way, against the argument that Berkshire has a great moat and good enough diversification. But doesn't this amount to protecting against earthquakes while an index protects against everything, including earthquakes? This bothered me enough to make me undertake a careful scrutiny of my largely intuitive view that Berkshire is in some ways superior to an index.
The main thrust of the Swedroe article was that both the Vanguard S&P index fund (MUTF:VFINX) and his own DFA large cap value fund (MUTF:DFLVX) beat Berkshire on a straight comparison of annualized stock price returns on a five-year basis through 2013. On a ten and fifteen year basis, Berkshire did beat the Vanguard index fund, but the DFA large cap fund won on all time frames. I did a little closer examination of the numbers, however, and found that my intuition about them was right. The surprising outperformance of the DFA fund over ten and fifteen years came from its very large outperformance over the last five years, and a slightly smaller outperformance over the past five by the Vanguard fund was what kept it close on the ten and fifteen year time frames. I'll let the reader file this fact away, because I plan to return to it later.
Even if Berkshire had outperformed by a point or two annualized, it would be necessary to consider risk adjustment. One may argue that there are armed drones hovering above the market at all times which may fire unpredictably and strike individual companies far more often and more destructively than meaningful strikes against the market as a whole. Thinking statistically, you can say that Berkshire was lucky not to get hit by such a drone strike, but that the risk of one is nevertheless higher than the risk to the market as a whole.
Or is it possible that Berkshire is in some respects less risky than the market in the aggregate and has structural advantages not available to the market as a whole? This is the idea I want to examine - the collection of experiences and ideas that have caused me and might encourage others to use Berkshire as substitute for part of an index allocation. The term I want to use for it is dhandho.
What is dhandho and how does it apply to Berkshire Hathaway?
Mohnish Pabrai gave us the word dhandho in his book The Dhandho Investor. Pabrai is a hedge fund manager and a Buffett admirer who actually bid for and won the Buffett charity lunch in 2007. He explains that dhandho is the Gujarati word for, literally, "endeavors that create wealth," but which means more broadly low risk-high return, or "Heads I win; tails I don't lose much." Dhandho is a very interesting investment concept.
For young people or others just starting out, it can be trying to shoot the moon with a small stake which they are willing to lose. For older folks like me, it can mean getting a pretty darn good return, good enough to meet your own goals, without putting what you already have at too much risk. Index funds are pretty good for this purpose, as Buffett once again said in this year's annual letter. Larry Swedroe's DFA funds are also quite good. I use Berkshire for a significant part of what would normally be equity index allocation in full knowledge of its single company and key man risks. It is, in my view, safer than the market as a whole over a full economic cycle and most sorts of extreme conditions.
The unusual market volatility over intermediate time frames has led to a number of distortions in measuring returns over the last fifteen years. Many readers may have noticed that highly volatile funds enjoyed a jump from the bottom to the top over the past year, and sometimes over as short a period as a quarter. If you look at a simple chart, you can easily figure out the reason. Measurement from relatively high market peaks in early 2008 gave way to measurement from deep troughs in later 2008 while markets ploughed ahead through 2013. This same comparison tended to drag down the performance of steady and conservative companies and funds on a comparative basis. Berkshire is such a steady and conservative company, a mega-cap with consistent cash flow, and a stock with historically low beta. It held up very well in the last market collapse until the very end, which makes for a tough comparison with the end of 2008.
Market price is always an imperfect measurement of value, but this is especially true of Berkshire, where there is a much better measure. Buffett has always suggested ignoring price at a given moment (with the one exception of comparison to his buyback threshold) and looking at book value as a series. Book value is a much steadier and more reliable measure of the worth of a company like Berkshire, and it gets steadier every year as the relative importance of its public stock portfolio shrinks. The value of an index is also grounded in a very steady series of increasing sales, book value, cash flow, earnings, and dividends, but the market puts widely varying public values on these time series. If the last fifteen years have only one lesson, this should be the lesson. Market price is much more volatile than value.
What does this mean for Berkshire? In short, Berkshire does much better when measured over periods which include major downturns and looks less good at or near the top of market and economic cycles. Buffett allowed himself to sneak this fact into his 2014 letter by comparing Berkshire to the S&P since 2007 rather than 2008 - thus a full market cycle. Critics have already jumped in to say that he moved the goalposts from five-year comparisons with the S&P, but he had already acknowledged that Berkshire was not going to beat the S&P over a backward-looking five years. This may say more about the valuation and forward prospects of the S&P than about the performance of Berkshire, however. Buffett has also projected that Berkshire will beat the index in the future, though not by as much as in the past. Buffett doesn't make many "forward-looking statements," so when he does, they are worth thinking about.
But Wait, There's More to Berkshire's Dhandho, a Lot More
At the bottom of the markets and the economy, Buffett is like the rabbit in the briar patch (I think he actually said something about a sex-starved guy at a cathouse, but I'll leave that note in parentheses). Near the bottom of the last crackup, he made a cluster of those deals that people who were wetting their pants at the time now describe as taking unfair advantage of businesses with problems - Goldman (NYSE:GS), Dow (NYSE:DOW), GE (NYSE:GE), and Bank of America (NYSE:BAC). You can do this if you are Berkshire and keep your head. He also bought Marmon. Not far off the bottom he made his famous "all-in" bet on Burlington Northern, much criticized at the time, but now a foundation of the new Berkshire. (See my earlier piece on this transformation.) Here's the major point: at the bottom of the last crackup, many leading American companies were scrambling around for capital to stay afloat; Berkshire was employing capital at high rates of return.
Then there is the matter of buybacks. Buffett's view of buybacks has been very interesting. At first he didn't do them, but thought about them. Then he realized that he really like companies that did them, if their reasons were right. Then he realized that Berkshire gave an especially good measure for buybacks - book value - and was, of course, the company he knew best. He knew also that he didn't have to buy below book to be accretive for his shareholders, he just had to buy below a conservative estimate of intrinsic value. He dipped his toe in the water at 110% of book.
I remember a few weeks when the market wasn't very steady, and Berkshire was at 110% of book, somewhere in the low 90s on BRK.B. Every time it bobbed down below that number, it bounced back up like a cork. Then along came a deal to buy out a large owner at a fair price, and Buffett realized that the fair price in both directions was more like 120% of book. This was the last stage of his evolution on buybacks (not quite as major as his evolution from fair businesses at great prices to great businesses at fair prices).
Now in the 2014 annual letter, Buffett says that he will buy back Berkshire shares at 120% of book "aggressively," meaning that at this price, it beats anything he can buy on the public or private market in meaningful size. That's a big statement. So what is book now? Last year, we learn, it increased more than 18% to about 135,000 on the A shares. This tells lots of things, truly loaded with dhandho. To start with, 120% of book is about 162,000. Store that number away and keep it in your head for a year. The A shares are now trading at 173.7K. This implies - check my figures on this - that if Berkshire falls 6.7%, Buffett is an aggressive buyer. If this is really a floor on Berkshire price - and there is admittedly some room for argument in the fine print - it certainly takes in the "low risk" side of dhandho. And one other point: does anybody think the S&P in the aggregate will buy itself back accretively at the next market bottom?
What about upside? Well, the S&P trounced Berkshire's book value increase this year 30% upside to 18% upside. Berkshire's stock price, of course, did better, but let's use Buffett's measure of true value. The S&P went up 30% in price while Berkshire's hard value - value - went up over 18%. Is there anybody out there who thinks the hard value of big cap corporate America went up 18% last year? In the long run the value of corporate America goes up something like 3 or 4% plus inflation plus paying you dividends, now around 2%. That doesn't get you close to 18%. Well, Berkshire doesn't do that well every year, and won't in the future, but I think it's a good bet they beat the aggregate increase regularly, and do it with less risk. Dhandho: Heads I win; tails I don't lose much.
You can also make a number of qualitative arguments for preferring Berkshire to an index. It is reasonably well-diversified; at least, as the annual letter mentioned, it has 8 and a half subsidiaries which would be in the S&P 500 as stand-alone companies. Its subsidiaries tend to be best of breed in many sectors. Its sectors reflect areas where Berkshire thinking has been ahead of the rest of the world - railroads and what they are going to transport being one, the new virtue of capital-intensive, return-assured businesses being another. Most Berkshire managers are exceptional (I know, there was Sokol, but I think we allow Berkshire one individual who was like the world outside).
On the other hand the index-plus of Berkshire doesn't let you participate in Facebook (NASDAQ:FB), Twitter (NYSE:TWTR), LinkedIn (NYSE:LNKD), etc. That's too bad. Or maybe it isn't. I'm prepared to live without the alpha increment they are likely to provide to indexes over an extended period.
Conclusion: The Present Moment May Favor Berkshire
Before leaving the subject of Buffett's annual letter I have to say a word about his investment advice, which included not just his recommendation of index funds but the story of a farm he bought some years ago. I also bought a farm - pine trees - in recent years. It has some risks like fire and pine borers. It comes with a few of the problems of being an absentee owner (as you always are with stocks); a family started burying its dead there a couple of decades ago when another family member owned it, and it's hard to know whether to smile about it, take an action, or let it go. But I like that farm, which has some family memories and traditions wrapped up in it. It's the kind of investment which Buffett describes in his farm, which pays no dividends (although we get hunting right payments and a bit from thinnings). It just keeps plugging away and doing its job, adding value from the ground. Wasn't Buffett's farm sort of a metaphor for Berkshire Hathaway?
I admit it. My investment goals are somewhat conservative at this point, and I do indeed hold Berkshire for much of the allocation I might normally put into indexes. (If I sold my Berkshire, the capital gains taxes would produce a very negative anti-dhandho, and that is actually true for almost anybody who holds capital gains in excess of about 10% of cost.) I think there is a place for both an index and Berkshire-as-index and a time for emphasizing the one or the other.
If the market had another significant bear, I would keep my Berkshire but put everything else that didn't involve taking big capital gains into the Vanguard index ETF. It should do quite well enough over the rest of my life. If I were a betting man, though (and obviously I am not), I would make you something like Buffett's index-versus-hedge fund bet that over the next five years Berkshire will beat the index.