The real secret to Buffett's success...

The real secret to Buffett's success... Berkshire isn't the same business it used to be... Why investors should be worried...

Fair warning, gentle reader... today's Friday Digest is, in all likelihood, a waste of your time. Unless, that is, you'd like to know the real secret to Warren Buffett's almost unbelievable investment track record over the last 50 years. In my view, understanding this secret is bar none the most valuable thing you can ever learn as a common-stock investor.

But of course, it's Friday. And I'm sure you're really busy... So just file this away where you keep all of the things you're going to read some day... eventually... when you get around to it...

As I mentioned yesterday, I'm writing a book about Buffett, titled Warren's Mistakes. I know... the world doesn't need another book about Buffett. There are dozens of good books out already. (The two best are The Snowball by Alice Schroeder and Buffett: The Making of an American Capitalist by Roger Lowenstein.)

I'm writing my book because the mainstream media and most of the financial community fail to realize how much Buffett's investment style has changed since 2000. And not for the better, either. In today's Digest, I'd like to show you what lies at the center of Buffett's incredible results and explain exactly why his more recent results have been lackluster.

AQR Capital Management is one of the world's largest and best quantitative hedge funds. Its founder, Cliff Asness, got a PhD at the University of Chicago. He studied under the same group of knuckleheads I wrote about yesterday – the "market is perfectly efficient" crowd. But when he studied the market carefully, he found patterns – thousands of anomalies – that could be exploited for profits.

If you read Steve Sjuggerud's work, you'll recognize the strategy that Asness discovered. He found that both value strategies (buying stocks when they are significantly underpriced relative to their assets or earnings) and momentum strategies (buying stocks that are going up or selling stocks that are going down) worked. But the best strategy was combining the two approaches, buying cheap stocks after they began to go up. Or, as Steve puts it, Asness found that buying stocks that were "cheap, hated, and in uptrend" was the best way to beat the market.

While Asness wasn't the first market analyst to adopt this approach, he was the first researcher who took the time to assemble enough high-quality data to prove it worked empirically. By the mid-2000s, Asness was managing almost $40 billion in capital, utilizing various computerized strategies that distributed investments among hundreds of stocks that met his criteria. (Likewise, Sjuggerud has also built computer systems to aid his investing. His True Wealth Systems track record outperforms his own human-selected recommendations in True Wealth.)

I bring up Asness because it's important to note that not everyone that learned finance at the University of Chicago believes a bunch of worthless nonsense... but also because you need to understand there's probably nobody in the world more qualified to pass judgment on Buffett's track record.

Asness has been the world's leading market empiricist since the early 1990s, when he built investment bank Goldman Sachs' first quantitative investment fund before founding AQR in 1997. Asness is a genius at studying how the market actually works, instead of simply describing how it ought to work.

In 2013, Asness sponsored a "deep dive" into understanding how Buffett was able to produce such outstanding results over such a long period of time. The authors of the study I cited yesterday (Frazzini and Kabiller) were employees of AQR. They worked for Asness. They looked at all the stocks and mutual funds that have traded since 1926 and found that nothing beat Berkshire Hathaway. That is, Berkshire has the highest "Sharpe ratio" among all possible publicly traded equity investments.

For those of you who aren't finance geeks, the Sharpe ratio is a simple measure of a stock's annual return over its volatility. Put simply, Buffett has earned roughly twice the returns he should have been able to make given the volatility of Berkshire's stock. While that's incredibly impressive – the best record in the entire stock market – it's still not enough. Doubling the return of the stock market only gets you to about 16% annualized returns... not the 20% annualized Buffett has earned at Berkshire. So how did he do it?

The real secret Asness discovered is leverage. What? How is that possible? Berkshire is rightly famous for being a fortress of financial stability. The company currently holds around $250 billion in cash. It has no net debt. Over the long term, the company has averaged total debt (short- and long-term) of around 20% of book value.

Nevertheless, Asness' team reports, "We estimate that Buffett applies a leverage of about 1.6-to-1, boosting both his risk and excess return in that proportion." This leverage was created by using the "float" from his insurance companies. That's the money that was paid in insurance premiums but not yet paid out in claims. Berkshire currently holds $84 billion in float. Berkshire's insurance subsidiaries have a 12-year record of profitable underwriting, so Buffett has used all of this capital for free. It's this leverage that explains the rest of Berkshire's "alpha" – of the excess returns Buffett has earned.

Buffett isn't the only investor who figured out that leveraging safe stocks can produce outstanding results. In the 1940s, while serving as an insurance regulator for the state of New York, Shelby Davis figured out that some insurance companies could compound their book value at market-beating rates. These rare firms, with exceptional underwriting skills, proved that they could increase their net worth, year after year, regardless of the economy.

In 1947, Davis quit his job. Starting with just $25,000, he set to work as a private investor, exclusively buying high-quality insurance stocks. Within 15 years, he was a millionaire. At his death in 1994, he was worth nearly $1 billion. His average annualized gain was 24% – matching Buffett's. Davis produced this result without the benefit of controlling an insurance company. He was merely an investor in the common stock. But he used full margin at all times. This greatly amplified his returns and greatly increased the volatility of his portfolio. It nearly wiped him out in the 1974 bear market.

The Asness team saw the same kind of volatility in Berkshire's shares. Berkshire fell 51% from peak to trough in the 2008-2009 financial crisis. It saw a 49% decline in the bear market of 2000. It fell 37% during the "Black Monday" stock meltdown in 1987. So the Asness researchers asked the key question: What kind of stocks are good enough and safe enough to be leveraged successfully over long periods?

What they discovered won't surprise you. Here's how they described Buffett's picks: "He buys stocks that are safe, cheap, and high-quality (meaning stocks that profitable, stable, growing, and with high payout ratios)." Pay special attention to the last criteria (high payout ratios). We'll come back to this attribute in a minute... By using this description, the Asness team is withholding something critical.

Using this information, the Asness researchers programmed their computers to run simulations "buying" large numbers of companies that matched Buffett's criteria. By doing this, they were able to build portfolios that were highly correlated to Buffett's portfolio. Using leverage on these portfolios, they were able to produce results that matched Buffett's, even though their portfolios were far more diversified and made up of different stocks. As the Asness team concluded...

We find that the secret to Buffett's success is his preference for cheap, safe, high-quality stocks combined with his consistent use of leverage to magnify returns while surviving the inevitable large absolute and relative drawdowns this entails. Indeed, we find that stocks with the characteristics favored by Buffett have done well in general, that Buffett applies about 1.6-to-1 leverage financed partly using insurance float with a low financing rate, and that leveraging safe stocks can largely explain Buffett's performance.

The key to finding companies that are good enough and safe enough to leverage is to look for highly capital-efficient businesses. These are simply companies that do not require much additional capital to grow. Please, if you haven't yet, read my initial recommendation of Hershey. (We've "unlocked" the issue for you right here.)

I'm not a lucky "coin flipper." And I'm not using Hershey as my example retrospectively. I told my readers at the time of my original recommendation that I believed Hershey would go on to be one of the best investments I'd ever recommended. And that's what has happened, even though I recommended the stock just months before the worst bear market of our lifetimes. Even with the worst possible timing, this stock didn't decline more than 25% from its highs... and it has produced excellent returns for our portfolio.

These are exactly the kinds of stocks that Buffett made a fortune buying, leveraging, and holding forever. And the thing that unites all of these stocks – financial stocks like Wells Fargo, branded consumer-products companies like Gillette, and insurance firms like GEICO – is their almost magical ability to grow sales, earnings, cash flows, and dividends without correspondingly large investments in their businesses.

Asness' team doesn't spell it out this completely. But trust me, it is this particular quality – capital efficiency – that matters most. It's what they really mean when they write "high-quality." It is capital efficiency that allows companies to set high payout ratios (that is, they pay out a larger-than-normal percentage of earnings via dividends). The most common hallmark of capital-efficient companies is a noticeable lack of capital expenditures.

The most striking thing about Berkshire Hathaway today versus the years before 2000 is that its focus on capital-efficient businesses has almost completely disappeared. Here's an anecdotal example... In his annual letters to shareholders, Buffett has frequently explained his focus on capital-efficient stocks by pointing to the success of See's Candy. He did so in this year's letter (published last Friday)...

To date, See's has earned $1.9 billion pre-tax, with its growth having required added investment of only $40 million. See's has thus been able to distribute huge sums that have helped Berkshire buy other businesses that, in turn, have themselves produced large distributable profits. (Envision rabbits breeding.)

This is classic Buffett. This is his quintessential investment, much like Hershey is my quintessential recommendation. You never have to worry about companies like these going out of business. And you're a fool if you ever sell them. They are virtual cash machines, creating profits while requiring almost no additional capital.

But consider how different this business is from the lion's share of investments Buffett has made since 2000. His "powerhouse five" of wholly owned companies were all purchased after 2000: railroad firm Burlington Northern, regulated utility MidAmerican Energy, chemical giant Lubrizol, diversified industrial firm Marmon, and Israeli manufacturer Iscar.

These companies represented massive investments on behalf of Berkshire. Burlington Northern alone required a 6% dilution of the stock and $16 billion in cash (a total value of $44 billion). These are good businesses. They are not capital-efficient. Their need for additional capital to grow eats up most of their earnings. Berkshire earned $12.4 billion with this group of businesses this year. It will spend $15 billion on capital investments this year to grow, with most of this capital going back into the powerhouse five.

Buffett is going to spend $6 billion on the railroad alone this year. That's more money than Berkshire made from all of its other wholly owned non-insurance businesses outside of the powerhouse five.

In summary, with the purchase of large heavy industrial businesses, Buffett has transformed Berkshire from an insurance float-funded collection of the world's highest-quality businesses into a global conglomerate of stable businesses with average economics. The question for investors to ponder most carefully is if this new collection of businesses will weather the next major economic storm as well as Berkshire has in the past. My intuition? Absolutely not.

When you look at Berkshire's cash flows and capital expenditures, this fundamental change in the business really jumps out. Berkshire had so few capital expenditures historically that prior to 2000, it didn't even include capital expenditures as a line item on its statements. (You have to look in the footnotes to find the numbers, which are tiny.)

Prior to 2002, Berkshire spent less than 20% of its cash earned from operations on capital expenses. (Capital expenses are "lumpy" – with big charges in some years and almost none in others. To measure this, we use a five-year rolling average.)

But since 2002, these investments back into their own businesses have grown massively. They now average 42% of cash earned from operations. This means that although Berkshire's earnings may be growing (and fueling a matching rise in book value and stock price), the quality of these earnings has significantly declined. Most important, the growth in capital expenses means the business is producing far less free cash (as a percentage of profits). And that means there's less cash for Buffett to invest. So Berkshire will almost surely see a big drop in the rate that it can compound its net worth.

Now... what does all this mean for investors? First, it's clear that Buffett isn't doing nearly as good a job as he has historically with capital allocation. Berkshire's book value isn't compounding at nearly the rate it has historically. (It has underperformed the benchmark S&P 500 index in five of the last six years, and seems likely to do so again this year.)

Buffett explains this by making a circumspect argument (in my opinion) that his wholly owned portfolio of companies is not marked to market, so Berkshire's book value doesn't fully reflect the value of these companies' increasing intrinsic value. That is true. But it wasn't meaningful until Buffett stopped being a great allocator of capital.

Think about See's Candy. Sure, See's is still valued on Berkshire's balance sheet at its acquisition cost of $25 million. And yes, See's is still worth a lot more than that. But Berkshire's balance sheet has captured all of the $1.9 billion in retained earnings that See's produced. And because Berkshire hasn't historically paid a dividend, this capital has produced tens of billions in book value for Berkshire via the earnings of other companies and the shares of other companies Buffett purchased with this capital.

You don't need to follow all of the accounting. Just remember this... Buffett is a smart man. He chose what his benchmark would be – increasing book value by more than the S&P 500. For decades, he said that if he couldn't do this, he isn't doing his job and that Berkshire should either replace him or begin paying dividends.

After never losing to the S&P in any five-year period before, Buffett has now lost for essentially six years in a row. That explains far more about him deciding to change his benchmark than the accounting complexities. So... investors should be careful about extrapolating into the future the returns that Berkshire has made in the past. Doing so will almost surely cause significant disappointments.

Here's another interesting way to see what's happening with Berkshire...

This chart shows the Asness group's "Buffett index" – the large number of stocks that match Buffett's historic investment criteria (cheap, high-margin, and capital-efficient). This index (in the green) represents the performance of the kind of stocks Buffett has historically invested in. That's why the blue line – the actual performance of Buffett's portfolio – tracks it so closely.

The Buffett index is going to show some outperformance over time because it is only a paper portfolio. It doesn't pay taxes or transaction costs. Also, the index is going to do a little better because it is far more diversified than Buffett's actual portfolio. Even so, those factors don't explain why the relative outperformance of the Buffett index has increased so much since the late 1990s.

There can be no doubt that the Buffett of the last 15 years isn't doing nearly as good a job as he once did. He has had dozens (or even hundreds) of opportunities to buy incredibly high-quality businesses at dirt-cheap prices over the last 15 years. My newsletters have recommended plenty of them: Tiffany, Starbucks, Hershey, Harley-Davidson, eBay, Disney, and McDonald's, to name just a few.

But Buffett didn't buy any of these great businesses. Instead, he spent $44 billion on a railroad. That's 44 times more capital than he invested in Coca-Cola. He also invested $15 billion in IBM, which he specified by name in his 1996 letter as one he would never buy. Few investors realize how critical these decisions will become over time.

And that's the bigger question: Will a portfolio dominated by slow-growing, highly regulated and competitive businesses with large, ongoing demands for capital investment be able to withstand an economic storm? Maybe. But it's no longer the sure thing it used to be.

By the way, I know that some readers of my Friday Digests are in a position to ask Buffett about these concerns (or even pass along this letter to him). Although Buffett has never responded to me directly, I know that my work has reached his desk on several occasions and I've corresponded with his secretary before. If you decide to pass this note on to Buffett – or if you speak to him about this letter – please let him know I'm still interested in interviewing him for my book. I think he would like the first chapter, which we published in yesterday's Digest.

New 52-week highs (as of 3/5/15): Becton Dickinson (BDX), ProShares Ultra Nasdaq Biotech Fund (BIB), Bristol-Myers Squibb (BMY), Anheuser-Busch InBev (BUD), Blackstone Group (BX), Cempra (CEMP), WisdomTree Europe Hedged Equity Fund (HEDJ), SPDR S&P International Health Care Sector Fund (IRY), ProShares Ultra Health Care Fund (RXL), and Varian Medical Systems (VAR).

We've gone on long enough in today's Digest, so we're skipping the mailbag. But please send your questions and comments, praise and criticisms on my Buffett writings of the past two days to feedback@stansberryresearch.com.

Regards,

Porter Stansberry
Baltimore, Maryland
March 6, 2015