On Berkshire Hathaway

Warren Buffett’s Berkshire Hathaway fascinates academics, investors and the public. For dyed in the wool efficient market theorists, Berkshire Hathaway’s performance is a six sigma slap[1] in the face. For value investors, Buffett’s success is proof that the methodology pioneered by Graham and Dodd works.

To hear young value investors analyse equities is to hear the Gospel According to Buffett: different mouths open but the same words come out. Clutching their copies of Security Analysis, the next generation of brilliant value investors tell us: “I never buy with leverage. I look for “castles with moats”, businesses with “sustainable competitive advantage”. Much of the media coverage of Berkshire Hathaway makes the implicit assumption that Buffett’s success can be attributed to this set of ideas--a kind of analytical prescience about competitive dynamics.[2]

This paper argues that this assumption is wrong. Through an examination of the academic literature and industry results, I demonstrate that there is no evidence that “moat investing” leads to outperformance. I trace the source of these ideas to Michael Porter and show how and why his framework has no predictive value. 

 Instead, it is Buffett’s clever use of leverage[3] to purchase cheap, safe, quality companies that has driven its compounded annual growth rate of 20% over 52 years.[4]

 Berkshire Hathaway: A Brief History

In 1839, a man named Oliver Chace founded a textile manufacturer named the Valley Falls Company. Valley Falls grew steadily in the following decades, growing revenues and production capacity to keep up with demand.

Through a series of acquisitions and mergers, Valley Falls would become first Berkshire Fine Spinning Associates, then Berkshire Hathaway. Oliver Chace’s small textile firm had become a “giant of New England textiles”.[5]

Yet by the mid 1950s, the New England textile business was an anachronism. Berkshire Hathaway’s first seven years of existence as a corporate entity saw its net worth shrink by 37% and 9 of its production plants liquidated.[6]

Intrigued by a stock price that was less than half of book value, a young investor by the name of Warren Buffett decided to buy shares in hopes of a quick exit.[7] The long term prospects of the company were obviously poor, but the price of Berkshire Hathaway was simply too cheap to resist. In December of 1962, Berkshire Hathaway stock was selling at $7.50--a 30% discount to its per share working capital of $10.25.

Buffett soon received an opportunity to exit. The president of Berkshire Hathaway, Seabury Stanton, offered him to buy back Buffett’s stake at $11.375 a share. Buffett replied that he wanted $11.50 per share instead. Stanton agreed, but then subsequently shaved an eighth of a point off the price in official documents. Irritated by this maneuver, Buffett ignored the revised Stanton offer and began to purchase more Berkshire Hathaway shares to gain control and fire Stanton. 392,633 shares later, Buffett was victorious. The Buffett era at Berkshire Hathaway had begun.

In the following years, Buffett and his colleagues would direct the cash from Berkshire Hathaway towards purchasing other companies to diversify its business away from the suffering textile business. Today, Berkshire Hathaway is a juggernaut with $242 billion in revenue and $700 billion in assets.[8]  It owns everything from GEICO to Dairy Queen to significant portions of Coca-Cola and Kraft Heinz.[9]

 The Gospel According to Buffett

To those who ask how he transformed a struggling textile manufacturer to a multinational conglomerate, Buffett has many answers. Buffett tells us the key has been hiring excellent managers, building the right corporate culture, staying within a circle of competence, understanding the difference between price and value, closely studying the work of Graham and Dodd, buying companies with sustainable competitive advantages, never using leverage, always maintaining a margin of safety, never relying on investment bankers and so on.[10] Buried in this litany of answers is sometimes the mention of something called insurance float.

Buffett readily admits that much of what he describes as important are principles of security analysis as articulated by Graham and Dodd.[11] There is a reason after all that Buffett is known as Graham’s most famous student. Graham’s methodology has been examined ad nauseum. The Wall Street Journal writes:

“Graham's 1934 book, "Security Analysis," with David Dodd, is widely viewed as the urtext of modern value investing. The long-held idea is that some stocks trade significantly below an identified "intrinsic value" and can be bought at a discount, with a built-in margin of safety against a complete washout.”[12] 

Yet value investing cannot be the whole story. After all, Buffett has outperformed not only the market, but also his fellow value investors. Graham’s other students all received the same education in the principles of value investing. While Bill Ruane, Walter Schloss, Marshall Weinberg, Jack Alexander and Tom Knapp have certainly done well for themselves, Buffett’s results eclipse their’s by several orders of magnitude. What has driven this difference? Is it analytical brilliance? Managerial prowess? Luck? 

Buffett differs from Graham most crucially in his attraction to the idea of sustainable competitive advantage. In classic Buffett fashion, he has developed a folksy analogy for businesses with sustainable competitive advantage. In his words:

“I don’t want a business that’s easy for competitors. I want a business with a moat around it. I want a very valuable castle in the middle. And then I want…the Duke who’s in charge of that castle to be honest and hard working and able. And then I want a big moat around the castle, and that moat can be various things.”[13]

Buffett’s reputation as Graham’s star student at Columbia leads many to attribute these ideas to Security Analysis, but this presumption could not be further from the truth. In the 895 pages of Graham and Dodd’s text, the words “moats” “competitive advantage”, “barriers to entry” and “market power” do not appear once.[14]

Charlie Munger, Buffett’s long time partner and Vice Chairman of Berkshire Hathaway put it this way:

“If we'd stayed with classic Graham the way Ben Graham did it, we would never have had the record we have. And that's because Graham wasn't trying to do what we did...And so having started out as Grahamites which, by the way, worked fine—we gradually got what I would call better insights...And, by the way, the bulk of the billions in Berkshire Hathaway have come from the better businesses.”[15]

What Munger does not mention is that while “classic Graham” has been proven again and again in empirical tests to be effective[16], there is little evidence that the Buffett-Munger idea of moats creates better investment outcomes. The inclusion of “moats” into the investing orthodoxy has been driven not by evidence, but by intuitive analogy, Buffett’s reputation and the patina of academic rigor provided by Michael Porter.

Intuitive Analogy

Buffett mentions the word “insurance” 500 times but  “moat” only 19 times in his shareholder letters from 1965-2013. Yet the attention readers, professionals and enthusiasts have dedicated to the latter far eclipses the former. Reading about insurance is tedious and difficult. It involves understanding ratios of float to premium volume, risk management strategies for low probability events and role of float in cheapening cost of capital. In contrast to these abstractions, moats are simple and intuitive--one does not need a career in finance to visualise a castle. To the masses, Business Insider declares: A Key to Warren Buffett's Investing Strategy is Incredibly Easy to Replicate (MOAT). Retail investors have flooded into ETFs that invest based on identifying “attractively priced companies with sustainable competitive advantages”.[17] Yet how have these moat based investment products performed?

 A peek at the lifetime performance of this moat fund offered by VanEck finds it underperforming its index, exposing a performance differential of -0.58. For all the common sense and colorful analogies, this set of ideas appears to have no efficacy in guiding the investment decisions of this moat fund across this time horizon.  

The reality is that Buffett’s outperformance comes primarily from structural advantages (ie. cheaper cost of capital) rather than analytical ones. But one cannot arrive at this conclusion and discount Buffett’s moat concept based on the performance of a single ETF. Perhaps these VanEck analysts are simply applying these ideas incorrectly. To evaluate the empirical evidence behind moats and competitive advantage further, one must directly examine the work of Michael Porter.  

The Rise of Michael Porter

In the world of business strategy, Michael Porter is a giant. In 1979, he first articulated his now famous “Five Forces” concept in 1979 with How Competitive Forces Shape Strategy and again in 1998 with Competitive Advantage. These ideas have had broad resonance in schools and boardrooms. After Porter’s coinage of these terms, Buffett began to adopt them in public remarks and in his annual shareholder letters. In fact, Buffett admits:

“What we refer to as a “moat” is what other people might call competitive advantage.”[19]

Porter arrived at Harvard Business school in the 1970s and found the predominant method of instruction, case studies and SWOT[20] analysis unrigorous.[21] He set out to do better. Out of the chaos of thousands of cases, Porter identified Five Forces that determined the profitability of a business and provided competitive advantage: (1) the bargaining power of suppliers, (2) the bargaining power of buyers, (3) the rivalry among existing firms (4) threat of substitute products and (5) the threat of new entrants.

Drawing on his training as a Economics PhD at Harvard, Porter crafted his forces around evidence in a field of study called Industrial Organisation.[23] Industrial organisation at the time argued that industries with higher concentration (where a few firms dominate the market), had higher profits than those with lower concentration.

The influence of Industrial Organisation’s price-concentration studies on Porter’s Five Forces is palpable. For example, new entrants are a threat in Porter’s framework because they dilute industry concentration and thus decrease profits. Likewise, if few suppliers exist compared to numerous buyers, suppliers will possess bargaining power over buyers--the suppliers’ high industry concentration will lead to high supplier profits, while the buyers’ low concentration will result in low buyer profits. In this way, all five forces are natural extensions of the Industrial Organisation consensus on price-concentration before 1970.

Finally, rigor had arrived in the world of business strategy. Here, ostensibly, was undeniable proof that Buffett’s moat investing was on solid intellectual foundations.

Evidence Disowned

Ironically just as Porter’s ideas entered mainstream consciousness, Industrial Organisation as a field was busy disowning price-concentration studies. In his “definitive review of the cross industrial literature”, Schmalensee found in 1989 that cross-industry price-concentration studies were inherently flawed because of systematic problems with methodology.[24] At Harvard, Richard Rumelt put the “metaphorical nail in the coffin of Porter’s industry-centric view of strategy”.[25] In Rumelt’s own words:

“The ‘classical focus on industry analysis’ is mistaken, because these industries are too heterogeneous to support classical theory.” [26]

In plain words, industries are simply too different to generalise from, and there exists no evidence that higher industry concentration leads to higher profits. This reality is so widely accepted by the field today that researchers have essentially given up on building generalised, structural models. The standard Industrial Organisation textbook now concedes, “knowledge does not easily accumulate across industries”.[27]

Porter’s Five Forces, it turns out is no more rigorous than the SWOT analysis that Porter so disdained as a young researcher. There is not a shred of empirical evidence to suggest that targeting moats, or “sustainable competitive advantages” improves investment outcomes.

The True Drivers of Berkshire Hathaway Outperformance

In 2008, Buffett wrote:

“Whether we’re talking about socks or stocks, I like buying quality merchandise when it is marked down.”[28] 

How much of Buffett’s return can be attributed to following this simple rule? To find out, researchers at AQR translate this folksy aphorism and abstract concept of quality into a set of quantitative rules. They construct what is known as a “systematic strategy” that selects stocks based on simple financial ratios and test how they would’ve done in the past.[29]

In introductory finance, undergraduates learn the capital asset pricing model (CAPM). Stated plainly, CAPM contends that the expected return of an asset is related to its historical volatility.[30]

Modern academic finance has identified a number of “anomalies” that CAPM fails to explain. Academics empirically prove that these  anomalies exist by running regressions on historical stock data from CRSP/COMPUSTAT databases.[31] There are, in other words, other “factors” beside CAPM that account for these anomalies and predict superior returns.[32] The most famous of these “factors” are Value (Buy companies that are cheap), Quality (Buy companies that are profitable and growing), Momentum (Buy companies that have been doing well recently) and Size (Buy companies that are small).

The exact construction of these factors is involved. For the purposes of this paper we can think of the procedure roughly like this: take the entire universe of equities and rank them based on some simple financial ratio (i.e. for “Value”: Value of company/Earnings). Then, buy the top decile of companies and short the bottom decile. Much of the performance of investment managers, it turns out, can be reduced to these simple sorts.  

But what of Buffett? Can these factors explain Berkshire’s performance? Researchers at AQR run regressions and find that most of Buffett’s returns can be explained by the use of leverage to acquire exposure to Value, Quality, Low-Risk (“low-beta”) factors.[33] In their words:

“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.”[34]

These algorithms are not able to interview CEOs or call up Michael Porter to discuss competitive dynamics. They do not drink cherry cokes or look for moats. They simply find things that are cheap, quality and low-risk based on simple financial ratios and buy them.[35]

Yet the choice of these factor exposures are not enough on their own. Without the magic of Berkshire’s insurance float, Buffet’s performance would not have been extraordinary.

The Magic of Insurance Float

Despite Buffett’s insistence on avoiding the use of leverage, he himself is roughly 60% levered.[36]  He accomplishes this through his use of insurance “float”. Because the insurance business collects premiums regularly, but only pays out when and if a “bad” event occurs, Buffett has a substantial pile of cash lying around that he can direct as he pleases. Whereas others must pay interest to borrow money, Buffett gets paid to hold money through the insurance businesses he’s gradually acquired.

As the table above demonstrates, his average cost of funds is significantly lower than the United States government. In 60% of the years passed since 1976, he has had negative cost of capital. This structural advantage combined with his factor exposure is the true engine of Berkshire’s extraordinary performance.

Conclusion

Analyses of Berkshire Hathaway are necessarily constrained by the background of the analyst. The quants who identify the factor exposures and degree of leverage driving Buffett’s outperformance are careful not to overstep. They leave Buffett’s thoughts on competitive advantage alone and make no attempt at tracing the source of these ideas. Meanwhile the academic theorists who attack Porter and Competitive Advantage construct strong intellectual rebuttals to the Five Forces, but do not link the failure of these ideas to Buffett’s moats. This paper’s unites these two perspectives to better understand Berkshire Hathaway as a financial institution. To invest like Buffett is not to buy a VanEck MOAT ETF. Instead one must acquire an insurance business, run it successfully and use its insurance float as leverage to buy cheap, safe, quality companies. Survive the inevitable and painful drawdowns, stick to the strategy for 50 years, and a cult of personality awaits you.

Bibliography

"The Five Forces." Harvard Business School, Institute for Strategy & Competitiveness., https://www.isc.hbs.edu/strategy/business-strategy/pages/the-five-forces.aspx.

"VanEck Vectors Morningstar Wide Moat ETF." VanEck., https://www.vaneck.com/etf/equity/moat/overview/.

Archer, Seth. "A Key to Warren Buffett's Investing Strategy is Incredibly Easy to Replicate (MOAT).", http://markets.businessinsider.com/news/etf/warren-buffett-moat-etf-simple-explanation-for-how-he-invests-and-its-easy-to-replicate-2017-10-1005613232.

Asness, Clifford and Liew, John. "The Great Divide Over Market Efficiency." Institutional Investors., last modified Mar 03, https://proxy.library.upenn.edu:7829/article/b14zbgrj5pflsc/the-great-divide-over-market-efficiency.

Asness, Clifford, Tobias Moskowitz, and Lasse Pedersen. "Value and Momentum Everywhere." Journal of Finance 68 (3). 2013. http://pages.stern.nyu.edu/~lpederse/papers/ValMomEverywhere.pdf.

Buffett, Warren. 1993. "Economic Moats." Outstanding Investor Digest, Jun 30,.

Buttonwood. 2012. "The Secrets of Buffett’s Success." The Economist, Sept 29th,.

Einav, Liran. "Empirical Industrial Organization: A Progress Report." Journal of Economic Perspectives 24 (2): 145-162.

Frazzini, Andrea, David Kabiller, and Lasse H. Pedersen. 2013. "Buffett's Alpha." . http://www.econis.eu/PPNSET?PPN=775205737.

Graham, Benjamin and David L. Dodd. 1934. Security Analysis. New York: Whittlesey House, McGraw-Hill Book Co.

Hargreaves, Rupert. "Warren Buffett on the Importance of Moats." Nasdaq., last modified Mar 28, https://www.nasdaq.com/article/warren-buffett-on-the-importance-of-moats-cm767018.

Hou, Kewei,Xue, Chen,Zhang, Lu. 2017. Replicating Anomalies. Vol. w23394. Stanford, CA] : National Bureau of Economic Research.

Munger, Charles. "A Lesson on Elementary, Worldly Wisdom as it Relates to Investment Management & Business " Y Combinator., http://old.ycombinator.com/munger.html.

———. . Vice Chairman's Thoughts--Past and Future.

Porter, Michael. 1998. Competitive Advantage: Creating and Sustaining Superior Performance Free Press.

———. "How Competitive Forces Shape Strategy.", last modified March, https://proxy.library.upenn.edu:2959/1979/03/how-competitive-forces-shape-strategy.

———. . 2002. An Interview with Michael Porter.

Rasmussen, Daniel. "The Gospel According to Michael Porter " Institutional Investors., last modified Nov, https://proxy.library.upenn.edu:7829/article/b15jm11km848qm/the-gospel-according-to-michael-porter.

Schmalensee, Richard. 1989. Inter-Industry Studies of Structure and Performance. Vol. 2 Elsevier.

Schroeder, Alice. 2008. The Snowball : Warren Buffett and the Business of Life. New York: Bantam Books.

Warren Buffett. a. Berkshire Hathaway Letters to Shareholders 1965-2017 Explorist Productions.

———. . b. Berkshire--Past, Present and Future. 2015

Weinberg, Ari. 2015. "Can an Index Fund Deliver the ‘Value’?" The Wall Street Journal, Sept 8,. https://proxy.library.upenn.edu:2729/articles/can-an-index-fund-deliver-the-value-1441764704.

[1] (Munger 1994)

[2] (Archer 2018)

[3] Namely insurance float. More on this later.

[4] (Buffett, 2017)

[5] (Buffett, 2015) Past, Present and Future

[6] Ibid.

[7] Ibid. Buffett likens this investment to “picking up a discarded cigar butt that had one puff remaining in it. “Through the stub might be ugly and soggy, the puff would be free.”

[8] Capital IQ for holdings information

[9] Ibid. I’ve simplified the narrative in the interest of space. For a full depiction please see Alice Schroeder’s excellent The Snowball.

[10] (Buffett, 1965-2017)

[11] An exhaustive discussion of the principles of value investing are beyond the scope of this paper.

[12] (Weinberg 2015)

[13] (Hargreaves 2017)

[14] (Graham and Dodd 1934)

[15] (Munger 1994)

[16] (Asness, Moskowitz, and Pedersen, 2013)

[17] VanEck

[18] VanEck

[19] (Buffett 1993)

[20] SWOT stands for Strengths, Weaknesses, Opportunities and Threats. The idea behind the framework is for the MBA student to identify and list strengths, weaknesses and opportunities of businesses on a case by case basis.

[21] (Porter 2002)

[22] Harvard Business School, Institute for Strategy & Competitiveness

[23] In fact, Porter is the student of the student of the student of the father of industrial organisation: Edward Mason.

[24] (Schmalensee 1989)

[25] (Rasmussen 2017)

[26] Ibid.

[27] Ibid.

[28](Buffett, 2008) Annual Letter.

[29] (Frazzini, Kabiller, and Pedersen 2013)

[30] Compared to the market as a whole.

[31] A caveat: there is considerable evidence much of the empirical finance literature is a product of statistical shenanigans. (i.e. “p-hacking”) Few doubt the existence of the size, value, low-beta and quality anomalies though as they have been proven out of sample across geographies. For a detailed discussion, see Lu Zhang’s Replicating Anomalies.

[32] Academics have provided theoretical justifications for each of these anomalies. For instance low beta stocks are thought to deliver anomalous returns because most investors (mutual funds, etc.) are unable to lever them up to higher volatility. Instead, these investors must buy high-beta stocks, which causes high beta stocks to be relatively overpriced. A complete discussion is beyond the scope of this paper.

[33] AQR sells algorithmic, systematic strategies and therefore has a strong incentive to distort results in favor of quant investing. My assessment is that this risk is limited given that the “novel” factors they use to explain Buffett’s return, “Quality Minus Junk”(QMJ) and “Betting Against Beta (BAB) are widely accepted in academia with strong empirical and theoretical justifications.

[34] (Frazzini, Kabiller, and Pedersen 2013)

[35] Cheapness is usually defined by dividing some measure of firm value, (say enterprise value) by some measure of cash flow (say EBITDA). Quality is usually defined by measure of profit margins or earnings or asset growth. Low-risk is defined by a low beta. (i.e. historical volatility of stock price compared to market as a whole.

[36](Frazzini, Kabiller, and Pedersen 2013)

[37] Ibid.