Prior the existence of Berkshire Hathaway (BRK.A) (BRK.B) or its predecessor, the Buffett Partnership, there was the Graham Newman Partnership. Ben Graham was one of the financial wunderkinds of his age. His partnership posted annualized returns of about 20% from 1936 to 1956, far outpacing the 12.2% average return for the broader market over that time.
The problem with great men is that their errors can be equally great. One all-too-common mistake that value investors make is emulating Ben Graham's emphasis on tangible assets such as real estate and heavy machinery. However, what appears to a Graham-school value investor to be "safety" can also act as an anchor on the company's profitability, ultimately resulting in dead money for long term investors.
The logic of Wall Street is proverbially weak. It is hardly consistent, for example, to despair of the railroads because the trucks are going to take most of their business, and at the same time to be so despondent over the truck industry as to give away shares in its largest units for a small fraction of their liquid capital alone. - Forbes Magazine: "Is American Business Worth More Dead than Alive?" by Benjamin Graham (1932)
In February of 1932, the Governor of Michigan declared a banking holiday in February, triggering a bank run across and a panic in the stock market as credit dried up. Within months, hundreds of companies were selling for below their liquidation value. Graham, who had already lost 70% of the Graham Partnership's money, decided to back up the truck. With just $375,000 left, Graham started buying distressed companies for less than their net tangible assets. If he was wrong, he had no other recourse than shareholder activism and a proxy fight. But Graham was right.
It was impeccable timing, too, as the chart below illustrates.
Fig. 1: The Great Crash
Businesses logically are worth far more than net tangible assets when they can be expected to produce earnings on such assets considerably in excess of market rates of return. The capitalized value of this excess return is Economic Goodwill. - Chairman's Letter, Berkshire Hathaway (1983)
Graham's experience with the Great Depression led him to design a ruthlessly defensive approach to stock picking (Net-Net) that continues to appeal to a large swathe of the investing public today. Nonetheless, after losing money on a number of his early ventures, Buffett realized that his mentor's focus on Plant, Property & Equipment was inherently flawed.
The problem was inflation.
Let's say you're thinking of purchasing one of two companies. Company A (a no name brake-pads company) is valued at $160 million and also has $20 million in annual earnings, but requires $160 million in net tangible assets to generate those earnings. Company B (a brand name bubble gum company) is valued at $240 million with $20 million dollars in earnings, and requires $80 million in net tangible assets to generate those earnings.
Which company should you buy?
If you follow Graham's advice, the answer is A, as the company is selling for the breakup value of its tangible assets, while Company B is selling at x3 the value of its tangible assets. (For simplicity's sake, we'll assume that the Economic Goodwill of Company A is $0.)
If you follow Buffett's advice, the answer is B.
Why? Let's assume that both Company A and Company B are still in business 10 years from now. For the sake of argument, let's assume that both Company A and Company B have to double their prices over the next decade in order to keep pace with inflation and that production volumes remain flat. We'll also add a real touch of fantasy to our calculation by assuming that wages will increase roughly in line with inflation.
Now, fast-forward ten years: Both bubble gum and break-pads now cost roughly twice as much as before, but everyone is getting paid twice as much which means that purchasing power hasn't changed and so no one is complaining. (The actual cumulative rate of inflation from 1913-present by decade is given in Fig. 2 below.)
Fig. 2: Cumulative Rate Of Inflation (By Decade)
All else equal, both Company A will require re-investment in tangible assets, as will Company B. The value of Companies A and B have also doubled, to $320 million and $480 million, respectively.
If you bought Company A (Graham's pick), you will now have to cough up another $160 million, even though your net profits over the last ten years equal only $200 million just to maintain a company that, all else equal, produces only 25 cents per $1 re-invested every 10 years, or $40 million. Not exactly an enviable rate of return.
On the other hand, if you bought Company B (Buffett's pick), you only have to cough up another $80 million vs. $200 million in profits over the same time period. That's $1.20 in profit for every $1 re-invested in the company every ten years, or $120 million. Meanwhile, brand name Company B is now worth $160 million more ($480 million) than no-name Company A ($360 million).
Total Return on Investment over 10 years
Company A (Graham's Pick): $40 million. ($200 million in earnings + $160 million in valuation minus $160 million in re-investment expenses.)
Company B (Buffett's Pick): $330 million. ($200 million in earnings + $240 million in valuation - $80 million in re-investment expenses - $30 million in Goodwill amortized over 10 years of the 40 year period allowable by law.)
Now, you can certainly question the assumptions underlying our little thought experiment. For starters, Economic Goodwill is unlikely to be $0, however blasé Company A's business is. Company A is much more likely to receive local, state and federal subsidies or special tax breaks due to the industry it serves and probably has a higher barrier of entry in purely nominal terms.
However, none of these objections can paper over the fact that heavy machinery composed of various moving parts will ultimately break down however well maintained and must be replaced at a generally higher nominal cost than a company with a competitive advantage and comparatively less equipment to replace.
For example, Varian enjoys a narrow but deep moat due to a combination of limited competition in linear accelerator manufacturing and high switching costs within the oncology industry. Urban Outfitters continues to post impressive revenue and gross margin growth in a tough retail environment, has one of the most recognizable names in vintage and bohemian fashion, and appears to be one of the few retail outlets that is actually thriving due to the internet, rather than in spite of it.
To be fair, this flaw in Graham's approach is largely a product of his 2-3 year time horizon, and can therefore only be considered a "mistake" if you intend on holding companies long enough for inflation to be a significant factor.
However, there's a price to be paid by rotating stocks through a near term portfolio that goes beyond transaction fees, dividend hikes and taxes on capital gains: Namely, you miss out on all the benefits of being long (upward drift, positive black swans, long-term demographic advantages, et cet.)
If Warren Buffett's career at Berkshire is proof of anything, it's that paying a fair price for outstanding companies with brand loyalty, pricing power, historically high ROE and limited PP&E tend outperform middling companies with lots of plant, property, and equipment over time, even when the latter are being offered by Mr. Market at fire-sale prices.