The best business to own is one that over an extended period can employ large amounts of incremental capital at very high rates of return.
The quote above is a far cry from value investing as described in such canonical works as Graham and Dodd’s Security Analysis (1934). Those works tended to focus on asset rich companies – especially firms with easy to value assets like cash and working capital. The idea was to buy shares at a significant discount to intrinsic value. A lousy company with easy-to-value assets of $100 per share might still be a steal at $50. A big price discount was your margin of safety. Later Nobel-prize winning work on the “value factor” also was based on an asset test – companies with low price-to-book ratios. Let’s call this Generation 1 of value investing.
Decades ago, the asset approach worked well. As discussed here, investors strongly favored exciting growth stocks over those with plentiful assets but struggling operating businesses. Deep value stocks – called “cigar butts” by some – often traded at very steep discounts. For the investor willing to trawl through financial statements and be patient, it was almost free money. And, in fact, this is largely how Buffett invested in the 1960s when he ran a private partnership: buying (cigar butt) dollars for 50 cents. Berkshire Hathaway itself was a dying textile business when Buffett began to buy stock in 1962.
In the 1980s, Buffett broke ranks with Generation 1 value investors. A 50% discount on a cigar butt looks great, but if the asset value slowly melts away, so will the discount and the investment becomes dead money. Some stocks, unfortunately, are cheap and built to stay that way. Plus, discounts narrowed as more people figured out the deep value game. The free money appeal went away.
Buffett began to focus instead on investing in great businesses at reasonable prices. (Think: Coca Cola.) The fundamental leap: if intrinsic value grows, then the discount today is less important. A 20% discount to intrinsic value today may seem relatively paltry for a Generation 1 investor, but not if intrinsic value is expected to grow 10-15% per annum. Consequently, you get paid to wait: if the stock hasn’t budged in three years, the discount has widened to 40% and you can back up the truck.
This shift was a source of heated debate among diehard value investors in the early 1990s. Had Buffett betrayed the value canon? Then a funny thing happened: other value investors followed suit. To understand this, we need to segue to the LBO boom of the late 1980s. Armed with PCs and new finance theory on valuation (see Tom Copeland’s seminal work, Valuation), armies of investment bankers were taught how to value businesses with discounted cash flow models. Build detailed projections going out five-to-ten years with tons of assumptions about growth rates, margins, sales of non-core businesses, etc. – then discount those cash flows back to the present, and you have a good estimate of what a private equity investor might pay in a takeover.
Generation 2 of value investors used these tools to determine if a given company’s shares were cheap. Intrinsic value was now “private market value.” The new metrics were not asset based, but multiples of cash flow. Find a stock at a substantial discount, the argument went, and you should buy it and wait. Eventually, other investors would figure it out and bid up the price or someone would take the company private. Either way, you win. (The trick, of course, is that intrinsic value of an operating business can also go down. See a post on Valeant.)
There are a few important observations from this. A recent paper by hedge fund firm AQR concluded that much or all of Berkshire Hathaway’s staggering outperformance since the early 1980s is due to something called the “quality” factor. Quality investing is a smart beta strategy of investing in companies with attractive traits like growing earnings, high returns on invested capital and healthy balance sheets (see How Smart is Smart Beta?). Since “quality” companies have outperformed the rest of the market over the past few decades, it’s not surprising that the back-tested performance lines up with Buffett’s actual, live results. The truly notable thing is that Buffett figured this out in advance – decades before academics could “prove” it. Also, given the shift in value investing described above, “quality” looks a lot more like a Generation 2 value strategy than a truly unique factor.
But what worked before often doesn’t work going forward. Academic studies are good at explaining what happened, but too often fail to predict what will work. For instance, strategies based on value, momentum and small cap stocks appeared to perform brilliantly in the decades before the studies were published. Results since then have been very disappointing. Arguably, this moniker “quality” sounds better than “Generation 2 value factor,” since the latter begs the question of why the original one didn’t perform as planned. Given the army of investors who imitate Buffett today, it’s hard to argue that the markets are somehow unaware of the appeal of “quality” companies today.
Rather than exhaustively study the past thirty years, investors are better served by studying how proven, talented managers evolve and adapt in the present.
Take Berkshire Hathaway’s purchase of Deere and Kinder Morgan shares in the fourth quarter of 2015. Given that each company’s operating earnings have been nearly cut in half over the past three years, you can be pretty sure that these companies aren’t at the top of a “quality” screen today. While it’s far too soon to draw any big conclusions from this, it’s a safe bet that if struggling industrial companies offer the best value over the coming decade, Buffett will participate. Then maybe in ten years some academics will write papers on a new factor (“Cyclical Contrarian”?) and wrap it in a smart beta product with promises to perpetually outperform. That will be cold comfort for investors who were buying yesterday’s factor du jour after the smartest investors had moved on.