The recent and ongoing confrontation between Russia and Ukraine over Crimea, with its saber rattling and flurry of diplomatic activity, has uncovered a lot of confusion about the causes, motives, and ultimate outcome of this crisis, as well as its likely economic and financial consequences.
I cannot predict any of this. I was, however, able to make one high-conviction forecast early on: The crisis would trigger a bubble in experts on Russia, Ukraine, and Crimea. Indeed, the cocktail circuit, which is where people with more opinions than ideas meet to exchange peremptory platitudes, was promptly inundated with a tsunami of ready-made experts in history, Central European affairs, and geopolitical strategy.
Refreshingly, in a March 5 Washington Post editorial, former Secretary of State Henry Kissinger enumerated a number of principles to end the Ukraine crisis, but was careful to remind us, “…not all of them will be palatable to all parties. The test is not absolute satisfaction but balanced dissatisfaction” (my italics).
Such is the nature of forecasting.
Futility of forecasts
I once mentioned a 2005 study published by the London Business School and ABN AMRO, which showed that GDP growth is not a useful predictor of equity returns. In fact, if I remember correctly, the correlation between a country's stock market returns and the same country's growth in per-capita GDP is often negative: Markets and GDPs tend to move in opposite directions over the short and medium terms. Hence, not only do most economists fail to anticipate economic turning points – a well-known fact – but trying to predict stock market behavior on the basis of economic forecasts seems a futile endeavor in any case.
The same can certainly be said of stock market predictions based on geopolitical constructs. On October 16, 2008, Warren Buffet reminded readers of The New York Times:
In the 20th century, the United States endured two world wars and other traumatic and expensive military conflicts; the Depression; a dozen or so recessions and financial panics; oil shocks; a flu epidemic; and the resignation of a disgraced president. Yet the Dow rose from 66 to 11,497.
This was written after a 35-percent drop in the market, in the midst of the additional downward spiral triggered by the Lehman Bros. bankruptcy, and not too far from the March 9, 2009, start of the currently five-year-old bull market.
Investors were anticipating the end of the world; and once again, one of former Merrill Lynch strategist Bob Farrell’s 10 rules of investing was verified: “When all the experts and forecasts agree, something else is going to happen.”
Do we have a consensus yet?
The February 28 issue of Value Investor Insight reproduces a letter from Seth Klarman (Trades, Portfolio), the publicity-shy chairman of the Baupost Group and one of the world’s most successful hedge-fund managers over the last 30 years. I quote from the article, because I could not put it better:
“The stock market, heading into 2014, resembles a Rorschach test. What investors see in the inkblots says considerably more about them than it does about the market.”
Klarman goes on to lay out impartially the optimistic scenario, which he nevertheless tends to attribute to people who don’t know history or have not learned from experience. He then cites a fairly compelling list of speculative excesses: “A sceptic would have to be blind not to see bubbles inflating in junk bond issuance, credit quality, and yields, not to mention the nosebleed stock market valuations of fashionable companies like Netflix and Tesla Motors.” Finally, he concludes, “We’re pretty sure we’ve seen this movie before. Can we say when it will end? No. Can we say that it will end? Yes.”
“Those who cannot remember the past are condemned to repeat it” has almost become an investment platitude. And indeed, a sense of historical cycles has been a great assist to using one’s common sense. It has its limitations, however, as the inescapable Warren Buffett (Trades, Portfolio) reminds us: "If past history was all there was to the [investment] game, the richest people would be librarians."
If we can’t trust either history or expert predictions to guide our investments, the only reliable criterion left is valuation. But even that is not as straightforward as it sounds.
Being right is not what counts
One of my favorite refrains is that what is important in investing is not to be right, but how much money our decisions can make or lose. And the two do not necessarily go together. Much as I dislike comparing investing to a gambling activity, the odds in the stock market work in a similar fashion to those in horse-race betting.
Let’s assume a horse race with only two horses and 100 bettors each buying a $1 betting ticket, so that the total pot is $100. If 50 bet on horse A and 50 bet on horse B, the 50 winners will pocket the whole $100, or $2 per person. Each winner will have doubled their $1 “investment.”
But let’s now assume that horse A is an all-out favorite and everyone bets on him. One hundred bettors will share the total $100 pot ($1 per person) and will thus have won no money. In fact, by being right with the overwhelming majority, they will have lost money, since there are house expenses that, in horse racing, I am told are in excess of 15 percent.
A similar reasoning holds true for the stock market. When a majority of investors agrees on the attractiveness of a company and buys its stock, the price of that stock rises. In doing so, it reduces the potential gain that would result from continued operational success. It also increases the potential loss if the company disappoints. This is why, whereas investors often interpret a stock rise as confirmation of the underlying company’s attractiveness, it really means that the market is reducing your odds of making money and increasing the odds of your losing money.
Robert G. Kirby once remarked that the label “value investors” was somewhat misleading because it seems to imply the existence of another category called “non-value investors.” In fact, value tends to be in the eye of the beholder: Everyone professes to pay attention to it, but no one calculates it in the same way.
Phil Fisher, high priest of growth investing, once remarked, “The stock market is filled with individuals who know the price of everything, but the value of nothing.” For these poor souls, Warren Buffet cites Ben Graham, godfather of value investing: “Price is what you pay, value is what you get.” So, pretty much everyone agrees.
The usual pattern is that when hope reigns, stock prices rise faster (and farther) than value. But when reality takes over, even a very rosy reality, risks become more visible and prices tend to fall back toward value. The secret to success, therefore, is to buy a stock at a price below your estimated value for the company.
Importance of the batting average
In 1993, the National Bureau of Economic Research published a working paper (No. 4360) entitled “Contrarian Investment, Extrapolation and Risk.” The three authors, professors at prestigious universities, studied strategies of buying stocks with low prices relative to earnings, dividends, book assets, or other measures of fundamental value.
They concluded that their paper provided evidence that “value strategies yield higher returns because these strategies exploit the mistakes of the typical investor.” They further speculated,
Value strategies might work because they are contrarian to ‘naïve’ strategies followed by other investors. These naïve strategies might range from extrapolating past earnings growth too far into the future, to assuming a trend in stock prices, overreacting to good or bad news, or simply equating a good investment with a well-run company irrespective of price.
The interest of this study is to indicate that, by following the contrarian/value criteria identified by the authors, the odds of picking outperforming stocks should increase. But in my view, the question of whether this purely statistical discipline, as practiced by the authors, can be brought down to individual stock picking remains open.
The universe of stocks covered by the study included several thousand companies, the performance of which the authors ranked by deciles (10 groups of 10 percent each). If the total universe in the study comprised 5,000 companies, for example, the decile with the lowest price-to-earnings would include 500 low-P/E companies.
Nowadays, large, comprehensive databases and huge computing power make it relatively easy to construct and track such a large portfolio, so it is conceivable to manage portfolios that way. But I doubt whether it really is the value discipline that is vindicated in the study or whether the experiment benefited from a sample-size effect.
The ultimate contrarian decile
In the study, there was a decile that was about as contrarian as you can get: It was made up of the companies with the slowest growth in sales over the previous five years. Unexpectedly, it significantly bested, in subsequent years, the stock market performance of the decile made up of companies with the fastest past growth in sales. There are two explanations for this seeming paradox.
One is that companies that had achieved superior growth in the previous five years had created high expectations among investors; and in investments, higher expectations translate into higher prices. As I pointed out earlier, rising stock prices reduce potential gains and increase downside risk.
As it happens, in ensuing years, the companies with the highest past growth failed to grow materially faster than companies with slower past growth. The disappointed expectations caused these former fast growers to underperform in the stock market and, by default, the former slow growers outperformed. That was basically the authors’ interpretation.
Another explanation, which is personal, is that many companies in the slowest-growing decile may have been troubled companies on the verge of bankruptcy. Some may have eventually failed, resulting in a 100-percent stock market loss. But others may have survived and, possibly after restructuring or as a result of new strategies, returned to prosperity. From the depressed prices that had reflected low investor expectations, their stocks may have appreciated 200-300 percent or more, more than making up for the companies that failed or nearly failed. These strong recoveries would have allowed the decile to achieve a superior performance. I have no proof of this, but a strong conviction that it probably was a factor.
The computer can’t do all your work
The contrarian/value statistical (“quantitative” in modern lingo) approach is a time-tested method to narrow the field of candidates for investment to a still-large number of stocks that, as a group, have good odds of outperforming the general market. It is not a way to pick individual stocks, which may have very valid reasons for being cheap. Individual stock selection will still require serious financial analysis to determine the operational and financial viability of the cheap stocks in the large statistical sample.
Further, the quantitative approach as described in this paper tells you which decile is cheaper than the total universe. But when the whole universe is too expensive, as in 1972, 1981, 1987, 2000, or 2007, to cite only some extreme instances, all prices are vulnerable. This is what makes me a timid and very selective investor, these days.