Aftermath

by James Rickards

  Another trap laid by the choice architects is an all-or-none view when it comes to 401(k) elections. The choice architects are so certain their preferences are the right ones for you they engineer sign-up forms to steer investors toward maximum participation. That’s not necessarily the best choice. If a 401(k) plan has optional participation rates of 5 percent, 10 percent, and 15 percent of gross income (subject to a ceiling), you do not have to elect 15 percent. You could opt for a smaller amount (say, 10 percent), which leaves some discretionary income for investment outside the 401(k) bubble of Wall Street–sponsored funds. In a future crisis of the kind described in this book, these alternative investment paths disdained by choice architects may prove the best wealth preservation strategies.

  CHAPTER FOUR

  The Alpha Trap

  If we were all passive investors, there would be no mechanism to adequately value companies in the market based on their business, and therefore, it would be virtually impossible to trust the values for anything.1

  —Gerry Frigon, “What Would Happen If We Were All Passive Investors?” Forbes (2018)

  Alpha, Beta, … Omega

  Investors are told time and again, “You can’t beat the market.” This dissuasion is both a pillar of modern financial theory and the go-to marketing pitch for the index fund industry. Of course, the proposition is wrong; investors beat the market all the time. And those who beat the market aren’t just lucky, as the professors would have it; they know exactly what they’re doing.

  There are two ways to beat the market other than sheer luck—inside information and market timing. This was shown by Harvard professor Robert C. Merton and a collaborator in two seminal papers published in 1981 by the University of Chicago.2 Merton shared the Nobel Prize in Economics in 1997 for his contributions to the Black-Scholes options pricing formula. It would have been more fitting if Merton had won the prize for his market timing papers. There are serious deficiencies in Black-Scholes, not least of which is the idea of a risk-free asset. In contrast, Merton’s ideas on how to beat the market are nearly flawless and have stood the test of time.

  The claim that investors can’t beat the market is a colloquial form of the more formal efficient market hypothesis, or EMH, a theory most closely associated with economist Eugene Fama. This hypothesis, like most tenets of modern financial theory, is only loosely related to reality, yet holds a powerful sway over academic economists and their Wall Street brethren. EMH says that markets are highly efficient at incorporating new information into prices. If a company announces disappointing earnings, the market instantaneously marks down that company’s stock price to reflect the new earnings outlook. If an energy company makes a large, unexpected discovery of oil and natural gas, the market immediately boosts the price of that company’s shares. It’s simply the case that a single investor can’t benefit from the news in ways that beat other investors. For better or worse, all investors are in the market together and receive the same information at the same time. An investor can win or lose; she can’t outperform.

  Objections to EMH are too numerous to detail in depth here; an overview will suffice. If markets were efficient at incorporating new information as the thesis requires, there would be no flash crashes, panics, manias, or bubbles. Yet those events happen all the time. In 1987, the Dow Jones Industrial Average fell 22 percent in one trading day for no apparent reason. Liquidity crises occurred in 1994 (Mexico) and 1998 (emerging markets) based on fundamental trends that were on full display months before each crisis. Market participants ignored them. Bubbles occurred in 1999 (dot-com stocks), 2007 (subprime mortgages), and 2017 (bitcoin) based on greed and wishful thinking; there was nothing efficient about market pricing of those instruments. Behavioral psychologists have catalogued over 188 cognitive biases from the availability heuristic to the zero-sum effect, which induce irrational behavior relative to the robotic wealth-maximizing android required by EMH. Empirically, EMH lies in shreds even as academics and analysts continue to use it as a bedrock belief for forecasting.

  Yet long before the rise of behavioral economics in the 1990s, and the run of crises from 1987 to 2017, Merton had blown a hole in the EMH edifice. His key was the use of inside information; knowing what the market does not. With insights provided by inside information, an investor could buy ahead of price spikes or sell ahead of drawdowns—an exercise in market timing—and easily outperform benchmarks. Inside information and market timing are two sides of the same coin. Used together these twin tools leave EMH in the dust.

  The term “inside information” raises objections about legality and whether individual trading on inside information is not somehow cheating. The suggestion is that EMH is valid so long as rational actors don’t break the law. The truth is that almost all inside information is perfectly legal. Inside information is defined as material nonpublic information. It’s information important enough to affect prices, yet not known to the market as a whole. For inside information to be illegal, the information must be obtained in breach of some duty; it’s a two-part test.

  If you are on a corporate board of directors and are informed of a pending takeover of your firm, it’s illegal to trade your company’s stock based on that. As a director, you have a duty to keep the takeover information confidential. If you use that inside information to advantage in personal trading, it’s as if you stole a valuable corporate asset, no different than stealing a company-owned car. Of course, such trading happens all the time. Still, it’s illegal and some insiders go to jail. This trading meets the two-part test—purchase or sale of a security based on material nonpublic information obtained in breach of a duty.

  Yet most material nonpublic information is not obtained wrongfully. The information is generated through research and belongs to the party who created the information. Hedge funds use private satellite companies to obtain images of store parking lots taken from space. By comparing those images over time, analysts can ascertain if store traffic is up or down (assuming there’s little pedestrian traffic). If the hedge fund has information on average purchases by customer, average shoppers per car, and vendor margins, it’s even possible to estimate a store’s net income using the satellite photos in advance of any public announcement by the owner. Such information is material and nonpublic, but was not stolen or obtained in breach of a duty. The information was obtained through diligent research by the party who hired the satellite, and that party is free to trade on the information and usually does. It’s perfectly legal.

  The satellite story and many more like it are examples of investors using inside information. Buying the affected security just prior to an earnings announcement, when other public information has been fully incorporated into prices, is an example of market timing. Inside information and market timing used together can beat the market—one more example of how EMH fails; exactly what Merton pointed to in his 1981 academic papers.

  EMH exists in so-called weak, semistrong, and strong forms. The weak form tests your ability to beat the market using historical prices and returns only. Few analysts confine themselves to so little information; research just outside these narrow bounds should produce superior returns. The semistrong form takes into account historical prices and returns plus all other public information. That sets a high bar for investors who try to outperform. The strong form includes all information, historical, public, and private, including the satellite imagery noted above. EMH proponents call it the strong form because outperformance is almost impossible. Yet no single investor could possibly have all the private information; that’s what makes it private. It’s like saying EMH does not work in real-world conditions, but works fine in conditions that don’t exist. As a rule of thumb, whenever a grand theory is broken into subtheories (weak, semistrong, and strong) that’s a good indication something’s amiss with the grand theory. It’s safe to discard EMH as a guide to market behavior.

  Just because it’s possible to beat the market does not mean that most investors do so—they don’t. Th
ere is ample research that demonstrates that not only do active portfolio managers not beat benchmarks, they do worse. This research is the calling card of the index fund industry. Why give your money to active managers, pay higher fees, and underperform popular benchmarks when you can invest in a low-fee index fund, earn the market return, and not sweat the details? Since stocks go up over long periods of time, your portfolio should perform well, especially if you begin to invest thirty years before your planned retirement. You can ride out the market drawdowns, capture big gains on the bouncebacks, and retire on that sailboat or vineyard you’ve always dreamed about.

  Before deciding that index investing is superior, it’s important to understand why the data shows passive investing outperforms active investing. If you don’t know why, then your index fund idea is just a leap in the dark. The index fund industry will trot out EMH as an explanation, yet we’ve already seen that’s nonsense. Markets are not at all efficient; the reason lies elsewhere.

  Next, the passive-investing industry claims that their outperformance is due to lower fees and expenses. Index funds do have lower fees than active funds. If you’re just going to allocate investor money across the components of a popular benchmark like the S&P 500 Index, you don’t need armies of analysts making trips to the headquarters of issuing companies. All you need is a computer and an automated order entry system. Still, lower fees account for a relatively small portion of passive investing outperformance. Another factor must be at work.

  Begin with definitions of two key terms: alpha and beta. Alpha is a measure of return over or under a given index. If your index is the S&P 500, and it returns 10 percent while your investment returns 15 percent, the alpha on that investment is +5. If your investment returns 5 percent, the alpha is −5. Positive alpha indicates the investment outperformed the index on a risk-adjusted basis; it produced excess returns.

  Beta measures the volatility of an investment relative to an index. If an investment produces twice the return of an index, it has a beta of +2. If an investment return moves in the opposite direction of an index, beta is negative. An investment that falls twice as fast as the index rises has beta of −2.

  Alpha and beta are used together to assess portfolio performance relative to the risk taken to produce that performance. A gambler playing roulette with your portfolio can easily double your money by betting on red; if the ball falls on red you win the bet. Of course, if the ball falls on black or green you lose all your money. Some players win at roulette, but most lose. Roulette has negative alpha; returns do not compensate for the risks.

  A passive- or index-investment strategy seeks a beta of 1 (return matches the index performance) and an alpha of 0 (return is consistent with risk). The Holy Grail of investing is to have positive alpha and a beta of 1. That means you are taking market risk, yet getting an above-market return. Index managers who accomplish this attract more assets under management on which to charge management fees. Hedge fund managers with positive alpha charge performance fees that give them a piece of the action for superior performance.

  One huge analytic problem with alpha and beta (and portfolio risk management in general) is the proper selection of benchmarks and the concept of the risk-free rate. If you have a large and highly diversified portfolio of U.S. stocks, then the S&P 500 may well be a suitable benchmark. If a portfolio is sector specific, say, in technology stocks or mining companies, using the S&P 500 will produce meaningless measures of alpha and beta. Likewise, the calculation of alpha requires the use of a risk-free rate of return so that the excess return attributable to manager skill is isolated. The yield on one-year Treasury bills is often used as the risk-free rate. Yet risk-free is a misnomer. Treasury bills’ rates reflect inflation risk and a term premium for risk of default or nonpayment. Those risks are small, but growing. In theory, the true risk-free rate would be zero if a risk-free asset could be identified. Gold comes to mind.

  This brings the analysis full circle. Active investing can outperform passive investing using inside information and market timing as described by Merton. Markets are not efficient and offer ample opportunities for risk-adjusted outperformance measured by alpha. Yet with the exception of a few legends like Bruce Kovner and Dave “Davos” Nolan, active managers do not outperform. The reason has nothing to do with EMH or fees, the two reasons often cited by the passive-investment industry to sell their wares. The two reasons for active-management underperformance are behavioral psychology and a statistical concept called skew.

  A Bend in the River

  Ironically, the same phenomenon that causes markets to be inefficient—behavioral bias—causes active managers to underperform benchmarks. Active managers are people too. In particular, active managers are subject to confirmation bias—the tendency to emphasize information that supports an investment thesis, and to discard information that contradicts it. A related bias is anchoring—the tendency to stay attached to an investment thesis that is primed in memory or experience, and to resist change. Anchoring creates inertia that makes it difficult for an active manager to detect changes in prevailing market dynamics or the macro policy environment, and to adapt an investment strategy accordingly.

  There are myriad examples of highly intelligent active fund managers succumbing to behavioral biases, often due to a lack of cognitive diversity in decision making. Partners at Long-Term Capital Management, including two Nobel Prize winners and legendary fixed-income trader John Meriwether, tripled investor funds from 1994 to 1997, paid out most of the profits in a $3 billion all-cash distribution, then lost 92 percent of the remainder in a few weeks in August and September 1998. The LTCM partners failed to detect a gathering global liquidity shortage and deleveraging at competing firms. These failings were examples of anchoring in a previously successful investment process, and confirmation bias in discarding evidence of market stress coming from Asia. The billionaire hedge-fund manager Bill Ackman is another illustration. Ackman and his partners produced above-average returns throughout the 1990s and early 2000s. Then between 2013 and 2018, Ackman’s firm, Pershing Square, lost over $4 billion in one disastrous investment in Valeant Pharmaceuticals and $1 billion more on a short position in nutrition company Herbalife. Ackman’s fund lost 20.5 percent in 2015, 13.5 percent in 2016, and 4 percent in 2017 as these losing bets were unwound. A host of cognitive biases, including those labeled postpurchase rationalization and selective perception, played a role. None of the geniuses at LTCM or Pershing Square suddenly became dumb. Still, they did succumb to behavioral biases; indeed, the strength of their bias was amplified by prior success, a kind of “I can do no wrong” bias.

  Active managers who do produce alpha over long periods of time are those who do a better job of neutralizing behavioral biases. In my decades of experience in hedge funds, the best manager I ever encountered in terms of taming behavioral bias was Bruce Kovner, the fabled founder of Caxton Associates. Today, Kovner ranks 108 on the Forbes 400 list with a net worth of over $5 billion, and 372 on a separate Forbes list of global billionaires. From 1983 through 2012, Kovner ran Caxton Associates. Caxton averaged 21 percent per year net returns during his time there. At its peak, Caxton managed $14 billion, but Kovner had a practice of periodic multibillion-dollar distributions to investors to reduce Caxton’s size. This was done to maximize returns on a finite set of winning trades. Hedge funds charge management fees based solely on fund size, and performance fees based on alpha. In numerous funds, this dual-fee structure sets up a conflict of interest, where managers earn fortunes on a huge pool of assets even with mediocre or losing performance. Kovner did not believe in size for its own sake.

  I worked for Caxton in the early 2000s after my roller-coaster ride at LTCM in the late 1990s, and had many interactions with Kovner. Despite a congenial demeanor, he was ice cold when it came to risk management; he excluded cognitive bias from investment decision making better than anyone in the business. Interestingly, Kovner’s academic background includes studies at the Juilliard School of
Music; he’s an accomplished harpsichordist. He finds a resonance in markets not rooted in the artificial mathematical constructs of the quants.

  Kovner’s method was old school; not difficult to understand but extremely difficult to practice, because cognitive bias pulls in the opposite direction. Trade ideas began with fundamental and technical research and development of a thesis on the likely performance of the trade. Diverse perspectives were pulled into the discussion to make sure no critical factors were missed. If the trade passed muster, the trader would execute, using futures markets whenever possible to gain leverage and conserve cash. The cash conserved is invested separately to improve alpha on the trade.

  The key to Kovner’s success was tight stop-loss limits, one of the oldest yet most effective risk-management tools. If your trade lost money, you had to close it out, take the loss, and move on to the next idea. Loss limits varied by market. They might be as little as 1 percent in currency markets or 3 percent in stock and bond markets. For Kovner, a loss was nature’s way of telling you that you had missed something in your analysis. If the trade made money, you had a trailing stop, which meant the stop-loss limit moved with the market to make sure you did not give up all your profits if things went in reverse. For example, if you bought a stock at $40 per share, your initial stop loss might be $39 per share, down 2.5 percent. If the stock went to $50, the stop loss might be adjusted to $47, down 6 percent, but still a nice profit on the $40 purchase price. There were no preset limits on how much a trade could make, unless the opportunity cost of tying up cash was an issue. Yet exit strategies were always considered. This style of trading is summed up in the old Chicago rule, “Let your profits run and cut your losses short.” Kovner didn’t invent this system; stop-loss limits have been around as long as liquid tradable markets have existed. Others have used limits successfully, including the Commodities Corporation where Kovner got his start as a trader.