The experts told us that Brexit would never happen. They told us that Trump would never win the Republican primary much less the general election. These are only two blaring examples of a larger truth: experts are absolutely terrible when it comes to predicting the future.
The psychologist Philip Tetlock ran a 20-year-long study in which he picked 284 experts on politics and economics and asked them to assess the probability that various things would or would not come to pass. By the end of the study, in 2003, the experts had made 82,361 forecasts. They gave their forecasts in the form of three possible futures and were asked to rate the probability of each of the three. Tetlock found that assigning 33% probability to each possible alternative future would have been more accurate than relying on the experts.
I began my career at Bridgewater Associates in mid-2008 and went to Bain Capital in mid-2009 at the height of the financial crisis. I started my career at a time when the expert judgment of the financial community had been thrown into question.
I had to answer for myself an obvious question: if the future is unpredictable, how are we as investors to make decisions, since, ultimately, investing is about betting on the future? Answering this question led to the creation of Verdad and a new investing methodology, focused on systematically betting on unpredictability and betting against those who are too confident in their own predictions.
At a time when most investors are focused on reducing risk, the Verdad strategy embraces volatility — in 2013, the strategy was up 86%; in 2014, down 10%; in 2015, down 22%; and in 2016 up over 40%. The net result is astonishing out-performance of the market. But beating the market the Verdad way requires a change in temperament and expectations — seeking out risk and uncertainty and thriving from it.
The roots of modern investment analysis come from the 1930s, when a group of young scholars attempted to figure out why markets were so volatile and what had caused the great crash of 1929.
John Burr Williams, a young PhD student at Harvard, was at the vanguard of this project. Like many of his age (and of our own), Williams believed in the promise of technocratic governance — that the world’s problems could be solved by men of solid intellect meeting in wood-paneled rooms to weigh in on the matters of urgency for society.
Williams hoped for a future in which the hurly burly of the trading floor would be replaced by an atmosphere more akin to Widener Library. Bespectacled experts would gather in Cambridge, develop long-term forecasts for every company in every industry, and agree on the fair value of each company.
To translate these forecasts into a price, Williams invented the dividend discount model (now known as the discounted cash flow model): forecast the entire future, calculate how risky that future is, and, voila, you have a stock price.
“Gradually as men become more intelligent and better informed, market prices should draw closer to the values given by our theory,” he wrote optimistically.
But nearly 80 years later, with almost every bank and fund on Wall Street using Williams’ models, men are no more intelligent or better informed, and stock prices remain stubbornly volatile. We have more experts armed with ever-more-complicated dividend discount models, but the volatility and unpredictability of stock prices remain — as evidenced not just by the financial crisis of 2008 but so many market events this year.
Robert Shiller won the Nobel Prize for identifying the problem and conclusively proving that Williams’ dividend discount model was a failure. Shiller took historical earnings, interest rates, and stock prices, and calculated the true price at every moment of every stock in the market with the benefit of perfect hindsight. He found that doing this could explain less than 20% of the variance in stock prices. He concluded that changes in dividends and discount rates could “not remotely justify stock price movements.” Far from being able to explain the future, experts armed with historical data couldn’t even explain the past!
So what does explain the volatility of stock prices, if not discount rates and future earnings?
The answer lies in unpredictability. A stock’s price at any given moment reflects the median forecast of future earnings of all the traders of the stock. And, if you read many analyst reports, you know how many different forecasts there can be for a given company. When the future unfolds, it proves many of these forecasts wrong and only one right. Stanford professor Mordecai Kurz proved mathematically that incorrect forecasts could explain up to 90% of stock price volatility. These incorrect forecasts are what’s really driving the markets.
There’s an old joke about economists. A can of soup washes ashore. The physicist says, "Lets smash the can open with a rock." The chemist says, "Let's build a fire and heat the can first." The economist says, "Assume a can-opener..."
This is essentially the methodology most investors use: rather than saying “the future of these companies is completely unpredictable,” they say, “okay, let’s make some reasonable assumptions about the future based on what we know and historical trends to price this stock.”
The problem with assuming predictability is that it doesn’t work — and the track record of those who use this type of methodology is strikingly bad. A recent study found that 70% of actively managed funds have failed to beat their benchmark index and just 2.3% have delivered excess returns of more than 2.5% — and those are pre-fee numbers. Fees make the picture even worse!
Experts can’t predict the future — in fact, Shiller proved that their models cannot even explain the past. It should therefore come as no surprise that the active fund management industry that uses dividend discount models and other 1930s-era theory should have so thoroughly failed investors.
In order to beat the markets, then, we need a strategy that is built around unpredictability — that takes advantage of these systematic forecasting errors.
In 2011, I worked on a team at Bain Capital that was tasked with analyzing what worked and didn’t work about private equity and about Bain Capital’s own investment process. After examining 2,500 deals representing $350 billion in invested capital over 30 years — a data trove not easily available to investors outside the secretive PE business — we came to several startling conclusions.
About one-third of the deals analyzed accounted for more than 100% of profits (no surprise there) and the majority of the deals in the sample fell well short of the forecasts built into the financial models. The biggest predictor of whether a company would be a big winner or not was the purchase price paid. The dividing line seemed to be 7x earnings before interest, taxes, depreciation and amortization (EBITDA). When PE firms paid more than 7x EBITDA, their chance of success plummeted — regardless of how much managerial magic they threw at it. The 25% of the cheapest deals accounted for 60% of the profits. The most expensive 50% of deals accounted for only about 10% of profits.
In other words, all the fancy analysis and financial models performed worse than the simple rule “buy all deals at less than 7x EBITDA.” A simple quantitative rule worked better than expert judgment.
For me, this was a startling conclusion that suggested a need for a radical reappraisal of my career — and of my approach to investing. Rather than betting on stocks based on predictive models of the future, this data pointed to an alternative methodology: bet on stocks based on their characteristics at the time of investment. This methodology had the benefit of being much simpler and much easier to test than the fancy case-by-case analysis we were doing at Bain Capital. I could look at the history of the markets and discern, for example, whether betting on companies with low EBITDA multiples was a good idea or a bad idea, rather than having to analyze each company individually.
What's interesting about this approach is that it's supported by reams of data from academic psychology. The legendary scholar Paul Meehl claimed in 1954 that simple rules were superior to human judgment in predictive tasks. And, over 60 years later, a vast array of research has proven that he is systematically correct: algorithms make better predictions across a range of fields than do experts.
Since this study, I have been systematically evaluating different investment rationales — rules just like “buy all deals at less than 7x EBITDA.” My goal has been to establish some new principles of investing with universal application — an approach more akin to quantitative investment strategies than to the fundamental investment strategies so predominant on Wall Street.
I designed a strategy that took the characteristics of the most successful LBO deals from the database and screened for them in the public markets. I wanted to focus, as we did in our study at Bain Capital, on private equity. I wanted to focus on private equity for one simple reason: private equity is the one of the only forms of investing that has significantly outperformed broad stock indices over long periods of time. Over the past 20 years, private equity has returned 13% per year, relative to 8% for the S&P500.
There are a few things that distinguish private equity from the broader markets. PE deals have an average market capitalization of about $200 million versus S&P 500 companies at about $35 billion, PE deals are leveraged about 60% while S&P 500 companies are leveraged at only 3%. So, by investing in small, highly leveraged and cheap companies, I could quickly build portfolios that looked exactly like the best performing PE deals from our database.
Four years in, the evidence suggests the strategy works. But while investing in this way can be very profitable — our strategy has generated about 17% annualized returns since 2012 — it requires a change in temperament, an embracing of the volatility that scares others away, a celebration of unpredictability when what we most want is certainty, and a deep contrarianism to embrace businesses that other investors are most confident will perform poorly.
The Verdad strategy for profiting from the unknown unknowns is based on following three simple rules.
1) Other investors must be very pessimistic about the future prospects of the company — and this pessimism must be reflected in a very low valuation. The valuation multiple is a quantitative metric that indicates both the direction and the level of confidence of the experts, and an extremely low multiple means both strong negative sentiment and high confidence in that pessimistic prediction.
2) The company must be highly leveraged, with a large amount of debt on the balance sheet, just like private equity deals. The reason for this is that if the unpredictable happens, leveraged stocks will react much more strongly than unleveraged stocks (just think of a house: if you buy the house with all cash and the price increases 10%, you made 10%, but if you bought the house with 90% debt financing, and the price increases 10%, you made 100% on your money).
3) The company must have the creditworthiness and financial strength to continue servicing and paying down its debt and survive long enough so that unpredictable events can happen. The old saw is that markets can remain irrational longer than you can remain solvent. We need our companies to stay solvent long enough that they can see enough news cycles and have the highest chance of a positive surprise.
I have designed algorithms that quantify each of these rules based on an in-depth analysis of the history of the financial markets and rigorous testing of each rule and each quantitative range to ensure that the rule is statistically significant and would have applied over long periods of time. I rely on forecasting techniques that are proven to work, rather than dividend discount models and other failed technology.
And I implement these rules in the markets, buying and betting on companies that other people hate. I own second-tier hotel chains, Japanese propane delivery companies, declining coupon printing businesses, British pub chains, and a variety of other companies that would terrify you if you were to put all your money in them.
The brilliance of my strategy is holding enough of these terrifying businesses that my odds of having a few of them succeed is high enough to not only make up for all the losses but make a significant profit as well. So while many investors have portfolios of stocks that go up or down 10% or 20%, I have a portfolio of stocks that might go up 100% or down 40% — stocks that frighten other investors precisely because of their unpredictability and the wide range of potential outcomes.
One fascinating study looked at the impact of news events on companies. The study found a very interesting pattern: news events affected different categories of stocks very differently. For the stocks that investors were most optimistic about — the high-priced glamour stocks (think Facebook, Google, and Amazon) —news events had a high probability of negatively impacting the stock price. For low-priced value stocks which investors were most pessimistic about (newspapers, printing companies, etc.), news events had a high probability of having a positive impact on the stock.
In other words, as the future unfolded unpredictably, surprises disproportionately benefitted stocks that investors were pessimistic about and disproportionately damaged stocks that investors were optimistic about — thus proving that investors were systematically too confident in their judgment and that there is a way to systematically benefit from unknown unknowns.
By betting on unpredictability — and embracing volatility — investors can systematically profit, taking advantage of some of the most obvious failings of the broader market: too much hubris, too much faith in experts, too much faith in predictability, and an unwillingness to endure volatility.
This correspondence is being furnished by Verdad Advisers, LP (the “Firm”) on a confidential basis to investors in Verdad Leveraged Company Fund, L.P. (the “Fund”) and does not constitute an offer, solicitation or recommendation to sell or an offer to buy any securities, investment products or investment advisory services. This correspondence is being provided for general informational purposes only, and may not be disseminated, communicated or otherwise disclosed by the recipient to any third party without the prior written consent of the Firm.
The information, charts, and models presented herein contain data that was back-tested using the investment strategy of the Firm. Back-testing is done using Portfolio 123 on domestic data only. Investors are hereby informed that the Firm only began offering the given services after the performance period depicted by the information, models, and charts herein. The model performance results do not represent the results of actual trading but were achieved by means of the retroactive application of a model designed with the benefit of hindsight.
The projections or other information generated herein regarding the likelihood of various investment outcomes are hypothetical in nature, do not reflect actual investment results and are not guarantees of future results. Hypothetical performance results do not take into account the deduction of advisory fees, nor the reinvestment of dividends and earnings. Hypothetical performance results have many inherent limitations. No representation is being made that any account will or is likely to achieve profits or losses similar to those shown. There are frequently substantial differences between hypothetical performance results and the actual results subsequently achieved by any particular trading program.
This document may contain forward-looking statements and projections that are based on our current beliefs and assumptions and on information currently available that we believe to be reasonable, however, such statements necessarily involve risks, uncertainties and assumptions, and investors may not put undue reliance on any of these statements.
An investment in the Fund will involve a significant degree of risk, and there can be no assurance that its investment objectives will be achieved or that an investment therein will be profitable. The Fund-level performance presented herein reflects the reinvestment of dividends and other earnings, the deduction of all management fees, performance-based allocations, brokerage fees and other expenses applicable to the Fund. Investors will experience individual returns that vary materially from those illustrated in this presentation depending on various factors, including the timing of their investment in the Fund, the level of management fees and performance allocations, the effects of additions and withdrawals from their capital accounts. Certain of the performance information presented herein are unaudited estimates based upon the information available to the Firm as of the date hereof, and are subject to subsequent revision as a result of the Fund’s audit. Past performance is not necessarily indicative of the future performance or the profitability of an investment in the Fund. An investment in the Fund will be subject to a wide variety of risks and considerations as detailed in the offering documents of the Fund. The information set forth herein will be qualified in its entirety by the information set forth in the offering documents relating to the Fund.
References to indices herein are for informational and general comparative purposes only. Indexes are unmanaged and have no fees or expenses. An investment cannot be made directly in an index. The Funds consist of securities which may vary significantly from those in the indices referenced herein and performance calculation methods may not be entirely comparably. Accordingly, comparing results shown to those of an index may be of limited use. There will be significant differences between such indices and the investment program of the Fund. The Fund will not invest in all (or any material portion) of the securities, industries or strategies represented by such indices. References to indices are not intended to suggest or otherwise imply that the Fund will, or is likely to, achieve returns, volatility or other results similar to such indices. Indices do not reflect the result of management fees, performance-based allocations and other fees and expenses. . The S&P 500 benchmark is an index designed to measure the equity market performance of 500 large companies having common stock listed on the NYSE or NASDAQ.
The information in this presentation is not intended to provide, and should not be relied upon for, accounting, legal, or tax advice or investment recommendations. Each recipient should consult its own tax, legal, accounting, financial, or other advisors about the issues discussed herein.