In this exclusive article for the Edge, angel investor David Siegel argues that investors often believe that a strong track record points to skill, rather than luck in the markets. His six myths challenge many commonly-held beliefs of portfolio construction. He lays out three fundamental drivers that may have a better chance of powering portfolios than modern portfolio theory, ending with a description of antifragile portfolios that will interest most institutional investors.
All portfolios have at least one weakness. Many portfolio managers think they have their weaknesses covered, but as Nassim Taleb points out in his book, Fooled by Randomness, most managers build beautifully detailed models based on bad assumptions and make the mistake of thinking that their models reflect our complex world. Here are six common myths about portfolios and investor alpha that every family office and institutional investor should understand …
1. The Track-Record Bias
Would you put a portfolio together consisting of the highest-performing mutual funds of the last five years? You probably wouldn’t, because you know that these funds have more downside than upside. You know that the environment that powered these funds to success may not continue. The same goes for most hedge-fund managers. There are about 10,000 hedge funds investing around $2 trillion worldwide. Let’s suppose that 200 hedge funds have fantastic track records simply by being lucky, and perhaps 20 of those are actually good at investing. If you’re running a family office or making large investments, the only funds you will see are those with good track records. All of these funds will have a story – a story of cause and effect, showing with powerpoint slides and data that they know what they are doing. Managers with less than excellent track records won’t make it to your desk.
We know that hedge funds as an asset class do not outperform the broader markets. Assuming a few hedge fund managers actually have skill and can pick the winners, can you tell, going forward, who the winners will be? Almost all the research to date says you can’t tell skill from luck by looking at track records. Statistically, there simply have to be a fairly large number of funds with great track records and a good story – there always have been, and there always will be. Investors and hedge-fund marketers are fooled by this bias and are surprised when, sooner or later, the “absolute returns” stop returning.
Many track records are compared against a straw-man benchmark, like the S&P 500. The problem with many of the big indexes is that they are cap weighted, so companies with high valuations are over-represented. Equal-weight indexes tend to outperform cap-weighted indexes, and many mid- and small-cap indexes do even better. So if a fund has outperformed the S&P, better to ask whether it has outperformed a better index, after taxes.
2. The Smart Guys Can Pick the Winners
Buying public equities is fool’s gold. No one has been shown to be a consistently good stock picker, not even Warren Buffett. Sometimes he does, sometimes he doesn’t. It depends on the measuring stick and time frame you use. A portfolio of carefully selected public stocks is either going to go up or down. If it goes up, it gives you false confidence that you know what you’re doing. If it goes down, you blame external factors. Now that we’re coming to the end of an unprecedented period of government injection of cash into public markets, do you really think the strategies that have been successful in the past four years will outperform in the future? Is Warren Buffet a member of an elite group of superinvestors who know value when they see it, or has he just been lucky? Has Ray Dalio really built a cause-and-effect machine, or has he gotten lucky? In Dalio’s case, my guess is about half of each, yet he tends to take full credit for the good years. Long/short funds with active managers must get lucky to outperform the markets. And some do.
3. Successful People Know How the World Works
We learn much more from failure than we do from success. Success is a poor teacher. Anyone who has had success also has a story about how he earned it. We hear it all the time, from Ashton Kutcher to Donald Trump. Their message is: work hard, play fair, and watch for good opportunities. They don’t have any idea how many people out there follow the same advice and the opportunities don’t come their way, or they just don’t get as lucky. Luck plays a huge role in success. Few billionaires are willing to admit that if things had gone just slightly differently, they would not have 90% of their fame and wealth.
Let’s take a statistical look at wealth and luck. This article claims that 45% of billionaires are in the top 1% of cognitive ability, showing that billionaires are more than hardworking people who got lucky. They are also incredibly smart. Let’s break that down.
First, the other 55% of billionaires may be smart, but they aren’t in the top 1% – how did they get their money? By being almost super smart? Are these people ranked lower in total assets than the 45% who are said to be smarter? Seriously, you can imagine this correlation if you want to, but it’s more likely that there is tremendous variance here, and that, as Michael Mauboussin points out, you need both skill and luck to be above-average successful.
Let’s go back to the 45% of billionaires who got good grades and performed well on standardized tests. In the developed world, where the best schools are, there are about a billion people, so the top 1% pool represents ten million people. Accounting for children and non-business people, let’s take it way down to 100,000 – the top 1% of the top 1%. This must be a group of very smart people. There are about 1,600 billionaires in the world. 45% of that would be about 700. So, out of about 100,000 super-smart people on earth, 700 have become billionaires and 99,000 haven’t. Where is the cause-and-effect in that? Did all those other people end up doing very well for themselves but made under $1 billion, or is this author forgetting the base rate and measuring the wrong things?
One of my favorite examples is Richard Branson, who quite simply has gotten very lucky several times. Someone has to, and he’s the poster child. A great and fun guy, to be sure, and smart enough to hire people who execute well, but the vast amount of his wealth can be attributed to timing. He sold his retail empire to raise money for his airline at exactly the time when retail music was falling off a cliff and air travel increasing exponentially. There’s really no such thing as a timing genius. Timing is essentially doing what other forward-thinking people are doing and being the lucky one. Out of millions of struggling entrepreneurs, a handful come out with a few back-to-back trades that land them on the Forbes list. Richard Branson is one of them. If he were really good at timing, he could have made a hell of a lot more money than he has. Richard Branson has been the fortunate victim of a few positive black-swan events. Whoops – I should have said Sir Richard. How many hard-working people have been knighted for simply being in the right place at the right time?
It has been said that George Soros has been skillful in timing the markets, because he has made so many individual trades. But a power law applies – a small number of very big bets went his way and generated most of his cash. The vast majority of his trades could have gone either way without significant impact on the portfolio. Not surprisingly, in the few hugely successful trades, he had a systemic advantage. There is skill involved, but there is also a lot of luck.
4. Private Equity is the Place to Be
It’s getting harder and harder to get killer deals in private equity. There are so many PE shops scouring the globe that getting the right thing at a bargain price is wishful thinking. If the price is a bargain, it’s that way for a reason. Private equity is becoming more and more like public equity. The Internet is making this asset class more efficient every day.
One exciting development for people in private equity is that we’re learning much more about how to run companies than we ever knew before (and than they still teach in business school). By becoming lean and agile, most companies will become far more efficient and innovative, able to better keep up with today’s increasing pace of innovation. I have written a book about this and have a web site dedicated to business agility. Since most PE fund managers have very little operational experience, they should seek out people who can help them increase productivity, employee satisfaction, and customer wow using the principles of business agility.
5. Smart Venture Capitalists Beat the Markets
According to the Kauffman Foundation’s excellent report, We Have Met the Enemy, and He is Us, venture capital is a poor asset class. The people who think they can pick the winners are fooling themselves, as we have seen above. This article from CB Insights shows that, even though all the VCs think they can pick the billion-dollar companies beforehand, the data shows otherwise. In fact, recent studies have shown that venture capitalists have a strong bias toward funding good-looking white males.
It’s important for investors and venture capitalists to understand that a venture capital fund is not a machine that generates either companies or profits. A VC fund can only do so much. One thing they can do is get cash to entrepreneurs who probably can get it from other sources (they rarely fund companies that don’t look tasty, trendy, and backable). Another thing they do is give advice that the entrepreneurs can probably get from other sources. And a third thing the VCs do is help them find exits that are good for the fund (the sooner the better). Any venture capitalist who thinks he can steer his fund into the top quartile of returns, above the dividing line between 1st and 2nd quartile, doesn’t understand statistical variance. Once again, 25% of funds HAVE to be above that line, by definition, and several funds have found themselves above that line several years in a row, but cause and effect is dubious at best. It’s likely that in ten years of venture investing a fund will be able to point to two or three deals that contributed to most of the profits, and if you look at those deals carefully you’ll see that timing and luck had a huge amount to do with the outcomes. Bessemer Ventures has published a list of companies they passed on, any one of which would probably have outperformed their entire portfolio.
Believe it or not, there are ways to make money in venture capital, but not the way most VCs do it. You need to harness the power of convexity – making small investments that have big payoffs, capturing beta, rather than alpha. A few funds are already on this track, and a couple of them actually have the statistical understanding to make it work. More on that in a minute.
6. The Future Looks Like the Past
The next five years in investing are going to be nothing like the last five years. Mark Spitznagel has a compelling argument for our current stock market situation being another house of cards, ready for a significant correction. Why do we keep falling for the rosy scenarios when things are good and then believing the world is over every time markets collapse? Neither is true. Mean reversion is more powerful than prediction. If you believe in mean reversion, then you must admit that the road ahead is extremely risky. It’s impossible to predict the future, but it’s very likely that some huge corrections are in store, somewhere, at some time in the next several years. There is almost zero percent chance that the next five years will look like the last five. There is hidden inflation that isn’t reflected in the official numbers (we’re not really measuring inflation correctly to begin with). There is hidden unemployment. There are dangers in the increasing gap between ultra wealthy and middle class. In short, it’s the unknown unknowns that often show up as black-swan events and trash portfolios, even portfolios based on MPT.
Modern Portfolio Theory has disappointed many of its customers, and that’s because we don’t live in a world with normally distributed events. We live in a world of black swans, complexity, emotion, and surprises. Perhaps the biggest myth in portfolio construction for the past twenty years has been the belief that modern portfolio theory models the world and its uncertainties accurately. It doesn’t. It’s best to remember that all models are wrong – it’s a matter of how wrong – and we don’t get to find out until later.
To put these myths into a larger perspective, Doug Hubbard, author of The Failure of Risk Management, explains:
In response to the 2008 financial crisis, several of the major consulting firms and standards organizations have charged in with a variety of “solutions” for risk management, none of which is better than consulting astrologers. The worldwide financial system remains as interdependent, fragile, and poorly understood as ever.
Investor alpha is largely an illusion, a story told by people who have gotten lucky. I believe one should hope for the best but be prepared for the worst, taking the growing list of human cognitive biases into account. I think it’s worth looking at the following things as fundamental portfolio drivers:
1. Mean Reversion
You can capture the mean of a given asset class, or bet against people who are foolish enough to think they can beat it (fragile). This is a smart strategy, but you must understand its weakness, because every strategy has one. Mean reversion strategies are usually sellers or buyers of insurance. The sellers carry tail risk, which means that they can blow up suddenly if things go dramatically the wrong way. The buyers carry erosion risk, in which they pay premiums for years, waiting patiently for a correction. After five years of governments propping up markets, I would be more of a buyer than seller of insurance at this point. But you never know how far markets will go before they realize there is no visible means of support.
2. Systemic Advantage
A few investors have a systemic advantage, usually from their position in a complicated market. They often know their market better than anyone, and have a positional place in the market that gives them access to arbitrage and better-than-average deals. Systemic advantage breaks into these categories:
Structural advantage, where a group has access to the market infrastructure that others don’t. A good example is high-frequency trading, the way it’s done today, and which has recently been described by Michael Lewis in his book Flash Boys. Much of this high-frequency trading may turn out to be illegal, most of it is certainly unethical.
Positional advantage, where you function as a gate keeper or market maker. This is an advantage Soros had for many years. Simply by having visibility to everything and having everything go through your doors is an advantage. Another edge for Goldman Sachs.
Technological advantage, which gives you an edge over competitors. One of the original high-frequency traders is Thomas Peterffy, who built a machine that could type orders in faster than humans and who later introduced hand-held computers to floor trading. Most winning strategies these days use statistics and fast computers to find small discrepancies in prices.
Pricing knowledge. There are inefficiencies in mortgage markets, commodities, convertible bonds, and many others, where complex products require an investment in understanding price. This is different than waiting for things to go down. It has to do with sophisticated models that help find underpriced assets.
Information arbitrage. This has been the percieved edge of many event-driven funds, but in reality it is very rare to turn information into cash in real-time. There are many false signals. It’s difficult to know what information to act on. Even Roger Ehrenberg, one of the big quants on Wall Street, tried to start a purely algorithmic trading fund and closed it because they couldn’t get it to outperform.
Speed. In some cases, speed and agility can create an advantage. Quant funds that can execute quickly don’t need a structural advantage to make a profit, but the arms race forces ever faster execution as the competition catches up.
A track record of systemic trades is more meaningful than a track record of stock picking. It’s not as meaningful as the FX traders would like you to think, but since human judgment is removed, it has more chance for all the trades to add up individually and tell a story. Unfortunately, a given investor’s systemic advantage rarely lasts forever. Better to have more of them in your portfolio than fewer.
3. Antifragile Portfolios
You can own what most other people own, and then all of that will go down when the next crisis hits. Or you can build an antifragile portfolio – one that benefits from volatility and uncertainty. An antifragile portfolio is a bar-bell strategy that operates on the principle of convexity and diversification. It’s very hard to hurt a portfolio like this. It’s main weakness is dead calm, which, since all portfolios have a weakness, is probably the best one to design for.
I want to explain what an antifragile portfolio is in more detail. Let’s assume I have $1b to invest. I could invest in a bunch of indexes and ETFs, but the volatility would be high. I could try to invest in many of them, hoping to average out that volatility, but unfortunately, when the world ends, which it does roughly every ten years, everything tends to correlate. I could buy a combination of stocks and bonds – what I call a middle-of-the-road portfolio, but the wrecking ball tends to crush those portfolios on a fairly regular basis, because, once again, you can’t predict the future.
On the other hand, stock markets do go up over time, but when you get in and when you get out are the determining factors in performance. If you could go back in time and invest at the low point, that’s the way to do it. But would you put most of your money into a stock-and-bond portfolio today? I hope not.
So let’s start by taking 80% of the cash and putting it into “can’t go down” assets, like treasuries and certain kinds of rent/royalty-generating assets that tend to hold their value, even in difficult times. This part of the portfolio won’t outperform in good times, but it won’t go down when the shit hits the fan, either.
Now let’s look at the other 20% of the barbell, which amounts to $200m. For teaching purposes, I could buy $1m worth of gold and sell $1m worth of gold, and the result would be that I’d lose money on the transaction fees and make nothing on the trade. Furthermore, I could leverage that up both ways and the same thing would happen, though I would lose more paying the interest on the leverage. However, suppose I now buy $1m worth of options that gold will go up in the future, and $1m worth of options that gold will go down in the future. Let’s say each of these options pays off 10:1. Now, if gold goes up or down, I am guaranteed to make $8m in profit on a $2m investment, simply because of convexity – one side of the straddle will pay off 10x, while the other side simply expires worthless. Only in the case of a dead-calm market will I lose money.
Note that this is hypothetical – the perfect gold option doesn’t exist and would likely be too expensive to make it worthwhile. But there are options like this, both in the world of synthetic options and in the world of real options. It takes skill and experience to put a convexity portfolio together. If you imagine that the average convexity straddle provides a 5-to-1 net payoff, then you only need 20% of these bets to win to break even. The key here is to have many of them – more than a hundred – so that even if 70% of them go under, the remaining 30% will make up for the rest. With this portfolio, I don’t care what happens, and I don’t have to predict the future. The thing I’m most afraid of is dead calm, which, in the unlikely event that it happens over many years, will probably only generate a small loss anyway.
Real options are an excellent foundation for this kind of portfolio, and venture investing is a way to build a portfolio of real options, but it has to be done using a beta model, not an alpha model. There are others. It could be built as a FoF using only hedge funds, but that would require scale. To be antifragile, you need many asymmetric bets in many different markets. A few people are investigating this kind of portfolio, but few have so far built one.
Note that the 80/20 split is just for illustration – you need to build stochastic models to help understand what the best ratio is, and stochastic models must be built very carefully, with full knowledge that black-swan events are impossible to predict. A more conservative portfolio would have 90% safe, and a more aggressive one would have 70% safe. It’s important to understand that the convexity portion is less volatile and less subject to loss than most people think, while even traditionally hedged portfolios can be more volatile than their designers intended.
A barbell portfolio like this has a good chance of returning about 8-12% per year with some tax efficiencies, regardless of what happens to the markets. It’s market neutral and likes volatility.
The following chart, while not an exclusive taxonomy, helps us see in more detail what Taleb is talking about. It points out some of the differences in some of these terms and tries to put them in perspective:
Most large actively managed funds underperform. Luck plays a larger role in success than most people think. If failure is a good teacher, I’ve learned more than most. I’d invest differently now than I have in the past. I think the vast majority of wealth managers are following the herd and will go right over the next cliff when it comes. I think it’s better to design a portfolio by starting with the weakness, rather than the perceived strengths.
David Siegel is an angel investor and consultant in New York City, moving to Switzerland in June. You can find his work at www.businessagilityworkshop.com