Meb Faber on Smart Beta, Robo-Advisers, and Why You’re Overthinking Asset Allocation
Journal of Financial Planning: August 2015
Who: Meb Faber
What: Co-founder and chief investment officer of Cambria Investment Management, author of Global Asset Allocation and Global Value
What’s on his mind: “The basic idea of diversification is wonderful, but people often misinterpret it and mold it to their own biases in somewhat dangerous ways.”
Podcast: Listen to our podcast with Faber at FPAJournal.org
How does a guy with a double major in engineering science and biology end up a quintessential quant? For Meb Faber, the answer is: you have an innate interest in investing, growing up in a business-focused family that was always talking about investments and money. Then you work as a biotech equity analyst in 2000 as both the biotech and Internet bubbles were popping. You gravitate away from biotech and get more quantitative, researching methodologies, and eventually you start an investment management company. The rest is history.
As Faber says, “My career became my hobby and vice-versa.”
Faber, a well-known speaker, author, and blogger (mebfaber.com), is co-founder and chief investment officer of Cambria Investment Management, where he manages the firm’s ETFs, separate accounts, and private investment funds.
He’s also the brains behind The Idea Farm (www.theideafarm.com), where Faber himself curates the best research pieces that come across his desk into an email exclusively for subscribers.
The Journal recently sat down with Faber to find out what he thinks about smart beta, absolute return funds, robo-advisers, and more.
1. In your latest book, Global Asset Allocation, you say that while we’re busy paying close attention to our portfolio’s asset allocation, the greatest impact on our portfolios may be something we fail to notice altogether. What are investors failing to notice?
Investors in general—professional, individual, everyone—spend vast amounts of their time obsessing over their portfolios. And people are very opinionated on the topic of asset allocation, often accompanied by a zeal typically reserved for politics or religion.
Some of the most famous money managers in the world have publicly suggested various asset allocations for individual portfolios; everyone from Bridgewater’s Ray Dalio, who founded the largest hedge fund in the world, to Mohamed El-Erian and Rob Arnott. So we decided to go back and take a look at these allocations and demonstrate how those portfolios would have performed in various market environments over the past 40-plus years.
What we found is, they all do a pretty good job over time, and they all move up and to the right, both on a nominal and real basis. But what was surprising is that—and I exclude Permanent Portfolio because they have a lot in cash—if you look at all the allocations, they have an incredibly tight range of performance and risk statistics, meaning they’re very similar. And that was a fairly big surprise since the allocations were often wildly different. They took different paths to get to the finish line, but the returns were broadly similar over time.
So we said, all right, let’s pretend we have a crystal ball and can go back to 1972 and see which allocation was the absolute best allocation. It turns out to be the Mohamed El-Erian allocation.
It’s an endowment-style portfolio, and it’s not surprising that it has performed well because it’s pro-growth, and for the most part, the last 30 years have been great for markets. So El-Erian’s was the best performer, and Permanent Portfolio was the worst, but that’s partially because it’s really low volatility due to a very large allocation to bonds and cash.
Then if you said, we’re going to go back in time with perfect foresight and implement the single best allocation with the average mutual fund, which right now has an average annual management fee of 1.25 percent a year (and if you go back 20 years, it’s probably even more) then the drag of the management fee would have taken the returns from the very best allocation to almost as bad as the worst, which is a pretty stunning takeaway.
And then if you said, I’m going to have my money managed by an adviser, who, on average, charges 1 percent a year, and you then invest in the average mutual fund at 1.25 percent per year, then the combination of the two fees completely just decimates the performance. It takes the best-performing allocation and makes it far worse than the worst.
So people spend a lot of time thinking about the sexy, exciting part that draws all the headlines: What are my investments doing? How much should I have in gold? Are bonds in a bubble? What about Greece? These are likely the wrong questions to be asking. What they should be focusing on is the boring blocking and tackling of, how can I minimize my costs that I pay through not just management fees, but also through commissions and other sorts of trading costs, as well as taxes?
That turns out to be a much bigger influence on performance, though it is not as much fun to talk about at cocktail parties as your favorite stock tip. It’s tough to go on TV and talk about fee drag, but it turns out to be the most sensible advice when it comes to asset allocation.
2. What about financial advisers who are responsible for client portfolios—are they worrying too much about particular asset allocations in portfolios?
I think so, and it’s particularly timely now, as most asset classes—through indexing and ETFs and mutual funds—are very accessible for a super low cost. If you’re doing buy-and-hold investing and you have some global stocks, some bonds, and some real assets, the exact amount doesn’t matter that much; as long as you have those three ingredients, then the precise asset allocation becomes less material.
Many advisers may feel threatened by this realization, but I think it is a huge benefit for their business. It frees up the advisers to do all of the value-added services that really matter to investors, such as holistic wealth management, behavioral coaching, planning through trusts and insurance, tax management, and all of the services that people look for in an adviser. Think about all of the time you no longer have to spend worrying about how much you should have in gold or TIPS.
3. In your opinion, what’s the role of real assets in a portfolio?
We’ve always been a big fan of real assets, and if you look at the allocations from the book Global Asset Allocation, you notice that particularly in the ’70s, which was a high-inflationary period, a lot of the portfolios that did best had allocations to real assets, through gold, commodities, and real estate. And we think it’s a very reasonable allocation to include in your portfolio for that scenario, as well.
Now, the portfolios that were heavy in those assets during [the ’70s] also did worse in the ’80s and ’90s because those were dominated by equity and bond outperformance. But what’s most important is you have a robust portfolio that’s really not slanted toward any one environment, because the future’s uncertain.
Now, that having been said, at our core and since our founding, we are a trend-following shop at heart. So the managed futures-style allocations, which employ a long-short approach to commodities, but also the other assets of equities and rates and currencies, we think are a wonderful complement to a traditional portfolio that almost all advisers and investors are vastly under-allocated to.
I think you could easily make a case for up to a 30 percent allocation in trend-following-type strategies that are a wonderful diversifier to portfolios. 2008 is a great example of how they can help hedge down markets, particularly going forward, as historically, U.S. bonds have been a great hedge to stocks, but they may not be as good of a hedge now that they’re at such low rates. Many of these funds are fairly expensive, but I expect that to change in the coming years.
4. You introduced the word’s first no-management-fee ETF, the Cambria Global Asset Allocation Fund (GAA). What inspired you do to that?
We have said for a long time that investors shouldn’t be paying much for buy-and-hold strategies, because by definition, you’re not doing anything—you’re buying and holding. We decided we could put together this fund with no management fee that owns 29 underlying ETFs that cover the entire global marketplace, including stocks, bonds, real estate, commodities, and all-in has a total expense ratio of 0.29 percent due to the underlying ETFs.
We’re able to do this because three of the 29 funds are other Cambria ETFs, so we’ll break even hopefully pretty soon. We’re not going to make much money on this, but it’s a great service for people. Even if I was going to build it just for my mom and my friends to be able to invest in, I think it would be a wonderful allocation, but what we’re seeing is a lot of advisers and a lot of investors are allocating to it, because it is the lowest-cost option. Many individuals are using it for their core allocation, and advisers are often sweeping their smaller clients into the fund since, pass-through, it owns over 20,000 securities around the world with just one purchase. So we’re glad to get it out, and we look forward to launching a few more of what we call these zero-percent investable benchmarks, so stay tuned in the next couple years.
5. What’s your take on smart beta ETFs?
[Smart beta] is a term I try to avoid, though all five of our ETFs could be classified as smart beta depending on who you ask
Let’s use U.S. stocks as an example. I think market-cap-weighting indices is a great first step, but I think it’s suboptimal. You could weight stocks on almost any other variable and it’s going to beat the S&P 500, because market-cap-weight indexes have a natural bias, which is that their only variable is price. So it’s good in the sense that you own the market. But in times of bubbles and over-valuation—’99 in the U.S. is a great example, 2007 with China, 1989 with Japan—it can create a very top-heavy allocation with expensive stocks, which is not what you want.
You could do anything else; you could equal weight it, you could value weight it, you could weight it by letters of the alphabet, and that will likely out-perform the market-cap weighting. So value, momentum, a lot of the very common factors that people have known about for 30 to 100 years, we think make a lot of sense to tilt portfolios, but the main thing is just tilting away from market-cap-weighting. We particularly like the combination of value and momentum, and that is what our shareholder yield fund is looking to achieve with selecting a concentrated portfolio of stocks distributing cash through dividends and buybacks. Historically, that has done a much better job than market-cap weighting.
We like the idea of smart beta as long as you’re paying low fees. In many cases, smart beta is an excuse to charge more and to say you’re active or you’re doing something that’s super alpha-generating and proprietary, but really, most of them are formulaic and should be had for probably 60 basis points or less.
6. What about absolute return funds?
It’s rare to find people who believe in buy-and-hold but also believe in tactical asset allocation or absolute returns. And so you’re speaking to a manager that runs both styles of funds, and we have a sister fund to GAA that is a concentrated momentum and trend fund (GMOM ETF). It’s kind of like saying I’m both a Democrat and Republican all wrapped into one, but it’s just being honest about both approaches, and it becomes more, in my mind, a philosophical and emotional question.
The buy-and-hold allocations we present in our book [Global Asset Allocation], are fine, but there are certain times that you’re going to have long drawdowns, and there’s just no way to get around that. With buy-and-hold you have to wait and sit it out. Well, a lot of people are not designed emotionally to be able to withstand large drawdowns. It’s very painful to watch 50, 80, 90 percent of your net worth expire in any one asset class, or half of your entire portfolio disappear.
There are approaches like trend-following that we think can do a wonderful job of reducing volatility, reducing drawdowns, but psychologically are also a lot easier to digest for certain people if you have a market like the U.S. in the ’30s or the last two bear markets we’ve had in U.S. stocks.
The ability to have a system that reacts to that is a wonderful thing for a lot of people, but it comes with its own set of challenges such as having whipsaws when markets are going sideways. So let’s just say there’s no one easy solution for all investors.
So we’re a big fan of certain [absolute return] strategies with a caveat that it has to be something you could explain to your niece or nephew or middle-school-aged child and say, “This is why this works.” If it’s some screwy idea that’s simply a result of data mining, it’s going to be a lot harder for that to work in the future than if it’s something based on sound common sense. And as we mentioned before, be wary of costs.
7. You participated in a June 2013 Journal roundtable discussion on tactical asset allocation. At that time, you said, “Some of the takeaways from modern portfolio theory are OK, and others are really, really dangerous.” What did you mean by that?
I think the general tenet of modern portfolio theory is sound, and that is: you mix assets, you diversify, and the sum of the parts is better than any individual holding.
There are a few problems, however. One is that people can over-rely on the inputs; if you input junk, you’re going to get junk out. It’s also very susceptible over the time period you’re looking at. If you’re doing a mean-variance optimizer and you’re trying to spit out something based on five years of performance, it may be a completely worthless exercise.
Here’s a great example using managed futures. If you throw that into a mean variance optimizer, you end up with an allocation in managed futures of something like 50 percent because it’s a wonderful diversifier, it’s had great returns with low volatility and correlates to nothing, but no institution in the world allocates much to managed futures because it feels awkward. You’re lonely in the sense that it’s a career risk. So some may put 1 or 2 or 5 percent in, but I don’t think I’ve ever seen more than 10 percent.
So that’s an example of: if you’re not going to believe the math and result of what comes out, then what’s the point of even running the simulation?
There are other problems, too, such as it completely ignores valuations. In my book Global Value, I talk about some very simple metrics to look at when valuing global stock markets. There are certainly times when it’s unquestionably, in my mind, simple to be able to look at an asset class or a stock market and say, “That’s very clearly over-valued and it’s a bubble."
Looking at the U.S. stock market right now, it’s one of the most expensive markets in the world. I don’t think it’s a bubble. It’s not crazy, and it is not like it was in the ’90s, certainly not like Japan in the ’80s or even China in 2007, but it’s expensive. And that just means future returns will probably be low single digits, rather than high single digits that everyone is expecting.
But if you put U.S. stocks in your optimizer at an expected return of 10 percent a year, you’re going to have this massive allocation to U.S. stocks, when in fact, it’s much smarter right now to be allocating to foreign assets.
And that’s another one of the biggest mistakes people make in the U.S. and in other countries—the home-country bias of allocating. If you’re market-cap-weighting, half of your stock allocation should be in foreign stocks, but most people put around 70 percent in [domestic stocks] because it’s comfortable and that’s what everyone else does. We have an ETF that goes and buys the 11 cheapest stock markets in the world. We think it is a great idea (particularly now with global valuations much cheaper than the U.S.), but some can never get comfortable with owning countries like Brazil, Russia, and Greece. But like the old [John] Templeton quote, “Don’t ask me where things are the best. That’s the wrong question. Ask me where things are most miserable.”
So we think the basic idea of diversification is wonderful, but people often misinterpret it and mold it to their own biases in somewhat dangerous ways.
8. You’ve written that the largest drawdown will always be in your future. How important is this for advisers and clients to understand, and what implications could this have on a portfolio?
I sat down once with an investment manager, and he had this scale that showed a portfolio and he had me go through a little risk tolerance questionnaire. He said, “This is your expected return and volatility and drawdown.”
Because I’m a student of markets and have looked at returns going back 120 years I said, “This is interesting. What were you showing people in 2007? Because the drawdown on this portfolio would have been 20 percent up until 2007, and then a year later, the drawdown would be 40 percent. So someone who thinks the absolute worst-case maximum loss they would have is 20 percent just had a loss of 40 percent.” And he answered, “Well, we just readjusted the scales.”
The problem with that of course is that the person who thinks they have a very robust and time-tested portfolio that is only going to lose 20 percent in the worst-case scenario, just doubled and lost 40 percent, which is what most of the endowment-style or heavy-equity portfolios lost in 2008. That experience can be devastating not only financially, but also emotionally.
My point is: you need to be a student of history and markets so you understand what’s possible. Any one asset class or security can lose 80 to 100 percent; just look at Greece or coal stocks for recent examples.
In my favorite book, Triumph of the Optimists, you’ll learn that going back to 1900, yes, the U.S. had great returns, one of the best-returning stock markets in the world, but other markets like Austria would have had a zero real return for the last 115 years. Other markets, like China and Russia, shut down completely, and you would have lost 100 percent of your money.
You learn other concepts like the importance of real returns instead of just nominal returns, where bonds can be just as risky as stocks based on the slow chipping away of inflation. It’s hard because it’s a lot of work to learn the history of markets and have reasonable expectations, but mathematically speaking, yes, the largest drawdown will always be in your future for any asset class, security, or strategy—it’s a mathematical certainty if your time frame is long enough.
As far as how this applies to advisers, trying to build a portfolio that you can say, “These are our expectations historically, but also, this is what could go wrong and what it would look like,” is a really important exercise.
A simple exercise is: what do you think is going to happen if U.S. stocks and bonds decline at the same time? It’s rare, but what if they do? Do you have strategies in place that will mitigate that, or are you just counting on the future to look exactly like the past?
9. Do you think robo-advisers could ever replace human advisers?
The advent of the robos is a wonderful development for investors, as well as for advisers, because it frees everyone up to do other things.
We all know investors are their own worst enemy, professional and retail alike. They do dumb things consistently, over and over. They buy at tops; they sell at bottoms; they have all the behavioral biases that all of us have. I’m certainly guilty of many biases, and it is one reason I became a quant in the first place. So [robos] are a wonderful development if, like any technology, people know how to use it. I see the robos as a feature rather than a standalone company, and eventually every custodian will have their own robo offering, for free, and it will simply augment the human-adviser relationship.
The best part about having an adviser is you have that behavioral coaching, the ability to talk to someone or interact with them. Very few of us can have an automated investment solution that’s just whirring in the background and be completely emotionless and logical about it. It’s an easy thing to say, but a hard thing to do, especially when the inevitable monster drawdown shows up.
10. Where do you see things hashing out with the various robo-advisers? What do you think the landscape’s going to look like over the next few years?
A lot of these technologies have been in use in various formats for a long time. People have been doing tax harvesting since the ’90s with software. People have had automatic rebalancing software. It’s just that you’re seeing it packaged with a great user interface, and no qualms about that, because in a way, that’s more approachable, certainly to the younger generation, and [more] mobile-friendly as well.
But what I’ve always said is you’re going to have this race to the bottom on fees, and so you have a lot of the early robo-advisers come out around 25 basis points—the Betterments, the Wealthfronts of the world. The thing is, when a custodian or fund sponsor—so a Vanguard or a Schwab—wants to enter this space, they’re going to dominate it because they have the assets and the infrastructure, but more importantly, they have the underlying funds. You could say they are vertically integrated, and that changes the equation of running it as a business.
So Schwab, which, sure enough, launched a few months ago, uses their own funds and charges a zero-percent fee. No one else can compete with that that doesn’t have their own funds. And they’re now the largest automated investment solution; they’ve passed everyone else very quickly.
Vanguard is somewhat of a hybrid model, because they have advisers you can talk to, but they charge 30 basis points, and they’re about 10 times the size of the No. 2 offering by Schwab. They’re at something like $20 billion already, and they had to shut it down because they were growing too fast for a while. Now they’re reopened, and who knows, it’s probably $30 billion by now.
So you’ll see it shake out the same way the ETF space has, where there’s a handful of really big winners, but there’s probably room for 20 robo-advisers in some form. Every single wirehouse and independent RIA will use one. At the end of the day, it will become a commodity once everyone has the same technology.
Carly Schulaka is editor of the Journal. Contact her HERE.