John West: Why Most Capital-Market Assumptions are Unreasonable

What is a reasonable assumption for real returns?

A reasonable assumption is 2% to 2.5% over a 10-year time frame, per the way most people allocate their portfolios, which is a heavy dose of U.S. stocks and bonds and a light sprinkling of non-U.S. stocks and bonds and a dash of alternative strategies. That’s going to be with the standard 9-10% volatility you get with a 60/40 or most mainstream allocations.

How do most people react to those assumptions?

The biggest challenge is not that we’re going to get low returns from bonds. We have low yields and that means low returns; that’s pretty simple. People say, “Well, the low yields mean I can have a really high valuation for stock portfolios.” Our mainstream U.S. equity assumption is approximately 1% real. That assumes some mean reversion in valuations, let’s say from a Shiller P/E of 25 down to 20. If you don’t think that’s going to happen, that’s perfectly fine. But we have a good reason for that 1% assumption. Historically, when you’re in the top decile of valuations, 10 years out you will have lower valuations.

If you want to come up with a story or a rationale for why you think this time is different, then you must understand that you’re taking a chance. That means your stock return is not 1% real; it’s 3% real. With essentially 0-1% real for bonds, it’s still not magically getting you to 5% real. You can torture or move these things around a little bit, but the basic conclusion is pretty hard to deny.

The second thing that people say to us is, “Oh, your website doesn’t have illiquid, alternative strategies. It doesn’t incorporate alpha.” We touched on this in the piece you referenced. If your mainstream liquid asset strategies are going to get you 2% real, and that’s 80% of your portfolio, the 20% that’s left has to generate 12% real. How often have these alternative strategies, such as private equity, hedge funds or real estate, returned more than 12% real over a 10-year period? Outside of private equity and venture capital where mush of the last 30 years were dominated by the tech bubble run-up and where those things had some extraordinary returns, this doesn’t happen that often.

The other potential source of higher real returns is alpha. We think the first thing you should do is avoid negative alpha. Most people, when they seek alpha, end up looking back over time and saying, “I wasted a lot of returns.”

What we didn’t say in our research piece is that if you’re not happy with this, the first thing you should do is do no harm, accept those returns and adjust to the new reality, rather than trying to “alpha” your way into a higher return. Our experience, and I believe the experience of investors realizing lower dollar-weighed returns, means it’s really, really hard to pick superior managers ahead of time and stick with them over the long term. It’s implausible that’s going to make a meaningful amount of difference.

Are any asset classes undervalued to the point they can offer significantly higher returns over the next 10 years?

Non-U.S. equities, both in developed and emerging markets, offer returns from today’s level that are 5% real for developed ex-U.S. and closer to 7-8% real for emerging markets. Those are in U.S. dollars. You have to incur a little more volatility to get those, of course. But if you start building portfolios and incrementally move 10% away from U.S. equities into non-U.S. and emerging markets, you have a fairly meaningful impact on your long-term return expectation.

How confident are you in your 1% estimate for real returns for U.S. equities?

It’s important to separate the journey from the likely destination. The likely destination is the center point of 1% real for U.S. stocks. You can have a variety of different journeys. When you look back over time to when stocks were traded at similar valuations as today, you see that they can go on phenomenal four-year runs for the first four years of those 10 years. Then they give it all up in the latter half. In other periods, you give up a bunch initially, then you play catch-up and you wind up at 1% or 2% real. In other periods, you just plod along and muddle through.

We have no particular insight into what the journey will be. But we think that the likely destination is 1% real for U.S. equities, plus or minus. We know that we don’t have perfect forecasting accuracy, and that’s the reason why we show you a confidence interval. But when we look at that confidence interval and apply it to mainstream stocks and bonds, we can ask, “What’s the likelihood of 60/40 making a 5% real return over the next 10 years?” It’s a tick above zero. It’s essentially a rounding error, and that’s the reason why we had some fun with the Lloyd Christmas analogy and said, “So, you’re saying there’s a chance?” Yes, there’s a chance, but we shouldn’t bank on it as investors. [Ed. note: In West’s research article, he referred to the movie Dumb and Dumber, where Lloyd Christmas (Jim Carrey) is told that he has a one-in-a-million chance of dating an attractive woman (Lauren Holly). Christmas is unfazed and optimistic that he has “a chance” of such a relationship.]

What other questions are you’re hearing from advisors about asset allocation and portfolio positioning?

Those questions fall along two dimensions. The first is, “We understand your analysis. We get it. It’s intuitive. But how do I push my clients to take a stance that’s closer to your world view, and push them to diversify more substantially into non-mainstream assets to get higher returns?” They tell us clients are going to immediately look at a Vanguard balanced 60/40 portfolio, and as soon as it has one year where it does really well, it’s going to be really hard for them to have those clients stay the course.

One of the things we’re looking at is upgrades to our asset-allocation site. This will enable you to not only look at your portfolio from the standpoint of volatility, but also to pick a normal portfolio and measure your “maverick risk” (i.e. your willingness to be different) using tracking error to a normal portfolio. That’s something we anticipate putting into our tool so that advisors can have that very reasonable trade-off in those conversations with clients.

The other area that advisors are struggling with is what to do along the continuum of active management, smart-beta or factor-based portfolios and cap-weighting within their equity allocations. The vast majority of advisors have the bulk of their assets in equities. How do they make sense of those choices? We’re seeing a dizzying array of claims and data. One of the things that we’re thinking about is how we can add some context and transparency to that.

One of our papers that Rob Arnott and I co-wrote [along with Vitali Kalesnik and Noah Beck] was How Can “Smart Beta” Go Horribly Wrong?” The second and third pieces, which Rob authored with Vitali and Noah, culminated in Timing Smart Beta, which we released a couple weeks ago. That research should allow us to get some good insights into how do we build reasonable expectations, not for asset classes, but for strategies going forward. The key take-away of that research, and quite frankly of our asset-allocation philosophy, is don’t use the last five or 10 years to project the next five or 10 years. That’s not a reasonable conclusion, given the historical data. If anything, you should use the opposite; whatever has done really well, either from a strategy or asset-class perspective, in the recent past, is probably not going to do as well in the future.

What analysis have you done to look at the capacity of your smart-beta strategies? How much in assets can they accommodate?

Two things stand out when thinking about future returns on smart beta strategies. First, as you mention, we want to look at these strategies from the standpoint of how much in assets can they reasonably take. That’s a function of turnover in the portfolio and the liquidity of the stocks that they hold, from which you can get a pretty good sense of their capacity. I encourage people to first look at management costs and then to look at likely transactions. You’ll never see the transaction costs and they’re never in the disclosures. We have a pretty good sense of what they are going to be. We can’t do it to the basis point, like you can with management fees, but over a five- or 10-year period, you’re going to know the costs of these strategies with a high degree of certainty.

Second, we look at where our strategies trading relative to their history. The definition of a crowded trade is when so many people enter an asset class or a strategy that is trading so much richer than its long-term history. But then someone wakes up and says, “This is unsustainable.” That brings it back to more of a normal valuation. That’s the primary reason why, whether we’re talking asset classes, strategies or even active managers, you so often see that people bid up the price of those securities that are in those portfolios or asset classes to a point where they offer such low returns. Then you get mean reversion.

You have to look at these strategies through the lens of transaction costs. Then you must ask, “Are they cheap or rich?” That’s a good way to look at the future.

What do you see when you apply that analysis to your RAFI strategies?

We’re in the process of looking at valuations and coming up with return estimates moving forward, very similar to our asset-class work. You see a little bit of it in the work that Rob did with Vitali Kalesnik and Noah Beck on our equity research team. If you look at some of our publications over the last five years, we have put into those studies what the expected annual transaction costs are going to be for some of our strategies. Sometimes you see 30 to 70 basis points. You can get an ETF for 25, but you’re paying 70 basis points in expected transaction costs that you never see. Still, it’s coming out of your clients’ pockets, and you have to think about that.

Some big trends are affecting the active asset-management business, such as the DOL fiduciary rule, the influx of robo advisors, and the flow of funds into passive management. How do you see those trends affecting advisors?

I got in this business 21 years ago as an institutional investment consultant, which is someone who essentially does the same things that advisors do. My value-add proposition was asset allocation, manager selection and performance measurement. All of the factors you just mentioned mean that advisors really need to rethink how they can add value. Those that rely upon statements such as, “I’m a better asset allocator; I understand where the market is going more; I can pick better managers; I’m going to fire the managers before they get into performance trouble,” have to shift into understanding how to be a “long-term investor coach.”

The vast majority of advisors have client time horizons that are at least 10 years. An 80-year-old has an eight-and-a-half-year life expectancy. Everybody has a 10-year time horizon. What truly drives asset class returns over such horizons? What truly drives strategy returns? How do we keep our clients from making poor choices about abandoning strategies or their risk profile at times of market crisis?

This is what I mean by the advisor shifting to more of an educator-coach role, rather than someone who is outsmarting the competition to give their clients better returns. Smart beta, factor investing, and the overwhelming victory we’ve seen index funds achieve over the last 10 years over active management all lead to a structural shift in how advisors interact with clients.

Will the shift be permanent?

Cap-weighted indexes have notably dominated active managers over two distinct long horizons in my 21 years, in the late 1990s and today. Both of those were periods when growth stocks massively outperformed value. Growth stocks have higher multiples and therefore higher market capitalizations. Active managers tend to have their outperformance, cumulatively as a group, at periods when value and small size are winning. When growth and large mega caps are winning, active managers typically struggle.

Part of me wants to say this is cyclical in nature. But if people want to say value and small are going to come back, they can get that through factor and smart-beta strategies just as easily. I do think active managers will have a better run in the next five or 10 years than they had the last five or 10, but that doesn’t necessarily mean they’re going to get better flows. A lot of their traditional methods have been replicated by more systematic, lower-cost strategies.

What guidance are you offering relative to inflation hedging?

Inflation expectations have dropped from 2.5% on a 10-year basis to about 1.2% earlier this year. That’s an approximate 50% decline in inflation expectations. Commodities, energy stocks, TIPS relative to Treasury bonds, emerging markets, emerging-market currencies and high-yield bonds have all tended to have a little higher correlation with changes to inflation, and have been savaged relative to mainstream stocks and bonds. A lot of advisors are asking, “Do I need a little bit of inflation protection? It’s been so brutal to hold on to those types of strategies, but is now the time I should give up on them?”

The greatest mistake I’ve seen in my 21 years in the industry was in the late 1990s when people used recent performance to dictate their diversification policies. They gave up on value, they gave up on inflation and diversifying strategies. A lot of advisors are struggling with how much they should put in to things that can do better in a reflationary environment versus what we’ve had, which is clearly a disinflationary environment for the last handful of years. If you go to our website and look at the strategies we think are likely poised to give you better returns, it’s all those strategies that have been left behind in this disinflationary cycle.

Don’t give up on your inflation hedges just yet. Fill and diversify a basket of them. They will be important at some point in the next five or 10 years.

More Schwab IMPACT Conference Topics >