Arnott On True Elements Of 'Smart Beta'
Rob Arnott discusses the development of an investment phenomenon he helped to found.
[This interview originally appeared in our July issue of ETF Report.]
Rob Arnott, founder and CEO of Research Affiliates, is widely regarded as one of the luminaries behind smart-beta indexing. Global assets tied to his Research Affiliates Fundamental Index strategies stand at more than $130 billion, and he has pioneered partnerships with the likes of FTSE, Russell, S&P Dow Jones Indices and Citigroup.
But Arnott has not always had an easy ride, and has faced criticism from industry veterans from the likes of Burton Malkiel and Jack Bogle. He says smart beta is just a "fun expression" and investors shouldn't get too excited about it. Arnott also thinks fundamental indexing has "tremendous" investor benefits in asset classes like fixed income and commodities.
Is the ETF industry focused on providing what investors want, or what they need? And which is more important?
It seems to be a herky-jerky fashion of always chasing what investors want, and that's always changing. And because what customers want is sometimes a good idea and is sometimes an awful idea, the awful ideas die and the good ideas stay. So you see a steady, slow migration toward the gradual dominance and embrace of those good ideas. You could liken it to the evolution of the species on the planet.
One evolution is when we saw Jack Bogle introducing the index fund targeted at the general public, which ultimately spawned the entire industry of indexation. Was the marketplace clamoring in the mid-70s for a product that generates broadly average returns at a low fee? Of course not. What the marketplace was clamoring for was a hot product that shows brilliant recent results that they hope will continue. People are hoping for future success, and they are paying for that hope. You don't know what the future returns on any product will be.
So when Bogle introduced the product, it was a slow slog, and early adoption was very sluggish, but in the end, they beat most active managers, partly because these funds had better performance before fees than the average active manager, but mostly because fees took away what little success those active managers could deliver.
[Indexing] has since become an investment management behemoth.
Is combining smart-beta factors a kind of one-stop shop that you just mentioned?
No, that's a one-stop shop within equities. Unfortunately too often [products] are invested in mainstream bonds and equities. It's unfortunate, because right now, overall developed-world stocks are between fully priced and expensive, or between fully priced and very expensive. And as for bonds, my goodness, we're seeing some incredibly low yields around the developed world. With that in mind, what's the real return you can expect from a two-asset-class model?
You seem to suggest the industry is led by client demand. Yet I get the feeling many providers are either one step ahead of their customers or are throwing a lot of mud at the wall and hoping some sticks.
We rarely look at what customers want. Our approach is driven by how much value you can add to the industry and to the end customer. And how can we structure and package that value in a fashion such that customers may actually want to buy it?
For us, customer demand is a last stage of product design, but for most money managers, it's step No. 1. That's why we tended to be early in the offering of products that turned out to be big successes, but boy was it a long slog as they gained traction.
Now our global tactical asset allocation strategy (TAA) is a $50 billion suite. People thought: "Wow, that's not bad in just over 10 years," but I have been in global TAA for over a quarter of a century.
If you think of the Isaac Newton quote, "I see far because I stand on the shoulders of giants"—you could say fundamental indexing was built on the shoulders of three to four decades of work in quantitative equity management, carried out by thousands of people. This work peeled back the layer of understanding how markets work. It prompted me and my colleagues to realize that if markets are inefficiently priced, all you have to do is sever the link with price and weight the portfolio in a way that ignores price—this is how we will add value.
Does smart beta now look like what you thought it would at the start?
I think the label was invented in 2009 by Towers Watson, which has led to people shoehorning whatever they are doing into the label "smart beta." But the label was invented, in large measure, thanks to the Fundamental Index, and in effect, it broadened the arena to embrace other products that sever the link with price. We have hundreds of money managers saying, "We do smart beta too," without understanding what the label means. That label is desperately in need of a definition, as without it, it has no meaning.
In my view, any strategy that deliberately and carefully severs the link between the price and its weight in your portfolio and seeks to capture the many advantages of the index—replication, the ability to test an idea, scalability, capacity, low turnover, span of the broad economy—if it captures most, not necessarily all, of these elements, and severs the link with price, then it's smart beta.
So, equal weight, simple idea as it is, fits that definition. It's the same with minimum variance and low volatility.
But factor tilts that anchor on cap-weighting, and remain fixed to cap-weighting and have tilts, are not smart beta. They are quantitative factor tilts and they still have price as part of the scheme.
Smart beta is a turn of a phrase. It is as Bill Sharpe says—a distortion of the terminology he invented. He says alpha is not beta. He says beta is neither smart not stupid, it just is. It's the same with alpha: It's just a measure. And he's absolutely right. But it's a fun expression. View it that way. Don't get too excited about it.
Investment expert Burton Malkiel told ETF.com in a recent interview that the RAFI ETF is sold as a way to avoid overpriced stocks in the market and that "simply isn't true."
Burton Malkiel has served on our advisory panel a number of times and I have huge respect for him.
Obviously I disagree with him, and it stems from the fact that he's one of the fathers of efficient market hypothesis. Where we disagree is that when a stock is expensive, he would presume that it is highly likely to be fairly and justly expensive, and where a stock is inexpensive, he would say it's likely to be deserving of that low price.
I say it might be expensive because it fully deserves it, or because it's overpriced. Fundamental Indexing recognizes that if the good news is always already in the stock price, it's not going to help you, but if it's overpriced, well goodness, being underweight that stock is going to help you, and there aren't a lot of expensive stocks that are underpriced. That's where we get the advantage.
The success from an investment perspective is tremendous but from a product perspective it is more limited.
Let's look at fixed income. If you're a bond investor, you're a lender. Why on earth would you want to lend according to the debt appetite of the borrower? Why would you do that? So you're saying it doesn't matter how much Greece wants to borrow; you're happy to lend it, whereas no matter how little Australia and Sweden want to borrow, you're also happy with that.
Now you can measure bond indexes in terms of GDP, population, size of company and excess of resources—Australia has a huge land mass for resources—and create a composite for debt service capacity; then you fundamentally weight the debt you're willing to take on from that company or country. You go back historically and you find that works; in fact, it worked by 80 basis points a year among investment-grade sovereign debt.
The benefit with commodities is even more tremendous. Commodity indexes are mainly priced based on volume or open interest or other measures that are directly linked to price, and the higher the price of the commodity, the greater the weight in the index.
The Dow Jones RAFI Commodity Index, introduced last fall, weights the commodities in proportion to their five-year average open interest dollar volume, then equal-weights the three sectors, and as a result, doesn't get sucked into building weights by what has soared recently. It chooses the contract month to be the one with either the greatest backwardation or the least contango. Again, look at history: This is a very simple idea that has added 900 basis points per annum back to the year 2000.
There has been a lot of press recently around Vanguard founder Jack Bogle's views against ETFs. What do you think of his views on the market?
Jack Bogle is a wonderful human being, a friend and mentor, and one of the heroes in the business. He is also a stubborn curmudgeon, who doesn't like new ideas that are not his own. And that's OK. He describes ETFs as giving kerosene to an arsonist, as it allows intraday trading.
We don't see much difference: If you're stupid, you can really destroy [your portfolio] just as easily with either daily or intraday trading. We think ETFs give customers what they want. And if they use them badly, oopsie. If they use them well, they're wonderful tools.