Rob Arnott: Factor Investing Is 'Overhyped'

Q&A February 26, 2019 at 02:04 PM

Share & Print

Rob Arnott, chairman of Research Affiliates.

Sometimes corporate marketing can seriously wear out the potential golden-egg-laying goose. Take factor investing. It's "overhyped," "oversold" and marketed as "a panacea." So says Robert Arnott, chair and founder of Research Affiliates, who, in an interview with ThinkAdvisor, explains reasons for factor investing's bad performance last year.

Indeed, he discusses the three biggest risks examined in a paper he's written with colleagues, "Alice's Adventures in Factorland: Three Blunders that Plague Factor Investing," due for April publication in the Journal of Portfolio Management.

Arnott, 64, a pioneer in unconventional portfolio strategies, including tactical asset allocation and tax-advantaged equity management, is known for testing the conventional wisdom. In so doing, he often discovers opportunities for profitable investing.

He has been dubbed the "Godfather of Smart Beta," referring to strategies meant to rebalance portfolios on the basis of characteristics others than market cap. But "smart beta" is a term, Arnott argues in the interview, that has been co-opted merely as a marketing tool ("…pretty soon [it] was attached to everything under the sun.")

He also talks to ThinkAdvisor about the multiple market bubbles he sees — Twitter, for instance — and offers his long-term forecast for the U.S. economy, and stock and bond markets. Best place to invest now? Emerging economies, he maintains.

Arnott is currently researching the risks of government deficit spending posed by adherence to the faddish "modern monetary theory," which he explores in the interview too.

His recently published research papers include "Can Momentum Investing Be Saved?" ("So-called experts on momentum have generally delivered terrible returns to their clients.") and "A Backtesting Protocol in the Era of Machine Learning," about AI.

Arnott founded Research Affiliates in 2002 to focus on innovative approaches to active asset allocation, portfolio construction and other quantitative strategies. Based in Newport Beach, California, the firm, with more than $175 billion in assets under management, develops investing strategies distributed by partner firms, including BlackRock, Pimco and Schwab, in a range of formats, such as ETFs and mutual funds.

In August 2018, Arnott stepped down as CEO to increase his focus on research and portfolio management. He manages the Pimco All Asset and All Asset All Authority family of funds and the Pimco RAE suite of funds.

Chicago-born, he graduated from the University of California, Santa Barbara with a B.S. summa cum laude in economics, applied math and computer science. He has published more than 100 journal articles, about a dozen of which have picked up prestigious awards.

ThinkAdvisor recently interviewed Arnott, on the phone from his Miami office. One area of discussion was target-date funds. He says the industry has the strategy, well, essentially backwards.

Here are highlights of our conversation:

THINKADVISOR: You've made a career of testing the conventional wisdom. What's the upshot?

ROB ARNOTT: It's fun looking at ideas and trying to discern what works and what doesn't. Wherever there's a gap between conventional wisdom and the real world, that's a profit opportunity for investors.

You're not a cheerleader for mainstream stock-and-bond portfolios. Rather, you favor investments like commodities, REITs, emerging markets, high-yield bonds. What do you like right now?

I see emerging economies as the place to be. The best markets out there seem to be emerging markets stocks; the worst seems to be U.S. small-cap growth. The spread between the expected returns of those two is nearly 10% a year for the next 10 years. That's enormous.

What, for example, makes emerging economies so attractive?

They'll have a soaring population of mature adult workers in their 40s and 50s who'll be setting money aside for retirement.

Do you think emerging markets will beat the U.S. stock market this year?

I have no clue. But I'd bet very long odds that emerging markets investments will beat U.S. stocks on a 10-year horizon and very likely a five-year horizon.

You've just written a paper saying that understanding the risks of factor investing is "essential before adopting [that] investment framework." What are the three main risks?

The biggest risk is that many factors aren't real. They've worked historically, but that doesn't mean they'll work in the future. The second risk is of losing your edge, and alpha gets arbitraged away. The third risk is that factors, by dint of being popular, can become expensive so that you're paying a premium for the stocks that meet the factor criteria. By paying that premium, you wind up with a very expensive portfolio where the alpha has disappeared.

What do you think of factor investing broadly?

It has merit. There are some interesting ideas out there, but I think they're overhyped and oversold to an alarming extent. Every factor is suspect. If you use multiple factors, you diversify your risk a little bit. But it's been sold as a panacea.

How many factors are there?

Anywhere from hundreds to thousands. Some big ones are momentum, quality, size, value. There's a whole other array that looks at degree of leverage of the company, volatility of earnings, dividends, buybacks — and the list goes on and on. The thing is that every single one of these works in back tests. But all that tells us is that people publish only the ones that did work. It doesn't tell us that factors work.

Why did factor investing perform poorly last year?

One reason [lies] in construction and implementation. How many of the academics who designed factors have actually traded stocks for a living and know the potential ugliness of careless trading? That's a key part of the issue.

What's another cause?

The factors themselves were identified as a consequence of aggressive data mining. If you mobilize the entire academic community to search for factors because that's the easiest way to get tenure and you discover a factor that hasn't been published before — great! You become the expert on that factor and get tenure! So there's a powerful incentive to look for relationships that may or may not have merit.

That doesn't sound scientific.

All too often in the factor world, people did the research, found something that seemed to work and then created their hypothesis — they concocted behavioral explanations.

You've been dubbed the "Godfather of Smart Beta." What are your thoughts now about these strategies?

I didn't coin the expression, smart beta. The credit for that goes to [the firm] Towers Watson. Smart beta used to mean something — strategies to break the link between a stock's price and its weight in the portfolio so that you're rebalancing alpha. But it doesn't mean that anymore.

Why not?

Because the industry has commandeered the expression: It became a very powerful marketing tool. A lot of people realized: "If I call my strategy 'smart beta,' I can have more conversations and make more sales." So pretty soon "smart beta" was attached to everything under the sun.

Can smart beta be used in factor investing?

Early on, the factor investing community embraced the term. But factor investing doesn't break the link with price. Factors are, in most cases, cap-weighted. Momentum is the antithesis of smart beta because if the price goes up, you want to buy more of the stock. Well, that may make investment sense, but it doesn't hew to the original definition of smart beta.

Is there a way, though, that smart beta strategies can help investors in today's market?

As long as the market is chasing growth stocks and pushing growth to one new high after another, it won't work. Rebalancing is a strategy that works, on average, over long periods of time. The market is still rewarding growth stocks in general. So it seems like the bubble remains alive and well.

What's your definition of a bubble?

Anything where the price can be justified using a standard valuation model — like discounting cash flow — only if you use implausible assumptions and secondarily, the marginal buyer doesn't care about valuation models. Twitter, Spotify and Tesla are clearly bubbles. Amazon, Apple and Microsoft clearly aren't.

Do you see any other bubbles?

There are a lot of bubbles. Out of the Russell 3000, to the best of my recollection there are about 80 stocks trading at above 10 times sales, let alone 10 times profits. Ten times sales is a pretty extreme valuation.

How long can bubbles last?

Bubbles can continue longer than you can remain solvent. So bet against them by not owning them. But — you might not want to short-sell because something that's extremely expensive can become vastly more expensive before it eventually turns.

What's your forecast for earnings this year?

I would strongly expect an earnings slowdown. Wall Street loves to predict lofty earnings growth year after year. But the simple fact is that when you have earnings already at a record share of GDP, it's tough to push those earnings higher. Rapid earnings growth comes off recessionary troughs. And we're nowhere near that. So I expect earnings to be, at best, disappointing this year — decent odds that they might actually be negative.

What's your forecast for the U.S. economy?

I think that sometime in the next two to three years we'll have a recession. I see an imminent recession in Europe, if there isn't one already underway. So Europe could pull us down into a slowdown, possibly even a recession. I don't see the Fed being the proximate cause.

Are you worried about flattening of the yield curve, which is usually a predictor of recession?

A little. If the yield curve inverts, then I become [more] worried. Inversion has predicted pretty much all the recessions and slowdowns of the last half-century with no false signals. But my guess is that there'll be no interest rate hikes this year at all. That will allow the yield curve to remain positively sloped, and the Fed will not wind up punishing the economy with unduly high short rates.

What do you predict for the stock market for 2019?

One-year forecasts are a parlor game. It's not a serious part of what we do. We'll have a bear market — I don't know when. I think that U.S. stocks are arguably the most expensive in the world. The U.S. economy is very solid, but there's no room for disappointment at today's share prices. So I look 10 years down the road and see a whole cohort of baby boomers retiring — probably half already have — but the other half will be retiring and will want to sell financial assets in order to buy goods and services.

What's your outlook for bonds?

I'm not pessimistic about bonds. I think emerging markets bonds are low-hanging fruit for those who want less risk. I would go with broad diversification. The ones that are cheapest and have the highest bond yields and lowest valuation multiples are cheap because they're scary. In order to do well, all they have to do is exceed expectations that are set very, very low.

Is there any other conventional wisdom you're challenging these days?

The notion of target date funds. The trillion-dollar target date industry is built on a foundation of a strategy that was never tested. The notion of starting with extremely high equity allocations when you're young sets up young adults for the risk of having a triple whammy: If there's a recession and a bear market, the value of their defined contribution plan tumbles. They [may] lose their jobs. Then they'll have to pay a tax penalty to get access to the money they need just to get by between jobs because most of them don't have other savings.

How should target date funds be reworked, then?

People ought to start conservative and become more aggressive as they get older and have more resources outside their 401(k). Once you have serious assets in your portfolio, you can absorb the risk. I think the correct profile is to start off very conservative: 20% or 30% equities, ramp up fairly quickly as you go through your 30s, and reach peak allocation of 70% or 80% when you're in your late 50s. Then ramp down only for the last six or eight years. That would leave people with more money in retirement and with surprisingly less risk.

What else are you researching?

I'm doing work on the macro economy. There's a very popular fad in the economics profession called "modern monetary theory" — MMT — which says you can print as much money as you want [in order] to spend as much money as you want, and that it doesn't hurt the economy. That's pure rubbish!

Why?

I've done research that shows when you engage in deficit spending, not only does it not help in the year when you engage in it, but it also doesn't help a year or two later. If you have a debt burden that's too high, it slows down economic growth in a very measurable way.

So what are proponents of MMT missing?

The narrative of modern monetary theory ignores the very basic reality that the more the government spends, the more resources they siphon away from the private sector — and the private sector is where almost all innovation and economic growth starts. So MMT is a severely misguided idea.

What's your opinion of the way President Trump is managing the deficit?

The deficits are exploding. That's bad. Deficit spending is dangerous. Trump has gone bankrupt — how many times [in his private businesses]? Well, we don't want the nation to do that.

Has Trump done anything good as president?

Some things are pretty good. Regulations of the private sector aren't expanding. They're not being eliminated, but the rate of growth of regulations has stalled. So companies can actually initiate projects without worrying about getting on the wrong side of some regulation they haven't even seen. That's good.

— Related on ThinkAdvisor: