AQR Style Premia Alternative: A Review

This analysis originally appeared in the September 2015 Mutual Fund Observer, but I'm republishing it here for completeness's sake. As always, comments welcome.

Objective and strategy

AQR’s Style Premia Alternative, or SPA, strategy offers leveraged, market-neutral exposure to the four major investing “styles” AQR has identified:

Value, the tendency for fundamentally cheap assets to beat expensive assets.

Momentum, the tendency for relative performance in assets to persist over the short run (about one to twelve months).

Carry, the tendency for high-yield assets to beat low-yield assets.

Defensive, the tendency for low-volatility assets to offer higher volatility-adjusted returns than high-volatility assets.

To make the cut as a bona fide style, a strategy has to be persistent, pervasive, dynamic, liquid, transparent and systematic.

SPA offers pure exposure to these styles across virtually all major markets, including stocks, bonds, currencies, and commodities. It removes big, intentional directional bets by going long and short and hedging residual market exposure. As with all alternative investments, the goal is to create returns uncorrelated with conventional portfolio returns.

SPA sizes its positions by volatility, not nominal dollars. In quant-speak, risk is often used as short-hand for volatility, a convention I will adopt. Of course, volatility is not risk (though they are awfully correlated in many situations).

SPA’s strategic risk allocations to each style are as follows: 34% each to value and momentum, 18% to defensive, and 14% to carry. Its strategic risk allocations to each asset class are as follows: 30% to global stock selection, 20% each to equity markets and fixed income, and 15% each to currencies and commodities. There is a bias to the value and momentum styles, perhaps reflecting AQR’s greater confidence in and longer history with them.

Risk allocations drift based on momentum and “style agreement,” where high-conviction positions are leveraged up relative to low-conviction positions. The strategy’s overall risk target falls in steps in the event of a drawdown and rises as losses are recouped. These overlays embody some of the hard-knock knowledge speculators have acquired over the decades: bet on your best ideas, cut losers and ride winners, and cut capital at risk when one is trading poorly.

SPA targets a Sharpe ratio of 0.7 over a market cycle. AQR offers two flavors to the public: the 10% volatility-targeted QSPIX and the 5%-vol QSLIX.

Adviser

AQR Capital Management, LLC, was founded in 1998 by a team of ex-Goldman Sachs quant investors led by Clifford S. Asness, David G. Kabiller, Robert J. Krail, and John M. Liew. AQR stands for Applied Quantitative Research. The firm’s bread and butter has long been trading value and momentum together, an idea Asness studied in his PhD dissertation at the University of Chicago. (Asness’s PhD advisor was none other than Eugene Fama, father of modern finance and one of the co-formulators of the efficient market hypothesis.)

When the firm started up, it was hot. It had one of the biggest launches of any hedge-fund up to that point. Then the dot-com bubble inflated. The widening gap in valuations between value and growth stocks almost sunk AQR. According to Asness, the firm was six months away from going out of business. When the bubble burst, the firm’s returns soared and so did its assets. The good times rolled until thefinancial crisis shredded its returns. Firm-wide assets from peak-to-trough went from $39.1 billion to $17.2 billion. The good times are back: As of June-end, AQR has $136.2 billion under management.

The two near-death experiences have instilled in AQR a fear of concentrated business risks. In 2009, AQR began to diversify away from its flighty institutional clientele by launching mutual funds to entice stickier retail investors. The firm has also launched new strategies at a steady clip, including managed futures, risk parity, and global macro.

AQR has a strong academic bent. Its leadership is sprinkled with economics and finance PhDs from top universities, particularly the University of Chicago. The firm has poached academics with strong publishing records, including Andrea Frazzini, Lasse Pedersen, and Tobias Moskowitz. Its researchers and leaders are still active in publishing papers.

The firm’s principals are critical of hedge funds that charge high fees on strategies that are largely replicable. AQR’s business model is to offer up simplified quant versions of these strategies and charge relatively low fees.

Managers

Andrea Frazzini, Jacques A. Friedman, Ronen Israel, and Michael Katz. Frazzini was a finance professor at University of Chicago and rising star before he joined AQR. He is now a principal on AQR’s Global Stock Selection team. Friedman is head of the Global Stock Selection team and worked at Goldman Sachs with the original founders prior to joining AQR. Israel is head of Global Alternative Premia and prior to AQR was a senior analyst at Quantitative Financial Strategies Inc. Katz leads AQR’s macro and fixed-income team.

Frazzini is the most recognizable, as he has the fortune of having a last name that’s first in alphabetical order and publishing several influential studies in top finance journals, including “Betting Against Beta” with his colleague Lasse Pedersen.

Unlisted is the intellectual godfather of SPA, Antti Ilmanen, a University of Chicago finance PhD who authored Expected Returns, an imposing but plainly-written tome that synthesizes the academic literature as it relates to money management. Though written years before SPA was conceived, Expected Returns can be read as an extended argument for an SPA-like strategy.

Strategy capacity and closure

AQR has a history of closing funds and ensuring its assets don’t overwhelm the capacity of its strategies. When the firm launched in 1998, it could have started with $2 billion but chose to manage only half that, according to founding partner David Kabiller.

Of its mutual funds, AQR has already closed its Multi-Strategy Alternative, Diversified Arbitrage and Risk Parity mutual funds. However, AQR will meet additional demand by launching additional funds that are tweaked to have more capacity. As of the end of 2014, AQR reported a little over $3 billion in its SPA composite return record. Given the strategy’s strong recent returns, assets have almost certainly grown through capital appreciation and inflows.

Because AQR uses many of the same models or signals in different formats and even in different strategies, the effective amount of capital dedicated to at least some components of SPA’s strategy is higher than the amount reported by AQR.

Management’s stake in the fund

As of Dec. 31, 2014, the strategy’s managers had no assets in the low-volatility SPA fund and little in the standard-volatility SPA fund. One trustee had less than $50,000 in QSPIX. Collectively, the managers had $170,004 to $700,000 in the SPA mutual funds.

Although these are piddling amounts compared to the millions the managers make every year, the SPA strategy is tax-inefficient. If the managers wanted significant exposure to the strategies, they would probably do so through the partnerships AQR offers to high-net-worth investors. But would they do that? AQR, like most quant shops, attempts to scarf down as much as possible the “free lunch” of diversification. The managers are well aware that their human capital is tied to AQR’s success and so they would probably not want to concentrate too heavily in its potent leveraged strategies.

Opening date

QSPIX opened on October 30, 2013. QSLIX opened on September 17, 2014. The live performance composite began on September 1, 2012.

Minimum investment

The minimum investment varies depending share class, broker-dealer and channel. For individual investors, a Fidelity IRA offers the lowest hurdle: a mere $2,500 for the I share class of the normal and low-volatility flavors of SPA. Or you can get access through an advisor. Otherwise, the hurdles are steep: $5 million for the I class, $1 million for the N class, and $50 million for the R6 class.

Expense ratio

The I class for the normal and low-volatility versions cost 1.50% and 0.85%, respectively. The N classes costs 0.25% more and the R6 classes costs 0.10% less.

The per-unit price of exposure to SPA is lower the higher the volatility of the strategy. QSPIX targets 10% vol and costs 1.5%. QSLIX targets 5% vol and costs 0.85%. Anyone can replicate a position in QSLIX by simply halving the amount invested in QSPIX and putting the rest in cash. The effective expense ratio of a half QSPIX, half cash clone strategy is 0.75%.

Comments

Among right-thinking passive investors who count fees by the basis point, AQR’s SPA strategy elicits revulsion. It’s expensive, leveraged, complicated, hard to understand, and did I mention expensive?

To make the strategy easier to swallow, some passive-investing advocates argue SPA is “passive” because it’s a transparent, systematic, and involves no discretionary stock-selection or market forecasting. This definition is not universally accepted by academics, or even by AQR. The purer, technical definition of passive investing is a strategy that replicates market weightings, and indeed this definition is used by the venerable William Sharpe in his famous essay, “The Arithmetic of Active Management.”

I do not think SPA is passive in any widely understood sense of the word. In fact, I think it’s about as active as you can get within a mutual fund. And I also happen to think SPA is a great fund. Regardless of my warm feelings for the strategy, I consider SPA suitable only for a rare kind of nerd, not the investing public.

Though SPA is aggressively active, its intellectual roots dig deep into the foundations of financial theory that underpin what are commonly thought to be “passive” strategies, particularly value- and size-tilted stock portfolios (DFA has made a big business selling them).

The nerds among you will have quickly caught on that what AQR calls a style is nothing more than a factor, a decades-old idea that sprung from academic finance.

For the non-nerds: A factor, loosely speaking, is a fundamental building block that explains asset returns. Most stocks move together, as if their crescendos and diminuendos were orchestrated by the hand of some invisible conductor. This co-movement is attributed to the equity market factor. According to factor theory, a factor generates a positive excess return called a premium as reward for the distinct risk it represents.

It is now widely agreed that two factors pervade virtually all markets: value and momentum (size has long been criticized as weak). AQR’s researchers—including some of the leading lights in finance—argue there are two more: carry and defensive. They’ve marshalled data and theoretical arguments that share an uncanny family resemblance with the data and arguments marshalled to justify the size and value factors.

The SPA strategy is a potent distillation of the factor-theoretical approach to investing. If you believe the methods that produced the research demonstrating the value and size effects are sound, then you have to admit that those same tools applied to different data sets may yield more factors that can be harvested.

OK, I’ve blasted you with theory. On to more practical matters.

Who should invest in this fund?

Investors who believe active management can produce market-beating results and are willing to run some unusual but controllable risks.

How much capital should one dedicate to it?

Depends on how much you trust the strategy, the managers, and so on. I personally would invest up to 30% of my personal money in the fund (and may do so soon!), but that’s only because I have a high taste for unconventionality, decades of earnings ahead of me, high conviction in the strategy and people, and a pessimistic view of competing options (other alternatives as well as conventional stocks and bonds). Swedroe, on the other hand, says he has 3% of his portfolio in it.

How should it be assessed?

At a minimum, an alternative has to produce positive excess returns that are uncorrelated to the returns of conventional portfolios to be worthwhile.

However, AQR is making a rather bold claim: It has identified four distinct strategies that produce decent returns on a standalone basis and are both largely uncorrelated with each other and conventional portfolios. When combined and leveraged, the resulting portfolio is expected to produce a much steadier stream of positive returns, also uncorrelated with conventional portfolios.

So far, the strategy is working as advertised. Returns have been good and uncorrelated. In back-tests, the strategy only really suffered during the dot-com bubble and the financial crisis. Even then, returns weren’t horrendous.

Is AQR’s 0.7 Sharpe ratio target reasonable?

I think so, but I would be ecstatic with 0.5.

What are its major risks?

Aside from leverage, counterparty, operational, credit, etc., I worry about a repeat of the quant meltdown of August 2007. It’s thought that a long-short hedge fund suddenly liquidated its positions then. Because many hedge funds dynamically adjust their positions based on recent volatility and returns, the sudden price movements induced by the liquidation set off a self-reinforcing cycle where more and more hedge funds cut the same positions. The stampede to the exits resulted in huge and sudden losses. However, the terror was short-lived. The funds that sold out lost a lot of money; the funds that held onto their positions looked fine by month-end.

AQR is cognizant of this risk and so keeps its holdings liquid and doesn’t go overboard with the leverage. However, it is hard for outsiders to assess whether AQR is doing enough to mitigate this risk. I think they are, because I trust AQR’s people, but I’m well aware that I could be wrong.

Bottom line

One of the best alternative funds available to mutual-fund investors.