In pursuit of returns and diversification, growing numbers of investors are turning to alternative risk premia (ARP) strategies, which offer access to well-known sources of excess return potential – such as value, carry and momentum.
As this space is still outside the mainstream and rapidly evolving, the most common questions posed by investors considering ARP strategies can be challenging to answer: Which strategies should I allocate to, and in what amount? How important are strategy design and implementation? And how do I ensure appropriate risk management?
In the following Q&A, PIMCO strategists Brad Guynn and Ashish Tiwari provide insight into these questions.
Q: More than $300 billion of assets were invested in alternative risk premia strategies globally as of 2016.1 What’s driving their adoption?
Tiwari: Alternative risk premia strategies seek to provide investors with systematic exposure to well-known sources of excess return, such as value, carry, momentum and risk aversion, obtained across all major asset classes. These risk premia have been utilized by active managers, including PIMCO, for decades to seek a complementary source of alpha alongside traditional alpha sources (such as bottom-up selection and sector rotation).
They are fast gaining popularity as stand-alone investments because portfolios combining multiple alternative risk premia are among the few investment options that offer not only high return potential but also meaningful portfolio diversification. Generally lower fees and better liquidity terms than traditional hedge funds further add to their appeal. Finally, investors also value the transparency of these strategies, which may be easier to understand and monitor than traditional, qualitative investment strategies.
Q: How should investors determine which ARP strategies will best meet their objectives?
Guynn: It can feel overwhelming to sort through the cornucopia of potential alternative risk premia strategies.
More than 300 alternative risk factors have been identified in academic literature, according to a paper by Campbell R. Harvey, Yan Liu and Heqing Zhu titled “… and the Cross-Section of Expected Returns.”2
Other publications have cast doubt on the performance of many alternative risk factors. In their working paper, “The History of the Cross Section of Stock Returns,” Juhani Linnainmaa and Michael Roberts show that out-of-sample results for many equity-based alternative risk premia are significantly less compelling than (or even contrary to) in-sample results.3
When contemplating an ARP allocation, we believe investors should be highly selective. First, they should eliminate strategies that lack a reasonable economic explanation for their existence or a long history of persistent returns. Recently discovered factors should be approached with healthy skepticism. Second, investors should consider limiting allocations to only those strategies that can potentially deliver positive returns after making conservative assumptions for transaction costs. Finally, investors may want to focus on strategies that are truly devoid of traditional market exposures and therefore able to deliver truly diversified returns.
Attention to these key considerations may help investors avoid suboptimal outcomes and achieve their objectives for allocating to these strategies.
Q: How do I differentiate among strategies that target the same alternative risk premium?
Tiwari: Many strategies claim to target the same factors, but there is significant variation in how managers approach strategy design, implementation and overall portfolio construction. Even small differences in approach can lead to widely divergent outcomes. We believe strategy design, more than any other factor, will drive manager performance.
For example, foreign exchange (FX) carry strategies typically take long positions on higher-yielding currencies and short positions on lower-yielding currencies. As such, naïve implementations of this strategy are generally long emerging market currencies and short developed market currencies, thereby introducing significant latent correlations with equities and commodities. With appropriate adjustments and hedging, these undesired correlations can be lowered, thereby improving the portfolio’s diversification potential while continuing to harness the return objectives of the strategy.
Q: Will including more markets in my investment universe always lead to better diversification?
Guynn: Not really. A thoughtful approach to deciding which markets and securities to include is key and will depend on the rationale underlying the alternative risk factor being targeted.
Trend-following strategies offer a good example of why more is not always better. These strategies seek to generate returns by taking long or short positions on securities based on price trends in various asset classes. Over the long run, they have typically delivered attractive returns, but have often done so through infrequent large wins balanced against an accumulation of many smaller losses.
In order to both diversify portfolios and increase the probability of capturing these large moves, many managers try to maximize the number of markets where the strategy is run. However, this “kitchen sink” approach may result in allocating to highly correlated markets. Furthermore, shifting toward less-liquid instruments may also increase transaction costs without significantly diversifying the underlying exposures.
So while expanding the footprint is desirable, it’s more important to broaden the portfolio into uncorrelated markets to realize the desired benefits of an expanded universe.
Q: Are there benefits to solutions that integrate different alternative risk premia strategies?
Tiwari: We absolutely believe there are intuitive reasons why integrating certain strategies may lead to improved outcomes compared to implementing them on a stand-alone basis.
For example, value and carry are both sources of returns realized over a longer horizon. Intuitively, it is more desirable to own a cheap asset and wait for the price to revert to fair value while it pays you a positive income or carry than to do the reverse, when waiting incurs a negative return. This is often the case in commodities when the futures curve is in contango and the passage of time leads to a negative return as the higher futures price converges with the lower spot price. Thus, integrating value and carry allows for dynamic allocation between them in proportion to their relative attractiveness.
Developed market rates, for example, trade in a narrow range of real yields as a result of the unprecedented monetary stimulus over the past decade. The carry, as a result, is near zero. In contrast, value metrics signal more divergence across these markets. An integrated signal allows a portfolio targeting these factors to tilt toward the more attractive signal at a given time (e.g., value in today’s environment). (We elaborate on this concept in detail in a recent paper by our portfolio management team, “The Carry and Value Pendulum.”)
Q: Do transaction costs have meaningful implications for strategy design and implementation?
Guynn: Transaction cost considerations are particularly important for shorter-horizon strategies, for which trade frequency and portfolio turnover are high, and less-liquid strategy implementations, where positions may be quite expensive to enter or exit.
Trend-following and momentum-based strategies are examples of short-horizon/high-frequency strategies. The largest trend-following investors typically target longer moving averages, in part to reduce transactions costs. While using longer-horizon signals has recently led to outperformance, shorter-horizon signals tend to offer better diversification characteristics because they are able to catch the reversal of a trend faster.
Transaction costs are also a consideration for equity strategies that create factor portfolios by going long and short on individual names. In periods of market stress, the cost of shorting single-name equities can rise quite substantially and erode returns. As a result, many practitioners choose to go short via index futures, which substantially improves the liquidity and transaction cost profiles of these strategies.
In addition, large, diversified managers may be able to leverage their size and scale to secure better terms and execution with counterparties. Moreover, diversified lines of business allow larger managers to combine systematic trades with those of unrelated or more qualitative strategies. This makes it harder for trading counterparties to identify systematic trade flow, reverse-engineer signals and anticipate future trades – reducing the risk of falling prey to front-running.
Q: Is it possible to time exposure to ARP strategies?
Guynn: Timing allocations to specific ARP strategies is highly challenging, for two key reasons. The first is that these strategies experience idiosyncratic cyclicality that may be uncorrelated with traditional risk assets and difficult to anticipate. The second is that these strategies are uncorrelated with one another, making it more challenging to create value through active timing of exposures.
Nevertheless, certain strategies lend themselves to forming forward expectations that may be used to tilt risk exposures over time. For example, one may reasonably estimate the expected carry from certain positions or estimate the divergence of a security’s price from fair value. Therefore, strategies that target value or carry may lend themselves to forecasting.
Informed tilts away from simply equalizing risk allocations can be a prudent form of risk management. For example, as I mentioned earlier, the narrow dispersion of interest rate levels across developed markets today has led to compressed risk premia in rates-based value and carry strategies. To achieve an equal risk allocation with other strategies in the portfolio, investors would need to take on above-average exposure to rates-based strategies, but for less potential reward. As a result, in today’s environment it makes sense to fade exposures to rates-based value and carry ARP strategies and favor ARP strategies with a more attractive potential risk-reward.
Q: Where should alternative risk premia strategies fit within my strategic asset allocation?
Tiwari: First off, it’s important to understand that these strategies are similar to allocations to “beta” strategies in that they represent systematic exposure to sources of risk premia that undergo their own cyclicality. They will have their own ups and downs, albeit uncorrelated with those of traditional market factors. For this reason, investors need to observe ARP strategies over a full market cycle before forming conclusions about their performance.
In terms of how to classify an allocation to alternative risk premia, we observe investors taking a multitude of approaches to fit them into their policy allocations depending on the risk and return objectives of the ARP allocation. We think investors should consider funding allocations to alternative risk premia strategies from three sources of capital:
Alternatives/hedge funds - ARP strategies offer similar risk and return characteristics to hedge funds. Exposures to alternative risk premia may often be gained through cheaper and more liquid implementations than typical for hedge funds.
Fixed income - Low yields across high quality bond sectors mean the diversification offered by allocations may come at a steep opportunity cost with respect to returns. ARP strategies may thus represent a potentially higher-return diversifier. In some cases we’ve seen investors using ARP strategies as an overlay to a portfolio of high quality bonds.
Pro rata strategic allocation - Many diversified ARP strategies target returns in the mid- to high single digits. This is in line with the characteristics of many institutional portfolios. A pro rata allocation to a diversified portfolio is a common practice we’ve observed among institutional investors.
An interesting trend that has started to take hold in the U.S. is the use of alternative risk premia strategies as a component of a larger “risk mitigation” bucket. ARP strategies are combined with allocations to long-duration bonds and trend-following strategies in order to enhance portfolio diversification without the significantly lower returns that might be expected from more traditional diversifiers, like high quality bonds.
Q: How has the ARP space evolved over the past several years, and how should investors approach manager selection?
Guynn: This ARP market began to take shape in 2012, as investors sought to gain systematic exposure to many of the same structural opportunities that historically were available only to hedge fund managers and other sophisticated asset managers.
Now, five years later, we see three main styles of ARP managers emerging: quantitative equity managers, commodity trading advisors (CTAs) and macro managers.
Quant equity managers extend the insights of Fama-French equity factors into non-equity asset classes. Not surprisingly, their overall risk allocations tend to retain a bias for strategies implemented within an equity universe, with allocations as high as 70%+.
CTAs leverage their expertise in managing high-frequency-based strategies, such as momentum and trend-following, and offer exposure to other alternative risk premia and potential enhancements. Accordingly, they tend to retain a significant exposure to trend-following strategies as the main source of risk and returns.
Macro managers have long relied upon value, carry, momentum and risk-aversion concepts as fundamental parts of their macro style toolkits. These managers tend to be more balanced across asset classes, leading to relatively more exposure to rates, FX and commodities-based strategies than quant equity managers or CTAs.