SUMMARY
- An analysis of historical U.S. stock market cycles and concurrent trends in Third Pillar asset classes (real return, high yield bonds, emerging markets) reinforces the view that diversifying investments tend to shine when mainstream stocks struggle.
- When the current market cycle turns, Research Affiliates believes the All Asset strategies have the potential to minimize losses and to give diversification in the very market environments in which investors may need it most.
- Macroeconomic variables, especially global economic growth and inflation regimes, are top-of-mind considerations for altering strategic positioning within the All Asset strategies, to ensure they are responsive to shifting conditions that ultimately drive investment opportunities. Research Affiliates’ macro forecasting models assess trends and conditions within individual countries and across regions.
In this issue, Rob Arnott, chairman and head of Research Affiliates, discusses how stock markets and other asset classes tend to perform across market cycles over the long term, and looks at what this means for contrarian investment strategies. Jim Masturzo, head of asset allocation at Research Affiliates, discusses how macroeconomic models to forecast growth and inflation inform investment positioning. As always, their insights are in the context of the PIMCO All Asset and All Asset All Authority funds.
Q: How do you define a full market cycle, and what gives you confidence in the All Asset strategies’ long-term return prospects across cycles?
Arnott: Market cycles are often difficult to identify until after the fact. What’s more, there are multiple approaches to identifying the beginning and end points of a cycle. At Research Affiliates, we use a simple, sensible definition of a full market cycle: the span from one previous peak in cumulative total market return, to a subsequent higher peak, with an intervening decline of at least 20%. This approach generally ensures that a complete cycle captures both bull and bear market conditions. Using monthly U.S. equity returns since 1926 (as measured by the S&P 500 and, prior to 1957, the S&P 90), we identify seven peak-to-peak spans that include a total return drawdown of at least 20% from the previous market peak and that are followed by a rebound leading to a new, higher peak.
Before we turn to our All Asset strategies’ prospects today, let’s first examine U.S. stock market cycles over the last 95 years, and compare the latest cycle with that long-term history. Figure 1 shows the cumulative performance of the S&P 500 Index and a basket of Third Pillar assets1 (where available) over each of the full market cycles since 1926. Panel A displays the entire peak-to-peak span. We then dissect the full cycle period into its two component phases: the initial drawdown period from the first peak to the trough (Panel B), and the subsequent rebound from the trough to the new peak (Panel C).
Image Pop Up
Over full peak-to-peak cycles since September 1929 (the first peak of the measurement period), U.S. stocks as measured by the S&P index delivered annualized returns ranging from 2.0% (September 2000 – October 2007 cycle) to 15.4% (September 1987 – August 2000 cycle). On average, U.S. stocks returned nearly 9% per annum over a full cycle, which typically lasted 11 years.
Now let’s consider how Third Pillar assets generally fared over these cycles. Bear in mind that our observations begin in 1973, the first year when returns are available for at least three of our Third Pillar markets: high yield bonds, commodities, and real estate investment trusts (REITs). In market cycles over the last 46 years, a basket of Third Pillar assets on average returned 12.4%, exceeding U.S. stocks by approximately 2.5% per annum.
When we consider patterns of Third Pillar outperformance over market cycles, it’s evident that these diversifying assets tend to win when mainstream stocks falter. In Panel B, notice that from the peak-to-trough levels of past cycles, U.S. stocks fell around 20%–30%. Concurrently and relative to stocks, Third Pillar assets largely shrugged off these drawdowns, losing no more than 5% and delivering gains in two of the cycles. These observations are consistent with the truism that diversifying Third Pillar markets tend to shine when mainstream stocks struggle.
The current cycle, beginning with the 2007 peak just prior to the global financial crisis, was different. The drawdown for stocks was worse than most, falling by half, which works out to 41.4% per annum. The drawdown for Third Pillar strategies collectively was almost as bad. However, our All Asset and All Authority strategies fared far better during the market drawdown because we entered the financial crisis with a deeply defensive posture; All Asset All Authority’s inverse S&P 500 position as a risk hedge was particularly beneficial at limiting downside participation. In the subsequent recovery, Third Pillar markets have lagged S&P returns by half, instead of the 20% haircut typical in previous cycles. Most of this shortfall has occurred during the post-taper-tantrum period (after May 2013), in which we’ve seen the S&P 500 outpace most other liquid markets by a wide margin. Of course, this is what makes the Third Pillar markets collectively so very cheap today, especially compared with U.S. stocks.
The current cycle we’re in today will not last forever. As Keynes once said, “Trees don’t grow to the sky.” Mean reversion is relentless in the long run. When this cycle eventually peaks and turns downward, we believe that All Asset investors will be handsomely rewarded. Our investment approach has long relied on a variety of mechanisms in pursuit of meaningful long-term real returns. By design, our strategies tap into a deep toolkit of liquid alternative asset classes that offer diverse and diversifying risk premia, sources of structural alpha2 potential, and PIMCO’s decades-long track record of adding value across multiple asset classes. We then seek additional incremental return from tactically contra-trading across assets and trading against the markets’ most extreme bets.
When – not if – the cycle turns, we believe the All Asset strategies have the potential to minimize losses and give diversification in the very market environments in which investors may need it most.
Q: How are macroeconomic variables integrated into the allocation process of the All Asset strategies?
Masturzo: Along with asset fundamentals, macroeconomic variables are an important part of the allocation process within the All Asset strategies. In particular, global economic growth and inflation regimes are top-of-mind considerations for altering strategic positioning within the funds, as different asset classes respond differently to the different combinations of high/low growth and high/low inflation environments. Only by incorporating a perspective on these regimes into the models can the All Asset strategies be responsive to shifting conditions that ultimately drive investment opportunities.
Painting with a broad brush, let’s consider a few examples of asset class opportunities within four combinations of environments using qualitative categories of high/low economic growth and high/low inflation. Note that while these are not the only asset classes to consider in these environments and other dynamics can arise in any individual regime, Figure 2 is intended to provide theory-based context to each environment.
As with our other models within the All Asset strategies, namely our modeling of global asset class return expectations, we estimate both growth and inflation expectations at the individual country level. By considering over 20 individual countries in our models, we capture cross-sectional nuances which would often be missed with a U.S.-centric or global top-down approach.
Let’s begin with our inflation models, before turning to our economic growth models. Our models are based on recent inflation trends within countries combined with our confidence in each central bank’s ability to reach its publically stated inflation target (see the August 2019 All Asset All Access for details). Instead of diving deep into the design of the model, let’s focus on how we use the model’s output to inform allocation decisions within the All Asset strategies.
The primary usage of our inflation model is to forecast the future trajectory of nominal yields across countries. Although we focus our portfolio allocations on real returns, consistent with a stated objective of the All Asset strategies, nominal variables can and do have a large impact on the valuation of some assets. As we have discussed recently (see the January 2020 All Asset All Access), our forward-looking return models capture both cash flow expectations and expected changes in prices. As the change in price of a nominal bond is affected by nominal yields, our view of inflation – and, more importantly, our view of changes in inflation from today’s levels – is paramount to determining this value. To put a finer point on this, our expectations of real rates largely help us determine allocations between nominal government bonds and other assets, while our expectations of inflation changes directly affect our choice of nominal government bonds versus U.S. TIPS.
The other major macroeconomic variable considered within the All Asset strategies is a growth estimate. Like inflation, growth is measured at the individual country level and aggregated into the regions that map to our investment opportunity set. Our growth model, which focuses on a country’s business cycle, captures growth relative to recent trend growth, which has been shown to affect asset returns (see the October 2019 All Asset All Access for details). Because it’s impossible to know whether growth is above or below trend in real time, our model uses a set of indicators to estimate the probability of below-trend growth in the next one to two quarters. For example, our most recent estimate in the U.S. is of a 67% probability of slowdown (33% probability of acceleration) in the coming three to six months. Taken in isolation, this reading would indicate a preference for U.S. assets that tend to do well in low-growth environments and, at times and depending on global growth estimates, a preference for non-U.S. assets.
In actuality, our growth estimates are not taken in isolation. Instead, they are just one component of the All Asset strategies’ process and are used to refine our expected returns of assets based on their expected response to growth and slowdown regimes. By adjusting long-term expected returns in this way, the model increases responsiveness to short-horizon factors such as lower (higher) fair value P/E ratios3 for equities that are more natural in slowdown (growth) periods. Similarly, our growth model provides adjustments for expectations of future yields, the default and recovery rates that are crucial in pricing credit assets, as well as cycle-aware fair value commodity and currency price estimates.
Asset class volatilities and cross-correlations also vary during growth and slowdown periods. When economies are slowing, asset volatilities across the risk spectrum often rise, while correlations, on average, also rise. Our business cycle model also injects these dynamics into the allocation process by adjusting the risk metrics used in the All Asset strategies’ optimization processes.
Overall, our asset allocation approach remains grounded in strategic expectations, as shorter-term regime dynamics can be noisy to estimate and may lead to unnecessary turnover in the portfolio. That said, when readings on various regimes flash red, our strategies are capable of positioning themselves accordingly. As such, complementing our long-term expectations of risk and return with adjustments for expectations of growth and inflation regimes is essential, allowing the All Asset strategies to take advantage of the dynamics that take place during different regimes, for the benefit of investors.
The All Asset strategies represent a joint effort between PIMCO and Research Affiliates. PIMCO provides the broad range of underlying strategies – spanning global stocks, global bonds, commodities, real estate, and liquid alternative strategies – each actively managed to maximize potential alpha. Research Affiliates, an investment advisory firm founded in 2002 by Rob Arnott and a global leader in asset allocation, serves as the sub-advisor responsible for the asset allocation decisions. Research Affiliates uses their deep research focus to develop a series of value-oriented, contrarian models that determine the appropriate mix of underlying PIMCO strategies in seeking All Asset’s return and risk goals.
1 The basket of Third Pillar asset classes in Figure 1 represents six diversifying, real-return assets: real estate investment trusts (REITs), high yield bonds, commodities, emerging market (EM) local bonds, EM equities, and U.S. long Treasury Inflation-Protected Securities (TIPS).
2 Structural alpha strategies seek to exploit certain market inefficiencies; examples include targeting interest rate differentials or financing bond positions through reverse repurchase (“repo”) agreements or forward settlement transactions (i.e., deferred settlement).
3 The price-to-earnings or P/E ratio is often used to assign value to a company. It is the ratio of a company's current share price to its earnings per share (EPS).
DISCLOSURES
Investors should consider the investment objectives, risks, charges and expenses of the funds carefully before investing. This and other information are contained in the fund’s prospectus and summary prospectus, if available, which may be obtained by contacting your investment professional or PIMCO representative or by visiting www.pimco.com. Please read them carefully before you invest or send money.
©2020, PIMCO.
© PIMCO
More Active Management Topics >
Image Pop Up
