A relatively new way to construct portfolios says many investors aren’t truly diversified.
The future delights in humiliating seers.
In 2007, who would have predicted that the rich world would reprise the Great Depression? And yet here we are: The U.S. government can run annual deficits amounting to 9% of GDP, triple the monetary base to $2.6 trillion in a few years, and still issue 30-year bonds yielding 3%.
The surprising frequency of the “unthinkable” happening suggests we are overly confident in our ability to see the future. The “risk-parity” approach to portfolio construction is a powerful way to combat this mistake.
Pioneered by Bridgewater Associates, modern risk-parity portfolios adhere to a simple principle: Balance portfolio exposures across all the major economic scenarios by volatility. The hope is that such a portfolio will perform well in all economic climates— and indeed, risk-parity strategies have.
The strategy is gaining influential adherents, mainly among institutions, but hasn’t caught on with advisors and individuals. Will this strategy work in the future? Could a client apply it? I think the answer is, tentatively, yes to both questions. To understand why risk-parity works, we have to revisit a common fundamental misconception of portfolio construction.
The typical investor thinks of assets as being like indivisible elements, with distinct characteristics. In contrast, the risk-parity approach begins with the observation that asset classes can be described in much the same way atoms can be described as combinations of electrons, protons, and neutrons. This isn’t a modern insight, but one that’s been around academia for decades. The risk-parity application is relatively new, however.
According to Bridgewater, the fundamental particles in the risk-parity view of the world are inflation and economic growth. A 2012 paper (“The Risk in Risk Parity: A Factor Based Analysis of Asset Based Risk Parity,” by Vineer Bhansali, Josh Davis, Graham Rennison, Jason Hsu and Feifei Li) confirmed this observation, finding global growth and global inflation could explain the majority of the behavior of a variety of asset classes.
Standard portfolio construction advice does not lead to risk-balanced portfolios.
The typical 60% stock/40% bond portfolio’s volatility is 90% determined by equities, which are reliant on stable inflation and economic growth. What happens when inflation ticks up or growth slows? Standard portfolios fail.
The Four Economic Configurations
Economies have four main configurations, characterized by combinations of rising/falling inflation and rising/falling economic growth. Changes in inflation and economic growth can explain the gross behavior of almost any asset class.
Naturally, in each phase of the business cycle, different assets are king. Stocks do best when the economy is growing and inflation is falling, coinciding with the recovery phase after a recession; bonds do best when the economy is tanking and inflation is falling, coinciding with the downward leg of a conventional recession.
These are sensible relationships. Stocks rise in anticipation of increased earnings growth, and bonds rise when inflation or interest rates fall. For the most part, it doesn’t really matter what kind of stocks or bonds you own—whether they’re growth/value or corporate/Treasuries— each asset class will largely obey its relationship to economic growth and inflation.
This is not to say valuation doesn’t matter. It’s still the beating heart of long-run returns. When valuations get stretched to extremes, prices can snap back to more normal levels, regardless of what inflation or economic growth does. In other times, valuation’s effects are more subtle, operating as a “drift term,” all but imperceptible in the day-to-day volatility. The Warren Buffetts of the world can ignore volatility for years on end and focus on maximizing return for some distant future, but most investors don’t have that luxury. How they get to their destination matters, too.
The conventional, equity-dominated portfolio is mostly a bet on economic growth and stable or declining inflation. Such a portfolio is geared to do well half the time. This is tolerable if the bad months are randomly interspersed with good months. However, history suggests that economic scenarios are clumpy, sometimes lasting decades, lulling investors into only preparing for the risks that have showed up in recent memory.
Investors failing to hedge their bets against all four economic scenarios are implicitly betting on their prescience and exposing themselves to big tail risks. This seems foolhardy, given that in the past 40 years we’ve had inflationary recessions (1970s), a deflationary recession (2008–09), and disinflationary booms (1980s and 1990s). The practical outcome of risk-parity thinking is to add more inflation hedges to the conventional 60%/40% stock/bond portfolio, reducing the volatility that equities contribute.
The Original Risk-Parity Strategy and Beyond
Harry Browne’s Permanent Portfolio is probably the most well-known risk-parity-like strategy. Browne, an author, politician, and investment analyst, kept it dead simple: 25% allocations each to gold, long Treasuries, equities, and cash, roughly balancing one’s risk exposures across all four economic configurations.
While it is crude compared with newer volatility-targeted risk-parity strategies, the Permanent Portfolio has stood the test of time. Over the period 1973–2012, its Sharpe ratio, defined as return above cash divided by its standard deviation, was a respectable 0.48, better than the 60/40 portfolio’s 0.4. The Sharpe ratio understates its advantage, because the strategy had smaller drawdowns, less fat-tail risk, and held up well when you needed it to.
Thanks to the index-fund revolution, Browne’s portfolio is cheap and easy to implement. I fear, however, that investors expecting the Permanent Portfolio and its variations to repeat last decade’s performance are in for disappointment. Browne’s strategy is reliant on now richly valued hedge assets, gold and Treasuries. Investors can realistically expect only low-single-digit or zero real returns from today’s valuations.
I believe Browne’s formulation can be improved without sacrificing much in the way of simplicity, low costs, and robustness if we’re willing to make a few active bets.
Rather than keeping 50% of the portfolio in Treasuries and cash, I opt to go active with PIMCO Total Return ETF
In other words, PIMCO Total Return can be seen as a superior diversified fixedincome vehicle, rather than a pure-play on the market-timing powers of one man. The fees of the institutional share class and ETF versions of Total Return are arguably low enough to make the diversification benefit itself worthwhile.
There’s a good case for low-cost active management on purely theoretical grounds. The market-cap-weighted indexes reflect the bets of the “average” investor, but the “average” investor in this world includes foreign central banks pursuing macroeconomic stability mandates, pension funds and insurance companies hedging liabilities, sovereign wealth funds, and of course the Federal Reserve itself. There is certainly room for active management to add value by front-running these non-economic actors—again, provided you’re not paying a lot for it.
The other major bet I make with my take on risk-parity is low-volatility equities. This twist has a two-fold advantage.
First, it allows me to ramp up equities, which are undervalued to the hedge assets, without making the portfolio too reliant on economic growth.
The second is that low-volatility equities have historically returned about as much as the broad market, but with—obviously— lower volatility. This return pattern has plausible economic reasons, meaning we can reasonably expect it to persist over the long run. (This so-called low-volatility anomaly deserves a discussion of its own, but we’ll leave that for another day.)
The final change I make to the Permanent Portfolio is the use of an actively managed broad commodity mutual fund. While ETFs are near and dear to my heart, no commodity ETF right now is competitive with the best-of-breed, institutional-share-class mutual funds. PIMCO Commodity Real Return Strategy
In other words, Commodity Real Return is a leveraged bet on inflation.
The ability to employ leverage is critical feature of risk-parity strategies. Indeed, leverage may be the reason why risk-parity should work. In an ideal world, where anyone can borrow at the risk-free rate with zero risk, Modern Portfolio Theory says that investors should invest in the best risk-adjusted portfolio and, if unsatisfied with the portfolio’s expected return, lever it up.
Needless to say, the world is hardly ideal. Leverage is either outright banned by regulation or perceived as too scary and something to be avoided by many investors. It has its own risks. Many investors resort to concentrating their assets in equities and equity-like assets to achieve their return targets.
We are observing such a shift today. Investors are shifting from cash to bonds to dividend stocks to boost their yields.
The result of widespread leverage aversion is that low-volatility asset classes offer higher risk-adjusted returns, which can be exploited by investors willing to use leverage— that is, risk-parity investors.
Several risk-parity mutual funds attempt to earn higher returns by applying leverage to balance the volatility contributions of a variety of asset classes, including commodities, corporate bonds, and currencies. The biggest “true” risk-parity funds are Invesco Balanced-Risk Allocation
Even if you don’t fully trust these newfangled risk-parity strategies, it’s still worthwhile to stress-test portfolios to see how sensitive they are to all four economic scenarios and perhaps to reconsider whether avoiding leverage is worth the cost.
Note: In the table “Percent of Months in Each State, 1928-2012,” inflation is defined as year-over-year CPI growth above the 12-month moving average of year-over-year CPI growth; disinflation is when current year-over-year growth is below its 12-month moving average. Growth and recession are defined by the determination of the National Bureau of Economic Research’s Business Cycle Dating Committee.