March has been an eventful month. Republicans gave up on their attempt to replace Obamacare and the UK formally triggered Article 50 to initiate the Brexit process. However, the real news is the Fed raised its benchmark interest rate a quarter point to 1%. The Fed has made a shift in thinking and will raise interest rates faster and more aggressively than previously thought. It’s important to note that the Fed has not been this far behind the curve in the past 45 years. As shown in the following chart, the Fed historically tightens monetary policy when the unemployment rate touches the Non-Accelerating Inflation Rate of Unemployment (NAIRU) level or (full employment).
The NAIRU figures prominently in the Phillips curve, which is a relationship that incorporates a temporary trade-off between the unemployment rate and inflation. According to the Phillips curve, an unemployment rate that is below the level identified as the NAIRU (that is, a “tight” labor market) tends to be associated with an increase in inflation. Conversely, an unemployment rate that is above the NAIRU tends to be associated with a decrease in inflation. Tight labor (and product) markets have been one of the reasons for the Fed’s “preemptive strike” against inflation in the past. The strategy had worked particularly well during the Greenspan era, when price inflation stayed consistently low and stable. The current level and trend of inflation appear to be very similar to 1994, 1999 and 2004 rate hiking cycles (chart 2). It is the right time for a regime change at the Fed.
So far this year, short-term treasury yields have risen but long-term yields are basically unchanged — a phenomenon known as a flattening yield curve. A so-called flattening yield curve tends to occur when fixed-income investors are betting that the Fed is about to hike rates. This can lead to the sale of short-term notes which are significantly affected by changes in official rates. Note that the yield curve has been on a flattening trend since late 2013, when the Fed’s signaled a normalization of monetary policy.
Historically, changes to monetary policy have been process oriented. When the Federal Reserve raises or lowers interest rates, it rarely does so just once. As soon as the market believes the tightening (or easing) process is underway, the slope of the yield curve will shift downward. Higher short-term interest rates and a flattening yield curve act like a “tax” to leveraged bond strategies, significantly reducing their return potential. Take the carry and roll on a 10-year swap rate as an example:
A flattening yield curve disincentives banks and other financial institutions to take on more duration risk, and causes bonds to lose their “safe haven” status. The correlation between equities and bonds often increases during a flattening period. In fact, when the yield curve is inverted, the correlation between equities and bonds is consistently positive (Chart 3).
Bonds are a popular way to diversify risks due to their low/negative correlation with some of the other major asset classes, particularly equities. When the correlation turns positive, bonds can no longer deliver diversification benefits relative to equities. Strategies that are designed to rely heavily on using bonds as a diversifier will likely perform poorly, and risk parity is a typical example.
The risk parity approach to asset allocation allows investors to target specific levels of risk and to divide that risk equally across the entire investment portfolio in order to achieve portfolio diversification. For example, in a simple two asset class risk parity strategy, an investor equalizes risk exposure across bonds and equities, such that the risk contribution of bonds and equities to the overall portfolio volatility is the same. This means that lower risk assets (bonds) will have a higher allocation than higher risk assets (equities). Using S&P 500 and the Barclays US Aggregate Bond Index as building blocks, using data from February 1977 to February 2017, the risk parity portfolio on average allocates 73% of capital to Barclays Aggregate and 27% to S&P 500. To compare the strategy with a 60/40 portfolio (using the same two indices) at the same risk level, we also need to scale up the total risk exposure of the risk parity portfolio by a factor of 1.56x.
Here is the simulated historical performance:
In the long run, risk parity indeed outperforms 60/40 on a risk-adjusted basis. Since each asset class performs differently depending on economic conditions, by equalizing risk exposures to multiple asset classes the risk parity portfolio can weather a range of economic outcomes better than the 60/40, which will have more concentrated exposure to equities.
However, risk parity has its weaknesses too. In the same example, we separate the periods by the slope of the yield curve. During the 481 months of the sample period, the spread between 2-Year and 10-Year treasury yields is more than 100bps in 230 months, and for the other 251 months, the yield curve slope is equal to or flatter than 100 bps. As shown in the following table, the 60/40 strategy is a very consistent strategy. Its performance in steep yield curve environments is indistinguishable from the other periods. However, risk parity behaves markedly different in the two yield curve environments. During flat yield curve periods, risk parity underperformed 60/40 significantly in both absolute as well as risk-adjusted terms.
The current spread between 2-Year and 10-Year treasury yields is around 110bps. It’s reasonable to assume the spread will inevitably break downward through its 100 bps long-term average, after the Fed hikes 2-3 more times before the end of the year, as the market currently expects. Risk parity strategies will likely continue to underperform 60/40, just like what investors have seen in the first quarter of this year.
Investors in risk parity products, who still believe risk-based asset allocation is the primary determinant of long-term investment success, should consider alternative strategies. We believe strategies which can dynamically adapt to evolving conditions that may impact returns will do well in the current environment. Furthermore, we believe such dynamism can only be achieved through integrating multiple dimensions of risk, not simply volatility, at multiple time horizons.
The content contained in this report is produced and copyrighted by PhaseCapital, LP (“PhaseCapital”), an Investment Advisor registered with the Securities and Exchange Commission. Any unauthorized use, duplication, redistribution or disclosure is prohibited by law and can result in prosecution. The opinions and information contained herein have been obtained or derived from sources believed to be reliable, but PhaseCapital makes no representation as to their timeliness, accuracy or completeness or for their fitness for any particular purpose. This report is not an offer to sell or a solicitation of an offer to buy any security. The information and material presented in this report are for general information only and do not specifically address individual investment objectives, financial situations or the particular needs of any specific person who may receive this report. Investing in any security or investment strategies discussed may not be suitable for you and it is recommended that you consult an independent investment advisor. Nothing in this report constitutes individual investment, legal or tax advice. PhaseCapital may issue or may have issued other reports that are inconsistent with or may reach different conclusions than those represented in this report, and all opinions are reflective of judgments made on the original date of publication. PhaseCapital is under no obligation to ensure that other reports are brought to the attention of any recipient of this report. PhaseCapital shall accept no liability for any loss arising from the use of this report.Investments involve risk and an investor may incur either profits or loss. Past results are not necessarily indicative of future results. Barclays US Aggregate Indices are generally not investable indices and are used for informational purposes only.The content contained herein includes forward-looking statements, based on upon the current beliefs of PhaseCapital that are subject to risks and uncertainties. If underlying assumptions prove inaccurate or risks or other uncertainties materialize, actual results may differ materially from those set forth in the forward-looking statements.
WHERE SHOWN, SIMULATED RETURNS FOR THE 60/40 ARE CREATED USING WEIGHTS OF 60% EQUITY AND 40% BONDS USING THE STATED INDICES, REBALANCED MONTHLY. THE RISK PARITY ASSET MIX PERFORMANCE IS SIMULATED BY APPLYING RISK-WEIGHTED ASSET MIX WEIGHTS, WHERE RISK IS DEFINED AS VOLATILITY OF HISTORICAL MARKET RETURNS. WE USE ACTUAL MARKET RETURNS WHEN AVAILABLE AND OTHERWISE USE PHASECAPITAL’ PROPRIETARY ESTIMATES, BASED ON OTHER AVAILABLE DATA AND OUR FUNDAMENTAL UNDERSTANDING OF ASSET CLASSES. IN CERTAIN CASES, MARKET DATA FOR AN EXPOSURE WHICH OTHERWISE WOULD EXIST IN THE SIMULATION MAY BE OMITTED IF THE RELEVANT DATA IS UNAVAILABLE, DEEMED UNRELIABLE, IMMATERIAL OR ACCOUNTED FOR USING PROXIES. IN THE CASE OF OMITTED MARKETS, OTHER MARKETS IN THE SAME ASSET CLASS, WHICH REPRESENT THE VAST MAJORITY OF OUR POSITIONS IN EACH ASSET CLASS, ARE SCALED TO REPRESENT THE FULL ASSET CLASS POSITION. SIMULATED ASSET RETURNS ARE SUBJECT TO CONSIDERABLE UNCERTAINTY AND POTENTIAL ERROR, AS THERE IS A GREAT DEAL THAT CANNOT BE KNOWN ABOUT HOW ASSETS WOULD HAVE PERFORMED IN THE ABSENCE OF ACTUAL MARKET RETURNS. IT IS EXPECTED THAT THE SIMULATED PERFORMANCE WILL PERIODICALLY CHANGE AS A FUNCTION OF BOTH REFINEMENTS TO OUR SIMULATION METHODOLOGY (INCLUDING THE ADDITION/REMOVAL OF ASSET CLASSES) AND THE UNDERLYING MARKET DATA. THERE IS NO GUARANTEE THAT PREVIOUS RESULTS WOULD NOT BE MATERIALLY DIFFERENT. FUTURE CHANGES COULD MATERIALLY CHANGE PREVIOUS SIMULATED RETURN IN ORDER TO REFLECT THE CHANGES ACCURATELY ACROSS TIME.
It is expected that the simulated performance will periodically change as a function of both refinements to our simulation methodology and the underlying market data. HYPOTHETICAL OR SIMULATED PERFORMANCE RESULTS HAVE CERTAIN INHERENT LIMITATIONS. UNLIKE AN ACTUAL PERFORMANCE RECORD, SIMULATED RESULTS DO NOT REPRESENT ACTUAL TRADING OR THE COSTS OF MANAGING THE PORTFOLIO. ALSO, SINCE THE TRADES HAVE NOT ACTUALLY BEEN EXECUTED, THE RESULTS MAY HAVE UNDER OR OVER COMPENSATED FOR THE IMPACT, IF ANY, OF CERTAIN MARKET FACTORS, SUCH AS LACK OF LIQUIDITY. SIMULATED TRADING PROGRAMS IN GENERAL ARE ALSO SUBJECT TO THE FACT THAT THEY ARE DESIGNED WITH THE BENEFIT OF HINDSIGHT. NO REPRESENTATION IS BEING MADE THAT ANY ACCOUNT WILL OR IS LIKELY TO ACHIEVE PROFITS OR LOSSES SIMILAR TO THOSE SHOWN