Historically, bonds have played two primary roles in balanced investment portfolios. They have generally paid a higher level of current income than stocks, making them the go-to vehicle for those in need of current income. Additionally, they have provided a level of stability to portfolios due to their structure and relative safety (perceived and real). After a decades-long decline in interest rates, exacerbated by the global central bank response to the recent coronavirus pandemic, the income advantage of bonds has been significantly eroded (Exhibit 1). We believe however, there is still a substantial role to be played by bonds in your portfolio, particularly those issued by the United States Treasury.
Source: Federal Reserve Bank of St. Louis
In aggregate, the price volatility of the bond market has been significantly less than the price volatility of the stock market over time. This is primarily because bonds represent debt on the issuing entity’s balance sheet, and debts generally get repaid (more on that later). When you purchase a bond, you receive interest at the bond’s stated coupon rate until you either sell it or it reaches maturity, at which time you are repaid the principal. If your bond has a fixed coupon, and market interest rates increase while you own your bond, the price of your bond declines. This occurs because your bond’s coupon is now less valuable relative to current market rates, and a prospective buyer of the bond would demand a lower purchase price to make up for that loss in value. The reverse is also true when market rates decline (i.e. the price of your bond will increase), which is why bond prices move inversely relative to interest rates! Different types of bonds have different embedded risks however, so they don’t all react to changes in interest rates with the same magnitude.
A bond’s risk can fall into several categories (maturity, credit, liquidity, etc.). Credit risk is a measure of a borrower’s likelihood of timely debt service payments of both the periodic coupon (interest) and ultimate return of principal at maturity. Most bonds carry a credit rating issued by one or more independent agencies which helps investors to understand the degree of credit risk associated with the bond. Rating scales are letter-based, with a “AAA” rating signifying the highest grade, “AA” the next highest grade, and so on.
U.S. Treasury bonds are generally believed to be of the highest credit quality, being backed by the full faith and credit of the U.S. government. On the other end of the credit spectrum are high yield (aka junk) bonds issued by low rated (BB and below) companies. Everything else equal, yields across the credit spectrum are generally highly correlated with rating. The higher the rating, the lower the yield and vice versa. This reflects the need for investors to be compensated for taking on the additional credit risk associated with lower-rated issuers.
As interest rates have declined, particularly since the financial crisis, there has been an ongoing stretch for yield among income-oriented investors, leading to increasing amounts of investment dollars allocated to corporate bonds and the lower rated credit sectors.1 The mostly benign market environment since the financial crisis has generally rewarded this strategy, but the more recent market turmoil in early 2020 revealed this strategy’s weakness. The results below present evidence of this outcome.
We analyzed monthly total rate of return (TRR) data for several Blackrock exchange-traded funds (ETFs) that represent the major fixed income sectors. Our objective was to determine each ETF’s level of returns correlation with the broad equity market. We used the IVV ETF, which tracks the S&P 500 index, as a proxy for the equity market.
Each fixed income ETF and a brief description of its sector representation is listed here in descending order of credit quality:
IEF - iShares 7-10 Year Treasury Bond ETF
AGG - iShares Core U.S. Aggregate Bond ETF (proxy for the entire US high grade bond market including US Treasuries, mortgages and other structured products, and investment grade corporate bonds)
MUB - iShares National Muni Bond ETF (tax-exempt bonds issued by US state and local governments and their agencies)
LQD - iShares iBoxx $ Investment Grade Corporate Bond ETF (US corporate bonds rated BBB- and above)
HYG - iShare iBoxx $ High Yield Corporate Bond ETF (high yield/junk bond market)
We compiled monthly TRRs for each of these ETFs for the period October 2007 (the earliest point at which returns were available for all of the ETFs) to May 2021 and calculated how correlated those returns were with the returns for stocks. This period includes the financial crisis and resulting market crash, the long recovery period we enjoyed for most of the subsequent 12 years, and the 2020 bear market and recovery. Therefore, we feel the period measured represents more than a full market cycle.
The correlations for the entire period are shown below:
|IEF vs. IVV||-32.9%|
|AGG vs. IVV||3.2%|
|MUB vs. IVV||11.5%|
|LQD vs. IVV||33.7%|
|HYG vs. IVV||72.8%|
A positive percentage indicates positive correlation with stocks, and the larger the number, the greater the correlation. A negative number indicates that asset class actually moves in the opposite direction relative to stocks. The outcomes in the table are fairly intuitive. U.S. Treasuries, as the safe haven asset class, typically move in the opposite direction from stocks, deemed the riskiest of the asset classes we measured. Conversely, the high yield market moves directionally with stocks, which again seems intuitive since junk bonds represent the riskier end of the corporate debt market.
Given the extremes of recent market volatility, we decided to take our analysis a step further. We wanted to see how these correlations might shift during periods when stocks were doing either really well or very poorly. During the same measurement period (May 2007-May 2021) we analyzed correlations for only those months where the equity market TRR exceeded +5% or was worse than -5%. There were 23 months with a return exceeding 5% and 19 months with a return below -5% during the period, and these are the resulting returns correlations for those 42 months:
|IEF vs. IVV||-40.9%|
|AGG vs. IVV||11.7%|
|MUB vs. IVV||15.7%|
|LQD vs. IVV||51.6%|
|HYG vs. IVV||85.8%|
As can be seen from these numbers, during periods of extreme volatility in the market, all the positive correlations increased still further, and the negative correlations were more negative. Interestingly, in both sets of analysis, AGG (representing the broad investment grade bond market) was slightly positively correlated with stocks. Although this portfolio is roughly 70% allocated to bonds either explicitly or implicitly backed by the US Government, there are sufficient other exposures in the portfolio to end with a roughly neutral correlation with stocks. So what does this tell us, and how should it inform portfolio construction and the hedging of equity exposures in a balanced portfolio?
The quest for yield is a natural one, but it comes with a cost. Many individuals have bonds in their portfolios, including municipals, which they believe will provide stability to the overall portfolio if/when stocks go through a correction phase. However, as any form of credit risk is introduced, correlation to stocks increases, sometimes dramatically. As fund managers have stretched to increase advertised yields on their bond portfolios, they have increasingly turned to various forms of credit risk. These risks include the aforementioned investment-grade and high yield corporate bonds, as well as leveraged bank loans, collateralized loan obligations (CLOs), and lower rated municipal credits. As our correlation study shows, the higher the credit risk, the more likely a bond will go down when stocks go down, potentially compounding the negative returns experienced during the correction phase. In other words, the diversification and risk-reducing benefits received from these types of bonds tend to decline materially or even disappear at the exact time they are needed the most.
Even the investment grade market and municipal bonds are not much help during the most difficult market periods, as the chart below demonstrates. Exhibit 2 plots the price performance of the five ETFs in our correlation study during the violent equity sell-off of the first quarter of 2020, with IVV included as a proxy for the stock market. As indicated earlier, bond prices move inversely relative to interest rates, but they don’t all move with equivalent magnitude. Further, if enough credit risk is added to a bond portfolio, this risk may partially or fully offset the benefits received from falling interest rates during extreme market events.
Source: Blackrock, Portfolio Visualizer
As has generally been the case in prior periods of volatility, U.S. Treasuries did an excellent job of offsetting stock market risk during the most recent selloff in stocks. Interest rates remain near historical lows, which leads to the concern of how much lower rates can go and how much that might limit price upside for Treasuries going forward. While we do not forecast a particular future path for interest rates in the US, we do believe there is capacity for interest rates to move still lower than they are today. Several large, developed economies (Japan, Germany and France most notably) have issued significant amounts of debt at negative interest rates (That’s right, their creditors are paying them interest!), opening up the potential for rates in the US to follow a similar path. Only time will tell of course…
Shelburne, VT 05482
Burlington, VT 05401