What Would an Interest Rate Rise Mean for Bonds?
On February 27, 2013, the yield on the 10-year Treasury bond stood at 1.85 percent, an increase of 47 basis points (0.47 percent) since the low hit in July 2012.[i] However, the yield remains much lower than the 10-year average of 3.70 percent. With the December Core Consumer Price Index rising 1.9 percent year-over-year, inflation-adjusted bonds for the next five years are trading at a negative real yield.[ii] Yet, inflows into fixed income investment vehicles continue to outpace inflows into stock investment vehicles (which have seen net outflows since 2007). With bonds having experienced a three-decade long bull market, investors continue to overlook the risks inherent in bonds at their current low yields. In Jeremy Siegel’s op-ed in The Wall Street Journal in 2010, he wrote: “The last time interest rates on Treasury bonds were as low as they are today was in 1955. The subsequent 10-year annual return to bonds was 1.9%, or just slightly above inflation, and the 30-year annual return was 4.6% per year, less than the rate of inflation.”[iii] When Siegel wrote that piece, the yield on the 10-year Treasury was approximately 2.8 percent. Thus, the yield has fallen by almost 30 percent since then. Was Siegel’s thesis incorrect or just early? It is our view that Siegel’s forecast was just early. We believe that fixed-income as an asset class provides a poor risk-reward balance.
In 1981 the yield on Aaa-rated corporate bonds hit 15.5 percent. This was in response to Paul Volcker’s decision to combat double-digit inflation by raising the federal funds rate to almost 20 percent. Since that time yields have steadily declined for mostly all types of fixed-income securities. Because of the inverse relationship between bond yields and price, a decline in yields has caused an increase in the prices of bonds.
As the above diagram demonstrates, the convexity of a bond causes changes in yields to have a greater effect on the price of a (noncallable) bond at lower yields. Therefore, at today’s historically low yields, a slight increase in interest rates could cause a noticeable decline in bond prices. For bonds a security’s interest rate risk is approximated by a statistic known as duration. For example, if the duration of a bond was 5, this means that if interest rates rose by one percent, all other things being equal the bond would lose 5 percent of its value. Also, for larger moves in interest rates, duration tends to overestimate (underestimate) the decrease (increase) in price if interest rates declined (rose) because of the bond’s convexity. Thus, a bond’s duration is a simple answer to the question of how risky a bond is if interest rates rise. Although individual bonds are also exposed to other risks, including credit and liquidity, interest rate risk is the primary driver of price movements in the majority of bond classes. In order to attempt to quantify the amount of losses that an investor could experience on a bond holding, it is helpful to look at past experiences during which interest rates rose. In a study published by Welton Investment Corporation, the firm examined how seasoned Aaa-rated bonds performed when interest rates rose by at least 1.5 percent. Since 1920, this scenario has occurred eight times. The largest increase occurred in the mid-70s as interest rates rose by 7.6 percent from trough-to-peak. During that time period Aaa-rated corporate bonds lost 24 percent of their value.
There are a few things to keep in mind when viewing this graph. First, this shows the performance of Aaa-rated corporate bonds. Therefore, these corporations are considered the most creditworthy by Moody’s, and as a result their bonds would tend to perform better during downturns. Yet, as evinced by their past performance, even the highest of quality bonds can lose money. Second, the severity of losses is partially determined by the average coupon at the start of the uptick in interest rates. Bond returns are made up of two components: capital gains/losses and interest payments. Since the interest payments act as a buffer against losses, the smaller the interest payment (the lower the interest rate) the smaller buffer there is against receiving losses. For example, if the interest rate on bonds are on average 2 percent, it takes a much smaller increase in rates to begin to see losses than if the average interest rate on bonds was 5 percent. This is the reason why zero-coupon bonds that mature in 5+ years tend to be some of the most volatile bonds. Because an investor only receives his money back at maturity and does not receive any interest payments before then, the bonds are highly sensitive to changes in interest rates. Similarly, if investors are receiving smaller semi-annual coupon payments because rates are so low, then their bonds are also more sensitive to interest rate fluctuations. In its study, Welton Investment Corporation compared the annual coupon to the resulting capital loss. During the late 1960s, the interest rate on the bonds was initially 5.4 percent. Since the decline in bonds at their low was 24.3 percent, investors’ peak loss was 4.5x their annual coupon. During the late 1950s, the interest rate was initially 3.33 percent. Although investors’ peak loss was 15 percent, this was also 4.5x their annual coupon. Thus, in both periods the risk/reward for investors was the same. This has implications for today’s bond investors because rates are so low. The average yield of Moody’s seasoned Aaa-rated corporate bonds is 3.65 percent.[vi] If these bonds experienced a loss of 4.5x this annual yield, the capital loss would amount to 16.43 percent.
Another aspect to consider is the role of inflation on the returns of fixed-income securities. Historically, there has been a positive correlation between inflation and interest rates. Interest rates are typically increased if economic growth increases in order to ward off higher inflation, and interest rates are typically lowered if economic growth decreases in order to ensure that deflation does not occur. While economic growth is not necessary for inflation to occur (the 1970s was a time of stagflation), the two usually go hand-in-hand. During the 1970s, economic growth stagnated which created a dichotomy for bonds. An interest rate rise and economic stagnation caused corporate bonds to perform poorly. However, although interest rates rose, investors still purchased Treasuries for their perceived safety. Following is a table that displays the returns of Treasury bills and Treasury bonds from 1973-1980.
Annual Return T-Bills
Annual Return T-Bonds
After Inflation T-Bonds
As exemplified by the table, Treasury bills (which are issued with a maturity of one year or less) outperformed Treasury bonds. In a rising interest rate market, short term fixed-income securities typically outperform long term fixed-income securities. The reason for this is that investors have a shorter time to wait to receive their principal back, at which time they can re-invest the proceeds into newly issued fixed-income securities with higher coupon rates. For this reason short duration bonds have become popular lately as investors are worried over the possibility of rising interest rates and inflation. Despite their nominal growth of 7.34 percent over this time period, Treasury bills experienced an inflation-adjusted loss of 1.54 percent. Although this was the best return for any type of fixed-income security from 1973-1980, this still left an investor with less purchasing power at the end of 1980 than he or she had in 1973. Consequently, Treasury bills were not the panacea during this time period, and they most likely would have attained worse returns if economic growth accompanied the higher interest and inflation rates.
One type of fixed-income asset class that was not prevalent during the rising interest rate environment of the late 1970s was high-yield (non-investment grade) bonds. Prior to that time period, most non-investment grade companies did not have access the public markets as most investors considered them to be too risky. Instead, these companies received financing primarily through bank loans. Mainly all of the non-investment grade corporate debt that was publicly traded belonged to “fallen angels” – companies who were once rated as investment-grade but had declined to non-investment grade. In the mid-1970s, Michael Milken, who worked at the investment bank Drexel Burnham Lambert, came to the conclusion that a portfolio holding a diversified group of high-yield companies could outperform a portfolio containing only investment grade companies. His conclusion led to the junk bond boom of the 1980s. Since that time a secular decline in bond yields has occurred. As a result there is little evidence on how junk bonds could perform over the long term if interest rates rose.
Over the past three decades, one fact regarding high yield bonds that has become apparent is their relatively high correlation to equities. Correlation is the measure of the linear relationship between two variables. Correlation can range between +1.00 for perfect positive correlation and -1.00 for perfect negative correlation. For example, if two stocks or sectors had a correlation of .90 then that means their prices have moved in tandem 90 percent of the time. Historically, U.S. high yield bonds have a correlation of .62 with the S&P 500 Index. In comparison, U.S. investment-grade corporate bonds have a correlation of .21 with the S&P 500 Index, and U.S. Treasury bonds have a correlation of -.31 with the S&P 500 Index.
Despite their high correlation with U.S. equities normally, this correlation has tended to increase during times of market volatility. In 2008, the correlation between high yield bonds and stocks averaged .88 – both sold off heavily.[viii] The S&P 500 Index ended 2008 down 38 percent while U.S. high-yield bonds fell by an average of 25 percent. Much like the equity market though, high-yield bonds have rebounded substantially since then. Historically, the success of high-yield bonds has been more of a function of the performance of the overall economy than a function of the level of interest rates. If this relationship were to hold in the future, the performance of high-yield bonds should reflect the global economic outlook. If interest rates were to rise because the economy was continuing to improve, then high yield bonds should do well. However, if a 1970s scenario – high interest rates, but low economic growth – was to occur, high yield bonds would most likely be one of the worst performing asset classes.
Another type of popular bond is a Floating Rate Note (FRN). Instead of paying a fixed rate of interest, a FRN pays a floating rate that resets periodically with rates linked to a particular interest rate index. These types of bonds were first issued during the 1970s in order for companies to generate investor demand for their bonds in spite of the high interest rate volatility. FRN yields are typically stated as being a certain number of basis points (1 basis point = .01 percent) above a certain index, such as the 3-month Treasury bill rate, the Federal Funds Rate or the 6-month London Interbank Offered Rate (LIBOR). The interest rate is reset periodically depending on the index. For example, a FRN based on the Federal Funds Rate would have a coupon rate that is reset daily because the Federal Funds Rate is an overnight lending rate. The interest rate that is used to determine the coupon is based on the rate on a date set by the bond indenture (contract). Because the interest rate of the bond is reset periodically, there is less interest rate risk in a FRN than with a fixed-rate bond with the same maturity. Morningstar estimates that the duration of the average FRN is 0.45.
One downside is that there is typically more credit risk associated with the companies that issue floating rate notes. This is because many floating rate notes are bank loans. Since corporations with the best credit usually obtain financing by accessing the public capital markets, the majority of companies that use bank loans are below investment grade. Therefore, in times of economic hardship the typical non-investment grade company experiences problems paying off debt as revenues decline, and the value of their floating rate debt decreases. In 2008, floating rate bonds lost an average of 29 percent.[ix] For the period February 1992-April 2011, the correlation to both short-term and long-term Treasuries was negative, while the correlation was 0.74 to high-yield corporate debt and 0.42 to U.S. stocks. Consequently, the performance of floating rate bonds will most likely depend on both the change in interest rates and future economic growth.
With nearly all types of bonds trading near record highs, this leaves little room for increases in their prices. If interest rates do revert to historical means (approximately 4 percent on the 10-year Treasury), then the majority of bonds would most likely perform poorly. Using historical data as a guide, here are our estimates as to how each type of bond could perform under different economic scenarios if yields were to rise from today’s current levels.
Interest Rate Rise – How Each Type of Bond Could Perform Based on Historical Data
Type of Bond
Economy Does Well
Economy Does Poorly
|Investment Grade Corporate Short Duration|
|Investment Grade Corporate Long Duration|
As demonstrated by the table, we believe that most bonds would probably not perform well in these scenarios. Yet, if bonds were to underperform, where should investors put their money? We believe that equities and real estate are viable options. Historically, equities have been the best asset class during times of modest inflation because companies have the ability to pass along moderately rising costs to consumers. Also, many corporations with strong balance sheets and cash flows have been wisely issuing fixed-rate debt in order to lock in historically low interest rates. By investing in the stocks of these same companies, an investor may increase the probability that his or her portfolio could succeed in a rising interest rate environment.
*The opinions and forecasts expressed are for informational purposes only and may not actually come to pass. This information is subject to change at any time, based on market and other conditions and should not be construed as a recommendation of any specific security or investment plan. The representative does not guarantee the accuracy and completeness, nor assume liability for loss that may result from the reliance by any person upon such information or opinions.
[i] Zeng, Min. Ten Year Treasury Yield Tests 2%. 28 Jan 2013. The Wall Street Journal. 31 Jan 2013. <http://professional.wsj.com/article/SB10001424127887324329204578269640390484444.html?mg=reno64-wsj.>
[iii] Siegel, Jeremy. The Great American Bond Bubble. 18 August 2010. The Wall Street Journal. 31 Jan 2013. <http://professional.wsj.com/article/SB10001424052748704407804575425384002846058.html?mg=reno64-wsj.>
[iv] Sherwin, Ryan. Bond Education: Understanding a Callable. 27 April 2010. Credit Unions. 31 Jan 2013. <http://www.creditunions.com/articles/bond-education-understanding-a-callable/.>
[v] When Bonds Fall: How Risky Are Bonds if Interest Rates Rise? 2012. Welton Investment Corporation. 31 Jan 2013. <https://www.welton.com/uploads/insight/Welton-When_Bonds_Fall_%28Visual_Insight_Series%29.pdf.>
[vi] Seasoned Aaa Corporate Bond Yield. 31 Jan 2013. The Financial Forecast Center. 1 Feb 2013. <http://www.forecasts.org/interest-rate/moodys-corporate-bond-yield.htm.>
[vii] Jones, Kathy A. High-Yield Bonds – Extra Income, But Added Risk. 5 June 2012.
[viii] High Yield Bonds and Their Correlation to Equities. Dec 2008. Millares Asset Management. 4 Feb 2013. <http://www.artisanmg.com/millares/pdf/millares_583023.pdf.>
[ix] Swedroe, Larry. Floating-rate Note Funds Should Be Avoided. 18 Nov 2011. CBS News. 5 Feb 2013. <http://www.cbsnews.com/8301-505123_162-57326911/floating-rate-note-funds-should-be-avoided/.>