Theory and history suggest that for bonds, changes in interest rates are the dominant risk factor.
Most of the rest of the risk is explained by credit risk.
History also shows that in times of greatest stress, credit risk and liquidity risk can dominate interest rate risk.
These characteristics can help think about what role bonds should play in a portfolio, and how it might make sense to think about asset classes within the category of bonds.
Interest Rate Risk and Credit Risk Matrix
To a first approximation, we can think that there are four distinguishable asset classes within bonds. These four classes are the four combinations duration and credit risk. They are:
- Short duration, low credit risk
- Short duration, higher credit risk
- Long duration, low credit risk
- Long duration, higher credit risk
Or, in a matrix:
Short Duration Long Duration
Low Credit Risk 1 3
High Credit Risk 2 4
Historically, it has generally been the case that in times of market stress, such as the global financial crisis, bonds with higher perceived credit risk have also displayed signs of liquidity risk. Thus, it may be that credit risk does not fully capture the true downside risk of less than “riskless” bonds.
If the purpose of bonds in a portfolio is to reduce risk, especially in times of stress, there seems to be a strong case to favor low or “no” credit risk bonds.
Based on historical observations, it may be that for bonds to play the stabilizing role in times of extreme stress, only Treasury-type bonds (i.e. bonds issued in the currency by the issuer – which means government bonds of the currency issuer) have behaved as expected. As noted earlier, for example, during the Global Financial Crisis, even AAA rated corporate debt traded at extremely wide spreads to treasuries, and with extremely wide bid-offered spreads.
Viewed in this light, there is a case that for the vast majority of American investors, who measure results in US dollars, only two bond asset classes need to be considered:
- Short term US Treasuries
- Long term US Treasuries
We might further refine these two asset classes by looking at the short end and the long end of the yield curve.
Short Term Treasuries
The short end of the US Treasury debt market is dense with offerings. Various maturities of T-bills are issued almost every week. Notes and bonds are issued every quarter.
The money market rate is frequently represented by the 90 day or 3-month Treasury Bill. The short end of the note/bond market is frequently represented by the 2-year note.
Over the past approximately 50 years, the average yield curve has usually been upward sloping. That means that yields rise as the length of the maturity rises.
On average over this time, the 2-year note has yielded about 75 basis points (3/4ths of 1%) more than the 3-month Treasury Bill. (Today that spread is close to zero).
Thus, at the short end, for most investors it may (on average) pay to hold 2-year notes rather than 90-day T-bills. The duration of the 2-year note tends to be close to two. But while this is a multiple of the duration of the 3-month bill, the absolute number is still small, and on average the additional yield has been worth it.
Long Bonds
The long end of the US Treasury bond yield curve is represented by 10-year notes and 30-year bonds. Over the past half century, the yield on the 30-year bonds has averaged about 36 basis points (36/100ths of 1%) greater than the yield on the 10-year notes.
Thus, on average, the expected return on 30-year bonds has been slightly higher than the expected return on 10-year notes.
However, the size of that return advantage is small, particularly when compared to the additional interest rate risk of 30-year bonds.
For example, at current interest rates, the duration of the US Treasury 30-year bond is about 16, while the duration of the 10-year bond is about 8.
Thus, everything else equal, for most investors the 30-year probably doesn’t offer enough expected return to justify the additional interest rate risk.[1]
For the reasons outlined above, we believe that for most US based investors, the only bond asset classes that need to be considered are short- and long-term US Treasuries, which we believe can be simplified to 2-year notes (short term) and 10-year bonds (long term).
Other Bond Asset Classes?
There are other types of bonds that are frequently mentioned as constituting separate asset classes. Primary among these are:
- Municipal (“Muni”) Bonds
- Junk Bonds
- Corporate Bonds
- Foreign Bonds
The case in favor of each of these is some promise of extra return over similar term “riskless” bonds.
Municipal Bonds
When the income tax was introduced in 1913, state and local governments had enough clout with Congress to get a tax-exemption for interest on state and local bonds. These bonds, on which the interest is exempt from federal taxation, are generally known as Municipal bonds.
Because the interest is tax free, for many taxable investors, muni bonds offer a better after-tax yield.
The default risk or credit risk of muni bonds can be hard to assess. A further concern with muni bonds is that, especially if you live in a high tax state, there is a strong tax-saving incentive to own the bonds of that state.
Though it has not occurred in recent years, many US states have in the past defaulted. In the 1840s, eight states defaulted. That is over 25% of the states there were at the time.
More recently, Puerto Rico had the largest and most recent muni-bond default history in 2016-17, defaulting on over $70 billion of debt.
Cities and counties default too. Detroit defaulted on about $20 billion of debt in 2013. The California city of Stockton defaulted on over $2 billion of debt in 2012. And in a warning to investors tempted to lend to the governments where they live, Stockton’s solution to the default was to raise taxes.
In 1994, Orange County, California, one of the wealthiest counties in the United States, defaulted on its bonds, after mismanagement caused a $1.7 billion investment loss. In 2011, Jefferson County, Alabama, defaulted on $3.1 billion in bonds.
We believe that if investors are going to allocate a portion to muni bonds, the best way to do it is through extremely wide diversification across multiple states, jurisdictions and issues.
Junk Bonds
Junk bonds are also called “High-Yield” bonds. The term refers to corporate bonds that are rated below investment grade.
The question here is whether to consider Junk bonds an asset class. The table below[2] shows the correlation of returns between Junk bonds and other asset classes. As you can see in the table, the returns on Junk bonds have had much higher correlations with the returns to stocks than with the returns to bonds.
Based on this, and other considerations, we believe that Junk bonds should be considered as a subclass of equities, and not of bonds.
Investment Grade Corporate Bonds
In the 1990s I first heard the metaphor “picking up pennies in front of a steamroller” from Bill Dunn, the founder of Dunn Capital Management.
The phrase graphically refers to the use of an investment strategy that has the expectation of producing a large number of small gains, with the rare risk of catastrophic losses.
Depending on one’s investment objectives, investing in investment grade corporate bonds might be a version of picking up pennies in front of a steamroller.
The reason is that during “normal” times, investment grade corporate bonds offer small yield advantages over equivalent duration treasury bonds, and are highly correlated with treasuries. But during periods of stress, the correlation can break down. In practice, this often means that corporate bonds perform poorly, sometimes even going down while treasuries are going up.
This graph from the CME[3] illustrates—the blue line—the correlation between corporates and treasuries. The two red circles indicate times of great financial and market distress: the global financial crisis and the covid crises. For many investors, times such as those are why they hold bonds. Bonds that fall when treasuries are rising will tend to be huge disappointments.
For taxable investors in high tax states, taxes can reduce or eliminate the “pennies” part of the corporate bond yield advantage. US Treasury bond interest is exempt from state income taxes. Corporate bond interest is not exempt. So, for example, if a corporate bond yields 6%, and is subject to state income tax at 10%, that corporate bond yield is equivalent to a Treasury yield of 5.4%. In other words, the corporate bond has to yield an additional 60 basis points just to offer the same effective yield as a Treasury bond.
One further complication is that many corporate bonds are callable by the issuer. This is equivalent to the bond investor being short a call on the bond. Being short that call both reduces the expected return on the bond (compared to a non-callable bond) and makes the bond harder to analyze.
Foreign Bonds
From the perspective of a US dollar-based investors, there are two possible meanings to “foreign bonds.” One perspective is to consider the dollar-denominated bonds of foreign issuers as “foreign” bonds. Another is to consider non-dollar denominated bonds, regardless of the country of the issuer.
We believe that dollar-denominated bonds, whether issued by foreign governments or corporations, should be viewed largely the same as US corporates. Be aware that there might be meaningful institutional differences, such as the timing of coupon payments (Eurobonds tend to have annual coupons) and potential tax withholding issues for non-US issued bonds.
Currency Risk
Potentially more interesting for US dollar-based investors are non-dollar denominated bonds.
Such bonds introduce the element of currency risk. For example, for a US dollar-based investor, owning a 10-year corporate bond denominated in Euros adds the risk of changes in the Dollar/Euro exchange rate to the interest rate risk and the credit risk.
Some studies suggest that over the past 20 years, the addition of currency risk to a bond portfolio has doubled or more the volatility[4] of certain bond portfolios.
Hedging
However, it is possible, though not easy for an individual investor, to hedge the currency risk. Hedging involves adding an offsetting position in the currency, typically through futures, forwards or swaps. For example, if a US dollar-based portfolio is long $10 million of Euro-denominated bonds, that portfolio could sell short $10 million of Euros.
Recent (i.e. this century) historical evidence suggests that such hedging would have to a significant extent eliminated the currency-induced volatility of a globally diversified portfolio of bonds. And by spreading the interest rate risk across a number of markets that are less than perfectly correlated, the overall volatility of the portfolio might be less than that of a comparable (in duration and credit terms) US dollar-only bond portfolio.
This table[5] shows the correlation of monthly change in yields for ten-year bonds in six different major currencies:
Hedging vs. “Active” Hedging
Properly speaking, a currency hedged bond portfolio would always be hedged. The hedge can never, or almost never, be perfect. But the goal of currency hedging is to remove the currency risk from the investment. Some managers offer what they call “active hedging.” Active hedging is another term for betting on the currency markets. In our view “active hedging” is not hedging at all, it is speculating on currencies.
Currency Hedge Bonds in the Real World
In the ETF universe today, it seems that there is only one choice for a currency hedged bond fund. That is the Vanguard Total International Bond fund, ticker BNDX. The fund has been in existence since 2013. During that time to the end of February 2025, it has returned an average annual return of 2.45%.
Perhaps the closest US-only comparison fund is the Vanguard Total Bond Fund, BND. Since the inception of BNDX, BNDX has produced slightly higher returns than BND, and with slightly lower volatility, over most of the time period.
Next steps
To speak with us further, call us at 703 437 9720, email us back to set up a time to talk, or click here to request an advisor guide.
[1] The natural owners of 30-year bonds are investors like insurance companies that have very long duration liabilities. For owners with long duration liabilities, long duration assets help manage their overall interest rate exposure.
[2] Source: https://www.tiaa.org/public/pdf/enduring_case_for_high_yield_bonds.pdf
[3] https://www.cmegroup.com/insights/economic-research/2024/corporate-bonds-risks-returns-vs-equities-treasuries.html. The CME used to be called the Chicago Mercantile Exchange.
[4] https://www.pgim.com/fixed-income/blog/case-going-global-pictures
[5] Ibid.
Leave a Reply