Bonds Have Bear Markets Too

May 05, 2016

This article is part of a regular series of thought leadership pieces from some of the more influential ETF strategists in the money management industry. Today's article is by David Haviland, managing partner and portfolio manager of Beaumont Capital Management, based in Needham, Massachusetts.

When you look back to examine what has happened to an investment in the past, you must be careful about the length of the data set used and the bias it can create. Historical bond returns are a classic example.

Using only the last 35 years to analyze bonds and bond returns renders a data set so full of declining interest rate bias that any conclusions drawn about the future are bound to be askew, or worse.

The U.S. interest-rate increases started in earnest in the late 1960s with the effects of the Vietnam War; the massive consumer spending caused by the baby boom; and the 1973-74 OPEC oil embargo, which helped drive up energy prices 400% in just two short years.

Inflation skyrocketed in the U.S. and the then-current Federal Reserve Chairman Volcker had to impose a form of economic "chemotherapy"—he raised interest rates so high that they stifled any further inflation and, for a time, most economic growth.

Returning To 'Normal'

This economic chokehold was designed and destined to return our inflation and interest rate environment to "normal." Yet interest-rate normalization has taken the better part of 35 years. The full cycle is over 50 years. Could we have succumbed to this rate bias and set ourselves up for an unforeseen bear market?

Consider that on a total return basis (price movement plus coupon return), the Barclays Aggregate Bond Index has, since inception, only had negative returns in three calendar years: 1994 (-2.92%), 1999 (-0.82%) and 2013 (-2.02%).

Due to these modest declines, many have drawn the conclusion that bonds are "safe" and never lose a lot of money. This, of course, is not true. This index, which is the financial industry's most popular bond benchmark, did not exist at the beginning of this 50-year cycle. It's too new to tell a complete story.

Think about it: Our industry's main comparison for bond returns is completely subject to the declining interest-rate bias noted above.

For this reason, we will focus on the 10-year U.S. Treasury as our representative bond proxy, as it has a much longer data set that transcends the 50-year bond "supercycle."

The main reason there were no large total-return bond losses in the last 35 years is because the bond's coupons were so high due to the Volcker Fed's rate increases, they covered up the declining price movement.

Yes, higher coupons bring lower bond duration, but let's use a simplistic example. If a 10-year U.S. Treasury lost 10% in a given year while the coupon was 8%, then the yearly total return would have been -2%. But now that the 10-year U.S. Treasury yield is below 2%, that same 10% price decline would bear a -8% total return. That can come as quite a shock to retirees who have poured trillions into bonds to produce income to fund their retirement.

Figure 1

(For a larger view, please click on the image above.)

The above chart of the 10-year Treasury, which goes back 100 years, draws a completely different picture as to what historical interest rates have been and what a "normal" rate range might look like. The Treasury has spent more time with its interest rate between 2-5% than it has above 5%. Is this normal? Perhaps.

Size Matters

The size of our current U.S. Federal debt—$19 trillion—is far greater than at the end of World War II, when it was $251 billion. Yes, as a percent of GDP, these amounts are similar, at around ~110% of GDP, but what if there is a sudden surge to sell? Size matters if bondholders decide to sell all at once. It all comes down to a simple premise: Is there an incremental buyer for the bonds for sale, and at what price?

The charts below are from the Wall Street Journal. They illustrate other reasons that bond liquidity, due to the size of the bond markets, may create additional risk to bond prices. In short, as rates rise, losses mount, and selling pressure increases.

As it does, it will likely become harder and harder to find an incremental buyer for the amount of bonds for sale.

Figure 1

(For a larger view, please click on the image above.)

The Wall Street Journal pointed out that about 90% of bond returns were derived from the starting interest rate of each bond. With yields on 10-year Treasurys just under 2%, what are reasonable return expectations going forward if interest rates stay range-bound? What if interest rates rise for any reason?

Calculating bond losses involves a simple equation:

The duration of the bond (bond ETF or fund) x interest-rate increase = expected loss.

A modest rise in interest rates—let's use 2% in a year—will bring a 20% price loss to a U.S. Treasury bond with a 10-year duration. With the current coupon just under 2%, this would bring a total return of -18%!

It's conceivable that interest rates could rise 2% over 12 months if inflation heats up and/or the Fed raises rates. The last time the Fed raised rates in 2004-2007, it increased the federal funds rate by .25%, 17 times.

And there's also the issue of sustained bond losses over longer time periods. Between 1955 and 1959, we saw the 10-year Treasury lose money four out of five years:

Bonds can also be subject to credit and default risk, rating downgrades, liquidity risks for certain issuers, and reinvestment risk in lower-yielding bonds. High yield or junk bonds often tend to have return streams more similar to equity than other bonds. International bonds must account for political/military risks and currency fluctuation.

As a general rule, the higher the coupon, the greater the risk.

We all know interest rates are low across the globe. As the chart below shows, of all the government bonds issued and outstanding today—including all maturities—almost two-thirds have a yield below 1%, and more than one-quarter have negative yields.

In addition to this global phenomenon, as the chart below illustrates, real yields in the U.S. are also turning negative. The last time this happened was during the 1970s, when the Fed was trying to check inflation.

Figure 1

(For a larger view, please click on the image above.)

Complacency and ignorance do not bode well in our industry. Today we too often read about the "safety" or "relative safety" of bonds, and we feel investors have been led to believe they will not lose a lot of money in bonds, ever. That's not the case.

It is almost certain that rates will one day rise substantively. We hope the investing public and our industry is ready to deal with the potential fallout.

Contact Beaumont Capital at (888)-777-0535, or by email at [email protected]. See here for a full list of disclosures.

Find your next ETF

Reset All