Bursting the Bond Bubble Babble

History is on our side

The last secular bear market in bonds ran for 30 years, ending around 1980. What were the results? Nominal returns were up, not down. In many sub-periods, even real returns were up.

In a previous article, I analyzed retail bond mutual funds (adjusted for survivorship bias), bond index returns, Treasury bonds and a hypothetical discounting model. Each told a corroborating story: The worst bear market for bonds in modern history resulted in positive total returns for bonds.

Likewise, there has been similar good news since 1980. Chart 1 shows returns for 7-to-10-year and 30-year Treasury bonds for the five periods when rates rose by more than 2% and one recent period where the Federal Funds rate rose 4.25%. These were the six most difficult interest rate environments for investors in the last 30 years.

Chart 2 shows an equally-weighted portfolio of the three major fixed-income –sub-classes: Treasury bonds, corporates, and mortgages, plus the Barclays US Aggregate Bond Index during the same periods. Nowhere in these numbers do you find a financial debacle. Rising interest rates have been no match for well-diversified investors.

Two periods experienced negative returns. Returns will not always be positive in a rising-rate environment, nor will they be better than when rates decline. But investors did not experience the disasters about which some warn.

Why? Wall Street Journal reporter Katy Burne recently wrote, “The yield on the (Barclays U.S. Corporate High Yield index) has lost more than a percentage point this year, in a sign of hefty demand for income-producing securities.” She correctly identified important bond price variables – supply and demand – that are not part of any formula. Bonds prices are subject to supply and demand in the same way as stocks are.The demand for bonds will rise, not fall, when interest rates go up. This counter-balancing force tempers bond price drops even as rates rise.

Too much focus on the Federal Funds rate

Nominal yields on 10-year Treasurys rose only 61 basis points from 2003 to 2006 (the last row of returns in chart 2), but the Federal Funds rate increased by 4.25% (consisting of 17 quarter-point moves).

Since the Federal Funds rate is at the center of much of the commentary about rising rates, it is instructive to show what happened the last time it went up. Many investors worry that when the Federal Reserve fires the Federal Funds starting gun, all of the other horses will dash, with intermediate- and long-term rates rising apace. This did not happen the last time short rates went up. It took almost three years for 10-year Treasurys to realize that the race had begun, as shown in in Chart 3. The total return of the index portfolio was 8.03% (as shown in chart 2) during one of the steepest Federal Funds increases in history.

How did the Federal Funds rate rise so dramatically without affecting bond yields? One reason is that the Federal Funds rate is the only rate that government sets. Other interest rates are set by market forces. During slow economic periods when loan demand is relatively low, banks are reluctant to push up interest rates. Banks would rather maintain loan volume at a thinner margin than jeopardize business by raising rates.

Why the dismal warnings about rising rates?

The industry and financial media, perhaps unintentionally, misstate the effects of rising rates. One reason is simple and partially excusable: There is ambivalence toward historical data. This goes under the general heading of the oft-repeated mantra, “It’s different this time,” and is rooted in the academic environment where theory reigns. For example, the distinguished mathematician Fischer Black wrote in 1993: “I find theory to be far more powerful than data when we’re trying to estimate expected return. When I read an empirical paper, I usually seek out the theory section and ignore the tables.” Later, Bill Sharpe said: “I have concluded that I may never see an empirical result that will convince me that it disconfirms any theory.”

Theoretical constructs such as the capital asset pricing models and arbitrage pricing theory are preferred to data-based observations. Theories are important. Their study builds analytical tools to explain the dynamics of money management. And studying the past is no more reliable than theory. But theories do not necessarily have better track records than track records. Our predictions of the future should be informed and tempered by historical perspective. We need both data and theory.

Some in the investment industry make wild assumptions about what actually happened during past rising-rate cycles. Instead of positing steady long-term increases in rates – the historical precedent – models assume rate spikes. Thus, an analyst might say, “If rates go up by 3% in just one year, the unlucky owner of a 30-year bond would lose 47%,” without pausing to consider that the 30-year Treasury has rarely gone up 3% in one year. Bond pricing is too complex to be explained in a single formula that omits supply and demand, fails to recognize that less attractive alternative investment options may cushion prices and/or does not account for credit quality changes.

Few bother to ask who buys 30-year Treasury bonds. I have worked in the investment industry for over 30 years and have never met anyone who actually owned such a bond. They are primarily owned by U.S. and foreign governments, institutions and the Federal Reserve, not by individuals.

Below are the returns of the 7-year and 30-year Treasury bonds for the worst two markets (measured by 30-year Treasury bonds) in the last 40 years. Civilization can survive a 10% short-term drop in short or intermediate bonds, even if the 30-year Treasury drops 18%.

Chart 4

Could marketing be a reason?

Another reason why the effects of rising rates are misstated is less justifiable. There is an unquenchable thirst for yield now, in any form. This is natural when bond rates are at historic lows. In response, the investment industry is looking everywhere to boost yields, without looking closely at theory or data. For example, the Wall Street Journal recently reported that investors are seeking additional yield by investing in high-yield bonds, emerging-markets bonds, dividend-paying stocks, real-estate investment trusts (REITs) and master-limited partnerships (MLPs).

How did these five sub-classes perform in 2008?

Chart 5

Compare those returns with traditional income-producing assets for the same year.

Chart 6

Nevertheless, the industry responded to the perceived yield appetite. Since 2011, more than 30% of new issues of income-objective exchange-traded funds were in those five “non-traditional” sub-classes. Only three income-objective ETFs, roughly 3% of the new issues, were fully diversified bond portfolios in more than two fixed-income sub-classes with no stocks, derivatives or leverage. Additionally, Morningstar found that 309 bond mutual funds owned stocks at the end of December – the highest number in at least a decade.

Elsewhere, a recent IndexUniverseblog post highlighted five multi-asset Income ETFs that have “come of age.” The writer emphasized that these are “here to stay,” but 60% of the assets held by those funds are in assets other than fixed income, including MLPs, equities, real estate, American depository receipts and alternatives.

When I entered the investment industry 30 years ago, fixed annuities and cash were used as bond proxies. Now we are to believe that equities, real estate and MLPs are suitable substitutions.

What about shortening maturities and diversifying instead?

What is a safer alternative than allocating to non-fixed-income asset classes? The 10 years ending Dec. 31, 2012, was the first since the 1950s that the returns on Treasury bills underperformed the inflation rate. T-bills no longer can be expected to outperform inflation.

If we are entering a long-term period where fixed-income returns will be between 2% and 3%, as Bill Gross has suggested, we can manage it. Income investors are more interested in preserving principal and lowering volatility than equity investors. They will trade the chance of potential higher returns for a more predictable performance. Laurence Siegel of the Research Foundation of CFA Institute said, “People think bonds are safer than stocks and you’re supposed to get your money back. … If people lose money in bonds it’s going to hurt.”

Investors should shorten maturities and diversify solely among fixed-income asset sub-classes. The chart below illustrates returns during the five toughest periods of rising interest rates in the last 30 years, measured by 10-year Treasury yields, and the recent period when the Fed Funds rate increase by 4.25%. The last category, “12-Asset,” is an equal weighting of each of the 12 major taxable bond categories. Returns were higher in shorter-duration and diversified portfolios than in single-asset long-term allocations.

Chart 7

3 assets: Index portfolio of Barclays US MBS TR Index, Citigroup Broad Investment-Grade, Citigroup Treasury 3-7 year, equally-weighted. Source: Morningstar Advisor Workstation

When less is not more

With fixed-income investing, more is more – the more disparate assets, the better.

No one puts stocks into one category, nor should we with bonds. Morningstar recognizes nine style boxes for stocks, not including international equities. Bond allocations should be comprised of at least as many sub-classes.

Start with three major sub-classes: corporate, Treasury and mortgage-backed bonds (as I did with the “index portfolio” above). Then, subdivide into the 12 major taxable bond sub-classes: agencies, bank loans, cash, convertibles or preferreds, emerging markets, high yields, international corporate, international government, investment-grade corporate, mortgage-backed, Treasury inflation-protected securities and Treasury bonds. Now you have a truly diversified bond portfolio.

How did this 12-asset, equally-weighted portfolio perform? The index version returned a cumulative 85.63% compared to the Barclays US Aggregate Bond TR index’s return of 63.57% for the 10 years ending April 30, 2013. Further, there was a surprisingly wide dispersion of returns, even when interest rates were low. In the 12 months ending April 30, 2013, shown in Chart 8, the performance ranged from -2.5% to 15.47%. The correlations between the 12 asset sub-classes were .28, .33 and .35 for the three-, five- and 10-year periods through April 30, 2013, respectively.

If the industry were correct in building income funds with so few fixed-income assets, then we would expect to see 80% to 100% correlation between these 12 categories of bonds. This may surprise investors who believe that the only way to diversify a bond portfolio is to add stocks, alternative assets, or leverage. But it is a good surprise, because the diversified bond portfolio reduced long-term volatility and acted as a hedge in rising-rate periods.

Chart 8 – Returns for 12 fixed-income sub-classes for the year ending April 30, 2013

If we are moving into a higher interest-rate environment, the safest, easiest and most reliable way to generate income and total returns for conservative to moderate investors is to diversify into all of the available bond categories. This strategy is preferable to adding securities or strategies that could jeopardize investors’ principal, buying power and comfort.

Andrew D. Martin is president, co-founder and portfolio manager of 7Twelve Advisors, LLC.

Direct investment in an index is not possible. Indexes are unmanaged portfolios representing different asset classes, with varying levels of associated risk.

* (Agencies: Barclays US Agency TR, Bank Loans: S&P/LSTA Leveraged Loan TR, Cash: Fidelity Instl MM Fds Money Market I, Convertibles/Preferreds: BofAML All Convertible All Qualities, Emerging Mkts: JPM EMBI Global TR, High Yields: Barclays US Corporate High Yield TR, Int'l Corporate: BofAML Global Brd Mk Ex US Dollr TR, Int'l Gov’t: Barclays Global Treasury Ex US TR, Inv Grade Corporate: Barclays US Credit TR, Mort Backed: Barclays GNMA TR, TIPS: Barclays US Treasury US TIPS TR, US Treasurys: Barclays US Treasury yr TR – Morningstar Advisor Workstation)

Read more articles by Andy Martin