Mistakes Bond Investors Make
Let's examine some of the typical mistakes we make in our fixed-income portfolios, and then we'll turn to best practices in this sector.
Employing Managers Who “Cheat”
Given the extraordinary efficiency of the bond markets, especially on the taxable side, it's surprising to observe how many bond managers claim to have outperformed the indexes. Closer examination almost always discloses, however, that the managers have exposed their clients to significantly more risk than was embedded in the index, and it was the added risk that explains the outperformance, not the managers' skill. Adjusted for risk, these managers will usually be found to have underperformed.
Let's look at how a typical taxable bond manager “outperforms,” by observing the fictional firm of Wily Bond Management.
Wily manages taxable bond accounts for institutional investors and for families who have private foundations or large IRA accounts. Its typical benchmark is the Barclay's Aggregate bond index, a very broad long-intermediate index. Over the past 10 years, Wily has outperformed the Barclay's Agg, and this fact is naturally trumpeted in its marketing materials. Looking at this record, many investors have hired Wily, and Wily's business is growing and profitable.
But a closer look at Wily's performance discloses what is not disclosed by the firm's sales staff: Wily has subjected its client accounts to considerably more risk than is contained in the index, and it is that increased risk, pure and simple, that accounts for Wily's so-called outperformance. In fact, controlling for the risk it has taken, Wily has actually underperformed the index by a considerable margin.
How does Wily take on more risk than the index? In at least three or four ways. First, Wily has exposed its clients to greater duration1 risk. Over the past 10 years the average duration of the Barclay's Aggregate bond index has been about 3.8 years, whereas the average duration of the Wily portfolios has been about 4.1 years. Over time, longer-duration portfolios will outperform shorter-duration portfolios, albeit with greater price volatility along the way. This is what has happened to the Wily accounts.
Second, Wily has exposed its clients to greater credit risk. The average credit quality of the index is AA, but the average credit quality of the Wily portfolios is A–. Though this may not seem to be a big deal, lower-credit-quality bond portfolios naturally tend to return more than higher-quality portfolios for the simple reason that investors demand more return to take on the increased risk of default.
Third, Wily has exposed its clients to greater optionality risk. Many corporate bonds are callable; that is, under certain conditions the issuer can redeem the bonds long before they are scheduled to mature. Naturally enough, companies tend to call their bonds when they can replace them by issuing new bonds at lower interest rates. For the unhappy investors, this means that we have received our money back exactly at a time when we don't want it back—because we will have to reinvest it at lower yields. Issuers of callable bonds pay a very small premium for the privilege of being able to call those bonds, and Wily has taken advantage of that by buying a far higher percentage of callable bonds than is represented in the index.
Finally, Wily has exposed its clients to benchmark risk in other ways—by owning more structured products (mortgage-backed and asset-backed instruments, for example) than are present in the index, by overweighting certain sectors of the index, and so on.
There is nothing inherently wrong with anything Wily has done with its client portfolios, except for the singular omission that none of it was disclosed to the clients. Thus, Wily has achieved outperformance against the index, but it has not achieved something far more important—risk-adjusted outperformance. Nor is this a hypothetical problem for investors.
If we look at Wily's aggregate performance over the past 10 years, we do, indeed, see long-term outperformance against the Barclay's Aggregate (though not risk-adjusted outperformance, of course). But if we look, instead, at each individual account Wily has managed over that time period, we observe problem after problem. Investors look at Wily's aggregate performance and hire the firm, looking no further. Things go along well enough for a period of time, but then suddenly the risk embedded in the Wily portfolios jumps out and bites us.
Perhaps it is an unexpected jump in interest rates that harms the Wily portfolios much more than we expected (and much more than the index was harmed, as a result of its shorter duration). Perhaps slowing economic conditions have caused several of the weaker credits in the Wiley portfolios to be downgraded or—horror of horrors!—to default. (The index will experience downgrades, too, but at a less frequent pace, as a result of its higher average credit quality.) Perhaps rates have fallen and many issuers have called their high-yielding bonds, forcing us to go out and buy lower-yielding bonds to replace them (the index will also have experienced calls, but fewer of them due to its lower optionality).
Yes, investors who stayed the course with Wily for the entire 10-year period will actually have beaten the index, albeit with many unhappy surprises along the way. But there will actually be few such investors. Many more investors hired Wily, were unhappily surprised by the risks that bit them, then fired Wily, poorer but wiser. We don't want to be one of those investors, and so we need to look much deeper before we hire bond managers who claim to have outperformed.
Perhaps because bonds are such a low-returning asset class, the temptation for an advisor is to try to juice up the sector in one way or the other. But there are good and bad ways to do this.
The good way is to find a niche bond manager who is actually doing something different from everyone else and who therefore stands a chance of producing real, risk-adjusted outperformance. If you go in this direction, make sure you are straight with the client about what the manager is doing and what the risks are.
The bad way is to hire a bond manager who cheats and hope he doesn't (and you don't!) get caught. Remember that even though bond mistakes typically don't result in losses as big as mistakes elsewhere in the portfolio, you can never forget that this is your client's sleep-well money. Even a smallish loss can infuriate a client who is taken unaware by duration, credit, or optionality risk.
Paying Too Much for Bond Management
Whereas some investors are being penny-wise and pound-foolish (by hiring brokers to build their bond portfolios or by doing it themselves), other investors are simply paying way too much to have their fixed-income portfolios managed. Keep in mind that the long-term expected return on an intermediate bond portfolio, net of tax (or using tax-exempt bonds) is going to be on the order of 5 percent. Yet, according to Morningstar, the average intermediate bond fund has an expense ratio of 1 percent and charges an average load of 0.83 percent. That's nearly 2 percent in the first year, or roughly 40 percent of the long-term expected return.
Even the great Bill Gross, probably the best bond manager who ever lived, has trouble overcoming the high expense ratios of the retail versions of the PIMCO bond funds. As Morningstar puts it, “fees are simply too high to recommend some [of PIMCO's retail] share classes—no matter how good the management.”2