This article originally appeared on Morningstar.com. All references are to US-based bonds and bond funds. Click here for a perspective on Australian bonds. 

In the early days of the coronavirus-driven market pandemonium, most high-quality bond funds performed exactly as one would have hoped: On down days for stocks, they gained a bit of value or at least held steady.

But as the market sell-off ground on, a disconcerting phenomenon unfolded. Despite a few days of head-scratching downdrafts, Treasury bonds stayed aloft amid a flight to quality, as previous examinations of asset-class correlations suggested would be the case. But high-quality corporate bonds fell in value, taking many bond funds down with them. Municipal bonds also dropped. And it almost goes without saying that lower-quality bond types--junk bonds and bank loans, for example--also plummeted, thanks to their economic and equity-market sensitivity.

Meanwhile, a related drama was unfolding among bond exchange-traded funds: Their prices began dropping below their net asset values (NAV). On March 12, 2020, for example, large bond ETFs like Vanguard Total Bond Market Index (BND), iShares Core U.S. Aggregate Bond ETF (AGG), and Schwab U.S. Aggregate Bond ETF (SCHZ) traded at discounts of 6.2 per cent, 4.4 per cent, and 6.3 per cent to NAV, respectively. (Those discounts have since closed.)

All that volatility had some market watchers questioning the worthiness of bond funds in investors’ portfolios, especially for retirees who are actively drawing upon their portfolios for living expenses. 

Some bond-fund critics took the short-term losses as a cue to assert that investors shouldn't hold bond funds at all, but rather stick with individual bonds if they really want safety. While an investor who buys and holds an individual bond to maturity will be made whole, assuming the issuer is creditworthy, that's not necessarily the case for bond-fund investors.

Time horizon is key

But most high-quality bond funds did largely deliver during the recent turbulence, provided investors weren't using them as cash substitutes.

True, some bond funds did lose money recently, and some funds with low-quality credits lost quite a bit. But there's a big difference between losing 4 per cent -- the average one-month loss for intermediate-term core-plus bond funds through March 31 -- and the 14 per cent loss of the typical large-cap blend equity fund over that stretch. When it comes to diversification, whether an investment has a gain or a loss during a given period matters, but so does the magnitude of that gain or loss.

Moreover, the recent market action underscores the importance of time horizon and anticipated holding period when deciding which bond funds to invest in, or whether to hold them at all if your time horizon is too short and you need certainty.

For near-term cash outlays--money for your bills during retirement over the next year or two, next semester's tuition payment, or a new roof you may need tomorrow--there's only one asset class that reliably stays positive: cash. That's why my bucket portfolios include a persistent allocation to cash, even though it's a drag on returns in upward-trending markets. It's not often that bonds lose money at the same time stocks do, but cash is there in case they do.

MORE ON THIS TOPIC: Navigating ETF discounts and premiums during turbulent times

Meanwhile, past performance would absolutely suggest that investors in bond funds should be prepared for periodic bobbles in their principal values. But that's OK, provided the investor's expected holding period is reasonable given the expected frequency, depth, and duration of those dips. For the safest bond funds, like short- and intermediate-term government-bond funds, losses have usually been brief and quickly recovered. For riskier bond types, losses have been deeper and it has taken longer to recover from them.

Digging into the data

To help assess whether a given bond-fund type is a good fit given your holding period and proximity to tapping your capital, rolling-period returns and maximum drawdowns provide a useful lens. The shorter the time period until you'll need to begin drawing on your money, the less you'll want to hold those funds in bond funds that have a history of experiencing frequent losses over a similarly short time period, especially if those losses have been sharp and sustained.

For example, short-term bond funds have posted losses in about 6 per cent of rolling 12-month periods since the early 1970s. The worst drawdown for short-term bond funds over the past 20 years came during the financial crisis, when the typical fund in the Morningstar Category lost a cumulative 7 per cent, and investors weren't back to break-even for another 15 months.

Short-term government-bond funds were generally safer. Interestingly, their losses over 12-month periods were somewhat more frequent than short-term bond funds; they landed in the red in about 7.5 per cent of rolling one-year periods. But their worst drawdown, 1.6 per cent, in April/May 2004, was much lower, too, and they recovered fully six months later.
Taken together, those data suggest that short-term bond funds (whether broadly diversified funds or government-focused options) aren't a cash substitute, and that investors should be prepared for drops in value that last for six months to more than a year. But such losses haven't been terribly frequent or deep, and the funds have readily recovered from them. In my model bucket portfolios, I use short-term bond funds as next-line reserves in case the cash cushion becomes depleted and a retiree has additional cash flow needs.

Meanwhile, intermediate-term core bond funds have experienced deeper troughs and a higher percentage of one-year rolling periods when their returns were in the red. That suggests, not surprisingly, that prospective investors hold them with an even longer time horizon in mind than short-term vehicles. Intermediate-term core bond funds have posted losses in 19 per cent of rolling one-year periods since the 1950s; for core-plus bond funds, it was 20 per cent. And the worst drawdowns for both categories were more painful than the short-term categories. In their worst drawdowns in Morningstar's database, intermediate core and core-plus bond funds experienced losses of 9 per cent and 10 per cent, respectively, during the financial crisis. Investors weren't back to break-even until the summer of 2009, 17 months after they started to fall. Meanwhile, the maximum drawdown for intermediate-term government bond-funds was much more shallow, about 3.6 per cent between May 2013 and August 2014.

Further out on the risk spectrum, data about high-yield bond performance illustrate why it makes sense to think of them as an alternative to stocks, not bonds, and maintain an accordingly long-time horizon. (I hold them in bucket 3 in my model portfolios, meaning that a retiree would have at least a 10-year horizon before needing to sell them for living expenses.) The typical high-yield bond fund lost about a third of its value during the financial crisis. The drawdown was also prolonged: High-yield bonds began to slide in June 2007 and investors weren't back to break-even until December 2009--roughly 2.5 years. In rolling three-year periods, high-yield bond funds posted losses about 12 per cent of the time; funds in the group had losses in 7 per cent of rolling five-year periods.

The limitations of past returns

Of course, past performance isn't necessarily prologue; for one thing, today's very low starting yields mean that bond funds have much less of a cushion against price drops than was the case during the financial crisis. But observing the frequency of magnitude of losses of various bond funds over history--and during the recent market shock, too--can help you determine where to slot them in your funding queue--or whether they're a fit at all.