Bond Worries? What Advisors Should Do
Many experts now believe bonds are riskier than stocks and warn of a bond bubble ready to burst. A better understanding of both the likelihood and impact of such a bubble would have clear implications for advisors, who must face two key questions: how much bond exposure clients should have, and whether it is prudent to shorten durations to soften the impact of rising rates.
A rate increase is widely expected: Despite media quips about QE Infinity, it is just not feasible for the Fed to continue buying its own intermediate- and long-term bonds forever.
The yield on the 10-year Treasury is forecast to rise nearly one percentage point from current rates to 3.47% by December 2014, according to a Wall Street Journal survey in October of 50 top economists. One economist, Tracy Herrick of Avidbank, predicted a 5.20% rate.
If those economists are right, bonds will perform badly in 2014. And thats on top of a bad 2013; the Barclays Aggregate Bond Index was down 1.10% year to date as of Nov. 1.
Christopher Philips, a senior analyst in Vanguards Investment Strategy Group, suggests advisors remain skeptical of these economist forecasts. As co-author of a 2010 Vanguard research paper on deficits, the Fed and rising interest rates, Philips pointed out the inaccuracy of those forecasts; the Vanguard advice back then, as it is now, is to beware of the perils of shortening duration and turning to cash. (When that paper was published, a Wall Street Journal survey had economists predicting a 4.24% 10-year Treasury yield by December 2010, an increase from 3.61% at the time of the forecast. In actuality, rates declined, finishing out the year at 3.30%.)
Neither the fact that these economists were directionally wrong in predicting the 10-year Treasury bond, nor this dismal forecast, is new. A 2005 study by professors at the University of North Carolina, titled Professional Forecasts of Interest Rates and Exchange Rates, finds those same economists predicted these rates far less accurately than a random coin flip would fare as a predictor.
HOW BIG A LOSS?
But what if the economists are finally right and rates do surge after the Fed eases its support for the bond market?
I asked Philips to run a stress test to see the potential impact of rate increases using the Barclays Aggregate Bond Index; as of the beginning of November, the index yielded 2.30% and had a 5.6-year duration. The following chart, If Rates Rise, shows Vanguards calculation of the impact of immediate rate increases ranging from one to 10 percentage points.
If the economists forecast is correct and rates rise about one percentage point, the intermediate aggregate bond portfolio will lose 2.8% next year. The three-point increase predicted by Avidbank would result in a 13.0% loss.
According to Philips, the largest historic increase in rates in any given 12-month period was 4.51% for the year ended July 31, 1982. Such an increase, if it occurred next year, would result in a 20.7% loss for the index. The Barclays bond index only lost 5.39% over that 12-month period, however, Philips notes because the higher rates then led to a much shorter duration than now.
Philips thinks rate increases of this magnitude are very unlikely because intermediate-term rates are mainly driven by inflation expectations, real GDP growth and a risk premium. Vanguards economic research reveals that the Fed drives only about 21% of longer-term rates, with the market controlling the other 79%.
Using a five-point rate increase (which would be the highest ever) as a stress test gives us a 23.2% loss in one year quite bad, but far less than the 37.0% loss of the total return of the S&P 500 for 2008, and only a bit more than the 20.5% loss on Oct. 19, 1987, aka Black Monday.
There is, however, an even larger insight to be gained from the chart. If a client doesnt panic and sell, the returns actually turn positive by year five. And by year eight, the total annualized return of 2.9% is higher than the 2.3% return the client would have received without any increase. In fact, even a 10-point increase ultimately results in higher returns in the long run although, admittedly, real returns would likely be much worse, because such a rate increase would probably be at least partially driven by higher inflation.
Theres a simple explanation for these higher total returns. Funds that follow this index, such as Vanguards Total Bond ETF (BND) or iShares Core Total U.S. Bond Market ETF (AGG), are laddered bond portfolios. Each month, bonds are maturing and reinvested at the higher rates. With a five-point increase, bonds maturing next month would be reinvested earning an average of 7.3%, rather than the 2.3% they yield now.
These Vanguard calculations use a couple of assumptions: that rates stabilize after the increase and that durations remain constant. In reality, rates could either increase or decrease afterward. Durations would likely shorten, however, making the returns a bit higher than shown.
Clearly, if rates do rise, a shorter-duration bond portfolio will outperform the intermediate portfolio in the short run. But investors who bought economists earlier forecasts of rising rates lost out. Over the three years ended Nov. 1, the total return of the Vanguard Short-Term Bond ETF (BSV) was 1.44% annually versus 2.79% for BND.
Vanguard ran a similar analysis on the short-term Barclays U.S. 15 Year Government/Credit Index, which at the beginning of November yielded 0.83% and had a duration of 2.64 years. Because the duration is less than half of the comparable intermediate-term index, it took no more than four years to achieve a positive return with rising rates. As was true with the intermediate-term bond fund, rate increases ultimately resulted in higher annualized returns.
But it would be a mistake to assume that clients total return will always be higher with shorter-duration bonds during periods of rate increases. The following Intermediate- vs. Short-Term Bonds chart shows the incremental return differential an investor earns in intermediate- versus short-term bonds (as defined by the two indexes). It assumes that rates increase in a parallel fashion between short- and intermediate-term portions of the yield curve.
The intermediate-term bond fund does worse during the first year whenever rates increase at least one percentage point. Yet the intermediate-term bond yields nearly an extra 1.5 percentage points over the short-term bond fund each year, which may compensate for the immediate loss from rising rates.
In fact, if rates rise by the one point predicted by economists, the break-even point for owning the intermediate-term bond index is only two years. Even the extreme stress test of a five-point increase would result in only a 0.4% lower return over 10 years for the intermediate-term bond. The implications are that, over the long haul, the client is likely better off taking an intermediate-term (versus short-term) interest rate risk.
IMPLICATIONS FOR CLIENTS
How does this all affect clients? One clear takeaway: Dont assume consensus forecasts about rising rates will be any more right this time than theyve been in the past. The impact of QE really is unknown, and the market, which usually outsmarts us, already understands the impossibility of QE Infinity.
If, however, the economists are right this time, then bonds are still likely to be far less risky than stocks over a one-year period. As Philips puts it, stocks are riskier in a day than bonds are over a year.
Also keep in mind that rising rates actually mean higher nominal returns for clients over the long run. Even a five percentage point increase delivers higher total returns by the eighth year. But only clients who stay the course will get those returns; the human tendency to chase performance is just as destructive for bonds as it is for stocks. Showing clients these charts may help them to stay the course.
Even if rates do increase modestly, dont assume that shorter-duration bonds will outperform intermediate-term bonds over a period of several years. As long as the yield curve is positive, interest-rate risk is mitigated over long periods of time.
Bond investing based on economic forecasts has long been a losers game. Understanding (and being able to explain) the potential short- and long-term impact of potential rate increases will continue to be a better way to help clients stay the course with a bond portfolio strategy.