The question has gone viral lately, courtesy of the recent rise in interest rates. The benchmark 10-year Treasury yield, for instance, edged up to 2.73% yesterday, the highest in nearly four years. The 2-year rate’s ascent has been even stronger, rising to a 10-year high on Tuesday.
The upward bias in rates has prompted two influential bond-fund managers – Double Line’s Jeffrey Gundlach and Janus Henderson’s Bill Gross – to advise that trendlines for yields could be signaling an end to fixed income’s long-running bull market. Earlier this month, Gross tweeted that “Bond bear market confirmed today. 25 year long-term trendlines broken in 5yr and 10yr maturity Treasuries.” Meanwhile, Gundlach opined that “3.22 percent on the 30-year Treasury yield “would end the bond bull market for good.”
The bond gurus may or may not be right, but the larger question is whether the basis for their analysis (studying trends in rates, which is the equivalent of prices for bonds, albeit in inverse terms) is valid? Do trends in rates matter for bond investing?
Reliable intelligence for monitoring the business cycle:
The US Business Cycle Risk Report
Not really, according to some financial advisors, at least not in the short-to-medium term. Cullen Roche at Pragmatic Capitalism dismisses such analysis, asserting that “lines on a chart” aren’t worthy of attention. Rather, inflation is the primary, perhaps the only relevant factor for making decisions about fixed-income investing. Roche writes that “drawing short-term trend-lines on charts is a form of short-termism that confuses short-term market trends for long-term secular trends in the real economy.”
In other words, long-term secular trends in inflation aren’t necessarily changing just because some short-term trends on interest rate charts were broken. This analysis is putting the cart before the horse in that the secular fundamental trends drive the lines on the chart and not vice versa.
Roche concludes that “secular macro trends” suggest that low inflation will endure.
This is supported by be the evidence showing that inflation tends to be reinforcing with momentum in the long-term. That doesn’t mean inflation can’t move a little higher from here (bond traders will overreact and do what traders do), but calling for an end to the bond bull market is essentially the same as saying that inflation is about to move much higher and that all of those big macro trends are about to reverse. That might be right, but a short-term trendline on a chart isn’t going to tell you that.
Is this and open-and-shut case? In other words, is your inflation outlook the only relevant variable for making investment decisions related to bonds? If the answer is “yes,” then yield trends (and by extension bond prices) are as useful as heaters in the Sahara.
Yet a deep pool of research tells us that dismissing momentum in matters of fixed income is hardly worthless. Granted, an investor with a long-term horizon may be inclined to take Roche’s advice with little or no fallout. There are many ways to successfully model and manage bond portfolios. But arguing that momentum isn’t on the short list for fixed income is going too far.
In fact, a widely read study in the Journal of Financial Economics a few years ago documented that time series momentum is robust over decades across all the major asset classes, including fixed income, and is linked to excess returns. “We document an asset pricing anomaly we term ‘‘time series momentum,’’ which is remarkably consistent across very different asset classes and markets,” according to the authors of “Time Series Momentum”.
Specifically, we find strong positive predictability from a security’s own past returns for almost five dozen diverse futures and forward contracts that include country equity indexes, currencies, commodities, and sovereign bonds over more than 25 years of data. We find that the past 12-month excess return of each instrument is a positive predictor of its future return. This time series momentum or “trend” effect persists for about a year and then partially reverses over longer horizons.
Since bond yields are directly related to bond prices, albeit inversely, it’s reasonable to extend any trend-analysis insight for performance to rates. What’s more, the “Time Series Momentum” study can’t be dismissed as outlier research. A 2013 paper from the New York Fed, for instance, also reports “sizeable excess returns” related to momentum in the Treasury market.
Momentum also exhibits traction for corporate bonds, a pair of researchers at Robeco, a money manager, noted last year in the Financial Analysts Journal. “We offer empirical evidence that size, low-risk, value, and momentum factor portfolios generate economically meaningful and statistically significant alphas in the corporate bond market,” the authors report in “Factor Investing in the Corporate Bond Market.”
There’s also evidence that “a simple momentum investing strategy on bond mutual funds” generates “attractive” results, notes Alpha Architect, citing research by CXO Advisory. A recent research paper from MainStay Investments finds a similar result, noting that a backtest of a fixed income momentum-style investment strategy “outperformed the Barclays Aggregate Bond Index, gross of costs. Accounting for transaction costs and fees associated with holding ETFs as representatives for the fixed income sectors, the strategy still would have generated higher returns than the Barclays Agg Index over the sample period.”
Does this mean that momentum-based analysis of the bond market is a short cut to easy profits? No, of course not. But it’s hardly rubbish either.
It’s fair to say that bond investors can do quite well by ignoring the momentum factor entirely, particularly if your investment horizon is five years or more. But suggesting that momentum strategies are irrelevant for bonds requires overlooking quite a lot of research that demonstrates an alternative conclusion.
The details matter, of course. Meantime, there’s room for debate about the insights via Gundlach and Gross. But there’s enough meat on the bond-momentum bone to warrant a serious study of trend analysis for rates and prices.
“The existence of momentum is a well-established empirical fact,” reminds AQR Capital Management’s Cliff Asness and three colleagues in “Fact, Fiction, and Momentum Investing” in the 40th Anniversary issue of The Journal of Portfolio Management. That’s widely recognized in connection with the equity market. For bonds, on the other hand, it’s still news in some corners.
thanks so much for pointing this out. Also, and I will try to find the research, currencies and rates are the most auto-correlated of all major asset classes, which would suggest trends and momentum most certainly matter in fixed income. While I am not willing to bet on lines drawn on a chart, I am willing to let academic research guide my decisions especially when real yields are still very far from historically normal.
and yet…Larry Swedroe and SPIVA point out that the vast majority of active bond managers, who could/did utilize momentum strategies, underperformed, both gross and net of fees, to say nothing of taxes, which a high turnover strategy like momentum would generate. The problem with all the backtesting and claims for momentum is that they usually don’t incorporate realistic fees, and none factor in taxes. 30 years of declining rates might have led to a “momentum” effect, but the superior results didnt materialize substantially in most funds, the future rate environment looks less predictable, though fund underperformance remains likely.