Long-term bonds usually pay a higher yield than shorter-term ones to encourage investors to lend for longer. But sometimes the so-called yield curve inverts, as it has now, and short-term bonds offer the highest yield. When that happens, it’s tempting to move money to short-term bonds, or even cash, to grab that extra yield. Now that the yield on cash is nearly 5%, why bother with long-term bonds offering 3.5%?
The answer is that for most bond investors, particularly those who own bond funds, yield is only one component of total return, the other being changes in bond prices. When bond prices decline, total return will be lower than the yield, a reality investors encountered last year when interest rates surged, sending bond prices lower (interest rates and bond prices move in opposite directions). Inversely, rising bond prices add to yield, all of which is confirmed by the fact that bond funds’ yield and total return almost always differ.
Knowing that, it’s possible to look back at historical yields and total returns to determine whether investors were better off with short-term or longer-term bonds during previous yield-curve inversions. The tricky part is that, for most bonds, multiple variables influence prices, and it’s hard to disentangle them. One notable exception is US Treasuries, whose prices are driven by changes in interest rates.
So I looked at how Treasuries performed during previous inversions going to back to 1953. I compared monthly yields for one-month Treasury bills, which are a good proxy for cash, with those of five-year Treasuries. In months when the yield on T-bills exceeded that of five-year Treasuries, I compared their subsequent one-, three- and five-year total returns to see which performed best.
I counted 66 monthly inversions during the past seven decades. T-bills won about 60% of the time over subsequent one-year periods. But over three and five years, T-bills won only a quarter of the time. So despite a lower starting yield, investors were more often better off with five-year Treasuries over longer periods.
It turns out that changes in interest rates have had a greater impact on subsequent total return than starting yield. T-bills won by a median of 1.4 percentage points over one-year periods and lost by a median of 2.4 and 1.3 percentage points over three and five years, all of which are multiples of T-bills’ median yield advantage during inversions of 0.4 percentage points.
The broader interest-rate environment mostly dictated who benefited from changes in interest rates. The past seven decades featured two vastly different interest-rate regimes. From the 1950s to the early 1980s, interest rates trended higher for three decades, climbing to high teens from near zero. In the ensuing four decades, interest rates trended back down near zero before climbing again last year.
The impact on yield bets was much different during each period. In the first, T-bills won most of the time over one year but only about half the time over three and five years, mostly because T-bills’ higher starting yield wasn’t always enough to sustain their lead when the yield curve righted and five-year Treasuries regained the yield advantage. Since the 1980s, however, declining interest rates have given five-year Treasuries a big advantage. They won two-thirds of the time over one year and every time over three and five years.
The tailwind of declining interest rates for five-year Treasuries was even more pronounced during the dot-com bust in 2000 and the 2008 financial crisis when the Federal Reserve dropped short-term interest rates to near zero, handing longer-term bond investors a windfall. Indeed, it’s reasonable to wonder after that experience if long-term bonds are preferable during inversions. If recessions closely follow inversions, as they have since at least the 1980s, and the Fed can be counted on to lower rates aggressively to fight recessions, longer-term bonds should continue to win after inversions despite a lower starting yield.
I ran the same analysis comparing five-year and 20-year Treasuries. I counted 169 monthly inversions this time, but the results were similar, although more pronounced, which isn’t surprising given that longer bonds are more sensitive to interest rates. The median difference in total return was even larger relative to the median starting yield. And while that greater interest-rate sensitivity was a bigger drag on 20-year Treasuries from the 1950s to the early 1980s, it also helped them win easily since then during five-year periods after inversions.
My takeaway is that reaching for yield during inversions misses the bigger driver of total return for most bond investors, namely the path of interest rates. Unfortunately, there’s no way to know the precise direction of rates, which also means there’s no way to know whether shortening maturity during inversions will pay, never mind the risk of ending up worse off. And if investors can’t bet on inversions reliably with Treasuries, it’s likely to be a hairier proposition with more complex bonds such as corporate and mortgage-backed debt.
Investors are probably better off picking a spot on the yield curve that matches their desired risk and return and staying there. It may not beat a lucky gamble on inversions, but they’re more likely to get what they sign up for.
Bloomberg News provided this article. For more articles like this please visit
bloomberg.com.
Read more articles by Nir Kaissar
