- Bond markets move in long – secular – cycles, with at least five since the 1870s
- After a bull market since 1982, we are now likely headed for a multi-decade bear market
- However, bond yields are unlikely to reach levels seen from the 1970s to the mid-1990s
Bond markets, like other markets, work in long cycles, Beyond the scope of the business cycle, there are also the very long secular markets. Based on the 10-year US Treasury, there have been five secular bond markets – three bull markets and two bear markets – since 1873.
The bull market from 1982 really ended in 2012. Since then, advanced economy bond yields have been artificially kept down primarily by central bank purchases, quantitative easing. However, with the exception of Japan, this has either ended or is coming to an end over the next year. This is one reason for expecting bond yields to trend up.
A second reason is the return, however slow and hesitant, of inflation. As labour markets globally tighten, that process is likely to accelerate.
Connected with that, central banks have seen the face of deflation, and they don’t like it. Even assuming that central banks remain independent inflation targeters, as long as the memory of threatening or actual deflation remains, they are more likely to accept an overshoot of inflation than an undershoot. (In other words, if 2% is the target, 2.5% is likely to be acceptable, while 1.5% is less so.)
Finally, the beginning of QT. While large central bank balance sheets have occurred in the past, actively shrinking the balance sheet is a new phenomenon. But one consequence will be a sustained upward pressure on bond yields. And, if central banks generally follow the Fed’s pattern – no sales but allowing some maturing bonds to run off and not be replaced – this pressure could remain for decades. (This obviously depends on the ultimate target size of central bank balance sheets, something that so far has not been clarified.)
However, as long as central banks do remain independent inflation targeters, the rise in bond yields is likely to be limited. Based on an inflation target of 2%, with allowance for acceptable overshoot as noted above; the fact that the real bond yield in the US tends to revert back to 2% over longer period; and some fluctuation, we can assume a 10-year Treasury yield over the very long period ending up between 4% and 6%. In this, the coming secular bond bear market is likely to be similar to the one from 1899 to 1921, when the yield on the 10-year US Treasury rose from 3.1% to just under 5.1%.
This previous secular bond bear market also offers interesting parallels from two other perspectives. For one thing, when it began, US equities were generally perceived to be overvalued. Although they were the better investment over the entire period, they needed to “rerate” – i.e., fall in price – first. For another, this was a period when the yield curve (yield on the 10-year Treasury bond less the Ninety Day Time-Money Rates on Stock Exchange Loans for New York) was generally negative. Crucially, while the yield curve inverted before recession – and there were seven during this period – it also inverted without being followed by an economic downturn, something of relevance today, when there is much concern about the flattening of the US yield curve.