The Bullish Cases for Bonds Are All Premature

There was a time, some 12 years ago, when I bowed to no one in my enthusiasm for bonds. Longer-dated government debt yields had surged after the global financial crisis. Yet for various reasons, inflation was likely to remain subdued, I argued. US Treasuries and even German bunds yielded about 3.5% at the time. Although headline inflation waxed and waned, it consistently came in below forecasts for the next 10 years.

For the past 18 months I have warned that inflation would surprise to the upside and investors should sell bonds in the US and Europe, which were, by any yardstick, insanely expensive. I also said rising bond yields would negatively impact most other financial assets. With yields having risen as anticipated, there is now a swelling chorus arguing that bonds are a buy. They’re wrong. Although yields on shorter-dated bonds might find some support for now, those on longer-dated bonds have much further to rise.

If some bullish arguments are better than others, others are just, well, bull. Into the latter camp goes the argument that yields are higher, so buy. Well, yes, they are higher. But the rise in yields doesn’t necessarily make them a good buy now. That yields on corporate bonds are their highest in 11 years is only true in the US and is irrelevant when inflation is the highest in 40 years.

Another claim by the bulls is that after the recent rout, putting longer-maturity, higher-quality debt in portfolios will help diversify them. It won’t. Returns on government bonds and equities were largely positively correlated for many years before the turn of the century. This meant there was little or no diversification benefit to holding both. In the late 1990s, after having spent many years fretting about inflation, investors started to worry about growth and disinflation. Correlations flipped the other way. They are now becoming more positive again and will become increasingly positive until markets stop worrying about inflation. Adding longer-dated bonds simply adds to portfolio risk.