The Problem With Bonds, According To Morgan Stanley

For the past couple of years, warnings that government bonds may no longer be a reliable diversifier have become more shrill.

For decades (literally), owning benchmark Treasurys was a (mostly) foolproof way to Teflon-coat a simple multi-asset portfolio. The vaunted 60-40 split was (and still is) the bedrock of starter portfolios, and the assumed negative correlation between stocks and bonds is a key underlying assumption for many a model.

Just as important, the idea that government bonds are less volatile than stocks is a foundational concept that informs strategies which invest unfathomable amounts of wealth.

All of these assumptions have been called into question in a world where yields grind inexorably lower and periodic bouts of tantrum-like behavior serve as harbingers of what some worry will be a more volatile future for previously stoic bonds.

Morgan Stanley takes this up in a new cross-asset note, which, while not particularly enthralling, is nevertheless worth a mention.

“As bond yields fall to historical lows on the back of recession risks, uncertainties over the COVID-19 situation and unprecedented, coordinated monetary policy stimulus, expected returns for government bonds have also fallen precipitously”, the bank writes, noting that 10-year USTs are expected to yield just 1.2% over the next decade.

The problem for multi-asset portfolios which rely on bonds for diversification is that in a world where benchmark yields have been driven to zero and below, the levels necessary to offset another double-digit equity decline are now deeply negative (figure below).

Worse, correlations are becoming less negative and realized rates vol. has demonstrated a tendency to spike.

“If the past year acts as a credible guide, low returns [on bonds are] likely to be accompanied by higher volatility, and less negative correlation and downside beta to stocks”, Morgan Stanley goes on to say, adding that “the correlation has in fact become less negative as UST 10Y yields fall even further to multi-decade lows… suggest[ing] the diversification benefit of bonds could be greatly reduced if yields remain at such low levels”.

(Morgan Stanley)

Again, these are familiar concerns. It’s the “diversification desperation” dynamic, and it could potentially become more vexing in a scenario where the combustible combination of massive supply (to fund virus relief) collides with better-than-expected economic outcomes to push long-end yields sharply higher before the Fed (and other central banks) has a chance to tamp them down.

It’s also worth noting that, in mid- to late-March, investors of all stripes began dumping Treasurys not because they didn’t like the safe-haven appeal, but rather because the situation was briefly acute enough to catalyze the wholesale liquidation of bonds and gold in order to meet margin calls and raise USD cash.

That, in turn, caused the Treasury market to “snap“, leading to a vicious risk parity unwind and, in the same vein, massive losses (and volatility) for traditional 60/40 portfolios.

Some readers will remember the following chart, which represented what many commentators long contended would never happen – namely, rapid risk parity deleveraging.

In any case, Morgan Stanley also reminds you that the decorrelation effect (i.e., “the incremental impact of adding an asset which has a lower correlation to stocks”) is actually less important than the volatility buffer.

“There is too much emphasis on finding an asset which is negatively correlated with stocks when actually a large part of the diversification benefit from bonds came from its relatively lower volatility and lower correlation”, the bank writes.

(Morgan Stanley)

The point there is obvious: It’s the volatility buffer that really matters, which means that if rates volatility were to reset structurally higher, the read-through for bonds as a diversifier would be especially poor on top of the self-evident problem illustrated in the first simple chart above.

Morgan goes on to conduct a search for possible alternatives to bonds, and that quest gets pretty adventurous at times (long staples and telecoms versus short the ruble and the rand, for example). But the key takeaway for the general investing public is simply that “low yields, higher volatility and less negative correlation are reducing the diversification benefit of bonds”.

Fortunately, you don’t need to trouble yourself with such things when stocks are prone to rallying ~45% in the middle of a global depression.