In some ways, the third quarter of 2019 was a microcosm of the environment since the financial crisis.
Rapid declines in equity markets turned out to be a buying opportunity given a relatively rapid rebound:
While bond yields, despite threatening to sell-off from their lows, didn’t reach previous peaks:
This pattern in bond yields has a familiar ring. In the US, the environment for much of the last thirty years has been that when yields have fallen dramatically, they have not subsequently returned to prior peaks (increasingly resembling a stylised ‘L’ shape in the near term, before falling further later on).
The only potential exception to this in the US, at least up to now, could be the period since the pivotal moment of mid-2016.
However, in Germany there is no such sign of a ‘yield floor.’ The propensity for gaps down in yield to prove sustainable is even more pronounced, with yields now below their 2016 levels.
This falling yield environment has been a challenge for investors who have positioned themselves for yields to revert to the levels of the 1970-1990s (this summer, Jim Leaviss referred to these short bond positions as ‘widow makers’).
However, with the benefit of hindsight, we can see that this is a case where naïve ‘mean reversion’ value strategies would have been flawed. And there are a couple of related reasons for this:
The Long term: the inflation regime
In the case of many global bonds rapid yield declines never reverted because investor anchoring meant that the starting yield was already too high.
Investors, perhaps scarred by the experiences of painful bond losses in the 1960s and 1970s (consider Eric’s anecdote at the six-minute mark here), in love with the equity story of the late 1990s, but also anchored to decades of inflation experience, appeared unwilling to acknowledge evidence that inflation had declined materially.
In this environment even short term yield declines which appeared ‘episodic’ did not go far enough.
Indeed, such anchoring is why long term trend following can work: investors are slow to adapt to a new regime because we need lots of evidence to shake us out of our mental models of how the world works.
The short term: portfolio properties
This very fact that inflation has been relatively benign in many parts of the developed world explains another factor in ever lower yields: the change in the correlation properties between equities and bonds.
Over the long term, falling yields have been associated with positive returns from both bonds and equities, since government bonds in part reflect the global discount rate which underpins the valuation of all assets. This partly supported the trend since the financial crisis for short term equity weakness to be buying opportunities (and feeds the narrative that assets are only priced as they are due to QE).
But just as important is the scope for bonds to be negatively correlated with equities in short term periods of growth fear, because the scope for inflation to act as a correlating force has largely disappeared. This is always an attractive property for investors, and particularly so when the desire to avoid short term volatility is pronounced. In this environment, the term premia (the reward you are paid for taking on duration) becomes very low, or even negative, meaning investors are willing to accept even lower yields.
Looking ahead from an episode perspective
The episode philosophy is often seen as an approach which entails being contrarian in the face of short term, rapid price moves.
This is indeed frequently true, but it is not a philosophy built on being contrarian for its own sake. Instead each case must be supplemented by considerations of the behavioural drivers which may underpin both prevailing valuations, and the short term moves themselves.
In the case of bonds over recent decades, it was investor anchoring to an out of date regime that proved more significant than shorter term phases of panic. In the case of equities since the financial crisis, prevailing valuations already had embedded within them a meaningful degree of pessimism regarding the growth regime. This meant that when already negative outlooks became even more so in short-term panics, it often proved to be an overreaction.
The analytical challenge for investors today is whether current bond pricing reflects a well-calibrated assessment of the regime with regards to growth and inflation. Figure 6 suggests that there is little sign that investors are ignoring evidence of structurally lower inflation in the US as they were previously (in fact, break evens might reflect the opposite) and it would be a bold call to suggest that inflation is likely to be lower than implied by pricing in Europe.
In Q3 there were signs that short term yield moves looked behavioural in nature. However, the fact that these moves haven’t unwound more quickly suggests that the market may have some fundamental motivations underpinning prevailing yields. It remains to be seen whether the pattern for developed bond yields resembles the ‘V’ of 2016, or the ‘L’ of much of the last 30 years.