8 min read 21 Aug 19
Summary: For those of us who weren’t around during the dot.com boom and bust, this period can seem like an historical oddity. It’s often used as a classic example of the foolishness of human beings and the ability of markets to make blindingly obvious mistakes.
From the vantage point of hindsight it is tempting to believe that we couldn’t possibly make the same errors.
And yet we live in a world where investors appear to be happily buying government bonds with a guaranteed negative return, amidst recent price action that certainly looks very bubble-like:
Will the investors of twenty years from now look at this period with a similar mixture of disbelief and amusement as we do the tech bubble, or is there something else going on? What signals should we be looking for to identify a bubble?
The starting point of all assessments of potential ‘bubbles’ (by which we mean price moves vulnerable to reversal and permanent loss) is valuation. Importantly, we need to consider not just the prevailing level, but also the extent to which that valuation can be said to have departed from some sense of ‘fair value.’ Clearly a negative yield is unattractive, but does that mean it is inappropriate?
To assess, it is helpful to consider recent bond price action in context. Consider the same chart of recent bund moves relative to the Nasdaq prior to its bursting in 2000:
Or relative to Bitcoin prior to its (last?) peak two years ago:
The point I am making here is not that we should be less worried about a Bund bubble because the price moves are not as large as these other examples. Rather, I am observing on the differential nature of valuation anchors across assets.
Assets without an obvious ‘valuation anchor’ (or a high level of ‘pricing model uncertainty’) can move far further into bubble territory than those where a sense of fair value is clear.
In equity markets the anchor should be related to the path of long term earnings delivery (together with interest rates, inflation, and some kind of ‘risk premium’); in the case of US tech stocks this appears to be true, with the tech bubble proving to be an aberration:
Valuation anchors are loosest when those future earnings are uncertain, hence the bubble emerging in a sector of new, fast-growing, industries. Bitcoin is even more extreme, it arguably has no anchor at all, meaning that price moves can become even more extended.
The situation is different for government bonds, where the anchor is far stronger. If those bonds are largely free from default risk, prevailing yields should be very closely aligned to rate expectations and those rate expectations are a manifestation of the dynamics of the real economy. In theory, fair value should be easier to assess and prices less likely to depart from it meaningfully.
Clearly a low or negative yield represents an unattractive financial asset, but that doesn’t necessarily mean that it is ‘wrong.’ If yields are genuinely reflecting the path of future rates then we can moan about the poor returns as much as we like, but that doesn’t make it a bubble.
However, since the end of April and especially over the past month, something else also seems to be going on. The behaviour of bond prices (or yields) have become the story itself, rather than a reflection of the economic backdrop (our observations on the ‘QE makes bonds a one way bet’ narrative can be found here).
Some have already discussed how these moves already meet some commonly held criteria for defining bubbles, but there are some other observations we can make from a behavioural standpoint.
Bubbles are often characterised by investors changing their rationales with a view to justifying current levels of pricing. In the tech bubble, those arguing that stocks were ‘expensive’ ultimately had to capitulate and buy into the ‘new economy’ narrative, or run high levels of career risk.
In the case of bonds, previously held views on equilibrium interest rates have been jettisoned as various commentators start to use ‘new era’ arguments to explain why very negative yields are justified and here to stay. See for example, this article on ‘The Non-Weirdness of Negative Interest Rates’.
Of course, many of these bond arguments are not new (remember ‘secular stagnation’?) and such rationales will always have some validity (just as the expectations of huge profits growth out of the technology sector in the US proved to be correct).
The warning signs come when we start making these arguments after prices have just moved rapidly and there are signs of groupthink. It suggests that investors have lost their sense of what fair value is and are changing the story to fit yield moves. Have investors really made a considered view of the regime, or is it just more comfortable to follow the crowd?
It is also important to look for possible emotional drivers of near term moves. Today, the speed of moves and the nature of commentary suggests a degree of confusion, while we are also seeing somewhat unusual arguments to make the case for holding certain bonds today: that Bunds or JGBs are ‘cheap’ when hedged back into US Dollars for example, or that investors need convexity and duration, rather than yield. Return chasing, and fear of missing out do seem to be prevalent:
A recent article did much to sum up the current allure of abandoning fundamental assessment in such periods. To quote: “While many human traders may question the wisdom of buying or keeping a bond that apparently offers a guaranteed loss, robot traders that monitor price moves have no such qualms.”
Quantitative strategies offer many advantages, but the above encapsulates the degree of ‘greater fool’ thinking that sees to have taken hold today, echoing the environment of 2016. It can also be the case, as Eric has pointed out, that quantitative strategies only serve to reinforce behavioural dynamics in markets.
Government bonds offer a challenge in that yields should be tied to an underlying economic reality: the path of rates in the real economy. These rates in turn will reflect the savings preferences of households, not the desired returns of bond investors.
This means that negative yields, while representing a poor investment could nevertheless be justified. However, there are enough signs today to suggest that recent market moves are not about a considered assessment about where prevailing rates will be over the long term, but the very human influences of short-termism and herding.
We are now at a point in time where M&G Prudential is hiring staff who weren’t even born when the tech bubble burst. It remains to be seen whether, when I come to retire, new members of staff will look at this era with a similar mix of disbelief and amusement.
The value of investments will fluctuate, which will cause prices to fall as well as rise and you may not get back the original amount you invested. Past performance is not a guide to future performance.