Could it be that professional investors are chasing year end equity returns to flatter calendar year performance? And does this set us up for a weaker January as those same investors unwind positions they might not really believe in?
As equity markets have continued to rally there has been a sense that markets are getting carried away with themselves, as Maria discussed last week.
Many have struggled to justify this price action, typically citing the Trump election victory as something to do with it, and questioning whether the gains are sustainable. More recently there has also been the suggestion that so-called “Santa Claus” rallies are typical when professional managers seek to boost calendar year performance.
The table below shows that, since 1971, median returns have certainly been higher in the last two months of the year in some major indices.
In the US around 70% of December returns since 1971 have been positive, while in Germany and Japan this figure is closer to 64%. These percentages compare favourably to the rest of the year: across the entire year 59% of US monthly returns have been positive, 58% in Germany and 57% in Japan.
The Santa Claus rally falls into the category of “seasonal anomalies,” along with “Sell in May and go away” (which you can also see in the table) and the “January effect”, whereby small cap stocks in the US outperform at the start of the year.
Some have characterised these trends as anomalies which challenge efficient markets theories. However it is also true that apparent seasonal anomalies could be largely random; over such a short sample it is not unreasonable for such patterns to emerge without an underlying cause.
Whether these anomalies are real or not, it is not a simple matter to make money out of them. Take the January effect; it was certainly not in evidence last year, while we may have already missed the boat for 2017.
A more reasonable explanation for market behaviour?
The problem with seasonal anomaly arguments is that they require a large number of investors to display correlated behaviour. It is unlikely that the incentive of professional investors to boost investment performance would do this in a predictable way (and without being arbitraged).
However, there is another explanation for the type of trending/herding behaviour which may be evident today in both equity and bond markets, and which touches on the motives of all investors. This is the concept of “pricing model uncertainty”, which uses the lessons of game theory and has been put forward by Horace “Woody” Brock. This model holds that when it becomes harder to value assets there is a greater propensity for the market to trend and “overshoot”.
Without a clear value anchor, investors are less likely to step in against the tide when markets move in one direction or another. This is one reason some have put forward to a tendency of trending behaviour in currency markets.
With potential changes to the prevailing policy consensus and a potential pivotal moment in the broader environment, it seems likely that pricing model uncertainty was already rising before the Trump victory. This, combined with the fact that growth expectations (and government bond yields) were coming from depressed levels, and the year has been characterised by significant shock to pre-existing beliefs, could well be a better explanation for price behaviour than the nature of the calendar year.