For many asset classes there has been nothing but a roundtrip, risk on before the Fed, replaced by risk off. However the momentum especially in say high yield, some dollar pairs (especially vs. EM) and to some extent equities appears to be headed towards revisiting what had been earlier sell-off lows in the China devaluation panic.
For some investors the Fed’s failure to raise rates is viewed as the culprit in the risk reversal, even though for many investors the Fed was viewed as being in a “dovish hold”, reflecting a more sanguine outlook for inflation as well as global growth concerns. The former view suggests that the Fed is still very much on track for hikes later this year and all will therefore be well as the start to normalization will instill confidence in the economic outlook and “remove” uncertainty that might otherwise be plaguing risk assets. And consistent with this it is the official view of our economist colleagues that the Fed will raise rates in December and then again twice more in 2016.
Yellen, in her speech late last week, specifically included herself in the group of FOMC members that expect to raise rates later this year and as such seemed to redirect the market’s perception of the September decision toward a “hawkish hold” rather than a “malign” dovish hold. The malign dovish hold is one in which the Fed delays lift-off because it is concerned about the vigor of activity domestically and abroad. The hawkish hold, in contrast, leaves rate hikes on the table for 2015, and is likely to keep downward pressure on inflation, risk assets, the general level of yields, and the slope of the curve.
The potential issue with Yellen’s position is that it is based largely on an entirely conventional view of inflation dynamics, whereby headline inflation over time is expected to revert to the more stable core measure. This entails assigning a greater “weight” to inflation survey data rather than traded market inflation compensation, posits that the long-run inflation trend is unchanged, and assumes no changes in the relationship between core and headline inflation. This is all conventional stuff, but remains exposed to risks that inflation dynamics have changed. For example, we have highlighted statistical evidence that suggests that headline inflation “Granger causes” core inflation rather than vice versa since the crisis in the US, with a similar and somewhat more robust effect evident in the European HICP data.
This result could be deceptively important in that, if it persists, it suggests that expectations of Fed hikes could be having a negative effect on the inflation metric that drives them. That is, Fed expectations strengthen the dollar, depress risk in general and commodities in particular, with lower commodities driving headline inflation lower.
This would be less problematic if headline inflation did not feed back into the core measure, as base effects would lead the decline in headline inflation to dissipate over time. The problem is that there is demonstrable pass through into core, with the risk being that “transitory” influences could in fact be depressing the longer-run trend. Even if this statistical behavior is a transitory cyclical effect stemming from the financial crisis, the reality is that 1) the dollar linkage demonstrates the relevance of the global economy, and 2) other major economies are in worse shape than the US economy.
…Risk assets remain negatively correlated with real yields – a signal distinct from pre-crisis behavior that we interpret as reflecting market concerns over a policy error. While it is possible that risk-off dynamics will be limited in scope and hence acceptable if not desirable, it is also possible that risk off could overly tighten financial conditions in a fashion that precludes, rather than reflects, a more balanced recovery domestically and abroad.
Remember that at the zero bound, central bank signalling becomes all powerful and the Fed has been signalling rate hikes for so long that it may well have already over-tightened markets!
