Last year, it was a recession story. This year, it’s a dollar/inflation story.
Deutsche Bank’s Stuart Sparks and Aleksandar Kocic have been hard at work lately hashing out the dollar nexus story and using it to construct a framework for understanding the mode of the curve and, ultimately, the likely path of Fed policy.
To be clear, Deutsche’s house view isn’t for rapid rate cuts, and neither are Sparks and Kocic calling for them. Rather, what the two have developed over the past six or so months, is a risk scenario that involves a kind of self-fulfilling prophecy, whereby, in the absence of a marked upturn in ex-US growth, and in the presence of a flat Phillips curve, the Fed may find itself forced to ease in order to avert importing disinflation via the currency channel.
I’ve discussed this at length – and I do mean at length. In addition to the linked piece above, we’ve featured at least three more long posts explaining the situation which, as Sparks put it late last summer, “smacks of inevitability”.
Here, in a nutshell, is the problem (this is from a December note by Sparks):
The fundamental risk for the Fed’s inflation narrative is that the flat Phillips curve suggests that even above trend growth might not generate enough domestic inflation to offset dis-inflation imported by dollar strength that is driven by that same above trend growth. In these circumstances, ironically, the net result could perversely be that above trend growth lowers inflation!
So, the Fed cuts rates, which supports the domestic economy, but in an environment where other economies continue to suffer disproportionately in the face of trade frictions (and now the economic fallout from the coronavirus), the dollar-positive implications of favorable US growth differentials outweigh any inflationary impulse from stronger growth due to the stubbornly flat Phillips curve.
And so, the US continues to import disinflation, making it even harder for the Fed to achieve its mandate, quite possibly tempting the committee to keep cutting rates in a self-defeating loop until it all dead ends at the lower bound (or below).
In his latest note, Kocic discusses this in the context of the curve which, you’ve probably noticed, has flattened anew in 2020, retracing Q4’s steepening. For example, the 3-month, 10-year continues to flirt with inversion, the 2s10s has erased nearly half of its Q4 re-steepening off the August inversion, and JPMorgan is concerned about the very front end.
“The path remains nonlinear as the market continues to trade in a mixed mode shifting from global growth slowdown and strong US economic numbers as the main theme in the first three weeks [of the year] to event driven worries about the depth and extent of coronavirus in the subsequent two”, Kocic writes, adding that “what sets the scene for this pattern is the dispersion of global growth and divergent monetary policies that support bull flatteners in a counterintuitive way”.
What he means by “counterintuitive”, is that stronger US growth creates a bull flattening impulse due to the dynamics outlined above, and also in keeping with appetite for relatively high-yielding US assets. The safe-haven appeal of US debt only magnifies the issue, as does the typically inverse relationship between the dollar and commodities.
Kocic brings in the epidemic factor, noting that “the coronavirus acts either to amplify these modes, if its effects are contained abroad and they become a source of capital flight to the US, or to attenuate them, if they threaten to spread out and affect the US markets”.
In other words, if the virus threatens the US economy equally, that would help mitigate the dynamics discussed above to the extent the economic divergence between America and the rest of the world narrows, or at least isn’t amplified to the benefit of the dollar.
The interaction between the trade-weighted dollar and the curve is illustrated below. The three lines are obviously last year’s insurance cuts.
Do take a moment to appreciate how this marks a departure with precedent.
Previously, any negative correlation between the dollar and the curve was due primarily to bear flattening and bull steepening, as a stronger economy raised the odds of tighter Fed policy (stronger dollar) and vice versa (weaker dollar as short rates fall in anticipation of easier policy).
As Kocic writes, the interaction between the dollar and the curve slope is now “a result of persistent dollar strength and its dampening effect on inflation expectations”.
In a sense, we’re reverting to the abnormal mode (or “QE mode”) of curve dynamics seen post-crisis. Shocks are now arriving at the backend again, this time courtesy of the dollar-inflation nexus. Here’s Kocic with a quick recap for those who need it:
Under normal circumstances, monetary policy shocks propagate from short to long end of the curve causing either bull steepening (rate cuts) or bear flattening (rate hikes). As a consequence, curve slope is negatively correlated with short rate. In addition, monetary policy shocks create volatility that is concentrated at the front end of the curve. In contrast, QE mode consists of bull flatteners and bear steepeners accompanied by low volatility across the curve as the short end remains anchored near zero and long rates constrained to a range.
But it’s too simple to suggest we’ve merely returned to the “QE mode” – or at least as we knew it before, viewed strictly in the US context. Here’s Kocic again:
In its original form, the essence of (domestic) QE is to created demand shock. By compressing the term premium of nominal curve real rate is pushed lower and, as the front end is anchored at zero, the response of the long end becomes a referendum on the success of monetary policy. In that environment, there is only one trade: Everything is correlated with breakevens. If demand shock is effective, breakevens widen and push real rates further down, which becomes reinforcing as low real rates stimulate more consumption and, in principle, raise price level. This is articulated as bear steepening of the curve and is synonymous with the risk-on trade. Conversely, failure to reflate the economy makes the Fed action futile with compression of breakevens pushing real rates higher and inhibiting consumption. This is the risk-off mode with bull flattening of the curve.
Now, however, we have to take account of foreign demand for the US long-end precipitated by sub-zero yields abroad. That effectively mutes the bear steepener from the traditional QE mode. The Fed must then cut rates in order to avoid ongoing bull flattening associated with the stronger dollar and lower inflation expectations.
The dollar’s role in all of this is pivotal and paramount. “The key control variable that separates current regime from the domestic QE is currency”, Kocic goes on to write, on the way to noting that “while domestic QE is erosive for USD, foreign QE is supportive for it [as] the more foreign QE there is, the stronger the dollar becomes and the more compressed the USD term premium, which reduces inflation further, but does not result in lower real rates in the US”.
Things get really – really – messy when you start to think a few steps ahead.
In the absence of a material upturn in ex-US economies (and/or a reinvigorated Phillips curve stateside), there may be no way out of the situation for the Fed. Indeed, anything they do could make things considerably worse.
If all of this already “smacked of inevitability” in Sparks’s original exposition of the problem, Kocic drives the point home emphatically. To wit:
What could happen if the Fed cuts rates in order to weaken the USD? A (temporarily) weaker dollar would imply a stronger EUR (or CNH), which would undermine the effects of foreign QE. As a consequence, the market would likely price in more severe (or extended) QE abroad, which would further compress their yields and create another demand for US assets and USD and, thus, offset the effects of the Fed action — we would be where we started. This would force another round of rate cuts in the US with repeated response abroad causing further rate cuts until US rates catch up with their counterparts abroad This is the troubling destination that poses the most severe long-term risk. If we reach the point of zero rates without being able to inspire growth (and some inflation along the way) either abroad or in the US, we would find ourselves in the disinflationary world with low growth and no policy tools to combat it.