Well, the debate is over.
Or at least with respect to the euro.
Oh wait, we forget tell you what was being debated. Let’s back up.
So there are two ways to tighten in the post-crisis, Keynesian Twilight Zone where ZIRP/NIRP operate alongside bloated balance sheets.
You can hike rates or you can shrink the balance sheet. Needless to say, there are some questions as to what the implications are for FX of the two tightening options.
Intuitively, you’d kind of be inclined to think that shrinking the balance sheet would be more FX-neutral than outright rate hikes. And indeed that was one of the arguments earlier this year for why, considering the Trump administration’s preference for a weaker dollar, the Fed should substitute balance sheet rolloff for a hike (or two, or three).
To be sure, this is one long ass debate. And one we’ll likely be having for years although, as accommodative policy is pulled back, we’ll start to get evidence one way or the other in terms of which tightening option does what in terms of influencing FX.
But for now, Deutsche Bank has done as much as you can do preemptively when it comes to proving that “not all tightenings are created equal.” Read more below.
Via Deutsche Bank
The empirical evidence – it’s all about the front-end
There is strong empirical evidence that the front-end of the yield curve is the more important driver of the euro than the back-end. What is more striking is that this relationship has only emerged since the financial crisis.
Up to the 2008-12 crisis, there is little empirical evidence that any particular part of yield curves drove euro crosses more than others. Between 1997 and 2012, EUR/USD and EUR/JPY were slightly more sensitive to the front-end; EUR/GBP slightly more to the back-end (Figure 1). Ultimately, however, the tenor did not matter much, and overall sensitivities were low: A 1% increase in yield spreads would strengthen EUR-crosses by around 3%.
But something happened after the financial crisis. The euro’s sensitivity to rates increased markedly as monetary policy became the key driver of exchange rates and the global search for yield intensified. Even more striking, however, is the twisting in the curve: the euro became far more sensitive to the very front-end of the curve (Figure 2). Extremely high betas to one-year yield spreads fall off rapidly further out the curve and flatten out beyond maturities of four years. If we used 2008 as the cut-off between the two periods, rather than 2012, the pivot would be even clearer.
There are interesting nuances in terms of when the structural breaks occurred after the crisis. For EUR/USD, the importance of short-end rates emerged with the Fed’s initial QE in 2008 and was then magnified when Eurozone rates turned negative in 2014 (Figure 3). For EUR/JPY, the dominance of the frontend arose much later, around the time when the BoJ’s final acceleration of QQE in 2014 coincided with Eurozone rates turning negative (Figure 4). Probably because ECB and BoJ negative rate policies continue to compete in global funding markets, the dominance of the front-end yield spread over the exchange rate remains intact today: the cross is almost three times as sensitive to the very front-end of the curve as to the long-term spread. There is a similar post-crisis pattern in EUR/GBP, which remains vastly more responsive to frontend spreads, though FX-rates correlations have generally broken down since Brexit.