Earlier this morning, I noted that the March (get it?) higher in US 2Y yields looks pretty amusing if you plot it against near record low German 2Y yields.
Indeed, as noted here on numerous occasions, there’s a push-pull dynamic between the US and Germany when it comes to global rates. Have a look:
Of course Schatz yields have blown out since hitting record lows late last week. Indeed, between hawkish Fed commentary, what counts as “upbeat” Trump messaging, and, importantly, rising inflation in Germany…
…German 2Y yields have bounced sharply and now sit at their “highest” (and I use that term very loosely) levels in a week:
“Bund futures slide from the open after hawkish comments from Fed’s Dudley weighed heavily on Treasuries and as the month-end bid fades, German inflation data improves and stocks rally sharply,” Bloomberg writes, summarizing the action, adding that “core curves are bear steepening, with 5s30s steeper by 2-4bps [while] the front-end of the German curve sells off, with continued liquidations of longs seen in Schatz.”
In turn, swap spreads across the 2-5y sector are building on recent tightening, dropping by 4-6bps.
There’s been quite a bit of discussion about what exactly it is driving the insatiable (until today) Schatz bid and in a new note, Goldman argues that richening in the German front end “appears to originate from the ECB’s ongoing QE.”
This isn’t an esoteric discussion. We need to know the extent to which the Schatz spread to 2Y French govies is an indication of investors trying to price redenomination risk around a prospective Marine Le Pen win. That is, we need to understand how closely we should be watching the German curve as a barometer of political risk. If this is all technical buying tied to the mechanics of PSPP (including buying below the depo rate), then it decreases German 2’s usefulness as a litmus test for political uncertainty.
With all of that in mind, consider the following from Goldman.
At auction yesterday, Germany sold around EUR 4bn two-year government debt (the ‘Schatz’, short-hand for Bundesschatzanweisungen) at near record low rates (-91bp, around 2bp through mid-market and some 60bp below EONIA). The price action in the Schatz has attracted increasing attention of late and commentators have associated it with a build-up of ‘EMU breakup’ risk.
Since the sovereign credit risk on triple A-rated Germany is considered de minimis, the yield on 2-year Schatz has historically closely tracked the expectation of policy rates. Indeed, regression analysis reveals that, since EMU inception, the coefficient tying the generic 2-year Schatz yields to 2-year EONIA (which captures the path for ECB official rates over the corresponding period) is very close to one, and statistically significant. Since mid-2016, however, the two financial instruments have increasingly decoupled (Exhibit 1) and consequently, the portion of the variance of Schatz yields captured by moves in EONIA has fallen precipitously (from close to 95% to around 60% using an 18-month rolling window with weekly data, as shown in Exhibit 2). While Schatz yields make new lows, EONIA is stable at levels prevailing since last November (-30bp). This clearly implies that other, larger forces are at play.
‘Scarcity’ effects: The stock of 1-2 year German government bonds is relatively small (around EUR 150bn) and, given Germany’s low deficit, not growing either. Starting in early 2015, the German Bundesbank has been buying government bonds for its QE program, absorbing all of the limited new supply and eating into the stock of securities held by private hands. To get a sense of how these twin dynamics (low supply from the Treasury and high demand from the central bank) could be influencing the pricing of the Schatz, we construct a flow variable defined as the difference between the 1-year ahead expectation of Germany’s public deficit (taken from the EU Commission) and the 1-year ahead expectation on the amount of German Bunds the Bundesbank will need to buy to satisfy its QE quota (using GS Economics central projections).
The variable, shown in Exhibit 3, tracks the net new supply of German government bonds available to the private sector across all maturities, taken as a percentage of GDP. As shown in Exhibit 3, over the crisis years (2009-10) the new supply of German bonds available to investors was abundant – equivalent to around 5% of GDP. In more recent years, however, the combination of improving public finances (Germany is forecast to run a small budgetary surplus over the coming year) and large-scale purchases of bonds by the Bundesbank is reducing the expected flow of German government securities available to private investors to the tune of 4-6% of annual GDP. Adding this variable to the regression alongside EONIA, we find that every 1% change in the ratio of expected net supply to GDP moves the Schatz by 1-2bp, all else equal.
EMU ‘breakup risk’: An increase in EMU breakup fears, and the potential for a reintroduction of national currencies, would lead to an appreciation of the Schatz as investors look for liquid assets that could be redenominated in Deutsche Marks. It is therefore quite possible that concerns around convertibility ahead of the French presidential elections could also be behind the price action in Schatz. To check this, we construct a measure of EMU stress by taking the ‘common component’ of 10-year yield differentials between Italy, Spain, France, Portugal and Germany (see Exhibit 4). When added to the regression alongside EONIA and net supply, our EMU stress proxy is statistically significant and has the correct sign, i.e., an increase in the systematic part of EMU sovereign risk leads to lower Schatz yields.
Goldman’s conclusion: “All told, the forces behind the sharp rally in the Schatz since the start of this year are hard to dis-entangle.”