Go on, you know you want to

"Go on, you know you want to."

How many times over the last few years have macro punters contemplating a sale of the euro heard that voice inside their heads?  From the sovereign crisis to Cyprus, from Fed tapering to ECB QE,  there have been a myriad of (apparently very good) reasons to sell EUR/USD, and yet here we are, knocking on the door of 1.40.  Like the devil on Larry Kroger's shoulder, the voice looks at the chart and says it again.

"Go on, you know you want to."

It's a siren song that's hard to resist, particularly now that the ECB has joined the chorus.  In the absence of a ship-mast against which to lash themselves, punters would do well to examine some of the relevant narratives to determine whether it's a wise course to give in.

1) The 'Japanification' story.   A key argument against selling the euro has been the impressive rise in the Eurozone's basic balance over the last several years.    Not only has the emasculation of domestic demand in much of Europe pushed the current account into a tasty surplus, but Eurozone banks' unwinding of foreign assets has maintained a healthy capital inflow as well.  This combined demand for euros has gotta push it up, right?

The evidence is sketchy.   Macro Man can find no stable and sustained relationship between the basic balance and the euro.  That does not mean that there is no relationship, of course, merely that it does not appear to be a (let alone the) hegemonic driver of the euro exchange rate.  Indeed, at some points in recent history there appears to have been a negative correlation between the two.   The voice is getting louder....




2) Positioning.  With so many different actors participating in currency markets, it can be quit difficult to get a true read on accurate positioning.  Fortunately, the sector with the quickest trigger finger also happens to be the easiest to model positioning for, namely CTAs via the IMM data.  The story there is one of modest (and declining) length.   With the 50 and 100 day moving averages 50-120 pips lower from current spot, it wouldn't take much to trigger a flurry of stops, as that community exits longs and begins to go short.   The siren's song, it's so beautiful....



3) FX Reserve managers.  One of the earliest literary devices in this history of this space was the bequeathing of the sobriquet "Voldemort" upon the PBOC and SAFE.  The rationale was simple:  by accruing absurd levels of FX reserves to maintain artificially weak exchange rates (and artificially high current account surpluses), and then trading/investing these reserves aggressively, Voldemort and like-minded institutions were exerting a malevolent influence upon developed financial markets. Over the ensuing years, the influence of FX reserve managers has waxed and waned.  Recently, China appears to have been at least moderately active diversifying the $100 bio+ that they spent in Q1 to weaken the RMB.

However, how durable is this moving forwards?  The stated purpose of widening the RMB band and pushing the currency weaker was to discourage hot money speculation (both domestic and foreign.)   For the time being, at least, it certainly seems to have worked; moreover, it seems unlikely that the domestic corporate sector will play as many over-invoicing games as they have in the recent past.  At the same time, China's current account balance has shrunk back the level (as a % of GDP at least) of a dozen years ago, back when Voldemort was a mere Tom Riddle...


It therefore seems likely that reserve accumulation should slow, perhaps dramatically...and with less reserve accumulation comes less diversification demand for euros.   Hmmm, Macro Man is reaching for his red sell tickets....

4)   Interest rates.  Rate differentials are the bread and butter of any study of exchange rate determination.  Applying them to EUR/USD appears to support the conclusions above.   Based on the historical relationship with the 1y1y spread, for example, it looks like the euro should be closer to 1.20 than 1.40.   It's an open and shut case, right?




Not so fast.  Readers may recall that there were other interest rates than 1y1y swap spreads (themselves a proxy for monetary policy) that were driving the euro exchange rate in 2011 and 2012.   A proper rate model should include some sovereign spread component.   Macro Man ran two iterations of a very simple two factor model, using a measure of short term rate differentials and the 5 year Spain/Germany spread as the input factors.  One uses 2010-11 as the in-sample period (the Trichet model), and the other uses 2011-2014 as the in-sample period (the Draghi model.)


The Trichet model is not dissimilar to something that Macro Man ran in real time in 2011.   What's interesting to note is that the model broke lower relative to the actual market starting in July 2011- almost exactly the time that Italy and Spain got properly sucked into the sovereign crisis.  While the model has consistently suggested that the euro is too high since, it has retained a reasonable degree of correlation with movements in the FX rate, even in the out of sample period.   What is very interesting to note is the seemingly inexorable rise in the model since Draghi's "whatever it takes" speech in the summer of 2012.  How does it look if we use Draghi as the in-sample?

 

Pretty darned good.  Encouragingly, the basic shape of the model barely changes from the earlier version with a completely different in-sample data set. This version merely corrects for the structural break that occurred in July 2011.  The message, as you can see, is that the euro is pretty fairly valued, and that the recent rise is completely justified. 

Now, one can quibble about using a model like this to try to pin-the-tail on an exact currency valuation.  Indeed, Macro Man received a first-hand education on the dangers of such a practice three summers ago.  Nevertheless, to his eye at least these models do a pretty darned good job in explaining the underlying direction of the euro's trend.  And they both agree that it is up.

As such, from his perspective selling euros, while potentially profitable around event risks like ECB meetings, does not carry a terribly large (or even positive) expected value on a more strategic basis.  From his perch, therefore, it appears that punters' time could be more profitably spent finding a better trade.

Go on, you know you want to.