Tuesday, August 17, 2010
As regular readers will know, Team Macro Man detest currency piss-taking, especially from the likes of Voldemort and the dark side of the force, but on certain occasions, agree it is useful as a policy tool, but only when the conditions are right. The $200bn Swiss hedge fund run by ex-Moore trader Philipp Hildebrand screwed up on this front, because the man himself forgot that there are circumstances under which intervention is successful, and if you forget them, no matter how big your ego is, you will lose. And that is why he is sat on $7bn (or thereabouts) of FX losses... maybe he should've done a stint at SAFE instead of Moore Capital.
But let's get to the point of this piece. There has been a lot of noise and jaw-boning about the Japan intervening in the Yen, with a very rare joint BoJ/Government statement about the risk it poses to Japan's economy, and the political debate around this appears to have reached a consensus that verbal intervention is no longer proving an effective deterrent against Yen strength. TMM is increasingly of the opinion that the bar to intervention in the Yen is now very low - not just because policymakers appear to have reached a consensus, but the key conditions that have been consistent with successful FX intervention in the past are now virtually all in place. The single remaining condition is a sharp upward move in the Yen of 2-3% which, given last week's low in USDJPY is very close, seems pretty likely to happen.
So what are the conditions? Looking back at past FX interventions, they have generally been successful if they have occurred when: (i) the FX market isn't pricing the relative economic outlooks of the two countries properly (whatever that means...), (ii) there has been a large "mis-valuation" of the real exchange rate, (iii) market positioning is large and viewing the trade as a one-way bet, and (iv) there has been increased momentum in the market moves.
As far as (i) is concerned, we can get a sense of the relative economic outlooks of the two countries by looking at a spread of real interest rates over the short to medium term (e.g. 5yr) because these will price in the relative policy paths of central banks, which themselves are a function of economic conditions. TMM wanted to use consensus economist forecasts historically, but getting that data has proved somewhat challenging - if they were wearing their tin foil beanies, they'd swear it was a cover up of their horrendous forecasting records! But on a serious note, the below chart shows the 5yr real rate spread between the US and Japan (brown line) vs. USDJPY (green line). [We used TIPS yields from 1997 for the US and Inflation Swaps from 2007 in Japan, prior to those periods, we used the 5yr bond yield minus CPI]. It's pretty clear that on this measure, USDJPY has diverged pretty significantly from the fixed income markets' views of the relative economic outlook, a condition that was not met particularly in early-2009, the last time the Japanese triggered a G7 statement on the Yen. That's a "tick" then.
Moving on to (ii), the Real Exchange Rate... the below chart shows the percentage deviation of the ratio of the Dollar to Yen Broad Real Exchange rates from its 5yr moving average. Looks to be about 15% undervalued. Another "tick"...
As FX punters will know, working out how players are positioned in the FX market is difficult, as we only have data for CTAs (via the CFTC - see below chart of Yen longs) and for Mrs Watanabe. Clearly CTAs are very long of Yen, as is Mrs Watanabe, and TMM gets the sense that traders more broadly are also long the Yen. While we can't be certain here (proxy measures of the OTC market positioning such as Risk Reversals are clouded by their risk-on/off correlation and Black Swan derivative hedging), TMM thinks (iii) is a "tick" too.
Finally, market momentum (iv)... for such an esoteric concept, it's possible to come up with many measures. For example, it's pretty obvious when there is a large 3day move, but in terms of the underlying momentum in the FX market, TMM particularly likes using the 3month, 5day skip momentum measure (see chart below). In recent days we have surpassed the "normal range" for this, but are not quite at the levels exhibited in March 2008 (as the PRDC crowd were being taught about gamma) or Q4 2008. The point here is that the bar is not particularly high for this to move higher to the point where policymakers would act. TMM is of the opinion that a relatively quick 2-3% move is all that is needed to trigger this condition... we'll call it a "half tick".
One of TMM's mates this morning suggested that a further condition is support from other nations' policymakers. It's a fair point, in that coordinated intervention has only failed on one occasion (the Louvre Accord), while single country intervention has failed on several. TMM cannot help but point to Voldemort et al who have successfully intervened for many years now...
We digress... TMM thinks the BoJ are getting pretty close to adding the letters "LLC" to their name.