Jackson Hole and ECB Going Forward
Instead of giving trade ideas for Friday specifically, I want to look at the broader picture and try and discern what are the ECB's practical options as QE inevitably takes on a different form. I believe JH will only be the beginning, as the amount of ECB trial balloons will surely accelerate into October. For context, I have written in the past that the ECB will likely remain persistently dovish this year despite improving data, and that the inevitable transformation of policy will not look like the Fed's playbook from '14. For me, there are two key reasons why this is still the case. First, from Draghi's perspective, the APP (asset purchase program) was of course meant to narrow the massive output gap, but the major policy initiative was a convergence in spreads and to stabilize funding and capital flows for the periphery. As I wrote before the French election where I advocated buying BTPs, budget deficits have come in, but PSPP (public sector purchase program) has covered most of the cracks. Spain can run a -5.5% budget deficit and borrow 10y money at 1.6, Italy can fund itself at bunds +180, only in this paradigm does "austerity" become a bit more tolerable, if existent at all. This is why the ECB "taper" will not look that similar to the Fed's; the delicate balance of measuring how much withdrawal the perif can take with an upswing in economic activity will lead Draghi to be a bit more creative in his final two years. I am aware that the current ECB statement language indicates a more traditional policy path, i.e. similar to Fed playbook, but I think going into the October meeting the market will have to adjust certain probabilities in rates across the curve and by effect in FX. I want to avoid such binary terms such as "hawkish" and "dovish" as I think these two terms overlook the significant nuance and complication that will go into Draghi's exit plan. My pontification is that the current perception of the ECB exit sequencing is wrong, but does pulling the front end less negative instead of steepening the curve through taper, constitute as "hawkish", my feeling is not.
With that background in hand, Long 100.125P in ERM8, and short bobls, OEZ7 with a -40bp stop. (I've seen an interesting put fly but would love see other interesting convex structures). What I like about this thesis is, it will only take a famous ECB trial balloon for it to pay off, even if the Fed '14 model is the ECB what practically takes place. Now, this idea in theory contradicts with my core theme of this year, persistent ECB dovishness on the basis of keeping periphery spreads super tight. In essence it does, but the fundamentals of this trade expression convey certain assumptions that I want to explore as they are likely being underestimated by the market. And, these two trades provide a very convex outlet for a multitude of these variables being assigned a greater probability by the market in the coming months.
There are three likely scenarios in my view:
Option 1) Hike instead of taper. In Q1 of next year, a 10 bp DFR (depo rate) hike could come to soften the blow to buba of a dovish APP reduction to 40B a month for 2018. OIS out until April of next year has less than 10% chance of rate move. Again, despite the current ECB statement, I think Draghi may look at pushing EONIA closer to zero as a way to keep his periphery spreads project alive and well and appease German hawks. If anything, pushing up STIRs in Euroland could prove to actually be quite stimulative, as the banking sector would certainly welcome it with loan growth already picking up over 2% in '17. Due to scarcity concerns, Draghi at the current rate of purchases can likely get to around June, so take APP down slightly to keep program for most of '18 at a minimum. In this quid pro quo, hawks at the buba get a hike and Draghi continues to fund periphery budget deficits, albeit at a bit of smaller pace. The fact is, from Mario's perspective, APP works while NIRP reviews are a mixed bag, why not keep the more effective policy tool going? Draghi can push EONIA on path back to zero as a much healthier form of tightening relative to making significant changes to APP. A switch in the exit sequencing looks to me as much easier way out for Draghi, and the market is not pricing it, 4bps in ERZ7-M8.
ERM8 100.125P, vol in reds is super cheap
Option 2) Draghi has made it clear to markets that only one thing matters: "Our mandate is neither growth nor employment, but price stability." Would it surprise me if Draghi says screw you to the buba, core HICP inflation at 1.2 is not high enough, and considering recent EUR move weighing on 5y5y inflation swaps, he could have a case that to get inflation to mandate, the current state of mon pol is still appropriate. Draghi can also say, Chinese credit growth is likely to slow next year, Euro econ runs on a lag to CH expansion, do I want to be withdrawing stimulus into that? Very easy to see how Draghi takes the most dovish way out, especially if pressures from German election do not really mount and inflation begins to roll.
Biggest risk to ECB exit trades, China
In prior posts I have looked at Chinese imports from Germany as a leading gauge for HICP inflation, and it has largely played out so far this year. However, I think this chart better encapsulates the lag from Chinese credit expansion onto German HICP inflation. Despite three year highs in copper and a pickup in the credit impulse, I still remain skeptical of the durability of the current expansion in China. The rally in industrial metals looks to be more driven by WMP issuance and inflows, as domestic stocks remain very elevated. However, if German inflation is operating on a few month lag to Chinese credit expansion, prices could still slow over the coming months, which gives Mario an easy way out and bunds trade back into the teens. This is likely the biggest non geo-political risk to the trade, but one can easily say, Mario will be too late to notice and the German pressure will force action. The beta of Euro rates to Chinese industrial expansion cannot be underestimated, so to counter maybe add a receive AUD leg onto the trade to better hedge out CH risk? This sort of deflationary risk also speaks to raising rates being advantageous over taper. If Draghi hikes DFR 10 or even 20 bps, he can easily stop if deflationary risks from China or oil are to reemerge. With taper, the Fed model suggests some sort of monthly auto pilot is most ideal, as big monthly deviations will cause increased vol in rates land. Another thing for Mario to factor in if he is skeptical of current upswing as Trichet's shadow hangs.
The other side of the Chinese Coin
So, if Chinese credit impulse indicates that HICP inflation in Germany should fall, the rise in the RMB suggests the exact opposite. To me, this is a powerful chart because as I have talked about, persistent weakness in the USD is a powerful global stimulant. I am sympathetic to the view that dollar scarcity will reemerge following a normalization in bill issuance, but if until then, the debt ceiling and persistently low inflation in the States keeps the dollar on the offer, that weak USD stimulant will be alive and well. Lots to balance here, especially when Chinese fundamentals have proved so difficult to properly judge as individual variables, let alone systematically.
Option 3) More consensus exit sequencing. Draghi likely extends QE slightly into 2018 but lays out plan of eventual path to 0, obviously not including reinvestments. Draghi acknowledges output gap is narrowing, inflation forecasts for 2018 rise slightly, and DFR hike first or second quarter '19. The biggest pull to move Draghi in this direction will not be economic, as I have tried to emphasize, to Mario's credit, he is a big picture guy, cyclical upticks will not shake him off his systematic and structural goals. Scarcity is the real problem for the QE forever camp, unless much bigger capital key deviations are tolerated, which is tough to envision, QE parameters will have to change. German WAM is slightly above its lows, but still down to near 5. In this case, Draghi is forced to lay out plan in many stages to begin taking monthly purchases to zero, as QE reaches its natural limit.
5y Bobls v 1y1y EONIA
In theory, given my thesis, paying rates on 2y schatz should be a better trade expression. The reason I prefer bobls is its flexibility to different scenarios. In my base case, ECB talks market into pricing in a DFR move more quickly and pushes off significant taper talk, can bobls still trade inside a rising 1y1y EONIA? The generality is, if the ECB can ever so slowly push EONIA closer to 0, bobls would have to follow suit. In case two, Draghi stubbornly continues APP close to its current form for most of 2018, I don't think bobls can trade inside depo again, especially given the pickup in domestic economic activity and HICP inflation. This is really sort of the obvious asymmetry, unless Draghi expands QE, I do not see bobls trading at a real rate of -200 bps. The other point is, if Draghi goes full out dove, he kills EUR before German current account can effectively adjust to trade weighted strength, this would in essence shoot higher inflation expectations and likely prevent a persistent rally in the belly out. Case three is likely the most self explanatory. Threats of more traditional exit sequence will crush the overbought belly of the German curve, look how bobls sold off following Sintra. The latter case, despite being third on my list of probabilistic outcomes, shows why the bobl short is more advantageous to schatz, as it has more beta the Fed model of taper.
EU GDP v Expected Real 5y Bobl (reals are a nebulous concept to me, but I used 5y5y swaps here)
Draghi and Inflation
Part of the reason I think the street and even ECB rhetoric has the exit sequencing wrong is because of inflation. Yes, inflation has accelerated meaningfully this year, even Euro super core inflation is picking up. However, in Draghi's "hawkish" Sintra speech he effectively told us real rates have to be zero. So while core HICP inflation has made progress at 1.2, does it really satisfy, "the close to but below 2%" mandate, maybe for Weidemann but not for Draghi. So inflation still needs to pickup and this move in EUR is not helping. The question is, if Draghi needs an outlet to appease the Germans, why does moving EONIA closer to zero have less of a negative inflationary impact than tapering? The answer may lie in a working paper from the ECB, which was released in June (http://www.ecb.europa.eu/pub/pdf/scpwps/ecb.wp.2075.en.pdf). The paper goes over a VAR based model which estimates the impact of APP on inflation and real GDP. The paper did not compare APP effects and impacts as opposed to using overnight rates, especially past the zero bound, but rather the models results are interesting with regard to the timing of inflationary impact. In effect the model attempted to quantify the lag through in the different transmission mechanisms, lending, FX, etc. What we learn is, there is a much higher higher residual time variance for inflation as opposed to real GDP. This means, by tapering, Draghi will risk messing with the positive lagged effects of the APP, especially on inflation, as they are still making there way through the system, even if some have peaked (VAR model suggests based off APP changes in '15, Q4 2016 and a 0.36% impact is the peak effect). Considering the general negatives we have seen with NIRP, it makes a lot of sense to me that taking the front end less negative will not have as big of an effect on inflation, as the bank lending transmission onto broader loan growth will likely be more positive. Can Draghi appease the buba and at the same time actually keep this uptrend in inflation, just may be, moving EONIA closer to 0 instead of big APP changes is that outlet.
EUR 5Y Forwards, Maybe let EUR do the Work?
In my opinion, Draghi's concern with the recent EUR may be a bit overstated, especially as it is everyones base case for a dovish JH speech. Yes, effects of rising trade weighted EUR negatively impacts growth and inflation, but, it disproportionately hurts German econ. We have already begun to see this in German survey data, PMI and ZEW starting to come in from very elevated levels, but also in the German stock market. Remember, since EMU inception, Germany has kinda taken current account surplus from others in the bloc to make their own monstrous surplus +8% of GDP. The contrarian in me says, well, Draghi is in a place where his priorities are systematic in nature (tighter perif spreads), and the biggest threat to him having to remove those safety measures is, German economic expansion.... The variable with the most beta to restoring some balance among the EU economies, i.e. leveling out German growth with an adjustment to current account, may be a slightly stronger Euro. Sure, Euro has effectively tightened some financial conditions in the EU, but the core effects look largely Germany bound, which actually prevents Draghi from having to rush on mon pol changes in the name of quieting the German hawks. In an Ideal world, Draghi knows relative weakness in EUR is a large boost to inflation expectations and trade across Euro Area, but at current levels it may actually smooth out some of the disparities in the continents relative growth rates.
Regime change in relative curves, German v US 2s10s
Something I've noticed over the past few months is that the US-German 10s spread trades tighter than the respective 2s spread. If you look at this series over more significant time frame, you will see this has not been the case for over five years. This paradigm shift does make sense given consensus outlook for the ECB and Fed respectively. Inflation in the US is rolling over and the ECB is about to taper, so 10s spread should tighten, which likely every macro HF has on given its positive c&r. With regard to 2s, yes, the Fed is towards end of hiking cycle but there is no way the ECB goes at rates before Fed is done, so 2s still trade super wide. I want challenge both of these assumptions and even suggest that with Fed SOMA changes to begin in October, this paradigm can revert back to its recent norm. In my mind the narrowing of schatz and US 2Y makes a lot more sense in my ECB framework than narrowing in the belly and out. Fed will be making conscious effort to turn duration lower for the sake of tighter financial conditions, while the ECB may go out of their way to keep duration, especially in perif, bid. On the flip side, the Fed is getting to the end of their tightening rope in o/n rates, while the ECB could be just beginning..... Short schatz long US 2s also works, especially given spec shorts in US 2s, or trade below.
ERZ7-M8 v EDZ7-M8, 9.5 bps
Fed Tightening Cycle has been more "Normal" than you think
While one wouldn't think it, on a relative basis, this tightening cycle has been pretty "normal" for a post Volcker cycle. In shadow rates (Wu-Xia above), you're at over 300 bps of tightening for this cycle if the Fed goes again in Dec, in my opinion a bigger than 40% chance. Also, if you asked the Fed at last year's JH, not sure Fischer would have told you the odds would be high that by year end '17 we would be this close to in policy rate to r*. However, given the calcs and weightings of prior months, inflation on YoY basis is stuck. The Fed has already tried to hedge future CPI readings by saying they will focus on sequential readings instead. Of course it is hard to say what the Fed path will be if Janet is replaced by Cohn in Feb of next year, which has contributed to the shallow path in reds, greens and golds, but with core PCE stuck near 1.5 and bank lending contracting, the writing is on the wall. This does not even take into account if SOMA changes begin to go awry in 2018, a very real possibility that the Fed and market is discounting in the name of a steeper curve and tighter financial conditions. But in my view, the latter will be taken care of by a normalization in Treasury issuance.
14bps, US 2s and Effective Fed Funds, Unless r* Moves Higher, the End is Near
US 2s inability to really sell off since Jan has been very telling. My thinking a few months ago was, this is the most obvious r/r short, 2s trade 5 bps outside IOER, 14 bps from effective Fed Funds, and they have great optionality on Fed surprising market with more hikes. It turns out I was not alone. People like to talk about market messages to the Fed, like, "what does it say that despite 3 hikes since the election and spoos are still at ATH's". My answer would be, look at the curve. 1s2s like 8 bps, and the 2s are about to invert with with o/n rates if the Fed goes again in Dec. This is either a massive mis-pricing or the curve is sending a message. I bet the latter and am emboldened by the fact positioning is so heavily skewed to the other side. I would say, going into JH this position is less appealing to me as Yellen could use Jackson Hole to remind the market who is in charge, but as a whole, I expect a big narrowing of US 2s German 2Y schatz spread.
Interesting dynamic between Fed and ECB could start in JH.
In the spirit of Jackson Hole game theory, it is fitting to end this post on something a little bit out there and not readily practical, but I here it goes anyway.
I think it is well accepted in the CB community that coordinated tightening, if justified, will have to be the name of game, no one is looking for a '94 type move in rates. This is where Draghi must think, when it's time for the process of real ECB normalization to begin, the Fed will have to play ball. It's simple, Draghi is looking at the Fed's expected policy path priced in ED$ and saying, they are nearing the end, especially as inflation has come down significantly the business cycle is very old in dog years (about to be second biggest post war expansion). Not hard to see, especially if problems arise from SOMA changes, that the Fed could be reversing course over the next 18 months. Well if that's the case, that creates a much cleaner buffer to hawkish policy shifts from the ECB, and in a sense, they did the same thing for the Fed. What happens to duration in USTs after QE is wound down and the Fed outlines dots path to neutral rate of +3 if the ECB doesn't come in and quash term premium? From Mario's perspective, why can't this play out in reverse, i.e. Fed keep some sort of lid on EGB term premium while ECB takes APP to 0.... This is a very out there, but its worth keeping in mind over the next couple of years. Mario could be thinking, its time for Janet to return the benevolence that ECB QE showed the UST curve in '14. Of course this would have massive FX implications that Draghi would want to avoid; it's a work in progress.
Short Sterling Rates still discounting BoE (L U7-M8)
Quickly touching on my BoE thesis from last post, I think it is more in tact than the previous presser would suggest. GBP/CHF long has moved but is basically unched over the month. While Carney's presser was interpreted as dovish, I actually thought he just went over justifications of a hike despite sluggish growth... Spare capacity receding, and if that process is to accelerate, OIS curve has to price in more and sooner hikes. Message from Carney, in my view was, ya economy is sluggish but if ECB moves the needle and EUR/GBP continues to shoot higher, I'll be forced to move. Reason I like pound swissy pair is, SNB in that situation will come out and likely say the opposite, that they are not following as geo-politics have kept CHF too strong this year. I believe I am right in saying that we will not be hearing from Carney at JH, but his deputy, Broadbent will be there, worth keeping an eye. Economy looks weak and FX pass through onto inflation is waning, but if Carney does not keep up, FX driven price instability will return.
Thanks for reading. My email is jonturek@gmail.com, please reach out. Have nice rest of the Summer.
Jonathan Turek
No comments:
Post a Comment