The Week A September Fed Hike Died? USD Bulls on the Run

This week’s dovish Fed minutes, disappointing US CPI, further oil price falls, growing Chinese uncertainties (Friday’s Chinese manufacturing PMI dropping to its lowest since March 2009) and sharp falls in global equity prices seem to have tipped the balance more decisively against a September 17th Fed rate hike. Market pricing has suggested around/less than a 50% chance of a September hike for a while, but that has now fallen below 30% according to OIS markets. Economists have been more bullish, with a recent Bloomberg survey reporting close to 80% expecting a September liftoff, but I suspect that we could now see a few shifting views. Hiking rates into current febrile market conditions, when there are ongoing doubts about the extent of inflationary pressures, would be a relatively brave FOMC call although it’s not completely out of the question given some OK US activity data and that a delay until December could also face challenging market conditions (see below).

Overall it’s difficult to see USD making much near-term headway against the majors. That would require a September Fed hike to be more realistically priced – requiring more evidence of inflation picking up, upside growth surprises and stronger Fed signals. But I’m dubious that next week’s main US data releases or potential FOMC comments at Jackson Hole will prove decisive. Of course, we’ve got another payrolls release before the Fed meets next, but the FOMC seem largely content with labour market developments. Moreover, USD upside also requires risk aversion drop back to more normal levels in order to reduce the recent supports for EUR (short squeeze of its funding currency status), JPY and CHF (safe havens). But recent elevated risk aversion has largely been driven by adverse Chinese developments (weak Chinese macro data, renewed Shanghai Comp falls and PBOC tinkering with CNY fix) and it’s a brave call to say that’s going to calm down near-term.  My prior is that things could well get worse near term before the PBOC returns with more decisive action, although you can question the efficacy given the apparent slowdown in growth dynamics and difficulties in transitioning to a more consumption-based system.

Moreover, Friday’s CFTC data (to 18 August) reported the sharpest unwind of aggregate USD long position since December 2014. Sentiment seemed to improve for most non-US currencies, with EUR and JPY shorts seeing the biggest unwinds and GBP net positions moving close to neutral.  And market movements last week suggest that those trends extended.

Indeed, Friday’s rise of EURUSD above 1.13 may cause further unwinds of EUR shorts as important technical resistance levels seem to be being approached.  And the longer-term hope of EUR bears of strong portfolio outflows, driven by negative EA rates, has yet to show up decisively: Wednesday’s data reported only a small uptick and the data are only to June. That said, a more substantial EUR rise could eventually prompt an ECB reaction (see here) although the trigger levels are likely higher (closer to 1.20?) and in any case the next ECB meeting isn’t until 3 September (when forecast downgrades and language tweaks are possible, notwithstanding Friday’s small EA PMI upside surprise which seems to have been a secondary factor in the EUR bounce after risk aversion). And option market derived implied correlations continue to suggest that over the next few months the USD will be a less important “independent” factor in the FX market than the EUR.


So within G10 dollar bulls will likely have to rely on USD strength versus commodity currencies (CAD, AUD, NZD, NOK) in the near term. Indeed, unlike EURUSD, the USD TWI has continued to be supported by declining non-US rate expectations (see chart). And that driver has been a more consistent factor in USD’s rise than the more erratic fluctuations in US rate expectations. So any further bad non-US news (i.e. further commodity price falls) would likely support this portion of the USD TWI (especially the Fed’s “broad” measure). With commodity prices down so sharply recently there’s no guarantee of further falls, but my prior that Chinese activity will continue disappointing near term (likely prompting further PBOC action which could eventually stabilise expectations) combined with no real signs of oil supply being reined in suggest that the risks are to the downside. But it’s interesting that option-implied upside USD risks (i.e. risk reversals) have fallen across a range of USD bilaterals, even including the commodity-exposed CAD.


In a similar (more amplified) vein continued USD strength against vulnerable EM currencies seems likely. Much of the uncertainty has been driven by negative Chinese news, which AxJ and Latam EMs are disproportionately exposed to (see here). And it’s difficult to envisage substantial positive news coming out of China near term (following Friday’s very weak Chinese manufacturing PMI). Here countries with strong links with China and either significant funding needs (current account deficits) or low inflation or seem particularly vulnerable – which the chart below indicates encompasses a wide range of currencies – as well as those with political difficulties. While a delay in Fed hiking may eventually buy them some breathing room (funding costs don’t rise as sharply) that hasn’t helped them in recent market dynamics.


Dovish FOMC minutes shift focus back onto inflation, with China concerns

Wednesday’s FOMC minutes were more dovish than expected, given the relatively minor tweaks in the 29 July FOMC statement. Indeed the addition of the qualifier “some” added to the requirement of “further labour market improvement” was taken to marginally lower the bar for a September liftoff. But the minutes revealed an apparently more divided FOMC with concerns encompassing low inflation referencing USD strength a number of times and worrying about Chinese developments.

On the hawkish side “Most judged that the conditions for policy firming had not yet been achieved, but they noted that conditions were approaching that point” and “Many participants indicated that their outlook for sustained economic growth and further improvement in labor markets was key in supporting their expectation that inflation would move up to the Committee’s 2 percent objective”.

But on the dovish side “almost all members indicat[ed] that they would need to see more evidence that economic growth was sufficiently strong and labor market conditions had firmed enough for them to feel reasonably confident that inflation would return to the Committee’s longer-run objective over the medium term” as some participants thought that “incoming information had not yet provided grounds for reasonable confidence that inflation would move back to 2 percent” and several participants “cited evidence that the response of inflation to the elimination of resource slack might be attenuated and expressed concern about risks of further downward pressure on inflation from international developments”.

The Chart below illustrates those concerns – the US Phillips Curve has been pretty flat in recent years, suggesting constraints on wage growth and inflation ticking up sharply in the coming months (see here for similar Eurozone analysis). Obviously tipping point effects cannot be completely ruled out, but the chart gives some credence to FOMC members’ concerns that the natural rate of unemployment may be lower than they’ve been assuming. And the weak US inflation dynamic theme was evident in Wednesday’s disappointing CPI data – which fits in with a range of indicators hardly screaming that FOMC can be “reasonably confident” inflation is heading back to target.


Returning to China, persistent PPI deflation there (-5.4% yoy latest and falling steadily) will concern FOMC given that Chinese imports account for around 20% of US goods imports (see here). That said, imports are less important for the US than most the rest of the G10 (see chart): the US economy remains relatively closed, albeit less so than in previous decades. By contrast, only 7½% of US exports go to China. The FOMC minutes nevertheless included concerns about USD strength, supported by widening rate differentials, putting downward pressure on inflation.  And a recent New York Fed  study reported fairly significant impacts of a USD rise on inflation and GDP, notwithstanding the US economy being relatively closed, again suggesting a cautious FOMC approach.


But markets seem to have scaled back Fed hike expectations in large part because of Chinese developments’ adverse impact on global financial markets, including on the S&P 500. And it would not be surprising if the Fed worried that hiking into such febrile financial market conditions would produce exaggerated equity price falls.  That could amongst other things undermine consumer and business confidence and hence stall the nascent recovery – with consequent risk of having to subsequently reverse the rate rise which would be a disaster for credibility.

Could Fed still raise in September? Definitely arguments but not that convincing if current fragile markets persist

Against that FOMC might not want, from a longer-term perspective, “Yellen put” to enter the financial vocabulary. And as I previously discussed the issue with waiting past September is that market conditions in December tend to be illiquid. So FOMC could conceivably reason that they’ll end up introducing volatility whether they move in September (assuming China etc worries persist) or in Decembe and conclude that “they might as well get on with it”. Personally, I don’t reall buy that argument: I think that they would be cautious in such circumstances.  But I wouldn’t rule out the 28th October meeting becoming active since it would give FOMC the extra data they’re looking for without the downside illiquid markets (organising a press conference seems trivial).

Overall a number of FOMC members seem to want to get away from emergency monetary policy settings, given that the US macroeconomy seems to be in OK shape (if not generating much inflation). Indeed, we’ve had some OK US activity data since FOMC last met: in-line labor market report, strong non-manufacturing ISM report (although the manufacturing index was weaker), generally strong housing market data and another solid retail sales print. And FOMC may well take comfort in Wednesday’s below-consensus +0.1% mom CPI release apparently being driven by a sharp 5.6% airfares, with core CPI printing in line with expectations (1.8% yoy). FOMC also place less weight on CPI than the core PCE deflator – so next week’s release of that will likely be more decisive. I’m not holding my breath for anything game-changing given the limited inflation apparent in the broad range of inflation/costs measures (see above).

But Fed staff also seem to be attributing the substantial fall in US inflation breakevens, which continue to be surprisingly sensitive to the fall in oil prices (see Chart and here), to declining inflation risk premia rather than underlying inflation expectations (see Chart taken from here). The kicker, however, is that models typically use household inflation expectations to tie down underlying inflation expectations and so they might not be telling FOMC that much new.


Next week’s US GDP release will likely be higher than the advance estimate (+2.3%), but traders may well treat it as ancient history and they took a glass half full approach to start (see here). So looking out for potential Fed policy utterances at next week’s Jackson Hole symposium could well be more interesting (particularly any hints on the weight the Fed is placing on ineternational and financial market volatility relative to solid US activity data) – Vice Chair Fischer’s contribution on Saturday will probably be the highlight. But St Louis Fed president Bullard (non-voter) argued last week that the Fed does not react to financial markets directly and he would ‘look through the decline in oil’ as well as being more sanguine than financial markets about the outlook for the global economy. Atlanta Fed President Lockhart (voter) is also up next week and he previously argued that there is a high bar for not hiking in September . NY Fed President Dudley (voter) is also speaking this week and Richmond Fed President Lacker (voter) is also due to speak about “The Case Against Further Delay” on 4 September.  Moreover, the 4th September payrolls data, particularly an earnings spike, could yet shift the debate back to a potential September rate hike if market conditions have stabilisd by then. But that seems like a big “if” at the moment and in any case FOMC already largely seem content about employment developments.

Of course, overall FOMC need to be forward-looking given the delayed effects of policy changes (hence why they want to be “reasonably confident” that inflation will head higher), so the recent below-target inflation data may be a bit of a red herring.  So the key issue is whether the apparent flatness of the US Phillips curve documented above holds in the more sophisticated models run by Fed staff.

Would an early Fed hike support USD anyway?

But there is also the meta question of whether a September Fed rate hike would in any case support USD.  Given current market pricing, a September rate hike would represent a major surprise.  And you’d normally think that the consequent wider interest rate differentials would support USD (continuing the story up until recently).  But an initial look at the history of Fed rate hike cycles suggests the contrary. I’m in the process of delving into this history more deeply (watch this space for more!). But one initial point is that the current significant role of risk aversion in supporting EUR was not apparent in previous Fed hike periods (EA interest rates weren’t previously negative), although the impact of risk aversion on JPY and CHF is longer standing. And an early Fed rate would have difficulty not exacerbating the risk aversion dynamic (its why FOMC may well delay), given everything else that is going on, thereby supporting those currencies. So perhaps USD is more likely to rise if the Fed doesn’t hike (and risk aversion dies down and the non-US world continues to underperform the US driving down non-US rate expectations)?  Certainly something to think about!


5 thoughts on “The Week A September Fed Hike Died? USD Bulls on the Run

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s