With everyone on tenterhooks about whether PM Cameron will manage to secure an acceptable deal at the EU summit, paving the way for a likely June UK EU referendum, I detail the impacts on markets thus far and the likely future propects. This follows my previous post where I correctly argued that growing Brexit uncertainties had increased the vulnerability of sterling and that volatility was likely to pick up (see also here for my take on general market volatility). In summary:
• Brexit impacts have been most apparent in GBP implied volatilities, with 6m vols rising to post-2012 highs.
• The Scottish referendum experience illustrates that GBP vols can rise 40-100% above G10 FX vols, so further GBP vol rises seem likely.
• Risk reversals have also moved to suggest substantially greater downside GBP risks, particularly against EUR and JPY where they are their most negative since 2009 and end-2011 respectively.
• Risk premia seem to have driven a large portion of GBP’s recent decline (TWI down over 4% since end-2015), with the largest falls occurring versus the Yen and EUR despite both the BoJ and ECB being in easing mode.
• The significant repricing of BoE rate hikes – now not expected until until early 2019 (!) – has also contributed to recent GBP weakness.
• The potential near-term GBP bounce on a successful EU deal (some resolution of uncertainty) seems smaller than the likely accelerated GBP weakness on an acrimonious end to the summit (unless the issue is kicked into the long grass).
• GBP seems likely to weaken further over up until June as investors have to price a Brexit scenario which would likely see EURGBP rise to 0.90-0.95. Given likely scepticism about opinion polls, EURGBP above 0.80 and a bigger GBPJPY fall, seem possible (with larger moves on heightened risk aversion).
• Brexit concerns will make it harder for GBP to be supported by strong macro data until June, especially given distant pricing of MPC rate hikes.
• A reasonably-substantial GBP bounceback would be expected on an eventual “remain” vote (EURGBP closer to 0.70) – both via uncertainty resolution and (relatedly) potential BOE repricing.
• Brexit concerns haven’t yet impacted Gilt yields, which have fallen at least as much as other government bond yields, and foreign gilt purchases have remained strong. But gilt term premia may well also rise in the coming months (10 year yields rose 25bp around the 2014 Scottish referendum), albeit less dramatically than FX.
• MPC are in no hurry to get started with rate hikes, with Carney’s conditions seeming some way off in an environment of flat core inflation, surprisingly weak wage growth and worries about second round effects (shared in other countries). So MPC likely won’t object to GBP’s recent decline and further limited moves.
• Near-term markets likely won’t pay much attention to any BoE attempts to get market pricing brought forward (the May Inflation Report is going to be interesting!).
Implied volatilities and risk reversals significantly impacted
The largest impact of the Brexit uncertainties has been in implied volatilities and risk reversals (i.e. growing market positiong for further GBP downside). Specifically, the chart below illustrates that 6-month implied volatilties of GBP against EUR, USD and JPY have risen to their highest levels since 2011. While the 6-month contracts cover the likely June EU referendum, the second chart below shows that both shorter and longer maturity implied volatilties are also at or close to their post-2012 highs. And it’s notable that the current levels exceed those both around the Scottish independence referedum and around the 2015 general election: markets are really pricing this once in a generation risk event as something special.
Of course, the generally stressed 2016 market conditions (with worries about oil price falls, Chinese prospects and policy, EA and a potential Fed policy mistake) have undoubtedly exacerbated the impact. Talk about choosing a bad time to have such a momentus decision on the table. Specifically, G10 FX implied volatility has ticked up sharply in 2016 (to its highest since around the January 2015 ECB easing), alongside higher equity market volatility (VIX) as equities have plummeted. So the ratios of GBP implied volatility to G10 FX volatilities are less unusual: GBP implied vols are “only” 5-15% higher than the G10 average (versus the EUR).
But it’s also notable that whereas both G10 FX volatility and the VIX have fallen back in recent days (helped by abating worries about banks, further ECB hints at easing in March and a mini oil price bounce) GBP implied volatilities have continued risng. That reinforces that Brexit fears make the UK situation special, in a bad way, at present.
And there’s another important kicker here. The Scottish referendum experience indicates that significantly-higher GBP/G10 average implied volatility ratios are likely in the months ahead as the referendum date approaches. Specifically, shorter-maturity GBP implied vols were 40-100% above the G10 average closer to the Scottish referedum approched. So if markets follow their typical tendency to get much more nervous as the event risk nears the recent volatility uptick seems likely to be a mere appetiser for a meatier GBP volatity spike.
Indeed, the generally-adverse global background means that G10 FX volatlity isn’t that likely to fall back substantially up to June: so a GBP-multiple of a large number is an even larger number. Fresh dovisness from the Fed, ECB and BoJ would help soothe nerves, although the market is increasingly questioning the efficacy of negative policy rates. But the event risk worries about a China devaluation, which would cause all hell to break lose, will be hard to shift given strong capital outflows and overcapacity/deflation problems. And what’s fundamentally needed is better macro data and more signs of inflation, which seems more likely H2 phenomena (at best).
The other major development is that the market has increasingly-priced risks of further sterling weakness, as put prices have risen relative to calls. Specifically, in line with my previous post, 6-month EURGBP risk reversals have moved sharply from pricing GBP upside in December to now pricing the greatest GBP downside since January 2009 (when GBP was depreciating sharply in the wake of the financial crisis, see my BoE Article). Slighly less dramatically, GBPJPY 6m risk reversals have fallen to their most negative since end-2011. The existing negative 6m GBPUSD risk reversals have also widened, although they remain narrower than around the general election (probably helped by USD’s general weakness on growing doubts about Fed policy). And given how unprecedented the EU referendum is these market prices of GBP downside risks seem likely to widen further over the coming months, especially is market risk aversion remains high because of global factors.
Continued GBP weakness, but not yet accelerating
By contrast, spot GBP has been suprisingly stable this week, with EURGBP generally trading in the 0.77-0.78 and the TWI actually up marginally. But GBP is substantially weaker since end-2015, with the TWI down over 4% (or a total of over 7% over the past 3 months). And it’s telling that the GBPJPY fall has been particularly sharp – down nearly 8% since end-2015 and 12% over the past 3 months – attesting to the impact of risk aversion effects. But EURGBP is also up over 5% since end-2015 despite the ECB ratcheting up the dovish rhetoric again (see here for my take on why it’s urgent that they act in March together with the constraints). The GBPUSD fall has been smaller given that Fed rate hike expectations have been pushed out alongside MPC expectations (see below). And those reduced expectations of Fed rate hikes explain why GBP has also generally fallen against EM FX despite heightened market risk aversion.
Overall, however, there’s little sign yet of GBP spot weakness accelerating, probably because traders are awiting the negotiations outcome. But it could also signal that traders are presuming that a deal will be forthcoming – given that it seems to be in the main players’ interest to achieve one (most EU members value continued UK membership, although smaller player who have more to lose in the near-term can still veto the deal).
And that would imply only limited near-term GBP upsides on a successful deal (some resolution of uncertainty), but larger GBP downsides should the EU summit ends acrimoniously (at the least further uncertainty as a deal is delayed, at the worst the prospect of Cameron campaigning for Brexit).
MPC rate hike expectations contribute to GBP weakness, but largely due to risk premia
The other, related, major development in 2016 has been a substantial pushing back of BoE rate hike expectations. Indeed, market pricing indicates that first MPC rate hike in now expected in early 2019 (!), not much sooner than the first ECB hike. I explained in my BoE Super Thursday preview that markets would pay more attention to the 9-0 MPC vote (which only 3 city economists agreed with me in anticipating), wage-price dynamics and Brexit uncertainties than MPC again forecasting above-target inflation at the end of their projections. Even though Carney et al have been clear that the situation doesn’t call for negative rates (they still expect the next move to be up), BoJ and Riksbank rate cuts have encouraged such expectations even if paradoxically markets have written off such measures as ineffective (but EUR has also come back down as the ECB once again stokes further easing expectations).
And this repricing of BoE rate hike expectations has contributed a little to GBP’s recent weakness, partly because other Central Bank rate expectations have also been pushed back (albeit less). The chart below illustrates that revisions to policy rate expectations account for under half of the EURGBP rise and even less of the GBPJPY fall. That relative ranking again indicates that risk premia considerations have been the larger driver of GBP’s 2016 fall.
MPC rate expectations are stretched but seem unlikely to reprice near-term
Yesterday, BoE’s John Cunliffe tried to push back against the recent substantial shift in market MPC expectations “I can’t see anything in the economic news over the last three-to-four weeks that would lead to a shift like that”. But such messages will have difficulty gaining traction while Brexit uncertainties persist (i.e. at least until mid-year), core inflation doesn’t “move notably towards the target” (one of Carney’s conditions) and wage growth remains surprisingly weak (failing another Carney’s conditions).
Indeed, while Carney and Broadbent slightly downplayed the potential for second round effects of low inflation back to wage bargaining Gareth Ramsay (Director of Monetary Analysis) recently described evidence of this from the BoE’s Regional Agents as “a real amber warning light for us, a really important message“. This is particularly important given that Q1 is the key pay bargaining period. And evidence from XpertHR is that pay settlement remained flat at 2.0% in February (see here) although the Q4 BCC survey reported a (suspiciously) strong number.
A further potential worry for MPC is that UK 5y5y inflation breakevens have continued falling, albeit with a small bounceback in recent days as oil prices have recovered. And even though they are the weakest since March 2015, that’s less dramatic than the record lows recorded by EA breakevens (which strongly suggest the need for decisive ECB action). Of course, MPC might well suspect that risk premia rather than underlying inflation expectations have driven such moves, and have models to parse them out. But US policymakers seem to be starting to doubt such models: St Louis Fed President Bullard this week mentioned the fall in US breakevens as a motivation for delaying rate hikes.
And a further likely dovish innovation is that Martin Weale’s MPC term ends on 31 July: it’s hard to think of anyone as relatively hawkish as him for the Chancellor to appoint.
Further GBP weakness likely: depends on risk appetite, positioning not major constraint
Overall, given the above my bias is that GBP will weaken further in the run-up to the (hopefully) June EU referendum. Certainly a substantial GBP strengthening seems unlikley in the next few months given the constellation of factors. But by how much could GBP fall? This will depend on:
(i) the general level of market risk aversion: larger GBP falls are likely if risk aversion remains elevated.
– China devaluation/slowdown risks represent barriers to substantial drops in risk aversion even with the Fed, ECB and BoJ likely dovish
– a June Fed hike can’t be completely ruled out (not my base case) which would likely rile markets (although calmer market conditions is a bit of a precondition to a Fed hike).
(ii) the probabilities market participants place on “remain” and “leave” scenarios
– the likely close nature of current opinion polls and the general scepticism about their accuracy after the General Election debacle means that investors may default to 50:50,
– this will especially be the case if the polls remain close as the referendum (hopefully) approaches.
(iii) GBP’s likely levels in the “remain” and “leave” scenarios
– “remain” could well see EURGBP fall back to closer to 0.70, particularly if the ECB has manages to engineer a market-surprising ease in March (I’m not yet convinced even though I think they should).
– “leave” would likely see EURGBP rise to the highs of around 0.90-0.95 experienced at the peak of the financial crisis (see my Bank of England Quarterly bulleting article here) .
(iv) Positioning effects and technical resistance levels
– CFTC data illustrate that market positioning is net short GBP, but not excessively so this doesn’t seem likely to be a major barrier to further GBP weakness up until June.
Putting that all together implies that EURGBP will likely rise to 0.80 and above in the next few months. I previously put the range at 0.78-0.80, but given that we’ve already risen close to the bottom of that range and GBP implied volatilities/risk reversals have moved so sharply the risks seem skewed to further upside. Indeed, the upside risks to that view are that Cameron can’t get an acceptable deal, that substantial parts of the Conservative Party and the media get behind “leave” and that it get an unassailable lead in the opinion polls. The downside risks are that market risk aversion retraces substantially, potentially helped by some form of global agreement at the forthcoming G20 meeting (unlikely given the large number of players and their competing aims), or that opinion polls turn absolutely decisively in favour of “remain”.
Short GBPJPY positions will, however, likely be more attractive than long EURGBP positions. GPPJPY has proved more influenced by risk aversion than has EURGBP (see above) and the market seems more dubious about the FX impact of further BoJ easing than that of the ECB. Although the bar is low for both, Draghi’s testimony this week and the ECB minutes both helped EUR retrace some of it’s recent strength (see here).
The dog that hasn’t barked: Gilt yields falling in line with others, strong foreign investor demand – but term premia seem likely to rise as referendum approaches
In sharp contrast to the sigificant FX movements discussed above the gilt market has been an relatively calm. Indeed, 10 year gilt yields have fallen in line with the 2016 global bond market rally associated with the strong market risk aversion. Specifically, the 46bp fall in 10 year gilt yields since end-2015 is virtually identical the 10y US Treasuries and actually 9bp more than the decline in bund yields. So no dog barking here.
Moreover, foreign demand for gilts has remained strong, at least up until December as covered by the BoE data, reducing fears about funding the UK’s large current account deficit. And that’s a very different picture from around the September 2014 Scottish referendum, when foreigners made net sales of gilts and other UK government securities (albeit only small ones).
But it’s also notable that gilts sold off somewhat around the Scottish referendum, with 10 year yields rising around 25 bp as the bond bull market was temporarily interupted. Interestingly, bunds and treaturies also sold off in this period – suggesting that the referendum risk was having wider impacts. Givem this, combined with the EU referendum being sigbificantly more important (and likely more finely balanced) than the Scottish referendum I anticipate that Gilt term premia are likely to build as we approach the likely June referendum. So gilt curve steepeners may also be attractive. That said, the action is likely to be less pronounced than in the FX market – with GBP shorts seeming to be the instrument of choice for trading Brexit concerns. And once again overall market risk appetite will be important (although perversely recent developments indicate that agilt sell-off could be less likely in a risk off environment).