Could lightning strike twice? FX impacts of US election cliffhanger: Trump win underpriced

This report analyses the likely FX market implications of the too close to call US election. Overall it seems that markets could, like the Brexit vote, be under-pricing lightening striking. But the dollar could rise irrespective of the result, although more immediately/obviously following a Clinton win, with vulnerable risk-positive currencies likely being most sensitive to the result. The main points are:

  • Clinton’s lead in national opinion polls has recently narrowed notably, to well within the margin of error, while model-based probabilities of a Trump Presidency have risen to around 35% (they encompass Clinton’s advantage of ‘safe’electoral college votes) broadly similar to the recent uptick in betting market odds of a Trump win.
  • But those metrics may not capture ‘shy’ Trump supporters, Trump’s potential ability to mobilise previously disengaged voters and Clinton’s potential difficulty mobilising previous (disenchanted) Obama supporters (although Clinton’s grass roots network is better). So the actual probability of Trump victory seems likely to be higher than the models imply and overall too close to call. I stick with my June argument that, in the febrile anti-globalisation environment, there could be Brexit echoes.
  • Given this, the FX market reaction to the increased possibility of a Trump win looks surprisingly small: (i) the fresh rises in USDMXN (and USDCAD) have been contained; (ii) the dollar index (DXY) is down only 2%; (iii) safe haven FX effects have been limited; (iv) EM FX losses have been small; (v) the sharpest moves have been in near-term FX implied volatilities and risk reversals, but they’ve been concentrated in USDMXN, while the VIX has also risen only mildly.
  • Such limited impacts in part reflect the ambiguous USD impacts of a Trump win, although I suspect that it would initially fall mildly before potentially rebounding relatively quickly. Trump’s protectionism has been equated to a preference for a weaker USD, while delayed Fed rate hikes/Yellen being replaced and foreign investor concerns about US institutions also argue for USD falls. But USD traditionally benefits in risk-off conditions plus Trump’s pledge on tax breaks for US companies repatriating foreign earnings ($2.3tn) and fiscal expansion plans ($600bn infrastructure) could be significantly USD-positive once knee-jerk reactions dissipate.
  • The economic/political uncertainties generated by a Trump victory would likely see a significant spike in risk aversion and FX volatility, although Trump has an incentive to try and calm nerves by clarifying . Overall, safe havens currencies (JPY, CHF) would benefit while risk-positive currencies (e.g. AUD, NZD in G10) would be hit.
  • Sigificant USDMXN upside is the most obvious consequence (likely exceeding the September high of 19.88), alongside likely USDCAD rises. But more broadly recently highly-successful carry trade strategies and EM currencies would be vulnerable (see Chart). Asian currencies reliant on exports (e.g. KRW, TWD) would likely be particularly in focus, as would EM’s with macro-political vulnerabilities (e.g. TRY). And USDCNY’s recent rise would likely extend given Trump’s 45% import tariffs and currency manipulator threats.
  • The adverse risk-off impacts on EM currencies of a Trump win could, however, be ameliorated by recent positive macro EM developments (e.g. shrinking current account deficits, inflation under control, improving terms of trade) and potentially delayed Fed rate hikes. But EM FX could recall the adverse Taper Tantrum experience if the recent Treasury yield uptick extends.
  • A Clinton victory would likely have more limited FX impacts. Recent moves would likely reverse as risk appetite recovers, with USD appreciating as a barrier to a potential December Fed rate hike is cleared. But FX carry trades and EM FX would likely continue to be supported given the likely slow pace of Fed rate hikes.

slide11slide6

Opinion polls have narrowed, betting markets have repriced for a significant chance of a Trump win

As is well-known, the national presidential election polls have narrowed considerably in recent days, in part due to the revelation of the FBI re-opening their investiation into Clinton’s emails. The latest Real Clear Politics average puts Clinton’s national poll lead at only 1.5pp (Clinton 46.5,% Trump 45.0%), down from nearly 7pp in mid-October (see Chart).

The US electoral college system means, however, that such national polls overstate the likelihood of an eventual Trump victory.  Around two-thirds of the 56 electoral college races are uncompetitive and such ‘safe’ states give Clinton a starting position of 160-180 electoral college votes versus only 120-145 for Trump (out of 270 required to win).  So Trump faces an uphill battle and the election result will be determined in the key swing states such as Florida (29 electoral votes), Ohio (18), North Carolina (15), Arizona (11), Georgia (16), Nevada (6) and Virginia (13).

slide1slide2

But the latest polls have turned more Trump-positive in some of those states: Trump looks to be leading in Florida (52% chance of winning according to FiveThirtyEight), Nevada (51%), North Carolina (52%) and Ohio (67%).  So the FiveThirtyEight model-based estimate of a Trump victory has risen to around 35%, up sharply from only 12% a few weeks ago (see Chart above).

Betting markets have also repriced for a lower probability of a Clinton win. PredictIt is now pricing a 73% probability of Clinton winning, down from a peak of 84% on 19 October (although this includes a recent minor bounce, see Chart below).

Opinion polls and betting markets have frailties: have markets learnt Brexit lessons?

Prior to the UK EU Referendum vote I argued that FX traders were myopically fixated on betting markets and hadn’t fully priced the impact of populist anti-globalisation and anti-establishment developments (see here). And immediately after the Brexit vote I argued that such developments meant that a Trump victory could well be more than a black swan tail scenario.  The recent narrowing in opinion polls and betting market odds reinforces those concerns.

Moreover, opinion polls and betting markets have several potential specific frailties in the current situation: (i) opinion polls may not capture ‘shy’ Trump supporters – rather like the ‘shy Tories’ in the 2015 UK general election; (ii) Trump may be able to mobilise a large number of previously disengaged voters – there’s looks to be a considerable well-spring of white collar workers angry at the perceived failure of the’political class’ and globalisation. (iii) Clinton may have difficulty in mobilising previous Obama supporters who will likely be disenchanted by the results of the past 8 years and uninspired by more of the same. That said, Clinton’s grass roots network is reportedly better organised than Trump’s. (iv) working class Trump voters may be less likely to participate in betting markets. So, as in the UK EU referendum, financial market participants risk just looking in the mirror if they myopically trade based upon them, rather than their providing accurate information on the likelihood of the Trump black swan event.

That said, the UK EU Referendum opinion polls ended up being fairly accurate – it was the betting markets which proved spectacularly inaccurate – and US election opinion polls have more established track records. Nevertheless, my take is that the above factors mean that the actual probability of Trump victory seems likely to be higher than the models imply.  It’s probably still below 50% (Clinton remains the favourite), but close enough to make the election too close to call and to consequently be worrying financial markets.

Muted FX market reactions so far: small safe haven impacts, even Vols/risk reversals moves narrowly focused on USDMXN

But given how close the election result appears, the associated FX market moves in recent days have been surprisingly relatively muted: the market looks to be under-pricing the black swan event of a Trump victory.

First, the dollar index (DXY) has fallen by only 1.7% over the past fortnight (see Chart), with broad-based losses. That fall seems to reflect Trump’s protectionist trade policies being equated to a preference for a weaker dollar, plus the risk of a Fed rate hike being delayed (see Chart) and/or Yellen potentially being replaced. Indeed, DXY seems to have become less sensitive to the Trump risk: past relationships with Clinton win betting market odds imply that DXY should be closer to 96 than 97.  By contrast, my previous arguments suggest they should have become more attuned to this risk.

slide3slide4

Second, while USDMXN’s fresh rises on Trump’s surge have generated media headlines the reality is that it has risen by a relatively contained 2% over the past fortnight (see Chart above). Indeed USDMXN has retraced a little in recent days as the Trump win betting market odds have fallen (see Chart above).

Third, safe haven FX effects have been limited.  The Chart below shows that falls in USDJPY and USDCHF over the past week/fortnight, at a maximum of 2.5%, haven’t been significantly larger than those of risk-positive currencies such as NZD (GBPUSD’s rise reflects the twin-whammy of the the High Court Article 50 ruling and MPC turning more neutral, as I expected).

slide5slide8

Fourth, EM FX losses have been small, and reasonably closely aligned to equity price falls (proxying global risk appetite, see Chart above). Specifically, the Bloomberg EM FX index is down only 0.2% over the past fortnight, and indeed is only 2.4% off it’s mid-August peak.  The chart below indicates that resilience is partly down to USD actually falling against several Eastern European EM currencies (CZK, HUF, PLN) which are little impacted by Trump’s policies.  But it also shows that USDMXN’s recent Trump-provoked rise hasn’t been the largest EM FX move (although USDBRL’s recent rise followed it experiencing the sharpest fall in 2016).

slide9slide10

Fifth, while there’s been larger moves in near-term FX implied vols and risk reversals, these have again been localised myopic.  That’s surprising given the potential global economic/political impacts of a Trump victory, which is starting to look less like a black swan tail scenario (see above). Specifically:
(i) USDMXN 1-week implied volatility has more than doubled in recent days, to a level more over three times it’s previous 2016 levels (see Chart above) as investors worry about Trump’s NAFTA, Mexican wall and immigration policies.
(ii) The story’s similar for USDXN 1-week risk reversals, with call prices rising significantly relative to put prices.
(iii) While many other 1-week implied volatilities have risen in recent days, they’re not particularly elevated relative to their previous 2016 levels (see Chart below). That’s encompasses USDCAD (also potentially impacted by NAFTA’s scrapping), export-orientated EM currencies such as KRW and the ‘safe haven’ currencies (JPY, CHF).
(iv) While 1-month and 3-month G10-USD implied volatilities have also risen in recent days, the increases are nothing out of the ordinary with volatility remaining well below it’s Brexit-driven June 2016 peak and China-driven August 2015/early 2016 peaks.
(v) The rise in equity market uncertainty (VIX) is a little more pronounced, but again remains below previous peaks.

slide11slide12

Trump’s policies would generate significantly more uncertainty/volatility than Clinton’s

Unhelpfully, Clinton’s and Trump’s main policies haven’t featured heavily in recent campaigning.  But a Trump win seems highly likely to involve potentially-dramatic changes and generate a period of uncertainty, investor risk aversion and financial market volatility. Importantly, Trump would likely face less barriers to enacting his (somewhat vague) proposals than Obama has faced or Clinton would likely face. Republicans seem highly likely maintain control the House of Representatives and FiveThirtyEight’s models indicate they’re 50:50 on also maintaining control of the Senate. By contrast, Clinton’s policies represent more continuations of Obama’s (which is part of her potential problem inspiring voters) albeit with a more left-leaning bias generated by Bernie Sanders’ impact, but could be stymied by the House/Senate.

Trump’s policies include: (i) Scrapping NAFTA and a general move towards protectionism, including China potentially being labelled a currency manipulator; (ii) Reduced immigration and forcing Mexico to build a wall; (iii) Reduced foreign military engagement, including less support for strategic partners and demanding greater contributions; (iv) A fiscal expansion – significant cuts in corporate tax rates allied to infrastructure spending – which would see US Treasury issuance will rise significantly; (v) Tax breaks on US companies repatriating foreign income, similar to the 2004 Homeland Investment Act; (vi) Criticism of the Fed, including potentially removing Janet Yellen as Chair; (vii) Scrapping Obamacare.

Severe initial risk off dynamics with a Trump win, but dollar may eventually rebound irrespective of the result 

Such sharp policy switches after a Trump win, together with the considerable uncertainty about many of the policy details (especially global geopolitical implications) would likely significantly raise market risk aversion. Volatility would consequently likely spike higher, supporting ‘safe haven’ currencies like JPY and CHF while risk-positive currencies (and equities) would be pressurised.

Consequently, there would likely be a sharp unwind of the recently-popular FX carry trades. Such trades have have been the most successful thematic approach in the low-volatility world of recent months – rising by around 8% since start-2016 according to the Goldman Sachs index (see Chart). SEK status as the weakest G10 currency over the past 3 months has been driven by it’s popularity as a ‘funding’ currency while the strength of NZD and AUD reflect their attractive returns in the low-yield world.

slide6

But the spike in investor risk aversion and market volatility means that EM FX would also likely come into stronger focus in a Trump win scenario, despite it’s recent resilience.  Indeed, it’s notable that EM FX trades have been the other major profitable FX trading theme of 2016 (see Chart above), obviously helped by the attractive carry on offer. But the greater risks of those trades   But I discuss below how several positive EM macro developments may ameliorate the EM FX sell off in a Trump win scenario.

A Trump win would, however, have ambiguous overall effects on the USD, which helps explain the limited impacts of Trump’s opinion poll/betting market surge I noted above. My suspicion is that a Trump win would initially be mildly USD-negative, as recent FX market moves extend, but that initial knee-jerk fall could potentially be reversed relatively quickly as Trump clarifies policies and investors focus on potentially more supportive factors.  

On the one hand, several factors indicate that USD’s recent weakness could extend should Trump win:
(i) Trump’s protectionist trade policies being equated to a perceived preference for USD weakness (even though he’s critical of alleged Chinese currency manipulation); (ii) The economic uncertainty, and consequent possible macro slowdown, leading the Fed to postpone it’s much-trailed December rate hike (increased expectations of a fed rate hike had been supporting USD before Trump’s poll/betting market surge, see Chart below);
(iii) Trump replacing Fed Chair Yellen could damage the credibility of US policymaking, potentially causing foreign investors to become more reluctant to hold US assets (demand a bigger risk premium);
(iv) Relatedly,  in the extreme Chinese authorities could potentially retaliate to Trump’s currency manipulation accusations by diversifying away from US Treasuries.

slide6

But there are also several reasons why USD could (eventually) rebound on a Trump win:
(i) USD traditionally benefits in risk-off conditions, even when their ultimate source is the US e.g. after Lehman Brother;
(ii) Trump’s fiscal expansion plans will support US demand, offsetting the potential private sector slowdown generated by increased uncertainty;
(iii) Trump’s pledge on tax breaks for US companies repatriating foreign-held cash could be lead to large inflows back into the US, as was the case after the 2004 Homeland Investment Act, representing a potentially significantly USD-positive once knee-jerk reactions die down and Trump clarifies policies.

By contrast, a Clinton victory would likely have more limited FX impacts. Recent moves would likely reverse as risk appetite is boosted, with USD appreciating as a barrier to a potential December Fed rate hike is cleared.  But carry trades and EM FX would likely receive fresh impetus, subject to country-specific factors, given the well-signalled slow pace of future Fed rate hikes and low end-point.

Macro Improvements may ameliorate the EM FX sell off on a Trump Win

I noted above how EM FX has thus far been relatively resilient to Trump’s recent opinion poll/betting market surge (obviously except MXN), with indeed EM FX thematic trades reporting the second-strongest performance in 2016. Indeed, there’s been some fairly stellar EM currency performances over the past 6 months, with ZAR up 15.1% and BRL rising (see Chart) reflecting the combination of attractive carry, positive political developments and beneficial terms of trade developments. But TRY’s 6% fall over the past 6 months illustrates how adverse political-economic developments can still even in the recent risk-friendly environment.

slide14slide15

And I argued above that EM FX seems likely to come into more significant focus in a Trump win scenario, in part as recently-popular FX carry trades are liquidated in a more risk averse/higher volatility scenario.  And a major concern is that there could be a repeat of the 10% EM FX sell-off in the wake of the May 2013 Taper Tantrum (see Chart). But several factors seem likely to ameliorate the damage to EM FX in a Trump win scenario.

The corollary of a Trump win may well be delayed Fed rate hikes, as elevated economic/political uncertainty adversely impacts US capex, hiring and consumption. So, unlike in the Taper Tantrum, US Treasury yields don’t look set to rise sharply after a Trump win (unless investors take fright at the prospect of higher Treasury issuance or foreign investors start doubting the US policy framework). So greater EM funding pressures from higher global yields seems like a less potent dynamic this time around.

EM macro conditions have improved in several important respects.  Such reduced EM macro fragilities/stress points may give FX investors greater confidence to maintain EM exposures. After all, the low G10 yield environment seems likely to be prolonged by a Trump victory. Obviously, country-specific factors, including political weakness (recently adversely impacting KRW and TRY), but it’s worthwhile highlighting several recent big picture positive dynamics (abstracting from the country by country minutiae).

First, most EM countries’ current account positions have improved over the past year (see Chart), thereby reducing their external funding needs and reassuring international investors. While some significant deficits persist, (5.7% of GDP in Colombia, 4.1% in Turkey, 3.1% in South Africa) it’s at least hopeful that the first two have managed improvements.  Of course, details of EM countries’ external funding needs (e.g. shares of foreign currency and short-term debt) matter to FX market participants and hence merit further investigation.

slide17slide16

Second, most EM countries have made progress in getting inflation under control (see Chart above), again helping buttress the credibility of their policy frameworks. The relatively hawkish turn by the Central Bank of Russia has been particularly noted by FX market participants.  But a number of other EM Central Banks have overseen similarly-significant falls in inflation. And the EM countries with rising inflation are generally those with low-inflation (i.e. it’s going in the right direction). The major exception is Turkey, with TRY consequently being the main EM under-performer (see Chart above) and to a lesser extent the Philippines (PHP has also struggled).

Third, most EM countries have experienced improvements in their terms of trade in recent months (see Chart below), driven by the recent commodity price rises. Conversely, Of course, commodity-importing EM’s (China, India, South Korea, Thailand) have suffered worsening terms of trade – but they appear modest relative to the gains.

slide20

That said, there are also some less positive EM macro developments. First, EM growth generally seems to be slowing, most broadly in Latin America. The WTO’s recent cut to it’s forecasts for world trade growth over the next couple of years (see here) isn’t helpful to export-orientated countries like Korea. Trump’s protectionist policies would likely exacerbate such weakness, increasing the vulnerability of Asian currencies such as KRW (also not helped by the recent political scandal).

slide18slide19

Second, a number of EM countries fiscal deficits have widened in recent months (see Chart above). Brazil is the biggest EM fiscal sinner by some way (9.6% of GDP deficit, up 0.6pp of GDP over the past year) but the new government has gained investor confidence that they have a plan to address the issue.  That, combined with a terms of trade improvement, has seen BRL be one of the star FX performers over the past 6 months (see Char above). Other high-deficit countries such as India, Russia and Turkey need to make sure that they retain investor confidence.  But it’s notable that Mexico’s fiscal position has improved by 2.2pp of GDP over the past year.

Advertisements

3 thoughts on “Could lightning strike twice? FX impacts of US election cliffhanger: Trump win underpriced

Leave a Reply

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

WordPress.com Logo

You are commenting using your WordPress.com 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