This piece considers the macro-financial implications of the Euro Area’s political risks, in greater market focus following Donald Trump’s election win (which I argued was underpriced), focussing on the 4 December Italian Consitutional Referendum. Overall, further politics-driven market volatility seems likely, even if political concerns eventually dissipate. The main points are:
- Financial markets’ previous under-pricing of 2016 political risks (see Brexit underpriced and Trump underpriced) has been replaced by high sensitivity to forthcoming European political risks – starting with the 4 December Italian referendum (and Austrian presidentail election re-run) before the 2017 Dutch, French and German general elections.
- Italian-German yield spreads have spiked to their highest since 2014 (see Chart) alongside the Trump-driven global bond sell off. And €/$ has fallen to 2016 lows below 1.06, alongside US-German yield spreads rising to multi-year highs, as the € has experienced broad-based falls since Trump’s win.
- The Italian referednum is about cutting the size/power of the upper house (Senate), to remove political gridlock, but has encountered concerns about excessive lower house excessive power (the Italicum ensures that the largest party receives 55% of lower-house seats).
- Opinion polls consequently give “no” a 5-8pp lead, although around 25% of voters remain undecided. Renzi’s likely resignation on a “no” win would potentially trigger an early general election which M5S seems likely to win (in a run-off) and hence call an Exitaly referendum which risks being backed by Eurosceptic Italians (see chart).
- The main consequence is heightened market sensitivity to €Area political news. But the risks are skewed to further periphery spread widening and € falls – alongside the political chaos after a Renzi resignation, markets will worry that French opinion could understate Le Pen’s chances in the second round (while Sarkozy’s race exit is positive, Fillon/Juppe are political insiders). So €/$ parity is the most likely for several years, although still constrained by likely slow Fed rate hikes even with Trumpflation.
- A December ECB QE extension (motivated by low inflation prospects) could counteract a periphery bond sell-offs, although the task is tougher than previously, but add to € weakness (although it’s not a done deal).
- EA political risks could eventually dissipate, leading to late-2017 € rebound. Le Pen faces an uphill task in the two-stage French electoral system (she’ll likely win the first round but, like her father in 2002, lose the run-off if socialist voters back Fillon/Juppe). AfD will also probably only win a few Bundestag seats. And several factors reduce Exitaly probabilities, including that any early election run under the old system would allow the Senate to frustrate M5S’s plans (should they win) and that the Italian constitution rules out Euro membership being decided by a referendum. But markets may not price such subtleties, instead knee-jerk selling BTPs, other periphery bonds and the € on adverse political news.
- The best case after a “no” Italian referedum vote could see much-needed Italian structural reforms scrapped and bank recapitalisation delayed, leading to continued macro underperformance and eventual further political unrest. Italian consumers are expecting rises in already-high unemployment (see Chart) and have suffered years of low wage growth. Moreover, Italian headline inflation again turned negative in October.
- Elavated political uncertainty also won’t be good for near-term €Area growth dynamics – investment and consumption could be delayed – potentially worrying given that recent steady growth has been driven by domestic demand.
- Paradoxically, a “yes” referendum vote actually increases the chances of M5S gaining greater power in 2018, although Renzi recently argued that further electoral reform is required irrespective of the referendum result.
Markets learning how to price political risks: EA Periphery Yields up, Euro down
I argued in advance of both the UK EU referendum and the US election that markets were underpricing the risks of Brexit and a Trump victory (see Brexit underpriced and Trump underpriced) as they myopically focussed on opinion polls and betting markets which failed to capture populism driven by anti-globalisation. But market participants seem to be finally learning the lesson that there are larger social-economic factors at work (although substantial questions obvioiusly remain about first whether the solutions proposed by Brexiteers and Trump will actually be implimented and second whether actually they’ll benefit their supporters if they are implemented).
So markets have become nervous about a sequence of forthcoming European political risks, consequently aggressively repricing EA assets. The 4 December Italian referendum (4 Dec, same day as re-run of Austrian election) is most immediately in view and hence the focus of this piece. But but the Dutch general election (March 2017), French election (May) and German election (September) are also wating in the wings given prominent insurgent parties with anit-Euro policies.
Specifically, Euro Area periphery yields have suffered the largest spillover effects from the US Treasury market sell-off following Trump’s election win, driven by his apparently growth/inflation-boosting fiscal plans (“Trumpflation”). Indeed, the chart below illstrates that Portuguese 10-year yields have risen by more than US treasury yields since Trump’s win. But the rise in Italian yields, though smaller, is more ntable because it’s been associated with the Italian-German (BTP-Bund) spread rising sharply to it’s highest since mid-2014 (see second chart, which also shows that the rise in the Spanish-German spread has been smaller/less unusual).
Bund yields themselves have risen only modestly – actually marginally less that JGBs despite the BoJ targetting 10-year JGB yields – so the Treasury-Bund spread has risen to multi-year highs (see Chart).
Given this combination of rising periphery yields and subdued bund yields it’s unsurprising that the € has suffered in the strong-dollar environment following Trumps’s victory (DXY has risen to the strongest since 2003; my pre-election analysis anticipated $ strength, as markets focussed on Trump’s expansionary policies). And the chart above highlights that since Trump’s victory there’s been a very strong correlation between €/$’s fall and the decline in the 10-year Bund-Treasury spread, after only a weak link in previous months. So the drivers of FX market dynamics seem to have shifted: my pre-election analysis highlighted how carry trades and EM currencies would be vulnerable to a Trump win.
But the Chart below makes the important point that EUR’s post-Trump weakness has been broad based. Specifically, EUR has only appreciated aainst a few vulnerable EM currencies (BRL, TRY, ZAR, MYR, HUF and PLN, with the latter two probably impacted by their geographical location) and JPY (reflecting the BoJ’s yield-targetting policy). And it’s notable that EURGBP most, as the post-Brexit risk premium that had been priced into UK assets (see here) became more mirrored by EA political risk premia.
That said, it’s notable that EURUSD has fallen to a 2016 low, breaking through 1.06 (closing prices). And further EURUSD falls seem likely as EA political risks rise in the coming months, the ECB likely extends QE and the Fed again starts hiking rates. Of course, FX traders have previously been disappointed several times in anticipating EURUSD parity on the policy divergence trade. And USD upside will still likely be contained by the continued slow pace of Fed rate hikes in 2017 (after the December Fed rate hike which is now 90%-plus priced) even with Trump’s more expansionary/inflationary policies (infrastructure/military spending, potential tariff impositions). But overall, given the boost to USD in risk-off conditions, the prospects of EURUSD parity (by mid-2017) look the best for several years.
The €TWI is notably more elevated than EURUSD (see Chart above), reducing the consequent boost to EA exports and inflation. Indeed, continued low inflation (with the September ECB inflation forecasts looking overly-optistic) provides a strong argument for the ECB extending QE by 6 months on 8 December. Plus the committees will report on the methods of ensuring he smooth implementation of ECB policies. That should hopefully overcome the technical issues which prevented a September QE extension. That said, December ECB QE extension isn’t a done deal (see below).
“No” seems likely to win Italian Referendum, setting fears of M5S and Exitaly
The Itailian referendum is about approving the Italian government’s constitutional reform to reduce the size/power of the upper house (Senate). It thereby aims to remove the structural cause of the political gridlock/instability which has plagued Italian politics. The lower and upper houses currently have equal power and can frustrate each other, while Italy has had more than 60 governments since WWII.
But the reform has encountered substantial concerns about giving the lower house excessive power, which could be potentially used by “fringe” parties such as Movement Five Star (M5S). Specifically, the previously-enacted Italicum law means that any party getting 40% of the vote automatically receives 55% of lower-house seats. And a second-stage electoral run-off occurs if no party gets 40% in the first stage. Combined with a weaker upper house, that would give the government unprecedented power (for Italy). Moreover, the referendum again gives disgruntled voters the chance to punish the government for Italy’s consistently-poor macro performane (see final section) or to lodge anti-establishment protest votes.
Opinion polls since September have consequently consistently indicated a “no” referendum win (see Chart): the latest polls give “no” around a 4pp lead amongst decided voters (52% v 48%). That said, around ,25% of voters remain undecided (or don’t intend voting) and of course my main thesis is to critique opinion polls. But this time the “populism effect” would seem to make a “no” vote even more likely than suggested by the polls.
Markets are then concerned about domino effects potentially eventually leading to Exitaly.
- First, Renzi seems likely to resign on a “no” referendum vote (having vascilated on this on 17 November he confirmed that he would take no part in see here), potentially triggering an early election (not officially due until 2018) or at least a temporary technocrat government (with potential uncertainties).
- Second, M5S could well beat Renzi’s PD party in such an election. While PD party is currently marginally ahead of M5S in opinion polls, separate polls suggest that M5S would win the two-horse second-round run off mandated by the Italicum law (see Chart above).
- Third, one of M5S’s main policies is to hold a referendum on Italian EUR membership (but remain in the EU).
- Fourth, opinion polls indicate that Italians are the most Eurosceptic Euro members (see charts below). On the one hand the 2015 Pew global attitudes survey found that 55% of Italians want to keep the € (40% want to leave), a small improvement from 2014. But the October 2015 EC Eurobarometer found that only 49% of Italians thought that having the EUR was good for the Italian economy while 41% though that it was Bad. So any Exitaly referendum seems likely, at best, to produce a close result – with Exiteers likely gaining courage/support from Brexit and Trump’s win (neither of which have had near-term adverse impacts).
There are some reasons to downplay the market impacts of such scenarios. After all the European economy and the EUR have, with ECB support, been relatively resilient to previous EA political uncertainties: Grexit risks, Portuguese political uncertainties, Spain managing without a government for several months. And markets stabilised relatively-quickly after both the Brexit vote and Trump’s unexpected win, after initial risk-off dynamics (although there’s clearly been a prolonged adverse impact on the £).
But uncertainty about the future membership of the third-largest EA member is arguably a more existential development. After all it calls into question the whole future of the single currency (either directly or do to after effects of Exitaly).
Several factors reduced the chances of Exitaly, but markets may not care
There are several barriers to all of the required dominos falling leading to Exitaly:
- Even if Renzi resigns after a “no” referendum vote, the remains of his government might not decide to call an early general election. An early election wouldn’t be in their interests if they don’t have time to reform the Italicum.
- Any (early) general election after a “no” referendum vote would be run under the old system, potentially allowing the (still equal power) Senate to frustrate M5S’s plans to hold a referendum on EUR membership.
- The Italian constitution rules out Euro membership being decided by a referendum. Of course, that may be a technicality – a decisive Exitaly referendum vote could then subsequently be legislated for by the (M5S-led) government.
- M5S policy of Italy leave the Euro but remaining in the EU is disallowed by European legislation. Again, this could be just a technicality – Italy could try and negotiate around the existing rules.
So markets may choose to downplay such technicalities and instead focus on their big picture existential concerns after a “no” vote (especially if Renzi resigns and an early general election is called). That would leadto a continued Italian (periphery) bond sells off and EUR weaknes, at least until the uncertainty is resolved.
Lack of structural reform and continued low growth the best case after a “no” vote
Even if a “no” referendum win doesn’t necessarily inexorably lead to Exitaly, it will likely have macro costs. In the near term the the rise in bond yields and political uncertainty will inhibit Italian capital investment and employment prospects.
Further out, the dropping/reversal of much-needed structural reforms to the Italian economy will be costly. If the PD party somehow remains in power, they will be much-diminished and will have likely lost their leading reform advocate in Renzi. In any case they would continue to face a powerful senate able to block surviving reform proposals. If M5S comes to power after an early election, even if they fail to get their proposed Exitaly referendum through a unreformed parliament their other policies are populist rather than pro-business. The consequence of abandoning structural reforms will likely be continued poor macro performance, sowing the seeds of eventual further political unrest.
December ECB QE extension could curtail bond market sell off but accelerate EUR fall
The EA’s low inflation, with core HICP not exceeding 1% for 12 momths (stuck at 0.8% for 3 months), underprins the argument for the ECB to extend their QE programme (APP) when they meet on 8 December. Indeed, the broad macro picture isn’t really any better than when I advocated a 6-month QE extension in September (see here although I also highlighted the risk of disappointment). And the latest ECB Meeting Accounts reiterated that inflation “continued to lack clear signs of a convincing upward trend”, and worryingly “wage dynamics appeared to have surprised on the downside”, plus past inflation projections had been subject to “considerable errors” and there were downside risks to growth prospects. Overall, the ECB’s September HICP forecast looks overly-optmistic (anticipating spiking to 1.3% in 2017 from October’s 0.5%).
Extending QE would now have the added benefit of providing some offset to the recent sharp rise in periphery yields. In October 2015 I argued that the ECB’s PSPP would help offset the rise in yields around Portuguese political uncertainties. The now-stronger bearish market trends after Trump’s win mean, however, that the ECB faces a tougher job trying to ensure that there’s not an “unwarranted tightening of monetary conditions”. Conversely, further ECB easing would likely add a little more momentum to the €’s recent decline.
That said, a QE extension is by no means guaranteed on 8 December, and will depend on intervening market developments, including the reaction to the Italian referendum result. Specifically, the ECB Accounts also noted that “A balanced communication was warranted….mindful not to trigger undue expectations in financial markets about future monetary policy action.” And they revealed concerns about side effects of low rates/QE on “the longer-term intermediation capacity of banks and other financial institutions had to be further examined”. Plus last week Weidmann argued that there was room to disagree over the degree of accommodation, although he did recognise that “Pronounced political uncertainty” is currently weighing on growth prospects. And Yves Mersch argued that stimulus should end as soon as possible, was critical of a “permanent commitment” to QE and that the updated staff forecasts could show “that we expect an inflation rate at the 2019 horizon that is very near to our price-stability goal of almost 2%”
Poor Italian Macro Performance: Required sructura reforms unlikely after a “no” vote
The poor Italian macroeconomic performance both provides Italian with plenty of motivations for lodging a protest vote against the Renzi Government and testifies to the need for the structural reforms which the electoral reforms aim to facilitate.
Specifically, real GDP growth has lagged the other big EA countries (see chart) – both during the crisis and after (where notably Spain’s recent strong growth burst hasn’t been mirored). And while the 0.3% q/q Q3 2016 growth (beating Germany’s 0.2%) provide Renzi with some limited ammunition, higher frequency indicators point to continued underperformance. Italy has the lowest composite PMI of the main EA countries (see Chart), which moreover has declined in 2016 whereas other countries have stabilised/improved.
Italian households have also had to contend with continued high levels of unemployment (11.7% versus Eurozone average of 10% and only 4.1% in Germany) and very low wage growth (latest 0.6% y/y). And Italian households have recently started expecting fresh rises in unemployment (see second chart below), hitting Italian consumer confidence. Italian industry is anticipating only meagre future employment increases (although there’s limited improvement here).
On top of this, Italy moved back into deflationary territory in October (HICP -0.1% y/y, versus +0.5% in the EZ overall) and core inflation recorded a record low of only 0.2 y/y (versus 0.8% for the EZ). While this has limited near-term benefits for Italian households (it limits the fall in real wages accociated with low nominal wage growth) it powerfully illustrates the difficulty facing the ECB in obtaning “clear signs of a convincing upward trend” in core inflation in Italy (as well as the EZ overall).
Finally, it’s notable that the World Bank ranks Italy as one of the worst OECD coutries for doing business – only Isreal, Chile, Luxembourg and Greece are worse. Moreover, Italy seems to be going in the wrong direction – it slipped to number 50 in the 2016 world rankings from number 45 in 2015. This is clearly an economy in urgent need of reform, with continued low growth likely should that not occur.