1. The referendum is not a make-or-break moment for Italian politics. Although important, near-term risks stemming from the referendum outcome may be overstated. A No vote would maintain the (perhaps not ideal, but wellknown) institutional status quo, although it could lead to government resignations. We reckon on balance the reform is positive for Italy’s future political stability and ability to implement reforms, but experts’ views, which are increasingly diverging on its merits suggest that investors should be cautious in drawing firm conclusions either way.
2. Opinion polls clearly point to a No outcome, risks may be skewed the other way. The average of 11 polls before the blackout started (Fri 18 Nov) showed support for No at 52.6% vs 45.6% for Yes (Figure 1). The 7pp-lead is just outside the margin of error of most polls (+/- 3pp). The trend over the past month has been in the favour of No, alongside a small decline in the share of undecided – and the most undecided are centre-right/right voters, whose party’s official position is for a No vote. However, (i) the share of undecided among those intending to vote remains high, at 20-25%; and TV programmes may have more influence on undecided voters than their party’s position.2 (ii) The ballot question is a clear Yes-answer for many Italians.3 (iii) Most polls are conducted via fixed-line phone calls; the only online poll to our knowledge by Termometro Politico (fieldwork: 17 Nov) showed a much tighter result (23.3% for No, 22.7% for Yes). Note there is no quorum for the vote to be valid. We keep our base case view that No is the most likely outcome and reckon this is probably also the base case for most investors. The probability of a Yes outcome, though, may be higher than generally assumed.
3. Changes to Italicum are key for Italy’s political outlook. We stress that the electoral law named Italicum, not the referendum, is what matters most for the probability of an anti-establishment/anti-euro party coming to power after the next general election. The Italicum (in force since July 16) hands a Lower House parliamentary majority (54% of the seats) to the party winning the national election by either garnering more than 40% of the first vote, or winning the run-off ballot. It strongly prioritises government stability at the cost of a significant concentration of power in one single party and reduced checks and balances – an increasingly risky feature in the new three or four-party political system. Current polls suggest the anti-establishment, anti-euro M5S political movement is likely to prevail in the run-off ballot (Figure 2). The referendum outcome may have an impact on how the Italicum will be modified. The law is currently under review by the Constitutional Court (CC) and a ruling is expected in early 2017. The prevailing view seems to be that the CC will likely reject some features of the Italicum, with the extent of the changes probably depending on the referendum result – i.e., deeper changes to compensate for reduced checks and balances if a Yes vote leaves the Lower House as the only house of Parliament (since the constitutional reform would significantly reduce the role of the Senate). However, we also flag the risk that a Yes vote could tempt an emboldened PM Renzi to push for few modifications to the Italicum in the hope of re-gaining control of the Lower House at the next election.
4. What happens if No prevails. PM Renzi will likely resign, but this does not necessarily imply snap elections, at least in the near term. Italy has had 65 governments over the last 70 years, but only 17 legislatures: it is pretty common to change government with the same parliament. If Renzi resigns, the President of the Republic, Mattarella will launch party consultations and probably appoint a new PM within a few weeks. If it is a ‘soft’ No vote (i.e. within say 45-55%), we would not rule out Renzi himself being re-appointed. 45% support for Yes with the entire political spectrum supporting No could still be claimed as a relative success for PM Renzi. If it is a ‘hard’ No, a superpartes/technocratic personality (like Upper House speaker Grasso or FinMin Padoan) is more likely to take on the PM role, possibly requiring slightly more time. A simple majority in both Houses of Parliament is needed to appoint a new PM, but to elect a super-partes/ technocratic government a broader parliamentary majority may be sought (with the most obvious candidate to support it being centre-right Berlusconi’s Forza Italia, Figure 3). The key mandate for any post-referendum government will be to change the electoral law before the next election, subject to the changes imposed by the CC. The sooner the electoral law is agreed, the higher the likelihood of early election in 2017.
5. Yes has its drawbacks too. If Yes wins, a strengthened government will probably stay in power until the next election in Feb-18. The Senate will take its new shape after the next election (and will require an ad-hoc electoral law to be agreed before that), while other parts of the constitutional reform will become effective immediately. However, risks of early election and of anti-establishment parties coming to power before 2018 may not dissipate entirely, as Renzi could be tempted to capitalise on the referendum outcome, as soon as a new electoral law is approved. Even though a Yes outcome would probably be more supportive for the medium-term structural reform outlook, we currently would not expect any major reform to be announced ahead of the 2018 election in any case.
6. Early election in 2017? Not our base case, but cannot be ruled out in either of the referendum outcomes. The next election is due by Feb-18. Even if it were to happen in 2017, the most important point is what electoral system will regulate it, especially if Yes prevails and the national ballot will elect only the Lower House. If No wins, agreeing a new electoral law (for both Houses of Parliament) within a probable grand coalition will likely take time, hence making election by summer 2017 somewhat difficult.
7. Irrespective of the referendum result, Italian political risk to stay high. The probability of an anti-establishment, anti-euro party gaining the largest share of the vote in the next election remains equally considerable, in our view. This risk will reduce the more the electoral system is close to full proportional representation (PR). Yet, such a shift would imply less political stability and higher chances of stalemates (like in the aftermath of the last election in 2013). M5S and the Northern League both have featured anti-establishment and antieuro themes centrally in their manifestos and together they garner more than 40% of the vote in recent polls (Figure 4). The latest Eurobarometer survey showed Italians’ support for the EU and the euro is among the lowest in Europe (Figure 5). Note that the Italian constitution prohibits referendums on international treaties, including EU/euro membership. Yet, an advisory referendum on EU membership was held in 1989, which in order to be called required a constitutional law passed by both Houses of Parliament.
8. Irrespective of the referendum result, no pick-up in reform momentum likely before next election. A No vote would add to the perception that Italy is unable to push through reforms. However, under either referendum outcome we reckon the scope for new reforms being approved and implemented in 2017 is rather limited. Agreeing a new electoral law will require a significant amount of political capital and compromises among mainstream parties. This could erode their support further ahead of the 2018 elections and foster antiestablishment sentiment further.
9. Economic impacts and Italian banks. The banking sector looks the most at risk in the event of increased political uncertainty due to (i) likely delays in the ongoing efforts to strengthen its capital position and (ii) banks’ still-high exposure to sovereign debt (ca. €400bn of securities only) (see Italian Banks: Fasten Your Seatbelts). However, we reckon that a super-partes or technocratic government may be more able to agree on public support for the banks and/or bail-ins than a non-technocratic, albeit strong, government. We estimate the negative effects of a No outcome on our already belowConsensus Italian GDP growth forecast (of 0.5% for 2017) are probably not large (probably no more than 0.1-0.2pp). Yet, we do acknowledge the risks of a more negative loop between banking woes and higher government bond yields which could result in much tighter financial conditions and lower GDP.
10. Rating downgrades might follow if political uncertainty persists for longer than we expect. Two rating agencies, Fitch and DBRS, currently have the Italian rating under negative credit watch (or negative review) partly because of the referendum-related political uncertainty. Fitch wrote in October that a “No vote would represent a political shock”. A downgrade from DBRS (from A Low) would increase the haircuts applied to Italian government bonds (and possibly other assets) posted as collateral for Eurosystem funding.
11. What, if anything, could the ECB do? Barring major financial market volatility, the referendum outcome is unlikely to sway the upcoming ECB decisions on 8- Dec, in our view. Many ECB Governing Council members would be very wary of reacting to an idiosyncratic political event. We currently expect the ECB to announce an extension of its asset purchase programme by six months at a reduced monthly pace of €60bn. But in the event of significant market turbulence, there is an increased likelihood of a 6-month extension at an unchanged €80bn/month pace. Further tensions in the banking sector could lead to a loosening in collateral requirements for assets to be used in ECB funding operations. In principle, the ECB’s OMT programme would also be available as a last-resort backstop facility, but only in case of existential Eurozone crisis/redenomination fears and only after Italy first agrees to an ESM bailout programme – highly unlikely, in our view, unless BTP yields were to rise close to 2011-2012 Eurozone crisis highs.
12. General political risks in Europe to stay high in 2017. The Italian referendum is only one of a series of political events which could deliver another anti-establishment/protest vote. The near-term political consequences are likely to be limited, especially in a country with a long history of government instability. However, political risk in Europe will stay high – the re-run of the Austrian Presidential election also on 4 December, the Dutch parliamentary election on 15 March, the French presidential election in April/May and the German election in autumn 2017 will be soon in focus.