Attention aux foules !

# 230 – 24 juin 2024

L'issue des élections françaises reste très incertaine. Les intentions de vote s'orientent vers les trois blocs principaux, car ceux qui sont attachés à des familles politiques plus petites veulent probablement faire compter leur voix, et le RN conserve sa position dominante. De plus en plus de sondages fournissant des projections de sièges rapprochent le RN du seuil de la majorité, mais à en juger par les enquêtes, un parlement sans majorité reste (de justesse) l'issue la plus plausible. Beaucoup dépendra du nombre de candidats qualifiés des partis historiques de gouvernement en troisième position au premier tour qui décideront de se retirer du second tour pour contenir l'extrême droite.

L'annonce par la Commission européenne que la France, ainsi que 7 autres Etats membres, est de nouveau sous " procédure de déficit excessif " (PDE) - bien qu'elle n'ait pas eu beaucoup d'impact sur les spreads jusqu'à présent - est une autre occasion de réaliser qu'une action corrective est nécessaire pour éviter une dérive majeure de la dette publique. C'est un point aveugle du débat actuel en France, organisé autour des propositions "moyennement dépensières" et "très dépensières" des deux forces politiques en tête des sondages. D'ailleurs, le fait que l'Italie soit à nouveau soumise à la PDE devrait nous rappeler que le "plan Meloni", souvent utilisé pour esquisser ce que pourrait être l'approche économique d'un gouvernement dirigé par le RN en France, n'a pas encore été testé pour un resserrement budgétaire potentiellement douloureux sous l'étroite surveillance de l'UE.

Les cours des actions des banques françaises se sont légèrement améliorés par rapport à leur niveau le plus bas du 14 juin. Heureusement, leur exposition à la dette souveraine est faible par rapport à celle de leurs homologues italiennes, par exemple. Pourtant, avant la grande crise financière de 2008, les banques françaises et italiennes avaient des niveaux d'exposition très similaires. L'expérience acquise durant les "mauvais jours" du début des années 2010 suggère que suivre la "voie italienne" et faire en sorte que les banques locales prennent en charge une part croissante de la dette publique locale, bien que compréhensible du point de vue de la stabilité financière, peut devenir économiquement toxique car cela évince les prêts au secteur privé et déprime la croissance tendancielle. Il existe une marge de manœuvre nationale pour absorber la dette française, mais l'utilisation de cette marge aurait toujours un coût macroéconomique.

Vous trouverez ci-dessous l'analyse de Gilles Moëc.

Latest Messages from the French Polls

The 10-year spread between France and Germany stood last Friday at close at 75 basis points (bps), 35 bps above the level seen before the announcement of the snap elections and almost unchanged over the week. The magnitude of the contagion effect to the periphery has not changed either. The Italian spread continues to oscillate around 150 bps, some 20bps above the pre-announcement level. This suggests that the decision by the European Commission to put France and 7 other member states – including Italy – under the Excessive Deficit Procedure had little immediate or direct effect over the market perceptions of these signatures. This was symmetric: Spain was spared - it was only found in temporary breach of the rules – but did not benefit from a particularly favourable treatment by the market (its spread stayed in the 85-90 bps range, up 15 bps, in which it has been pricing since the announcement). 

The move by Brussels was much anticipated. While the response of the national governments to this materialisation of the European disciplinary framework will matter in the months and years ahead (more on this later), investors did not have to wait for the EC’s analysis to understand that corrective fiscal action is needed. ​ In the case of France, the capacity to deliver such action will of course depend on the outcome of the elections. Candidacies in each of France’s 577 constituencies closed on Sunday 16 June, and while polls have probably become more precise as citizens now know about their local choices, they did not necessarily bring that much more clarity.

In Exhibit 1 we look at the latest voting intentions provided by the pollsters which had surveyed public opinion in the days immediately after the announcement of the snap elections by the President. Focusing on these sources helps us to get a sense of how voters are evolving now that alliances have been formed and candidates are known. Indeed, confusion was initially high, since the main centre-right party de facto split, with one component aligning on Rassemblement National, and there could be doubts as to the capacity of the broad left-wing alliance to effectively contain the number of dissident candidacies.

We draw two main conclusions. First, the three main blocks – far right, left-wing, and centrists– are all getting stronger support. If we take the average of the three pollsters, at the end of last week the three blocks together attracted 83% of the voting intentions, against 76% immediately after the announcement. Voters normally attached to the smaller political families are gravitating towards the main parties to make their vote count (the French expression for this is “vote utile”, literally “useful vote”). The “other left” – i.e., those on the left who refused the alliance with the far-left - and centrists non-aligned with Macron (DivL+C in the table) lost nearly 4 points. Voting intentions for the centre-right (CR in the table) fell as well, as some of their electors are moving to the candidates who allied themselves to RN. Voting intentions for Reconquete (Rec in the table) – another far-right party – also fell as its supporters are probably shifting towards RN.

Second, the left and the centre are making slightly more progress than the RN which however maintains a comfortable lead. Voting intentions for the left alliance were initially much lower than the total score of its components in the European elections (32%). It seems that the “Front Populaire” has been able to plug some of that gap, but not all. Some of the socialist supporters have probably shifted to Ensemble, the centrist alliance, which explains much of the improvement there (and gain relative to their poor showing in the European elections). The gains in voting intentions for RN are modest, suggesting that the common candidacies with a component of the centre-right in more than 60 constituencies and rallying Reconquete voters did not bring that much additional support. ​ Yet, given the likely number of three-way races in the second round, RN’s confirmed top position among the three blocks in terms of voting intentions puts them closer to an absolute majority on 7 July.

Louis Harris Interactive and Elabe are the two traditional pollsters which have already produced two sets of seats projections since the announcement of snap elections. They have both revised up their range for RN, with the upper bound now only 9 seats away from the majority threshold (see Exhibit 2). They have revised also up their forecasts for the number of seats secured by the left alliance – albeit still very far from the majority threshold – and revised down their forecasts for the Centre. Indeed, with a higher turnout, in many cases three rather than just two candidates will qualify for the second round by winning more than 12.5% of the votes of registered voters. This will benefit the RN. Indeed, traditionally, RN’s problem in both parliamentary and presidential elections is its difficulty to appeal to enough moderate voters between the two rounds to pass 50%. In 2022 there were only 8 three-way races in the second round of the parliamentary elections. This time, pollsters expect well north of 100 of them. RN’s difficulty to extend its reach will be less of a problem. Simply keeping their voters from the first round in many three-way cases will be enough.

Still, at this stage, the absence of a clear majority in the National Assembly remains (just) the most plausible scenario. A difference with the inferences we could draw from the first polls is the diminishing probability to see an alliance between the centre-right and the centrists in the assembly numerically stronger than the left alliance. Even if, as we discussed last week, they could get the support from moderate left socialist deputies, this would make appointing a PM from the centrist group less straightforward. This may be one reason why the latest “talk of town” in Paris is that rather than a political figure, the PM leading a “technical government” could be a technocrat – France aligning itself to the practice often seen in Italy to deal with similar situations.

A crucial and very uncertain variable in the outcome, irrespective of how voting intentions change in the coming days, will be the behaviour of candidates coming in third position in the first round with enough votes to qualify for the second round but with little chance to win. Parties for now are not communicating on this, but in practice there may be a myriad of local decisions in the constituencies rather than a top-down rule dictated by the parties’ headquarters, with some tactical withdrawals across moderate centrists, centre-right and socialist candidates, possibly greens, to contain RN’s progress. This continues to make seat projections very thorny. ​ ​ 

The manifesto war and the fiscal constraint

The debate on the economic platforms of the left alliance and RN which we briefly discussed last week continues. There seems to be a tactical difference between the two. RN has been retreating on its platform from 2022, while the Front Populaire (albeit with some nuances across its components) seems keen to hold on to even the most spectacular or controversial aspects of the manifesto it cobbled up hastily. We suppose that this reflects different strategic goals.

RN already has strong support among blue collars. The party probably calculates it can afford to let go some of its most progressive but expensive economic measures to extend its appeal to more moderate voters in the middle class, as long as it stands firm on its “fundamentals” (immigration and security). ​ There are limits to this approach though and some costly proposals remain, such as the VAT cut on energy consumption – which is also in direct breach of EU rules, since the 2006 directive set up a limitative list of products on which a reduced rate can be levied. On the other side of the spectrum, the left alliance still needs to solidify its traditional base. The French left has been for years facing more and more difficulty in attracting votes from blue-collar workers. College-educated people with fairly low-paying jobs – especially in the public sector – are the heart of its electorate. A lot of the prominent measures in their manifesto would benefit them, e.g., the 10% rise in public service pay, or the overhaul of the income tax system in a much more redistributive direction. The top marginal rate – above 410K per year – would attract a rate of 90% (a doubling from today), and the threshold beyond which taxpayer would start to lose relative to the current setup would stand at about EUR 42k per annum. The price the left must pay is that these proposals are weighing heavily on its capacity to attract moderate voters.

In any case, the ongoing debate between various shades of spendthrift agendas may look surreal given the underlying position of French public finances. The European Commission’s report detailing the reasons why France has – once again – been placed under the Excessive Deficit Procedure is edifying (link here). A point which we think is worth highlighting is the Commission’s Debt Sustainability Analysis (DSA), which presents a “no policy change” trajectory for public debt, i.e., assuming that the member state’s structural deficit (the deficit corrected for one-offs and the impact of cyclical conditions) does not change from its current level over the next 10 years. France would see its debt rise to almost 140% of GDP by 2034 (see Exhibit 3). Incidentally, the graph helps to understand why Spain was spared by the European Commission: even without additional efforts, Spain would stabilise its debt ratio. Finally, the graph also suggests that the “Meloni blueprint”, increasingly used in the market to sketch out what could be the policies of a RN-led government in France, has not yet been tested for a sustained, painful fiscal adjustment. If the current fiscal stance is not corrected and Italy fails to reap the benefits of the current flow of funds from Brussels to lift trend growth, public debt, in the Commission’s DSA, would reach 168% of GDP in 2034.

Any French government after 7 July will have to deal with these realities. Of course, one could argue that the Commission is too pessimistic with its assumptions for France’s GDP growth rate by the end of the decade (it falls back to 0.5%), but that is the result of the kind of potential growth the country would achieve without lifting productivity, in a context of slower contribution from labour supply in an ageing society. In any case, at least for 2024, the uncertainty triggered by the snap elections is already having an impact on economic activity, and hence tax receipts. The French composite PMI index fell further in contraction territory in June to 48.2 from 48.9 in May, a 4-month low. To quote directly from Markit, the survey source, “the overall strength of business confidence slipped below its historical average to its lowest since the start of the year. In some instances, companies expressed uncertainty towards the upcoming election result, as well as more broad geopolitical risks”.

True, much as the sovereign spread has stabilised last week, French banks’ equity price has rebounded slightly from a trough on 14 June, up 1.9% as of last Friday for the average of the three banks in the CAC40. Yet, when compared with the level before the announcement of the snap elections, the decline remains significant: 12.3%. We mentioned last week the findings of a Banque de France study which concluded to a visible impact of a decline in banks’ equity value on lending appetite. This matters particularly in France since the country exhibits one of the highest debt ratios of the corporate sector (c.150% of GDP), and contrary to the trend at work in the rest of the Euro area where disintermediation continues, banks have recently taken up a higher share of this burden.

Still, the magnitude of the transmission from pressure on the sovereign’ funding costs to banks should be a function of the banks’ exposure to government bonds, and from that point of view, French banks are in a relatively good position. In Q1 2024 sovereign bonds stood at 2.4% of French banks’ total assets, against 12% for their Italian counterparts. The so-called doom-loop between the sovereign and the banks is still far-away in France (see Exhibit 4). Yet, we think we should be wary of “boiling frog” phenomenon if the fiscal slippage continues, and banks are increasingly called up to absorb an excessive supply of government bonds.

We were surprised last week by the focus in the market commentariat on the bond auction organised – as scheduled – by the Treasury which was presented as the first “test” of France’s attractiveness after the announcement of the snap elections. The auction went well, with bids at roughly 2.5 times the amount collected by the Treasury, but if we mobilise our memory of the “peripheral crisis” of 2011-2012, auctions usually go quite smoothly even in times of stress. The issue is not there. Italy provides a good example of how gradually a local national system can become overly exposed to its sovereign. Usually, foreign investors are the first to stop investing in a “problem country”. They can be replaced by local banks which have a vested interest in keeping their sovereign afloat to avoid being engulfed in a systemic crisis – all the more since the capital charge they incur on those assets is small. When the European sovereign crisis struck, the exposure of Italian banks to sovereign bonds was on par with their French counterparts, at about 5% of their total assets. By the end of the crisis, the ratio had tripled. Ultimately, this was not a bad decision from a financial stability point of view, but it contributed to the stagnation of the Italian economy in those years, as banks’ capacity to lend to the private sector was increasingly crowded out by lending to the government. We are not there, not at all, but down the road, this is often how strong countries start struggling when fiscal policy is not under control: not necessarily a “sudden stop”, but a slow-grinding exhaustion of economic dynamics. 

The British counter-example

A few weeks ago, when surveying the coming British elections, we used the title “boring is making a welcome return”. Now that the main parties have issued their manifestos, the contrast with France is even starker. Our colleague Gabriella Dickens has just issued a thorough paper on the proposals and their challenges (see link here). In the economic realm we have two versions of mainstream policies facing each other, with only nuances between them. The fact that whoever comes to power after 4 July will have the smallest fiscal headroom – i.e., distance from the existing rules – of any modern government is clearly having an impact on the parties’ ambitions. It may be that the UK’s brush with populist policies since the Brexit referendum in 2016 – with scary consequences in the autumn of 2022 – is acting as a powerful vaccine.