Odds of a successful motion of no confidence against the French government are mounting. ​ Given the difficulty to build a proper majority around the budget, the government is very likely to invoke article 49.3 of the Constitution so that the budget would be deemed approved without a formal vote, but then the opposition will likely table a motion of no confidence. If the motion is successful – which takes an absolute majority in the lower house - the government must resign, with action limited to dealing with “current affairs” until a new Prime Minister and cabinet are appointed. ​

This could come as early as this week. An agreement was found last week between delegates of the two houses of parliament on the social security budget bill. This modified version is submitted to the National Assembly this Monday, where the government is unlikely to find a majority: the parties supporting the Prime Minister commanded a small majority of the two houses’ delegates, but not in the lower house alone. If the Prime Minister triggers Article 49.3 this Monday, the opposition groups have 24 hours to table a motion of no confidence. ​ If this motion– on the social security issue – fails, the same process will probably have to be repeated in Mid-December on the central government budget.

To work out the maths, two opposition parliamentary groups matter: Rassemblement National (RN) and the Socialists (PS). Without RN a motion of confidence cannot succeed. Without PS it is arithmetically possible but would require an unlikely perfect mobilisation of all other opposition and non-aligned deputies (abstention in a vote of no confidence is counted as a vote FOR the government). Let us look at their possible “reaction function” separately:

The main issue for RN which would emerge from triggering a collapse in the Barnier government is that it would dent their efforts to appear as a “responsible” actor since the dissolution. They have probably exhausted their margin of progression among those tempted by a protest vote and need to appeal to more mainstream voters to make further gains in the next electoral races. According to a Louis Harris poll released on Friday, while 53% of all respondents support a motion of no confidence, this falls to 27% for those leaning to the centre-right, who would be the natural electoral target of RN in the future. Similar results emerge from an Odoxa poll (54% support a motion of no confidence, but only a minority of centre-right voters). Yet, this must be balanced against the political cost for them to allow the budget and its unpopular measures to pass if they do not obtain enough concessions. ​

According to the Louis Harris poll, 59% of the RN-leaning voters support a motion of no confidence, and 67% according to Odoxa. The left will in any case table a motion of no confidence. If RN deputies do not support it, the left will have an easy angle against RN in any subsequent campaign: presenting RN as de facto part of the majority. In addition, RN can mitigate the risk of appearing as troublemakers ready to precipitate the country into unknown institutional territory – no new parliamentary election can be organised before next summer – by continuing to make the point they will support the “special law” in parliament which, under the Constitution, would allow tax to be collected in 2025 under the same conditions as in 2024. and thus avoid a US-style government shutdown if the budget bill fails to be adopted. In response to the government falling, the President could either re-appoint Michel Barnier, find another mainstream PM or resort to a technical government solution (i.e. appointing technocrats only to the cabinet) until it becomes constitutionally possible to trigger new elections. On balance, such configuration could play in the hands of RN in terms of future electoral fortunes, as they could calculate that any new “mainstream arrangement” would be even weaker than the current cabinet.

On the other side of the political spectrum, not voting a motion of no confidence would allow the Socialists to break their relationship with the hard left (LFI) which has been increasingly strained. Yet, this could cost them dearly electorally. They were big beneficiaries of the “single candidacy” pact of the left in July. Besides, their own electorate is leaning towards a motion of no confidence (58% according to Louis Harris , 63% according to Odoxa). In addition, breaking the left alliance would provide them with little immediate political gain: it would be difficult for them to participate to an extended pact with the centrists and the centre-right and join the cabinet. Their best option may well be to vote a motion of no confidence, and then push for a technical government (this is what K. Bouamrane, one of the emerging Socialist leaders, is explicitly calling for) as well as support a “special budget law” to avoid a shutdown. This would preserve their electoral capital for any new race. ​

In a nutshell, the objective interests of the socialists and RN may be moving towards pushing for a collapse of the Barnier government. There are two realistic options for the Prime Minister in his efforts to stop this – on which he is working on simultaneously. First, painting a very bleak picture of the consequences of a rejection of the budget, notably in terms of financial market reaction – Barnier made this plain in his latest interview on prime-time national TV – in the hope that public opinion will shift and change the opposition’s calculus. Second, offering enough concessions on the budget to opposition groups to give them a good reason to explain to their electorate why ultimately they would not support a motion of no confidence.

What is left of the initial budget bill and what is the alternative?

An issue with concessions is that the initial budget bill was already “touch and go” on its capacity to cut the deficit in a tangible manner next year. The bill seeks to bring the deficit from 6.1% of GDP in 2024 to 5.0% next year. This is based on an assumption for GDP growth in 2025 (1.1%) which is looking increasingly optimistic, judging by the latest high-frequency indicators (more on this in the next section). Our own forecast stands at 0.6%. A fairly robust heuristic is that a change of 1% in GDP growth affects the deficit to the tune of 0.5% of GDP. If we are right, then the deficit – assuming the measures contained in the initial budget bill pass entirely - would land at 5.25%. The concession on the electricity tax offered by the government last week would cost EUR3bn (c.0.1% of GDP), only very partly offset by an additional effort of healthcare spending (0.6bn). The government had already reduced the magnitude of its project to raise social contributions on salaries below twice the minimum wage, bringing the overall effect on the deficit down to 1.6bn from 4bn in the initial version. In a similar vein, the government has earlier dampened its plan to postpone the indexation of pensions to inflation until July – instead of January. All pensions would now be revised up by half the rise in consumer prices in January, the other half coming in July for those with small pensions, reducing the saving for the government by between EUR 0.5 and 0.8bn. So, between the possible adverse outcome on growth, and the three key concessions already offered, the deficit would probably land at around 5.5% of GDP in 2025.

The list of additional concessions demanded by RN by this Monday remains daunting. In an interview to Les Echos published on Saturday 30 November, Jean-Philippe Tanguy – the main RN spokesperson on budgetary matters – reiterated in no ambiguous words that his party still demands a full indexation of “all pensions” in January (which would lift the deficit by another 0.1% of GDP), which would be offset by more cuts in local government spending but also, and more crucially since it could only trigger a direct confrontation with the EU institutions, a reduction in France’s contribution to the European budget. The latter point is probably unacceptable to Michel Barnier. According to Marine Le Pen in a statement to Agence France Presse on Sunday, the “discussion with the Prime Minister is now over.”

What would happen if the government is forced to resign? As we discussed earlier, the most obvious option would consist in having the “current affairs cabinet” request from the lower house a vote authorising the rollover of tax into 2025. This would prevent a government shutdown and allow France to continue making it good on its financial liabilities. This would have some “interesting” ramifications though. Indeed, while the income tax brackets are usually indexed on inflation in France, it still takes a proper budget bill to activate this indexation. In case of a “roll-over,” there would be some “tax drag,” with households who so far had been exempt now caught up by the income tax, and a rise in the overall volume of receipts. Using the usual elasticity (we have some precedent to work on, brackets were frozen in 2011-2012), if inflation stands at 2% in 2025, the government’s income would rise by c.2.5bn, i.e. c 0.1% of GDP. On the spending side, in our understanding, expenditure would have to remain within the boundaries of the previous budget bill and could not be indexed on inflation unless the automatic operation of the law would mandate it. We can think of one big item: pensions. Indeed, it takes a budget bill NOT to index it, but in the absence of a new budget, it is the current state of the social security code which would be enforced, which would normally trigger an automatic indexation in January. Pensions stand for about one quarter of total government spending. So, overall, a nominal freeze in 75% of expenditure would mechanically reduce the overall spending to GDP ratio by 0.9% of GDP in 2025. Yet, by construction the discretionary savings/tax hikes planned for 2025, estimated by the government at 1.0% of GDP, would be lost. So, in a nutshell the automatic savings combined with the drift in income tax triggered by the “rollover” would be equivalent to the planned discretionary effort.

Such calculation is however very imperfect. First, it does not take into account behavioural reactions to the absence of the discretionary measures contained in the budget bill. For instance, there is no ex ante hard limit to healthcare spending. Delivering on the national objective on this matter (+2.8% , from 3.3% in 2024), which we already found very ambitious, is dependent on the enforcement of discretionary measures (such as transferring to the private sector some of the healthcare costs) which could not take effect without the budget bill, and the ensuing drift in spending could far exceed a “business as usual” trajectory. Second, the nominal freeze in government spending would likely be only transitory. Indeed, in our understanding, nothing would prevent relaunching a budgetary process in 2025 with a minimum agreement across parliamentary groups. It is unlikely that, socially, freezing spending in nominal terms could be palatable for more than just a few months. ​

More generally, rather than the immediate impact of a budget rejection on day-to-day spending, our main concern is with the overall trajectory for French public finances. The 2025 budget bill was supposed to be only one step towards a multi-year effort at curbing the progression of public debt. If it proves so difficult to get an adjustment of 1% of GDP across the line this time – a quite modest effort compared with what peripheral countries had to consent to in the midst of the sovereign crisis of 2011-2012 - ​ to the point that a government cannot survive it, then the chances of delivering anything meaningful next year would be materially lower, at least not without a political clarification.

The economic cost

We have already alluded to the government assumption for economic growth next year. The latest INSEE surveys continue to suggest that the level of confidence in the French business sector is lower than the Euro area average, but not massively so (see Exhibit 1) for the crucial services sector. This is no great performance in absolute terms, but probably enough to keep GDP growth in slightly positive territory. The significant drop in consumer confidence in November is however more concerning (see Exhibit 2). ​

It is back to 0.8 standard deviation below its long-term average, a more dramatic downturn than what can be observed in Germany, despite the constant noise around the demise of the country’s growth model. It seems that the benefit from disinflation – CPI fell from 3.1% yoy in January 2024 to 1.3% in November in France – has been offset by more generic concerns, with political instability potentially playing a role. Expectations on the “general economic situation in 12 months” have deteriorated significantly in November (from 0.6 standard deviations below the long-term average to 1.1).

This deterioration in households’ expectations may become self-fulfilling if this triggers another rise in the savings rate. It has already reached a level unseen since the late 1970s. Our savings tracker – which predicts the savings rate by using some components of the consumer confidence survey – points to yet another step up in Q4, hitting 19% (see Exhibit 3). The positive case about France – and the Euro area – rested on a rebound in consumer spending, as wage growth would exceed inflation, generating decent purchasing power gains. This can however be thwarted if these gains are entirely absorbed by another saving effort. The government’s forecast for 2025 were underpinned by such positive outcome. Indeed, the budget bill was built under the assumption real wages would rise by 0.8% in 2025, up from 0.6% in 2024, but also crucially that the savings rate would fall from 18.1% to 17.6%, giving way to an acceleration in private consumption to 1.3% from 0.7% in 2024. If this does not materialise, the consequences for the fiscal trajectory should not be ignored. Indeed, the GDP breakdown sometimes matters as much as GDP growth itself in terms of tax receipts: a weak domestic demand is detrimental (tax is not levied on exports….). ​

Who is going to pick up the tab?

The pressure on the French bond market has somewhat intensified over the last few days given the possibility the budget fails and the government falls, but the OAT-Bund spread has not widened that much further away from the 60-80 bps range in which it has been trading since the dissolution last June (it stood at 81 bps on Friday at close). This may be surprising since there are indications that international investors have become net sellers of French debt securities since the beginning of the summer. The French Treasury does not communicate in real time on its investors’ breakdown, but we can use the data from the Bank of Japan – Japanese entities are key players on the OAT market – as proxy. Net purchases tend to follow a seasonal pattern but comparing July-September 23 and 24 is illuminating: while Japanese investors have gone longer US treasuries, they also raised their participation to the German and even Italian sovereign bond markets (only marginally for the latter), they have divested from France (see Exhibit 4) to the tune of JPY 1,800bn. This is equivalent to 0.5% of the overall quantum of medium and long-term French sovereign bonds in circulation. ​

We do not have any precise enough data to tell us how other foreign investors are behaving, but we can look at the French balance of payments for broad indications. Between July and September, France recorded foreign inflows into “long term debt securities” (we do not have the private/public sector breakdown) of some EUR22bn. This was significantly lower than in July/September 2023 (EUR 62bn). There is however enough depth in the French domestic investor base to absorb such outflows without too much disruption.

Indeed, France’s high reliance on international investors for the funding of its public debt (54% of government debt is held by non-residents) coexists with a high domestic savings ratio. The corollary is that it is relatively easy to “replace” departing international investors with local ones in times of crisis. This is what was missing in Spain or Portugal at the time of the peripheral crisis 15 years ago: their banks were themselves in a sorry state and institutional investors were too weak to replace foreign players. From this point of view, France is closer to Italy. Still, precisely, while this reduces significantly the risk of “sudden stops” in the funding of public debt, in the medium term this is not without significant adverse macroeconomic consequences. A key risk in our view is a slow Italianisation of the French economy: in Italy, funding the government took an ever-growing share of the banks’ balance sheet, reducing the capacity to extend credit to the private sector. A textbook “crowding out” situation.

Counting on the ECB cavalry

Even if the widening of the spread since the dissolution reflects a higher risk premium on France, in terms of “fiscal mechanics” what matters is more the absolute level of yields, which commands the impact on the government’s debt servicing costs and hence the risks of “spirals”. From that point of view, France is lucky. Indeed, at 2.89% on a 10-year yield, Paris is paying less than just before the dissolution (3.10%).

The European bond market is at the moment resisting contagion from the US – the 10-year spread between Germany and the US has widened by 40 bps since early October. Expectations of weak real economy developments in the Euro area clearly play their usual role, but also the likely divergence in the inflation outlook. Last week we explored how the US dataflow was suggesting that, even before the likely implementation of inflationary policies by the incoming Republican administration, the Fed has to deal with accumulating evidence that a “resistance line” has emerged on inflation in the US. In the Euro area, conversely, good news continues to accumulate on the inflation front. Core inflation stabilised in year-on-year terms in November at 2.7%, while the market was expecting 2.8%. Base effects continue to make the monthly readings difficult, but it is reassuring that, when looking at the 3-month momentum, services prices have slowed down noticeably (see Exhibit 5). The ECB can comfortably continue to cut in December, which should continue to help anchor down long-term yields.