Marine Tondelier’s programme: what we were able to cost

Effects on households

Eight model households, annual disposable income, calculated with OpenFisca-France for only the measures the engine can simulate (8 measures). A zero difference means that these measures do not affect this household, not that the programme changes nothing for it. *: at least one simulated measure rests on a reading by the editors. The detail, household by household, and the comparison with the other candidates.

The public accounts

How to read these amounts. These are static costs and revenues, with no feedback effect, over a full year: nothing on growth, employment or the revenue that economic activity would gain or lose. A higher net cost does not on its own make a worse policy, nor net savings a better one: each programme rests on an economic model (demand-led stimulus, lower compulsory levies, consolidation of the public accounts…) that the cost schedule does not settle. GDP itself measures a volume of activity, not well-being or the usefulness of what is produced. Economic models and feedback effects.

Line by line

• LeverPositionEffect on the public accounts, per year
• Numbers and pay of civil servantsPay index point indexed to inflation, +10% for categories C and BAssumption by the editors. Partial costing. Coefficient calculated by the editors from Sénat, « L'essentiel » sur le PLF 2026, programme « Fonction publique » de la mission « Transformation et fonctions publiques » (page 2) and from DGAFP, rapport annuel 2026 sur l'état de la fonction publique, et Cour des comptes, note « La masse salariale » (2026): the position is marked as an assumption by the editors. Costed: the annual indexation of the public sector pay index point to inflation (common editors' assumption: 1.75% a year, 1.5 to 2.0%, five upratings from 2028 to 2032, i.e. a cumulative increase of 9.1% in 2032; reading key: a one-point gap for one year costs €1.53bn) and the 10% increase in the pay of category B and C staff, read as applying to their whole pay, all statuses (civil servants only at the lower bound). Not costed: the increase for « catégories A qui ont perdu en pouvoir d'achat » [“category A staff who have lost purchasing power”], not delimited. The 15% increase for teachers is costed under its lever; that of 10% for researchers has no lever of its own.−€27bnnew spending, −€42bn to −€20bn
• Wealth tax (IFI, ISF)2% Zucman tax above €100m, climate wealth tax (ISF) restoredAssumption by the editors. Partial costing. Reading adopted: « rétablir l'ISF » [“restore the ISF”] means the return of the 2017 ISF, whose entry threshold (€1.3m) is taken up by the Budget 2027 dossier (26 August 2026); the cost schedule gives its net gain relative to the IFI. The climate component, which modulates the tax schedule according to the carbon footprint of investments, is not costed (no rates): depending on how it is set, it may raise or lower this amount. The Zucman tax (2% minimum tax on fortunes of more than €100m) is costed at the upper end of the estimates, €15bn to €25bn (about 20), that of its proponents (Gabriel Zucman, National Assembly report no. 930 and Senate report no. 689); the administration has not simulated it; estimates that take account of exile, tax optimisation or the deduction of corporate income tax are much lower (€2bn to €5bn). Not costed: the extension of the exit tax (parameters not given).+€24bnnew revenue, +€18bn to +€30bn
• CSG (general social contribution)CSG made progressive and raised by one point for health insuranceAssumption by the editors. Coefficient calculated by the editors from Fipeco, L'impôt sur le revenu et la CSG: the position is marked as an assumption by the editors. One-point increase applied to all rates and all bases; progressivity is not costed. At the centre, the gain is net of income tax (12.5%, IPP), assuming the added point is deductible like the 2018 increase; the correction also applies to capital income subject to the flat tax (PFU), where the CSG is not deductible, which reduces the gain by at most €0.25bn. The additional 3-point increase in the CSG on capital income (p. 68) is recorded under the “flat tax” lever, with the increase in the flat tax that it accompanies.+€16bnnew revenue, +€16bn to +€18bn
• Reductions in social contributions on low wagesEnd of exemptions above 1.6 SMIC and on overtimeAssumption by the editors. Partial costing. Coefficient calculated by the editors from Groupe d'experts SMIC, Rapport 2025 (25 novembre 2025): the position is marked as an assumption by the editors. Reading adopted: the exit point of the general reductions in contributions brought down from 3 to 1.6 times the SMIC (minimum wage), the reduction remaining the same below that; €15.8bn was paid above 1.6 times the SMIC in 2024 (regime before the single reduction of 2026). Other reading, not adopted: a reduction redesigned to fall to zero at 1.6 times the SMIC, which would also lower the reductions between 1 and 1.6 times the SMIC (higher gain). Not costed: « nominaliser et désindexer » [“set in nominal terms and de-index”] their amount, whose effect depends on future increases in the SMIC. The end of the reduction in contributions on overtime is costed under the “overtime” lever.+€14bnnew revenue, +€13bn to +€16bn
• Legal retirement age62 immediately; 60 targeted, first for those who started early in arduous workAssumption by the editors. Partial costing. Coefficient calculated by the editors from COR, Rapport annuel juin 2026, Synthèse: the position is marked as an assumption by the editors. The targeted retirement at 60, with priority for people who started work early and in arduous conditions, is not costed. The funding provided for in the same chapter (p. 117) is not counted, as there is no lever in the grid: uncapped pension contribution raised by one point above the social security ceiling, old-age contributions raised by 0.2 point per year during the five-year term. The contributions on employee savings, provided for in the same place, are costed under the “Profit sharing (participation, intéressement)” lever.−€13bnnew spending, −€16bn to −€10bn
• Energy renovations per year500,000 high-performance renovations a year* Outside the range measured by the source: linear extrapolation (measured range: 91,374 homes only). Assumption by the editors. Coefficient calculated by the editors from Chiffres clés de l'Anah 2024 (DREAL Corse): the position is marked as an assumption by the editors. The « rénovations performantes » [“high-performance renovations”] are read as large-scale renovations. The amount is the cost of the target of 500,000 renovations per year once reached, at the current average public aid for a large-scale renovation (€26,100; lower bound: €11,800, average aid in co-owned buildings). This is not the budget announced by Les Écologistes: +€1.4bn for MaPrimeRénov' in 2027 (Budget 2027 dossier), removed on that account from their €10bn envelope under the “Additional public investment, per year” lever.−€11bn^{* (outside the range measured by the source: linear extrapolation)}new spending, −€15bn to −€4.9bn
• Research, share of GDP3% R&D « orientée vers l’intérêt général » [“geared to the general interest”] in 2032Assumption by the editors. Coefficient calculated by the editors from SIES, notes sur la DIRD 2023 (Flash juillet 2025, NI novembre 2025): the position is marked as an assumption by the editors. Target of 3% of GDP for all R&D, with no public share stated: read, as for all candidates, according to the Barcelona target, under which two thirds of the increase come from the private sector. One third of the increase from 2.18% (2023) to 3% is public: +0.27 points of GDP of public research spending in 2032, about €9.5bn. Business R&D is not public spending; « orientée vers l'intérêt général » [“geared towards the general interest”] does not set any other public share.−€9.5bnnew spending, −€10bn to −€7.6bn
• Additional public investment, per year€10bn for the climate from 2027Partial costing. Candidate’s amount, taken as is. €10bn envelope reduced to €8.6bn: MaPrimeRénov' (+€1.4bn) falls under the energy renovation lever. That lever costs the programme’s target, 500,000 renovations per year, in a full year and at the current average aid (€10.6bn): this is not the increase announced for 2027, and the sum of the two items is not the cost of the announced €10bn envelope. The other items (rail, social leasing, school buildings…) remain in the envelope and are not costed elsewhere. Overlap with “Prisons and sentences” examined: one is a cost, the other a gain: with no shares declared for this case, both amounts are added (the higher-amount rule only applies between two costs or two gains).−€8.6bnnew spending
• Teachers' pay+15% over five yearsAssumption by the editors. Coefficient calculated by the editors from Sénat, rapport spécial PLF 2026, Enseignement scolaire: the position is marked as an assumption by the editors. Overlap with “Numbers and pay of civil servants” examined: estimated common base zero, both amounts are added. No source says that an announced increase in teachers' pay is counted net of the general indexation of the public sector pay index point: by default, the common base is zero and the two amounts are added together (the teachers' increase comes on top of indexation).−€5.4bnnew spending, −€8.9bn to −€5.4bn
• 10% allowance on pensionsAbolish the 10% tax allowance for retirees+€5bnnew revenue, +€4.8bn to +€5.4bn
• New nuclear reactorsNuclear phase-out, end of the EPR2 and SMR programmesAssumption by the editors. Coefficient calculated by the editors from EDF, communiqué du 18 décembre 2025, « EDF shares its forecasted cost estimate of the EPR2 programme for €72.8bn » (copie d'archive du 13 octobre 2026, le site renvoyant une erreur 403): the position is marked as an assumption by the editors. +€4.4bnless State lending, on the debt, +€3.7bn to +€7.2bn
• Unemployment, max. benefit durationRepeal the unemployment insurance reforms of the last two five-year termsAssumption by the editors. Partial costing. Reading adopted: the repeal of the reforms of the last two five-year terms includes the 2023 countercyclical modulation, which sets the maximum duration. The other reforms targeted are not costed here (2019-2021 under the “Benefit amount and tapering” lever); the 2026 amendment on negotiated terminations (ruptures conventionnelles) is not costed either.−€4.2bnnew spending, −€4.5bn to −€4.2bn
• Production taxesRestore the CVAE for the benefit of local authoritiesAssumption by the editors. Partial costing. Reading adopted: restoring the CVAE (business value-added contribution) means at least maintaining it beyond 2030, the date of its scheduled abolition, at its current yield (€3.7bn). A return to its level before the 2023 cut would bring in more; it is not costed, as no rate is targeted.+€4.1bnnew revenue, +€4bn to +€4.1bn
• Net monthly SMIC (minimum wage)€2,000 gross from 2027, targeted support for very small firms, SMEs and associationsAssumption by the editors. Partial costing. Coefficient calculated by the editors from legisocial, Montant du SMIC 2026: the position is marked as an assumption by the editors. €2,000 gross converted to net with the contribution rate used by the cost schedule (20.84%), i.e. about €1,583. The support planned for TPE, PME (very small, small and medium-sized businesses) and associations (cut in the reduced rate of corporate income tax and in the payroll tax) is not costed: neither rate nor amount.−€3.9bnrevenue cut, −€5.3bn to −€2.6bn
• Profit sharing (participation, intéressement)Contributions on employee savings, profit-sharing and participation schemesPartial costing. The PERCO and the collective company PER (€0.2bn net in 2026), which Annex 4 classifies with supplementary pensions, are not counted.+€3.7bnnew revenue, +€3.7bn to +€4.9bn
• Funding of private schoolsPublic funding of private schools cut to 50%, conditional on social mixAssumption by the editors. Coefficient calculated by the editors from Assemblée nationale, rapport d'information n° 2423: the position is marked as an assumption by the editors. Reading adopted: the share of public funding (State and local authorities) in the resources of private schools under contract, 75% in 2022 (€10.4bn out of 13.8), brought down to 50%, tuition fees making up the difference at constant resources: €6.9bn of public money, i.e. €3.5bn less (2022 amounts). Static gain: pupils who moved to the public sector would carry their cost over to it. The social mix conditions have no cost.+€3.5bnsaving
• Announced savingsFewer subsidies to the private sector, CIR refocused, pooled public procurementAssumption by the editors. Partial costing. Coefficient calculated by the editors from MESR, « Le crédit d'impôt recherche en 2019 » (données semi-définitives): the position is marked as an assumption by the editors. Costed: the research tax credit « recentré » [“refocused”] on SMEs and mid-sized companies (PME and ETI), read as reserved for them: the claims of large companies (42% of the total in 2019) disappear, €3.0bn to €3.4bn. A cap or a lower rate for large companies would bring in less. Not costed: the cut in subsidies to private education (costed under the private school lever), the « verdissement » [“greening”] of the tax credit and the pooling of public procurement, with no amount.+€3.2bnnew revenue, +€3bn to +€3.4bn
• Electricity prices, European marketPrice decoupled from fossil fuels, lighter taxation of electricityAssumption by the editors. Partial costing. Coefficient calculated by the editors from CRE, rapport annuel 2026 à la Commission européenne (figures 22 et 23): the position is marked as an assumption by the editors. Reading adopted: « unifier le socle de l’accise pour le gaz et l’électricité » [“unify the excise duty base for gas and electricity”] by « diminuant la fiscalité sur l’électricité » [“reducing taxation on electricity”] brings the electricity rate for sites of 36 kVA or less down to that for households' natural gas, €16.66/MWh on 1 August 2026, with no increase on gas. Other reading, not adopted: a revenue-neutral convergence, like the one voted by the Sénat in November 2025 (−€3/MWh on household electricity, +€4.40/MWh on gas), presented as neutral for the budget. Not costed (partial costing): sites of more than 36 kVA, and the VAT cut on the standing charge, which falls under the VAT lever. Decoupling of the electricity price from that of fossil fuels: Target price: no budgetary cost as long as the gap with the market price is not compensated by the State; if it is, it becomes spending (benchmark: the 2022-2024 tariff shield, costed by the Cour des comptes).−€3bnrevenue cut, −€3.1bn to −€2.6bn
• Automatic payment, single social allowanceSolidarity at source: automatic payment of social benefitsAssumption by the editors. Partial costing. Reading adopted: automation brings take-up of the RSA and the in-work benefit to 100%, which is the upper bound of its effect. Only automation for the RSA and the in-work benefit is costed. The other social benefits and the protection mechanism against overpayments are not costed.−€2.9bnnew spending, −€3bn to −€2bn
• Minimum pensionMinimum contributory pension and ASPA raised towards the poverty lineAssumption by the editors. Partial costing. Coefficient calculated by the editors from Cour des comptes, Situation financière et perspectives du système de retraites, note 14: the position is marked as an assumption by the editors. Reading adopted, the same as for Jean-Luc Mélenchon and Fabien Roussel: total pension for a full career raised, once the target is reached, to the poverty line (€1,337 per month for a single person in 2024, Insee Première no. 2117, July 2026), starting from the target of the 2023 law (85% of the net SMIC, i.e. €1,256), hence +€81 per month of guaranteed minimum pension. The 85% target is an objective, not a guaranteed amount: the actual gap varies with careers; the poverty line, measured in 2024, will be higher when the target is reached. Not costed: the ASPA raised to the poverty line (no source read costs this increase relative to the law as it stands in 2026), the reform of agricultural pensions and the pensions of women in that scheme. Central value of the coefficient used: its upper bound, according to the cost schedule itself, overstates the cost (it assumes that all beneficiaries receive the full increase). External reference point, not counted: the Institut Montaigne costs raising the minimum contributory pension to the level of a SMIC of €1,600 net for full careers at €13.9bn per year by 2027 (an increase of €286 per month applied to €7.7bn of minimum contributory pension spending; 2024 legislative elections, Nouveau Front populaire page on pensions).−€2.8bnnew spending, −€4.2bn to −€1.5bn
• VATGreen VAT: reduced rates on essentials, increase on luxury and harmful goodsAssumption by the editors. Partial costing. Rates targeted by the Budget 2027 dossier: 20% for what harms the environment, 5.5% for what contributes to the transition. Costed: non-energy renovation raised from 10 to 20%, at the cost of the tax expenditure on works other than energy renovation (€2.2bn to €2.5bn, Voies et moyens). Not costed, for want of a published base: the cuts (rail and public transport, whose 10% rate is not a tax expenditure; organic products; repair of household appliances, bicycles, shoes and clothing; « première nécessité » [“basic necessities”]), and the other increases (takeaway food, not separated from eat-in catering; domestic flights; luxury hotels; ultra-processed products). The net balance of the reform may therefore be smaller, or even negative.+€2.5bnnew revenue, +€2.2bn to +€2.5bn
• OvertimeAbolish exemptions from social contributions on overtime+€2.3bnnew revenue
• RSA (minimum income), amount and conditionsRSA replaced by an autonomy guarantee at 50% of the net SMIC, without the 15 hours* Outside the range measured by the source: linear extrapolation (measured range: 0% to 10%). Assumption by the editors. Partial costing. Coefficient calculated by the editors from solidarites.gouv.fr, Revalorisation annuelle au 1er avril 2026: the position is marked as an assumption by the editors. Central value raised to the largest amount: cost acknowledged as underestimated, since linear extrapolation beyond +10% does not count new households becoming eligible; the central value of the coefficient (€0.17bn per point, updated to 2026 euros) is also its upper bound. Read as the RSA lump-sum amount raised to 50% of the net SMIC (€739 against €651.69), i.e. +13.4%; beyond +10% (increases from 2013 to 2017, the coefficient’s range), linear extrapolation, which does not count all newly eligible households (entry of new households not costed beyond +10%). Coefficient from the 2020 order updated to 2026 euros: €0.17bn per point (0.15 for households already receiving the benefit alone, lower bound). Not costed: indexation on the SMIC, a rule whose cost depends on future increases, and the abolition of the 15 hours of activity.−€2.3bn^{* (outside the range measured by the source: linear extrapolation)}new spending, −€2.3bn to −€2bn
• Benefit amount and taperingRepeal the reform of the ARE calculation, index benefits to inflationAssumption by the editors. Partial costing. Reading adopted: the repeal of the reforms of the last two five-year terms, « notamment celle sur le calcul de l'ARE » [“notably the one on the calculation of the ARE”], applies to the whole 2019-2021 reform (access, calculation, degressivity), as for La France insoumise. The indexation of benefits to inflation is not costed. Overlap with “Unemployment, max. benefit duration” examined: estimated common base zero, both amounts are added. The 2023 reform was costed on the law resulting from the 2019-2021 reform, and Unédic itself adds the two together (€2.2bn for 2019-2021, €4.5bn for 2023, €6.7bn together in steady state, February 2023, table A).−€2.3bnnew spending
• Medical deductiblesAbolish medical deductiblesAssumption by the editors. Reading adopted: the abolition covers the medical deductibles (franchises) and the flat-rate contributions (participations forfaitaires, €2 per consultation), which the text does not name, in view of the « prise en charge à 100 % par l'Assurance maladie » [“100% coverage by the Assurance maladie”] of essential care in the same proposal.−€2.1bnrevenue cut, −€2.3bn to −€2.1bn
• Wind and solar44% renewables in electricity consumed in 2030Assumption by the editors. Coefficient calculated by the editors from Délibération CRE n°2026-149 du 15 juillet 2026 (CSPE 2026-2027): the position is marked as an assumption by the editors. Reading adopted (common rule for volume targets): 44% of electricity consumption in 2030 according to the RTE (electricity transmission operator) reference trajectory, with which the PPE3 (multi-annual energy plan) says it is consistent (451 TWh in 2025, 580 TWh in 2035, i.e. 515.5 TWh in 2030 on a straight line), requires 226.8 TWh of renewables, against about 205 TWh planned by the PPE3: 21.8 TWh more to support, maintained in 2032. Other reading, not adopted: RTE's slow decarbonisation trajectory (505 TWh in 2035, 478 TWh in 2030), which would require only 5.3 TWh more. RTE's domestic consumption stands in for gross final consumption, which the PPE3 does not project.−€2bnnew spending, −€2bn to −€1.9bn
• Inheritance and gift tax+10% above €4m inherited, a further +10% above €13m, Dutreil cappedAssumption by the editors. Partial costing. Coefficient calculated by the editors from CAE, note n° 69, « Repenser l'héritage », décembre 2021 (page 3): the position is marked as an assumption by the editors. Reading: the contribution applies to the share received by each heir above €4m (10%) and above €13m (a further 10%), transfers combined. Gain of €1.8bn to €1.9bn, editors' calculation on the distribution of inheritances of the CAE (Council of Economic Analysis), static, with no behavioural effect. Not costed: the Dutreil brought down to 50% above €50m (share of transfers above this threshold not published), the end of the advantages of gifts before age 70, of split ownership and of the allowance specific to life insurance, and the taxation of unrealised capital gains above €4m. The €15bn announced concerns the whole reform, at cruising speed.+€1.9bnnew revenue, +€1.8bn to +€1.9bn
• Financial transaction taxFinancial transactions tax at 0.7%, broader base* Outside the range measured by the source: linear extrapolation (measured range: 0.3% to 0.4%). Assumption by the editors. Partial costing. Rate raised from 0.4% to 0.7%, costed at the yield of the last increase costed by the Senate finance committee (€600m for the move from 0.3% to 0.4%, i.e. €0.6bn per 0.1 point): beyond 0.4%, linear extrapolation, which ignores the fall in transaction volumes. Not costed: the extension of the base to other instruments, the lowering of the capitalisation threshold, European cooperation.+€1.8bn^{* (outside the range measured by the source: linear extrapolation)}new revenue
• Arduous workRestore the 4 arduousness criteria abolished in 2017Partial costing. Only the reinstatement of the 4 factors removed in 2017 is costed. The new criterion based on exposure to psychosocial risks and the possibility for each sector to replace the arduousness account with a list of arduous occupations entitling to early retirement (same proposal, p. 117) are not.−€1.7bnnew spending, −€2bn to −€0.3bn
• Personal housing benefits (APL)Housing benefit (APL) raised by 10%Assumption by the editors. Coefficient calculated by the editors from AN, rapport spécial PLF 2026, Logement et hébergement d'urgence: the position is marked as an assumption by the editors. −€1.6bnnew spending, −€1.7bn to −€1.5bn
• Violence against women and childrenComprehensive law against sexist and sexual violence funded with €3bnAssumption by the editors. Partial costing. Candidate’s amount, taken as is. The programme calls for a comprehensive law « dotée de 3 milliards d'euros » [“endowed with 3 billion euros”], without saying « par an » [“per year”] or « en plus » [“additional”]. Article 68 of the comprehensive law passed at first reading by the National Assembly (13 October 2026) schedules €3,000m of appropriations per year from 2027 to 2032: a total amount of appropriations, whereas existing spending on sexual and gender-based violence is €1.5bn according to the Government (€1bn from the State, €0.5bn from social security, sitting of 9 October 2026). Most cautious reading, the same for all: €3bn read as the total, i.e. €1.5bn per year more than current appropriations (partial costing). Read as a supplement, they would be worth €3bn; the Government estimates the cost of the law’s measures, in the long run, at between €3bn and €4bn on top of existing spending.−€1.5bnnew spending
• Income tax scaleDecrease for the lower brackets, 60% bracket above €250,000Assumption by the editors. Partial costing. Coefficient calculated by the editors from DGFiP, Impôt sur le revenu : nombre de foyers et montant de l'impôt net par taux marginal de taxation du barème, onglet 2024 (mise à jour du 7 avril 2026): the position is marked as an assumption by the editors. 60% bracket above €250,000 of taxable income per share (the scale applies per share), static: €1.0bn to €1.6bn, editors' calculation on the 61,476 households taxed at 45% (DGFiP (public finances directorate), 2024 incomes), in the absence of published data above €250,000. Not costed: the tax cut for the lowest brackets, with no amount; the overall cap on tax expenditures, with no amount (tax expenditures lever). High earners react to the marginal rate: static costing, upper bound.+€1.3bnnew revenue, +€1bn to +€1.6bn
• Family allowancesSame amount from the 1st child, return of the increase at age 14Partial costing. Only the return of the increase at age 14 is costed. The same amount from the first child is not costed for lack of a specified amount, nor is the simplification of the other allowances.−€1.3bnnew spending
• Flat-rate levy on capital (flat tax)CSG on capital income and flat tax raised by 3 points* Outside the range measured by the source: linear extrapolation (measured range: 12.8% only). Assumption by the editors. Partial costing. Coefficient calculated by the editors from Sénat, rapport général n° 108 (2017-2018) sur le PLF 2018, tome II, examen de l'article 11 (prélèvement forfaitaire unique): the position is marked as an assumption by the editors. Reading adopted: a single 3-point increase, the increase in the CSG on capital raising the flat tax (PFU) by the same amount (the CSG being part of it); two increases are not counted. The 3 points are costed on the flat tax base with the coefficient for its income tax component (€0.41bn per point, same base). The CSG increase on other capital income (property income, life insurance, real estate capital gains) is not costed: a position carries only one parameter, and the cost schedule gives about €2bn per CSG point on all capital income, flat tax included.+€1.2bn^{* (outside the range measured by the source: linear extrapolation)}new revenue, +€1.1bn to +€1.2bn
• Social housingSocial housing (HLM) relaunch: end of the RLS, reduced VAT on constructionPartial costing. The VAT cut on construction is not costed. Overlap with “Additional public investment, per year” examined: estimated common base zero, both amounts are added. The réduction de loyer de solidarité (solidarity rent reduction) is a rent cut imposed on social landlords, which the State recovers through lower APL; abolishing it is benefit spending (APL), not public investment.−€1.1bnnew spending, −€1.3bn to −€0.9bn
• Pupils per class, primary19 pupils per class in 2032, 12 in priority educationAssumption by the editors. Partial costing. Coefficient calculated by the editors from Sénat, rapport spécial PLF 2026, Enseignement scolaire (O. Paccaud): the position is marked as an assumption by the editors. Reading adopted: 19 pupils on average in primary school in 2032, as for Jean-Luc Mélenchon’s position. The “cap” reading and the 12 pupils in priority education are not costed.−€1.1bnnew spending, −€1.7bn to −€1.1bn
• Resources for the justice system9,000 judges, 10,000 court clerks and 7,000 attachés over ten yearsAssumption by the editors. Coefficient calculated by the editors from AN, rapport spécial PLF 2026, Justice: the position is marked as an assumption by the editors. 26,000 posts over ten years, half of them created by 2032.−€0.7bnnew spending, −€1.2bn to −€0.7bn
• Taxation of digital giants and multinationalsDigital services tax from 3% to 5%, extended to data centres* Outside the range measured by the source: linear extrapolation (measured range: 3% only). Assumption by the editors. Partial costing. Coefficient calculated by the editors from PLF 2027, Évaluation des voies et moyens, tome I (pages 24, 25, 29 et 30): the position is marked as an assumption by the editors. Digital services tax rate raised from 3% to 5%, costed at the average yield of one point at the current 3% rate (coefficient calculated by the editors from the PLF 2027 forecast): linear extrapolation, which ignores the reaction of the liable companies. Not costed: the liability threshold lowered to €5bn of turnover, the extension to e-commerce platforms and data centre operators, the strengthened reporting obligations.+€0.7bn^{* (outside the range measured by the source: linear extrapolation)}new revenue, +€0.6bn to +€0.7bn
• Prisons and sentences100% numerus clausus, moratorium on new prison places* Outside the range measured by the source: linear extrapolation (measured range: 0 to 15,000 places). Assumption by the editors. Partial costing. Coefficient calculated by the editors from AN, rapport spécial PLF 2026, Justice: the position is marked as an assumption by the editors. Reading adopted: the moratorium halts the places of the « 15 000 » plan on which work has not yet started (4,110 gross places at the study stage or with contracts awarded, Sénat, November 2025), i.e. about 3,000 net places at the ratio observed for the facilities delivered (73%, which removes replacement places), costed at the cost of one place (operation and investment); a decrease outside the measured range, linear extrapolation. Not costed: the occupancy cap at 100%, whose effect (about 21,700 prisoners above the number of places, to be given sentence adjustments) has no published estimate. Overlap with “Additional public investment, per year” examined: one is a cost, the other a gain: with no shares declared for this case, both amounts are added (the higher-amount rule only applies between two costs or two gains).+€0.3bn^{* (outside the range measured by the source: linear extrapolation)}saving, +€0.2bn to +€0.5bn
• University tuition feesAbolish the CVEC−€0.2bnrevenue cut
• Tax and social security expendituresTax expenditures capped, two homes at most; child maintenance tax-exemptPartial costing. Exemption of maintenance payments received for minor children, the payer keeping his or her deduction (the text does not abolish it): €121m a year according to the tax legislation directorate (direction de la législation fiscale) (2022 incomes). Not costed: the overall cap on tax expenditures per household, with no amount, and the limit of tax advantages to two dwellings, for want of a published breakdown of advantages by number of dwellings. Calculating property tax net of borrowing falls under the property tax lever.−€0.1bnrevenue cut
• High cost of living overseasAnti-trust action against monopolies, controlled margins, octroi de mer reformedPartial costing. Competition law and margin controls: regulatory measures. The reform of the quality-price shield and of the octroi de mer (dock dues) is not specified and is not costed.€0bnzero cost
• State grants to local authoritiesAbolish the DILICO levy on local authoritiesAssumption by the editors. The DILICO (local authority revenue smoothing mechanism) puts a share of local authorities' revenue in reserve (€740m in 2026, paid by the regions, the départements and the inter-municipal bodies), paid back 90% to the contributors in thirds over three years and 10% to the horizontal equalisation funds: everything returns to local authorities, and its abolition costs the State nothing in 2032 (if renewed, the year's levy is offset by the repayments from previous years; if not renewed, it no longer exists). The financial conference has no cost.€0bnzero cost
• Extra-billing by doctorsExtra fees capped at 50% of the tariff, banned for the lower-income halfPartial costing. Only the regulation is costed, at zero cost to public finances. The counterpart provided for in the same proposal (p. 111), « revaloriser le tarif de Sécurité Sociale des spécialités peu valorisées (psychiatrie, pédiatrie, gynécologie, dermatologie) » [“raise the social security tariff for undervalued specialities (psychiatry, paediatrics, gynaecology, dermatology)”], has a cost for national health insurance (Assurance maladie); it is not costed, for lack of an amount.€0bnzero cost
• Motorway concessionsGradually renationalise the motorwaysAssumption by the editors. Partial costing. Reading adopted: « progressivement » [“gradually”] means taking back each concession when it expires (2031-2036), under direct public management: zero static cost relative to the draft framework law on transport (toll revenue maintained). Other reading: early buyback of the concessions, €1.2bn to €1.6bn of interest a year (cost schedule option), the buyback price weighing on the debt. Same reading as for Jean-Luc Mélenchon and François Ruffin. Not costed: the investment in infrastructure that the takeover is to finance, with no amount.€0bnzero cost
• State medical aid (Aide médicale de l'État)Guarantee access to AME and PUMA, abolition of waiting periodsPartial costing. Access to the AME (state medical aid) is maintained (zero cost); the abolition of waiting periods (AME and universal health protection) is not costed, for lack of a source.€0bnzero cost
• Defence budget€6.4bn increase for defence in 2027, that of the current LPMAssumption by the editors. The dossier takes up the increase provided for by the updated military programming act (law of 16 August 2026: at least 2.5% of GDP in 2030), without its own level of effort: read as maintaining the current trajectory. The programme (pp. 66 and 193) provides for a new military programming act, with no amount.€0bnzero cost
• Support for UkraineLong-range weapons and air defence for UkraineAssumption by the editors. Military support with no amount or announced increase, for equipment that France already supplies (long-range missiles, air defence): read as maintaining aid at the current level.€0bnzero cost

Amounts announced by the candidate

The amounts that Marine Tondelier puts forward, with the direction given to them (“+”: gain for the public accounts, “−”: cost). They are never added to the totals or to each other. The gap with our calculation is only given when both cover the same scope: not for a position that is not costed, costed in part, based on an assumption by the editors, or whose amount is the candidate’s own, taken as is; and only when the announcement falls outside the range of our calculation (otherwise: “within the range”).

• LeverWhat the candidate saysAnnounced, per year
• Energy renovations per yearBudget 2027 dossier (26 August 2026, p. 8): « La rénovation des logements en augmentant MaPrimeRénov de +1,4 Md€ pour rétablir les 5 milliards prévus initialement en 2024 » [“Housing renovation by increasing MaPrimeRénov by +€1.4bn to restore the 5 billion initially planned in 2024”]Our calculation: −€11bn per year. No gap calculated: the announcement covers a different scope or a different year.−€1.4bncost to the public accounts
• Inheritance and gift taxBudget 2027 dossier (26 August 2026, p. 6): « A cette condition, le rendement pourrait atteindre 15 milliards d’euros par an en régime de croisière. » [“On this condition, the yield could reach 15 billion euros per year at cruising speed.”]Our calculation: +€1.9bn per year. No gap calculated: costing based on an assumption by the editors.+€15bngain for the public accounts
• Level of public spendingTable (p. 22) for 2026: « Transition écologique +9,1Md€, Réparation des services publics +6,1Md€, Effort de défense +6,2Md€, Contribution Union Européenne +5,7Md€, Protection sociale +6,3Md€, Recettes fiscales +28,5Md€, Maîtrise des dépenses -16,6Md€ » [“Ecological transition +€9.1bn, Repairing public services +€6.1bn, Defence effort +€6.2bn, European Union contribution +€5.7bn, Social protection +€6.3bn, Tax revenue +€28.5bn, Spending control -€16.6bn”]Position not costed. Amount to be read in the quote.—direction to be read in the quote
• Additional public investment, per yearBreakdown pp. 7-8: rail +€1bn, metropolitan RER €0.5bn, social leasing +€1bn, overseas territories +€0.4bn, biodiversity +€1bn, MaPrimeRénov' +€1.4bn, Fonds vert +€1.9bn, school buildings €2bn/year, civil protection €260mOur calculation: −€8.6bn per year. Candidate’s amount, taken as is.−€10bnnew spending
• RailBudget 2027 dossier (26 August 2026, p. 7): « la régénération et la modernisation du réseau de transport ferroviaire (+1Md€) » [“the regeneration and modernisation of the rail transport network (+€1bn)”]Position not costed. Amount to be read in the quote.—direction to be read in the quote
• Violence against women and childrenProgramme, p. 133: « Adopter une loi intégrale pour lutter contre les violences sexistes et sexuelles (VSS) dotée de 3 milliards d’euros. » [“Adopt a comprehensive law to combat gender-based and sexual violence (VSS), endowed with 3 billion euros.”]Our calculation: −€1.5bn per year. Candidate’s amount, taken as is.−€3bncost to the public accounts
• Adaptation to climate change« Budget 2027 » dossier, p. 8, section « Financer l'adaptation par un plan d'urgence »: Fonds vert « (+1,9 Md€) » [“(+€1.9bn)”], renovation of school buildings « 2 milliards d’euros par an » [“2 billion euros per year”] from 2027, firefighting air fleets « 260 millions d'euros » [“260 million euros”].Position not costed. Quoted, not counted.−€4.2bncost to the public accounts

What this costing does not measure

• Growth and employmentThe effect of the measures on activity and employment, and the revenue gained or lost as a result; it depends on the economic model used: models and feedback effects.
• Education, health, researchLong-term effects on human capital, productivity and innovation.
• InequalityThe distribution of gains and losses across the whole population, beyond the eight model households.
• Public servicesWhat new spending or savings change in the quality of the services provided and in access to them.
• Environment and climateEffects on emissions, natural resources and adaptation to climate change.
• Sovereignty and external dependenceEffects on the country’s dependence on other countries: energy, industry, defence, supplies.
• Interest rates on the debtThe effect of a programme on the terms on which the State borrows, which also depend on the make-up of the holders of its debt.