The 2027 presidential campaign is opening. Until the first round, it will dominate public space: polls, tactics, sound bites. Meanwhile, six transformations are reconfiguring French life, treated at the margins of debate as in previous campaigns.
This newspaper is taking the opposite approach. Each day until election day, an article will measure, backed by data, where one aspect of French society stands. What rupture is running through it, and what long-term choice it poses. The thread over 228 days: what a program guided by progress rather than conservatism would see, and see first.
A Democracy Desired, Institutions Discredited
In 2026, 22% of French people trust the presidential institution [21]. 20% trust the National Assembly, and 26% trust the Senate [21]. Since the Political Trust Barometer was created in 2009, CEVIPOF has never recorded figures this low. Attachment to democracy remains massive: 82% of respondents [22]. It is the perceived ineffectiveness of institutions that eats away at the legitimacy of any long-term reform.
This discredit has a precise cause: a decade during which politics treated them superficially. These transformations form a single system, where each piece reinforces the others.
Time in life is distributed more unequally. The productive machine is shrinking, public accounts are jamming. Artificial intelligence accelerates everything, climate sets the physical constraint, the world is closing in. Reading them separately amounts to missing all of them. The data shows what this gap costs, and why closing it is the central question on the horizon of 2027.
The Share of Life Expectancy Is Growing More Unequal
It all starts with the social body, and its most brutal measure: the number of years each person can expect to live. The poorest 5% and the wealthiest 5% do not live the same number of years [1]. The gap in life expectancy at birth reaches 9 years for women and 13 years for men [1]. Women among the wealthiest 5% live on average 17 years longer than men among the poorest 5% [1]. And this gap widened further between the periods 2012-2016 and 2020-2024 [1].
Aging amplifies this divide at the collective level. People aged 60 and over now represent 27.7% of the French population [9]. They accounted for 19.6% in 1994 [9]. The share of people aged 20-64 is projected to fall to 51.7% in 2070 under INSEE’s central scenario [10]. It stands at 55.3% in 2026 [10].
The number of dependent elderly people relative to people of working age would rise from 0.39 in 2023 to 0.52 in 2060 [9]. More care, dependency and pensions to finance, with a proportionally smaller working population.
These life expectancy inequalities result above all from inequalities in living and working conditions, more than from access to care [1][2]. The strain and exposure to physical and psychological risks, accumulated over an entire working life, accounts for most of the gap. Bruno Palier documents this mechanism as the neglected adjustment variable in social policies [30]. First loop of the system: repairing this divide requires massive health and dependency resources, at the precise moment when the productive machine financing them is losing ground.
The Productive Machine Shrinks While Reindustrialization Is on Display
Years of stated priority and 54 billion euros from the France 2030 program over five years have not reversed the trend. Industry accounts for less than 10% of French GDP in 2025 [16]. The balance of industrial sites has turned negative, with 65 net closures in 2025 [17]. France had recorded more than 100 net creations per year between 2020 and 2023 [17]. Nearly 20,000 industrial jobs disappeared in 2025, when the country was gaining 15,000 on average between 2021 and 2023 [17].
The causal diagnosis is known. Most of the shock stems from high interest rates and energy costs. This was established by the government task force on reindustrialization by 2035, led by Olivier Lluansi [16]. Before the invasion of Ukraine, European industry ran on Russian gas at around 20 euros per megawatt-hour; it has paid at global rates since.
France remains Europe’s least industrially dense country in creating wealth [19]. It has 200 robots per 10,000 industrial employees in 2024 [20]. Germany has 415, South Korea 1,012 [20].
Second loop: without an expanded productive base, the social model is financed by a shrinking economic fabric, as shown by public accounts.
Accounts Are Jamming: Record Debt, Inequalities at Their Highest
Public debt reached 3,460.5 billion euros in 2025, or 115.7% of GDP, exceeding the peak of 2020 touched during the height of the health crisis [12]. The Court of Accounts projects interest expenses that could exceed 100 billion euros by 2029, so much less room to finance national priorities [13]. France is the only country in Europe whose public finances continue to deteriorate over this period [12].
At the same time, income inequalities reached their highest level in thirty years in 2024 [15]. The mechanism is identified: primary inequalities, before redistribution, are among Europe’s highest. The Gini coefficient, which measures income inequality between households, reaches 37.0 in 2024 [14]. This level is partly due to an employment rate below the European average.
Redistributing more on a shrinking base produces less and less effect. Expanding the employment base and better remunerating work at its point of production changes the game. The distribution of progress’s fruits happens before redistribution. It happens in the structure of markets, governance of innovation, and labor rules. This is what Daron Acemoglu and Simon Johnson document [25].
This is where artificial intelligence tips the equation, both ways at once.
AI Accelerates Everything, Including the Shrinking Base
The first signal is already visible in the United States: 14% fewer new hires among 22-25-year-olds in jobs heavily exposed to AI in 2024 [25]. France has 200 robots per 10,000 industrial employees in 2024 [20]. Germany has 415, South Korea 1,012 [20]. This lag temporarily protects it, but permanently weakens it: the wave will hit a fabric with less prepared qualifications.
The double shock is formalized. AI is a general-purpose technology that modifies all production processes. And creative destruction is not guaranteed if gains concentrate at the top, analyzes Axelle Arquié [26]. Demography alone, at constant productivity for the period 2006-2019, would cause GDP per capita growth to fall [26]. From 0.72% to 0.48% per year.
Third loop: if AI amplifies productivity gains without a sharing policy, pressure on labor income intensifies. At the precise moment when financing the welfare state requires more of a salary base.
While this mechanism takes hold, two constraints bound the field of possibilities. The first is physical.
Climate Sets the Physical Constraint for the Coming Decades
Warming attributed to human activities reaches +1.3°C globally and +1.9°C for mainland France compared to the pre-industrial period [4]. Heat wave days have been multiplied by six compared to 1961-1990 [5]. We used to see 2 per year, now we see 13 [5].
The reference trajectory adopted by the state, TRACC, projects 2.7°C in 2050 for the mainland [4]. And 4°C in 2100 [4]. At this level, heat waves, droughts, forest fires, and floods become durably more frequent and intense.
Decarbonization, meanwhile, is slowing where political consensus believed the course was being held. CITEPA measures a 3.0% drop in emissions in 2024 and estimates 2.1% in 2025 [6]. The target of 270 Mt CO₂e in 2030 requires 4.7% reduction per year on average [7].
The good results of 2022-2023 came from increased nuclear production. Outside energy production, the pace slowed very significantly [8]. The High Council for Climate calls for relaunching decarbonization policies [7].
The bottleneck is known. Transport is the leading national emitter: 124.9 Mt CO₂e in 2024, or 34% of national emissions [6]. It will need to fall on average four times faster than between 2023 and 2024 to meet the 2030 target [7].
Buildings and agriculture follow similar trajectories. France said it wanted to decarbonize; the sectors that policy has not yet touched are heating and daily mobility for modest-income households. Fourth loop: this transition requires massive investment in a budget already under pressure, at the moment when the outside world imposes its bill.
The World Is Closing In and Autonomy Has a Price
At the NATO summit in The Hague in 2025, Allies committed to investing 5% of GDP in defense by 2035 [23]. The updated July 2025 National Strategic Review marks entry into a new strategic era, marked by increased risk of high-intensity warfare on the continent. France revised its 2026 military programming law: approximately 2.3% of GDP in 2027, 2.5% in 2030 [24]. The gap with the Alliance’s objective remains considerable, as the debt burden already exceeds 65 billion euros in 2025 [13].
Autocracies now form a semi-organized network that cooperates across ideologies to undermine the liberal world order. Anne Applebaum documented this in Autocracy, Inc. The post-Western world learns to compose with China while Europeans begin to fear America. Acting as autonomous strategic players, without counting on the American umbrella, becomes the elementary condition of sovereignty. Fifth loop: this autonomy costs, and it enters into direct competition with social spending in a budget under maximum pressure. Getting out of it requires explicit prioritization and an industrial defense strategy integrated with reindustrialization.
One Equation Alone: Who Pays, on What Base
The system is closed. Climate, technological innovations, aging, labor transformation, social balance, and geopolitical closure reinforce each other. An industrially weakened country cannot finance both aging, climate transition, rearmament, and technological competition. A country where 78% of citizens do not trust politics in 2026 [21] cannot conduct the long-term reforms these transformations require. All the loops converge on the same point of fragility: the architecture of social financing, inherited from a world where wealth was created primarily by human labor.
This assumption from the 1950s has been false for a long time, and AI will invalidate it faster still. When a company replaces ten employees with an automated system, its contribution to Social Security financing drops. If productivity gains concentrate at the top without a sharing mechanism, the contribution base shrinks at the precise moment when social needs—aging, retraining, health—increase [26]. The problem is architectural. It calls for an answer at the same level.
Change Who Pays When the Machine Works
The choice this newspaper will put up for debate is this: base employer contributions on the company’s added value rather than on its salary bill alone. A company that creates wealth contributes, whether it employs one hundred people or ten robots. Jurisdiction over rates and bases remains national, but the rules of the game change: reducing social contributions by substituting capital for labor ceases to be possible.
The trade-off is real, and this series will document it. Labor-intensive sectors—personal services, restaurants, crafts—would see their burden lightened where they employ many for little added value per person. Sectors with high capital and technology intensity—energy, digital platforms, automated industry—would contribute more. This shift finances aging and retraining without increasing debt.
Daron Acemoglu sets the condition for progress that benefits everyone: that citizens and institutions weigh on its orientations [25]. Changing who pays when the machine works is the most direct form of exercising this weight. This choice closes the possibility, open today, of collectively enriching ourselves from automation while shifting onto society the human cost it produces. That is what a progress-oriented program puts on the table, and what a campaign occupied elsewhere will leave in the blind spot.
Seeing Clearly, to Be Able to Choose: The Method of 228 Days
This series will adhere to one simple discipline. Each article will trace back to the causal mechanism before naming the lever. It will anchor itself in what primary sources publish: INSEE, Météo-France, CITEPA, the Court of Accounts, COR, OECD, CEVIPOF. These articles will also draw on the work that has built the frameworks to escape shortcuts: Olivier Lluansi on industry, Bruno Palier on work quality [30]. To these add Daron Acemoglu and Simon Johnson on the sharing of technological gains [25], and Hippolyte d’Albis on the economics of life stages [31].
The path is traced. Four weeks of state-of-the-art assessment, sector by sector, against the double reflex of “everything is collapsing” and “everything is fine.” Then the trajectories, read for each: what improves, what deteriorates, and for whom; then the money, the constraint, the choice, in the midst of budget winter. Then the levers, what works elsewhere and what could work here. Finally, as election day approaches, what is at stake.
Tomorrow, this newspaper opens the project with the living standards of French people, from the most modest to the most affluent. The day before the first round, it will close this series with what you need to know before voting, on the substance, without naming a candidate or dictating how to vote. Between the two, one article per day to map the real country. Here is the invitation: read France while the campaign talks about something else, and come to the vote armed with data rather than spectacle.
Sources
[1] INSEE, “From 2012-2016 to 2020-2024, the gap in life expectancy between modest and wealthy people increased,” INSEE Première no. 2085, December 2025, https://www.insee.fr/fr/statistiques/8679561 (accessed 15/08/2026).
[2] INSEE, “Life expectancy gaps between executives and workers: 5 years for men, 3 years for women,” INSEE Première no. 2005, July 2024, https://www.insee.fr/fr/statistiques/8220688?sommaire=8208818 (accessed 15/08/2026).
[3] INSEE-DGFiP-Cnaf-Cnav-CCMSA, “Living standards and inequality indicators,” Tax and Social Income Surveys, 2024, https://www.insee.fr/fr/statistiques/2491918 (accessed 15/08/2026).
[4] Météo-France, “Climate change in France: key figures in 2024,” 2025, https://meteofrance.com/le-changement-climatique/quel-climat-futur/changement-climatique-en-france-les-chiffres-cles-en (accessed 15/08/2026).
[5] SDES-Météo-France, Key Climate Figures France, Europe and World, 2025 edition, SDES, November 2025, https://www.statistiques.developpement-durable.gouv.fr/edition-numerique/chiffres-cles-du-climat (accessed 15/08/2026).
[6] CITEPA, Secten Report 2026 edition, June 2026, https://www.citepa.org/le-rapport-secten-edition-2026-est-en-ligne/ (accessed 15/08/2026).
[7] CITEPA / High Council for the Climate, “France’s decarbonization trajectory experienced a shift in 2024 that is incompatible with the goal of carbon neutrality by 2050,” July 2025, https://www.citepa.org/en/frances-decarbonization-trajectory-experienced-a-shift-in-2024-that-is-incompatible-with-the-goal-of-carbon-neutrality-by-2050-high-council-for-the-climate/ (accessed 15/08/2026).
[8] Ministry of Ecology, “The decline in greenhouse gas emissions continues in 2024 and 2025,” June 2026, https://www.ecologie.gouv.fr/presse/baisse-emissions-gaz-effet-serre-se-poursuit-2024-2025-france-doit-accelerer-sa (accessed 15/08/2026).
[9] Social Security Assessment, “Population Aging,” 2024 data, https://evaluation.securite-sociale.fr/home/retraite/1-5-vieillissement-de-la-populat.html (accessed 15/08/2026).
[10] INSEE, population projections 2021-2070, central scenario, cited by Crédit Agricole Economic Studies, June 2026, https://etudes-economiques.credit-agricole.com/fr/france-que-retenir-des-nouvelles-projections-de-population-de-linsee (accessed 15/08/2026).
[11] OECD, Employment Outlook 2025: France Country Notes, July 2025, https://www.oecd.org/en/publications/2025/07/oecd-employment-outlook-2025-country-notes_5f33b4c5/france_a4e153d7.html (accessed 15/08/2026).
[12] Court of Accounts, The State of Public Finances at the Start of 2026, February 2026, https://www.ccomptes.fr/fr/publications/la-situation-des-finances-publiques-debut-2026 (accessed 15/08/2026).
[13] Court of Accounts, The State and Outlook of Public Finances, July 2026, https://www.ccomptes.fr/sites/default/files/2025-07/20250702-synthese-RSPFP-2025.pdf (accessed 15/08/2026).
[14] FIPECO, “Income Inequality and Redistribution in 2024 in France and the European Union,” March 2026, https://www.fipeco.fr/commentaire/Les%2520in%C3%A9galit%C3%A9s%2520et%2520la%2520redistribution%2520des%2520revenus%2520en%25202024%2520en%2520France%2520et%2520dans%2520l%27Union%2520europ%C3%A9enne (accessed 15/08/2026).
[15] Inequality Observatory, “Income Inequality at Its Highest in Thirty Years,” 2026, https://www.inegalites.fr/spip.php?amp=&article1526=&id_mot=15 (accessed 15/08/2026).
[16] Montaigne Institute / Olivier Lluansi, “Reindustrialize: a Challenge, Solutions… and a Choice for Society,” April 2026, https://www.institutmontaigne.org/expressions/reindustrialiser-un-defi-des-solutions-et-un-choix-de-societe (accessed 15/08/2026).
[17] FO Metals / Olivier Lluansi, “Reindustrialization of France: Olivier Lluansi’s Unsparing Diagnosis,” April 2026, https://www.fo-metaux.fr/actualite/i/95835327/ (accessed 15/08/2026).
[18] Rexecode / METI / Olivier Lluansi, “Fifteen Competitiveness Levers to Reindustrialize France,” June 2025, https://www.rexecode.fr/conjoncture-previsions/veille-documentaire/document-de-la-semaine/quinze-leviers-de-competitivite-pour-reindustrialiser-la-france-meti (accessed 15/08/2026).
[19] Olivier Lluansi, “Reindustrialize to Live Better in France,” Futuribles, no. 465, 2025, https://shs.cairn.info/revue-futuribles-2025-2-page-39 (accessed 15/08/2026).
[20] IFR (International Federation of Robotics), robotization data by country, 2024, cited by dettedelafrance.fr (accessed 15/08/2026).
[21] CEVIPOF, Political Trust Barometer, Wave 17, OpinionWay, February 2026, https://www.sciencespo.fr/cevipof/fr/actualites/barometre-de-la-confiance-politique-cevipof-2026 (accessed 15/08/2026).
[22] CEVIPOF, Political Trust Barometer, Wave 17: Decryptions, February 2026, https://www.sciencespo.fr/cevipof/sites/sciencespo.fr.cevipof/files/BCP2026_Decryptages_fev2026.pdf (accessed 15/08/2026).
[23] NATO, “Defense Spending and the 5% Commitment,” 2025 Annual Report, https://www.nato.int/en/what-we-do/introduction-to-nato/defence-expenditures-and-natos-5-commitment (accessed 15/08/2026).
[24] Senate, Report on the 2026 Finance Bill: Defense Mission, 2025, https://www.senat.fr/rap/l25-139-38/l25-139-38_mono.html (accessed 15/08/2026).
[25] Daron Acemoglu and Simon Johnson, “Can A.I. Be Pro-Worker?”, The New Yorker, 2026 (accessed 15/08/2026).
[26] Axelle Arquié, “The Double Shock of AI: Employment and Taxation,” 2026, https://shs.cairn.info/publications-de-axelle-arquie–111394 (accessed 15/08/2026).
[27] Carl Benedikt Frey, How Progress Ends: Technology, Innovation, and the Fate of Nations, Princeton University Press, 2025, https://press.princeton.edu/books/hardcover/9780691232218/how-progress-ends (accessed 15/08/2026).
[28] Anaïs Voy-Gillis, For an Industrial Revolution, Presses de la Cité, 2025; Olivier Lluansi and Anaïs Voy-Gillis, Toward Industrial Renaissance, Lignes de Repères, 2020 (accessed 15/08/2026).
[29] Nicolas Dufourcq, France’s Social Debt, Odile Jacob, 2025 (accessed 15/08/2026).
[30] Bruno Palier, work on work quality and European social policies (corpus, accessed 15/08/2026).
[31] Hippolyte d’Albis, economics of life stages and demography, work 2024-2025 (corpus, accessed 15/08/2026).
