Workers attend performances four times less often than managers, and cultural policy changes nothing about it

France spends more than twenty billion euros in public funds each year on cultural policy. Yet a manual worker has three to four times fewer chances than a senior manager of attending a live performance in a year [6]. The choice to be made in 2027 is clear: redirect part of this budget toward human mediation and early education, the only levers that have shown an ability to durably change who gets access to what.

Half a century of measurement, a gap that won’t budge

The Ministry of Culture’s survey on cultural practices has existed since the early 1970s. Its sixth edition dates from 2018 [11]. Over this long period, the data confirms a rise in practices across most cultural sectors. That is the average.

Behind this overall progress, gaps between social categories have resisted every policy pursued since André Malraux. In 2022, 40 percent of people aged 16 or older had been to the cinema at least once [6]. Managers go 64 percent of the time, compared to 30 percent of manual workers [6]. For live performances, the disparity is even more pronounced: managers report attending nearly four times as often as manual workers [6].

Income gaps follow the same pattern. In 2022, 68 percent of high-income people reported going to the cinema in the twelve months before the survey, compared to 47 percent of low-income people [6]. For theater, 38 percent of high earners attended, against 14 percent of modest-income people [6].

Digital technology has partially narrowed certain gaps in home music or audiovisual consumption. But physical outings—theater, concerts, museums—remain a stable class marker over half a century [4]. For museum visits outside heritage sites, the gap reaches 20 percent for manual workers versus 67 percent for managers in 2020 [4]. Level of education is more discriminating than level of income. Economic capital opens the doors; cultural capital passed down through the family decides whether you step inside.

This mechanism—the intergenerational transmission of taste and familiarity with cultural institutions—is what Pierre Bourdieu analyzed in Distinction (1979). Cultural policy pursued since then has not changed this mechanism. It has sometimes gone around it, never dismantled it.

Three hundred twenty-four million euros to confirm what already exists

The Pass Culture illustrates the trap of the pricing lever. Generalized in 2021 to all eighteen-year-olds, extended in 2022 to fifteen- to seventeen-year-olds, it absorbs 324 million euros in budgeted spending for 2024, of which 244 million for the individual share [8].

The quantitative results are real. By late August 2024, 84 percent of eighteen-year-olds use the program, and 4.2 million young people have registered since 2019 [7]. But that is where the good news stops.

In 2024, 16 percent of potential beneficiaries—precisely the audiences least familiar with cultural practices—did not join the program [7]. Adding the 9 percent who downloaded the app without using it, a quarter of young people did not benefit from the Pass [7]. This missing quarter is precisely the target audience.

The Court of Audit formulates the diagnosis in its first assessment from December 2024 [7]. The individual share rests on young people’s freedom of choice. It does not counteract preexisting structural inequalities. The program accompanies young people already familiar with cultural practices. For those most distant from culture, it does not constitute a point of entry.

When credits expired in 2024, only 38 percent of users continued the activities they had discovered. 37 percent continued visiting the venues the Pass had introduced them to [7]. The financial lever does not hold without human mediation.

Too-high ticket prices are cited as a barrier by 54 percent of people who did not go to the cinema in 2024, and by 53 percent of those who did not attend concerts [7]. But the second barrier is lack of interest. That lack of interest, built over time by the absence of early familiarity, is not transformed by three hundred euros on an app.

The budget structure of the ministry amplifies the problem. The Ministry of Culture’s budget totals four billion euros for 2024 [8]. Roughly 35 percent of this envelope funds national institutions, and nearly half of those subsidies concentrate on six establishments: the National Library of France, Paris National Opera, Universcience, INRAP, the Louvre Museum, and the Centre Pompidou [8]. The audiences of these six institutions are overwhelmingly educated and affluent.

For 2026, budget forecasts announce cuts across all programs, from -1.9 percent for Press and Media to -9.1 percent for Heritage [10]. Pass Culture, meanwhile, absorbs a growing share of a budget under pressure.

In 2024, cultural spending by local authorities reached ten billion euros, of which 7.7 billion went to operations and 2.3 billion to investment [2]. From 2023 to 2024, this increased 0.4 percent in constant euros [2]. That is near stagnation, while several live performance unions report closures of subsidized venues.

Algorithms reproduce what they claim to open

The deepest structural break of the decade is architectural, not budgetary. In 2023, 82 percent of French people over 15 connect to the internet daily [3]. 87 percent of internet users consume cultural goods online [3]. Films lead at 55 percent of reported consumption, followed by music at 50 percent and TV series at 49 percent [3]. Digital culture is now the majority in terms of time spent.

Multiplying access does not multiply real choices. Algorithms used by major platforms obey a simple logic: maximize time spent on screen. This objective produces homogenization of recommendations and renders singular content invisible. On audiovisual platforms in Europe, American works occupy 61.2 percent of total viewing time in 2024 [16]. The apparent diversity of catalogs masks real concentration of attention on a limited number of titles.

Applications are designed according to well-tested neuropsychological principles: infinite scroll, constant notifications, randomized reward systems. Bruno Patino extends this diagnosis in The Age of Human Obsolescence (2026): algorithms optimize user behavior without their knowing it [12]. The expansion of access through digital means accompanies a reduction in the real diversity of consumed content. It also captures available mental bandwidth, which directly competes with active cultural practices: reading, attendance at performances, museum visits.

The platformization of culture captures the benefits of expanded access for the benefit of actors who control algorithms and the attention economy. Producers of local creation and local institutions lose visibility and resources. Daron Acemoglu and Simon Johnson formalize this mechanism in Power and Progress (2023): the powerful often use technological change to their own advantage [13]. Stéphane Grumbach observes that cultural recommendation thus escapes democratic institutions to fall under the control of a few private operators whose interests do not align with cultural diversity [14].

Current cultural policy treats these platforms as an additional distribution channel. They have become the dominant system of cultural recommendation. Their recommendation logic amplifies existing preferences rather than broadens them.

Redirect part of the budget for price compensation toward mediation and early education

For thirty years, French cultural policy has massively compensated through funding what it has not changed in transmission mechanisms. It has subsidized supply without altering demand. It has lowered prices without transforming early familiarity with works. Pass Culture is the most recent and most expensive example of this logic.

The main choice for 2027 is a reallocation. The idea is to move part of the credits currently allocated to price-compensation tools toward human mediation and early arts education. This transfer does not require a net increase in spending. The ten billion from local authorities and four billion from the ministry represent a considerable envelope [2, 8]. The constraint is first about allocation efficiency, not volume.

Three concrete reconfigurations follow from this.

The first consists in making arts and cultural education a structural priority of primary school. Familiarity with symbolic forms must begin in kindergarten to be effective. Interventions of long duration in primary school, measured by regular assessments by the Directorate for Evaluation, Foresight and Performance of the Ministry of National Education (DEPP), must benefit from a non-compressible share of the arts and cultural education budget. Middle school activities arrive too late when distance from the cultural world is already established.

The second reconfiguration concerns Pass Culture itself. The Court of Audit recommends reducing the amount of individual credit or targeting it to scholarship recipients and rural or priority neighborhoods [7]. The deeper reform would be to replace the model of free allocation with an accompanied discovery path. This path would mobilize local cultural actors as mediators, following the model of the collective share that involves teachers. In zones without cultural infrastructure or reliable digital access, the Pass loses all relevance without this prior work on access to venues.

The third reconfiguration concerns platforms. French law already requires 40 percent French works on video-on-demand services [16]. Imposing discoverability obligations—that is, a percentage of recommendations aimed at works outside dominant algorithmic trends—and metrics for exposure to linguistic and stylistic diversity under oversight of the French Authority for Audiovisual and Digital Communication Regulation (ARCOM) would extend this logic. The European Commission has identified this lever in recent work: editorialization of recommendations, support for translation, consolidation of metadata as levers for real diversity [16].

This transfer has real political cost. Cultural industries that currently benefit from the flow of consumption generated by Pass Culture will resist. Major national institutions whose appropriations form the heart of the ministry’s budget will object.

The data is there. In 2024, cultural, sporting, and leisure spending represents 7.0 percent of household spending in France, compared to 7.5 percent for the European Union as a whole [9]. Overall volume is not the problem. What is consumed, and by whom, is.

Sources

[2] DEPS, Ministry of Culture, “Dashboard of Cultural Spending by Local Authorities, 2019–2024,” 2026, https://www.culture.gouv.fr/espace-documentation/statistiques-ministerielles-de-la-culture2/statistiques-culturelles/tableau-de-bord-des-depenses-culturelles-des-collectivites-territoriales-2019-2024 (accessed 24/08/2026).

[3] DEPS, Ministry of Culture, “Key Figures, Statistics on Culture and Communication 2024,” 2025, https://www.culture.gouv.fr/espace-documentation/statistiques-ministerielles-de-la-culture2/publications/collections-d-ouvrages/chiffres-cles-statistiques-de-la-culture-et-de-la-communication-2012-2025/chiffres-cles-2024-de-la-culture-et-de-la-communication (accessed 24/08/2026).

[4] DEPS, Ministry of Culture, Philippe Lombardo and Loup Wolff, “Fifty Years of Cultural Practices in France,” Culture Études no. 2020-2, July 2020, https://ses.ens-lyon.fr/actualites/rapports-etudes-et-4-pages/cinquante-ans-de-pratiques-culturelles-en-france-deps-juillet-2020 (accessed 24/08/2026).

[6] Observatory of Inequality, “Cultural Outings Differ by Social Category and Income,” 2024 data, https://www.inegalites.fr/Les-sorties-culturelles-different-selon-les-categories-sociales-et-les-revenus (accessed 24/08/2026).

[7] Court of Audit, “First Assessment of Pass Culture,” thematic public report, December 2024, https://www.ccomptes.fr/fr/publications/premier-bilan-du-pass-culture (accessed 24/08/2026).

[8] Court of Audit, “Analysis of 2024 Budget Execution, Culture Mission,” April 2025, https://www.ccomptes.fr/sites/default/files/2025-04/NEB-2024-Culture.pdf (accessed 24/08/2026).

[9] Eurostat / INSEE, “Cultural, Sports and Leisure Spending in the European Union,” extraction June 2026, https://www.insee.fr/fr/statistiques/5350143 (accessed 24/08/2026).

[10] Senate, report on the Finance Bill for 2025, Culture program, https://www.senat.fr/rap/l24-144-38/l24-144-38_mono.html (accessed 24/08/2026).

[11] DEPS, Ministry of Culture, “Survey on the Cultural Practices of the French, 2018,” PROGEDO-ADISP, https://data.progedo.fr/studies/doi/10.13144/lil-1511 (accessed 24/08/2026).

[12] Bruno Patino, The Age of Human Obsolescence, 2026: JdP corpus.

[13] Daron Acemoglu and Simon Johnson, Power and Progress: Our Thousand-Year Struggle over Technology and Prosperity, Basic Books, 2023 (French trans. Pouvoir et progrès, Pearson France, 2024): JdP corpus.

[14] Stéphane Grumbach, “Digital Growth, Energy Needs, Environmental Impacts and Security Concerns,” 2024, https://shs.cairn.info/publications-de-stephane-grumbach–97659 (accessed 24/08/2026).

[16] Actualitté, “Facing Algorithms and AI, Protecting the Diversity of Cultural Expressions,” 2026, https://actualitte.com/article/133408/politique-publique/face-aux-algorithmes-et-a-l-ia-proteger-la-diversite-des-expressions-culturelles (accessed 24/08/2026).