Deindustrialisation, weakened export performance, and the vulnerability of economies to shocks in value chains underscore the case for strengthening industrial policy. Digital technologies, together with geopolitical and health crises, are increasing economies’ exposure to market concentration and supply chain vulnerability, while the climate transition requires investment, innovation and co-ordinated public policies. Against a backdrop of budgetary constraints, it is important to assess the effectiveness of existing support for firms (e.g. grants, taxes, regulations, repayable advances and other financial measures) and reallocate resources where they can be most effective at achieving competitiveness and resilience objectives. Enhancing France’s industrial policy calls for actions in four priority areas: i) maintaining price competitiveness through a better allocation of tax breaks and taxes; ii) strengthening industrial innovation and non-cost competitiveness through increased support for emerging technologies and disruptive innovation; iii) ensuring resilience to economic and geopolitical shocks by diversifying value chains, developing digital infrastructure and supporting decarbonisation while preserving competitiveness; and iv) strengthening human capital and its alignment with strategic needs.
OECD Economic Surveys: France 2026
Abstract
The increasing focus on industrial policy in recent years stems from concerns that economies are seeing their competitiveness weakened by what is seen as unfair competition, along with evidence of their increased vulnerability to external shocks. In particular, geopolitical and health crises have highlighted the vulnerability of value chains and the risk of over-reliance on a small number of suppliers and production areas for certain products. At the same time, the rise of digital technologies, as characterised by increasing returns and network effects, is likely to accentuate the risk of winner-takes-all outcomes. The implicit or explicit subsidies granted by some countries to their industries carry the same risk. Lastly, climate issues are imposing a rapid development of decarbonisation and adaptation technologies that require massive investments, a faster pace of innovation and better coordination of public policies.
Against this backdrop, many countries have implemented targeted strategic plans aimed at supporting innovation, securing value chains and strengthening competitiveness and economic resilience (for example, Made in China 2025 in China; Society 5.0 in Japan; K-Strategy in South Korea; the CHIPS Act in the United States; the European Chips Act and European Industrial Strategy in Europe). France offers some of the highest levels of public support to businesses. Industrial policy expenditure in France, defined as all direct public sector support granted to businesses with a view to promoting their economic development, including tax expenditures, grants and financing aid, was estimated at nearly 3% of GDP in 2023 compared to an average of 2.5% of GDP in the 17 OECD countries covered by the OECD industrial policy database (OECD, 2023[1]). Faced with the challenges of fiscal consolidation and the substantial financing required to meet the challenges of developing and scaling new technologies, it is essential to assess the cost-effectiveness of these measures and direct resources towards the most efficient and promising applications for the future of French industry.
The first section examines the challenges facing French industry, reviewing the causes of the decline in manufacturing and the main challenges for public policy. It also sets out the objectives, mechanisms, and scope of industrial policy examined in this chapter. The second section reviews policies designed to support cost competitiveness. The third section analyses policies aimed at strengthening non-cost competitiveness, focusing on innovation and barriers to technology adoption. The fourth section examines policies aimed at strengthening resilience to shocks, focusing on three major issues: securing value chains, capacity building in artificial intelligence, and the challenges for industrial competitiveness posed by the decarbonisation process. The last section analyses the challenges related to training and skills matching.
4.2.1. Responding to industrial policy challenges: rationale for policy action
Industrial policy has many objectives, which are traditionally linked to growth, competitiveness and economic sovereignty (Box 4.1). Nowadays, there are additional considerations related to value chain resilience, technological security, and the green transition (Millot and Rawdanowicz, 2024[2]; Criscuolo et al., 2022[3]). In sectors characterised by economies of scale, learning effects, and network effects, the advantages tend to lie with established producers, which can limit entry and increase technological dependence. Industrial policy therefore aims to strengthen or establish new comparative advantages, i.e. internationally competitive domestic capabilities. Public support for emerging technologies, such as artificial intelligence or biotechnology, is fully consistent with this approach of strengthening competitiveness.
In some cases, economic competitiveness may take a back seat, when the main objective is to strengthen resilience and achieve strategic goals. COVID-19 and Russia’s war of aggression against Ukraine have demonstrated the French economy’s vulnerability to shocks to global value chains, particularly regarding energy and strategic goods. Industrial policy therefore seeks to secure production capacities and diversify supply sources to address risks of disruption that are not factored in by the market, without necessarily achieving maximum efficiency. Similarly, decarbonisation policies may generate higher short-term production costs, which are unavoidable when viewed in the context of objectives to reduce emissions or adapt to climate change. Long-term public support may in that context be necessary, including in the form of regulation (e.g. the Carbon Border Adjustment Mechanism).
The ways in which industrial policy in France is implemented have changed over time (Box 4.1), reflecting a growing recognition of the need to combine several instruments and limit the discretionary selection of beneficiaries. In this sense, it contrasts with more traditional industrial policies, which are based mainly on targeted policies for specific sectors and, in some cases, on direct support for specific companies (the "choosing champions" approach). In contrast, direct aid is now generally conditional on specific technological objectives, supported by horizontal aid, i.e. without sector-specific criteria, aimed at correcting market failures (such as knowledge externalities, financial market imperfections, and coordination failures) and is governed by rules on open competition between projects (Tirole, 2015[4]).
Traditional industrial policy, implemented in the aftermath of the Second World War and through to the 1980s, was interventionist and aimed primarily at building a national industrial base centred on industrial champions. In emerging countries, it sought to shift productive specialisations when static comparative advantages led to a concentration in low-tech activities with limited returns (Hausmann, Hwang and Rodrik, 2006[5]). The objective was to create conditions, notably through trade protection and targeted sectoral aid, that would enable domestic companies to reach a critical mass and overcome high initial fixed costs, within a context of increasing returns.
In industrial policy since the 1990s, the state has mainly played the role of facilitator, with minimal intervention to address market failures (related to innovation, financing or coordination), and has pursued the broader objective of creating an environment conducive to innovation across the entire productive spectrum, without targeting specific sectors. Policy is mainly based on horizontal tools and emphasises the role of competition.
As of 2010, the new approach to industrial policy brought back the role of the state and the need for targeted intervention to address market or coordination failures, although without returning to the interventionism of the past. It focuses on market failures that are specific to network and platform industries, and which naturally push markets towards a dominant position or “winner-takes-all” model, characterised by increased concentration of market share and the associated profits. The purpose of government intervention is to steer the economy along specific trajectories that avoid the risk of dependency and it centres on specific technologies or missions (e.g. energy transition, digital capacities) in a context of increased international competition.
The scope of industrial policy has also expanded, and can no longer be limited to the manufacturing sector, its traditional focus. Services now make a meaningful contribution to value added and industrial competitiveness, particularly through business services, IT-related activities, and more recently, artificial intelligence. The expansion of the service side of industrial activities, with design, marketing and R&D remaining in developed countries while production is relocated elsewhere, is also blurring the boundaries between industry and services (e.g. Adidas or Apple, where design, development and marketing remain core activities, while production is largely outsourced) (Fontagné, Mohnen and Wolff, 2014[6]).
This chapter analyses French industrial policy, focusing on its effects on competitiveness, resilience and economic sovereignty. Industrial competitiveness is understood in the broad sense (i.e. including manufacturing, exported services and business services) and covers all measures that improve price and non-price competitiveness (i.e. innovation, human capital, work organisation, etc.) as well as framework conditions, such as access to financing and the competitive environment. Economic sovereignty is understood in the sense of strategic autonomy, i.e. a state’s ability to reduce its vulnerabilities to external shocks, while remaining integrated in an open global economy based on international trade rules.
Importantly, national industrial policy in France operates within a broader European framework. This requires close coordination with EU policies to ensure compliance with the rules of the single market, competition law, and state aid regulations. Aligning national strategies with EU priorities can also help capture economies of scale, enhance cross-border innovation, and strengthen competitiveness. It is also key to avoid conflicts or duplication with initiatives in other member states, ensuring that national measures complement rather than undermine broader EU objectives. Exploiting these synergies remains central to designing effective and coherent industrial strategies, although is outside the scope of this chapter.
4.2.2. France’s industrial competitiveness has declined
The deindustrialisation of manufacturing over the past 20 years has been more pronounced in France compared to the OECD average (Figure 4.1, Panel A). At the end of 2025, manufacturing accounted for only 11.2% of value added and 9.1% of salaried employment, down by almost one third compared to the early 2000s. While this trend has affected all OECD countries, the overall OECD share of manufacturing in value added and employment nevertheless remains higher, at 15% and 13% respectively in 2024, and the decline has been significantly lower. The deindustrialisation of manufacturing has stabilised since 2018, however, allowing the share of manufacturing in employment to remain relatively constant.
The decline of manufacturing in the economy has become a major public policy issue given the benefits of a broad industrial base for productivity and innovation. Due to their capital intensity, manufacturing industrial activities are on average associated with higher labour productivity and higher pay for low-skilled workers compared to service activities (Hausmann, Hwang and Rodrik, 2006[5]). Moreover, these activities can have local multiplier effects on employment due to upstream and downstream demand for services (Duranton and Puga, 2004[7]). Against this backdrop, France launched a Reindustrialisation Action Plan and the Territoires d'Industrie programme. These have helped to stabilise the industrial base, notably with the net creation of 450 factories since 2022, according to the Industry Barometer.
Nevertheless, the ability of industrial policy to reverse the deindustrialisation of manufacturing remains limited due to underlying structural factors common to all OECD countries, which have resulted in a restructuring of the productive framework towards services. An accounting framework that captures structural change breaks down manufacturing value added into three sources of demand — domestic final demand, net exports and demand for intermediate goods from other sectors — using the OECD’s input-output tables. This approach makes it possible to attribute the decline in the manufacturing sector to noticeable changes in these various demand components (Box 4.2; Demmou (2010[8]) and Kalantzis and Thubin (2017[9])). Based on this approach, 80% of the decline in manufacturing since 1995 can be attributed to structural factors, in particular higher productivity gains in industry, consumer preferences, and the shift in industry boundaries. Higher productivity gains in industry and stronger demand for services linked to rising living standards account for around 24% of the phenomenon. The redrawing of industrial boundaries and the growing importance of intangible services in the production process account for 58% of the decline. This suggests that industrial policy can only marginally influence the drivers of deindustrialisation, and that it needs to be framed with more specific objectives.
Note: Panel B: exports are adjusted for re-exports. Panel C: Data from the Standard International Trade Classification (SITC) has been aggregated "Agri-Food" = SITC 0, 1 & 4; "Raw materials" = SITC 2 & 3; "Chemicals" = SITC 5, with the balance mainly coming from perfumes and pharmaceutical products; “Manufactured goods by material (textiles, wood, iron, steel, etc.)” = SITC 6; “Road vehicles” = SITC 78; “Aircraft, spacecraft and ships” = SITC 792 & 793; “Other machinery and transport equipment” = SITC 7 minus SITC 78-SITC 792 & 793.
Source: OECD calculations based on data from the OECD Annual National Accounts database; OECD calculations based on data from the OECD Balanced International Merchandise Trade (BIMTS) database and UN Comtrade data; and OECD Balanced Trade in Services (BaTIS) dataset.
Although it does not account for the majority of the deindustrialisation process, the decline in France's trade position nonetheless explains a significant part of this trend compared to other European countries (Figure 4.2), highlighting the role of industrial competitiveness. The rise of emerging countries accounts for some of the downturn, which is reflected in a fall in French exports as a share of global exports comparable to the situation observed in other European countries. However, France has also lost ground relative to its European partners, with a decrease in its share of euro area exports (Figure 4.1, Panel B). Over the last decade, the trade deficit has averaged 3.8% of GDP. This structural deficit stems mainly from the manufacturing sector, particularly capital goods and automobiles, while the aerospace, chemical, luxury goods and, to a lesser extent, pharmaceutical sectors still show a surplus. The growth in service exports, driven by tourism, finance, shipping, and air transport, has not been enough to offset the losses recorded in industry, while France also has a structural deficit in new information and telecommunications technologies (Figure 4.1, Panels C and D).
An accounting model for structural change, based on work by the OECD and the Banque de France, has been developed and is based on a breakdown of manufacturing value added on the demand side. Domestic manufacturing output can be used for three main purposes: final domestic demand (household consumption and investment), intermediate consumption by other businesses, and foreign demand (exports). Trends in the composition of manufacturing output based on these three components are then simulated over the period 1980-2020 using OECD input-output tables, making it possible to link changes in the weight of the manufacturing sector to the following mechanisms:
Structure of demand and productivity: At given relative prices, rising living standards allow consumers to allocate a higher share of spending to satisfy less basic needs, in particular to consume services. This results in a decline in the share of industry in value added, which does not necessarily reflect a weakening of industry, but a natural evolution driven by household preferences. Furthermore, faster productivity gains in industry automatically reduce its relative prices, which also affect its share in total value added production. Overall, the resulting decline in the share of industry in value added does not necessarily indicate industrial decline but reflects a natural adjustment stemming from the combined effects of changes in relative prices and income.
Input-output linkages: the outsourcing of non-productive functions (maintenance, logistics, accounting, IT, etc.) by industrial companies to the services sector reveals a decline in industry statistically speaking, but in reality reflects an organisational transformation, with no real decrease in productive activity. On top of this comes an increase in intermediate consumption of services linked to the integration into global value chains.
Trade: greater exposure to global competition, particularly from emerging countries, reduces the competitiveness of certain industrial sectors and reinforces France's specialisation in less manufacturing-intensive segments. At a given level of specialisation, as global trade is dominated by manufactured goods, a deterioration in the trade balance, reflecting saving and investment decisions and competitiveness, also supports a decline in the share of industry in value added.
4.2.3. Non-cost factors explain France’s loss in export market shares
A decade ago, the downturn in France's trade performance was due in part to reduced cost competitiveness linked to relatively high unit labour costs. Since 2015, however, France has had some of the slowest growth in unit labour costs (Figure 4.3, Panel B). Studies by the Banque de France show that France's export performance relative to Germany, Spain and Portugal has improved since 2008 due to improved cost competitiveness, but that the remaining competitiveness losses now seem to be mainly due to non-cost factors (Berthou and Gaulier, 2021[10]) (Figure 4.3, Panel A). A recent survey on export competitiveness confirms that foreign importers perceive French products as having an unfavourable price-quality ratio (Rexecode, 2025[11]). While wage moderation and labour cost reductions (see next section) have improved cost competitiveness over the past 15 years, French manufacturers remain subject to strong price competition, particularly from emerging countries and some European partners. Data from the French Centre d’études prospectives et d’informations internationales (CEPII) indicate that approximately 60% of exports can be classified as medium or low quality (based on the difference between the unit value of the product and the global median), which is lower than in Spain but higher than in Germany (50%) (see Figure 9 in Cour des comptes (2024[12])).
Strengthening non-cost competitiveness could ease the cost pressures exerted by new players and enable France to consolidate its industrial base. This would require overcoming several challenges, in particular relating to business innovation, technology diffusion, barriers to the growth of small-to-medium enterprises (SMEs) and value chains based on a dense network of SMEs, and the development of human capital.
Note: Panel A: Export performance is measured by France's average annual logarithmic growth rates relative to the export growth of the country indicated, decomposed geographic orientation, into price, cost, and non-price contributions. A positive (negative) value indicates that French exports grew more (less) rapidly. Banque de France update from Berthou and Gaulier (2021[10]).
Source: Banque de France; and OECD Economic Outlook database.
Industrial competitiveness depends on the sector's capacity for innovation. At 1.4% of GDP, corporate R&D spending is around the EU average but far below that of the best-performing countries, which limits France’s capacity to innovate and compete in high value-added technology segments (Figure 4.4, Panel A). The patent rate for companies is also four times lower in France than in the five best-performing OECD countries. However, France has strengths in several emerging technologies, including artificial intelligence, quantum technology, cybersecurity, digital health, and carbon-free energy, supported by large-scale public programmes (France 2030, the Investment for the Future Programme (PIA), and European initiatives). However, this advantage remains fragile given the level of international competition and has not yet translated into productivity gains or the development of competitive industries. France is not ranked among the most active countries when it comes to patents in major disruptive technologies, such as robotics, quantum technology, and AI, a field dominated by the United States, China, Japan, and South Korea (Bellit and Charlet, 2023[13]; Bock et al., 2025[14]).
Note: Panel C: the boxplots indicate the maximum and minimum values among OECD countries.
Source: OECD Main Science and Technology Indicators database; Filippuci et al., (2025[15]); OECD (2024[16]); OECD calculations based on the OECD National Accounts Database and the Global INTAN-Invest database, https://global-intaninvest.luiss.it/.stat
The use of information and communication technologies (ICT) generates productivity gains that are essential for absorbing the higher costs associated with innovation, quality and differentiation. Artificial intelligence (AI) offers considerable growth potential, particularly for industrial firms. The OECD estimates that with fast adoption reaching 50% in ten years, matching that of mobile phone technologies, AI could increase annual labour productivity growth by 1.1 percentage points within ten years. The effect could be even more substantial with faster adoption, as anticipated for the United Kingdom and Germany (around 60%) (Filippucci et al., 2025[15]) (Figure 4.4, Panel B).
However, uptake of digital technologies remains relatively limited in France; nearly one in two companies (with 10 or more employees) in OECD countries already use cloud technologies, while the proportion of French companies doing so is half that number. France also lags behind in terms of the Internet of Things (IoT) and artificial intelligence, particularly among SMEs (Figure 4.4, Panel C). OECD analyses show that the limited uptake of digital technologies is primarily due to barriers to adoption (particularly in terms of skills, organisational capacity and financial constraints) that affect SMEs in particular. These barriers result in a stark contrast between a core group of technologically advanced firms (at the technological frontier) and the rest of the productive sector, and this has a negative impact on overall productivity performance (Andrews, Criscuolo and Gal, 2016[17]). These obstacles are examined in detail in the rest of the chapter.
On the other hand, France is also one of the countries that invests the most in intangible assets (16% of GDP in 2023, equivalent to that of the United States and well above that of the OECD average of 11% of GDP) (WIPO, 2025[18]) and is one of the best-performing countries when it comes to the use of big data (Figure 4.4, Panel C) (OECD, 2024[16]). This creates a paradox that can be understood by looking at the nature of these investments, characterised by a higher investment in organisational capital and a lower proportion of investment in R&D, and an insufficient integration of ICT technologies into production processes, which consequently limits their real contribution to growth (Figure 4.4, Panel D). Recent evidence also confirms that the gap in productivity growth between France and the United States can largely be attributed to ICT-intensive sectors as well as sectors that produce these technologies (Bunel et al., 2025[19]; Bock et al., 2025[14]). Countries with higher productivity growth, particularly the United States, have accumulated far more digital capital per worker than many European countries, including France, which increases the risk of a lasting divergence in productivity (Gal et al., 2026[20]) (Figure 4.5).
Real gross business investment by type of assets, index 2007 = 100
Note: Business sector in this chart covers all sectors excluding agriculture, forestry and fishing, real estate sector and sectors that are more likely to be public (sectors O to U from the ISIC Rev4 classification).
Source: Gal et al., (2026[20]).
French entrepreneurship is remarkably dynamic, placing it among the OECD countries with the highest number of new businesses; the three-year survival rate is also above the OECD average (OECD, 2025[21]). However, the growth of start-ups remains limited and the number of mid-sized firms is significantly lower than in other European countries, suggesting barriers to growth (Figure 4.6, Panel A). In 2020, France had around 5 000 mid-sized firms, compared with nearly 11 000 in Germany and 6 500 in the United Kingdom (Ben Hassine and Fotso, 2025[22]). The shortage of mid-sized firms restricts the development of competitive sectors and hampers integration into global value chains, resulting in a lower contribution of SMEs to exports and employment, with the productive base remaining dominated by large companies (Panel B).
Shortages of skilled workers are holding back the adoption of new processes, particularly digital technologies, which are essential for improving productivity and industrial competitiveness. OECD estimates suggest that human capital shortages account for around 20% of the productivity slowdown in France (Andrews, Égert and de la Maisonneuve, 2024[23]) (Figure 4.7, Panel A). Entrepreneurs report significant difficulties in recruiting the right skillsets, despite the easing of labour market tensions following the peak observed in 2022-2023 during the health crisis (Panel B). Recruitment problems also appear to be the main obstacle to growth for businesses according to the Eurobarometer on skills (European Commission, 2023[24]).
Note: Panel A: the figure shows the cumulative three-year rate of employment growth in new enterprises with between two and nine employees at the time of their creation that are still active and part of the target enterprise population three years after their creation ("surviving enterprises", excluding those that have left the market, those that do not report employment, and those with fewer than two employees three years after their creation). The OECD average of 8 countries includes Finland, France, Germany, Italy, Portugal, Slovenia, Spain and the United Kingdom. Panel B: "OECD" represents an unweighted average of the 31 OECD countries for which data are available.
Source: OECD (2025[21]); OECD database of structural business statistics by size class and economic activity (ISIC Rev. 4) and OECD database on international trade by enterprise characteristics.
Note: Panel A: the average is the unweighted average of the countries in the figure. The figure reports productivity over 1987-2005 against 2005 to 2022. Panel B: respondents were asked to indicate whether their company had experienced difficulties in recruiting staff in the last 24 months. Labour shortages are considered acute if all or most (as opposed to some, a few or none) of the job vacancies in the company were difficult to fill. OECD corresponds to the unweighted average.
Source: Andrews, Égert and de la Maisonneuve (2024[23]); Filippucci, Laengle and Marcolin (forthcoming) based on data from the OECD Global Forum on Productivity employer survey.
4.3.1. Social security contribution reductions – a cornerstone of French industrial policy – have had limited effects
Following the Gallois report (2012[25]), which recommended bold measures to boost competitiveness in response to France’s industrial decline, reducing labour costs became one of the cornerstones of French industrial policy (Figure 4.8, Panel A). Initially designed to combat unemployment among low-skilled workers, social security contribution reductions were expanded to meet competitiveness objectives from 2013 onwards. Building on general exemption measures covering low-skilled workers in place since 1993 (up to 1.6 times the minimum wage), the competitiveness and employment tax credit (Crédit d’impôt compétitivité emploi, CICE) introduced in 2013, followed by the Responsibility Pact, introduced in 2015, extended the scope of relief to 2.5 times the minimum wage (called the “health band” as it reduces health contributions) and then 3.5 times the minimum wage (called the “family band” as it is a reduction in family contributions), to better cover export sectors, particularly manufacturing (Figure 4.8, Panel B). Accordingly, the contribution rate at the minimum wage is 6.7%, including only unemployment and pension contributions, and then increases gradually until the exit point.
These reductions come at a significant cost, which has almost doubled in ten years, reaching around EUR 80 billion (EUR 78.9 billion in 2023 (2.8% of GDP), including EUR 27 billion for the health band, and EUR 10.3 billion for the family band (DARES, 2025[26])) (Figure 4.8, Panel C). These measures currently account for nearly 11% of private sector payroll costs (Cour des comptes, 2025[27]), and for 41% of total expenditure on employment and the labour market in 2023 (DARES, 2025[26]). This results in a marked bias of industrial policies in favour of employment-support measures, and in particular, reductions in labour costs (Figure 4.11, Panel A).
Unit labour costs, which were initially among the highest in the OECD, have significantly declined (Figure 4.3, Panel B). Wage moderation and large reductions to social security contributions for workers earning low incomes have pushed down labour costs at the minimum wage in France, while they have risen elsewhere in Europe (Figure 4.8, Panel D). On average, reductions in social security contributions have brought non-wage labour costs down from 46% to 32%.
The impact of the reduction in social contributions on industrial and export sectors has been limited. The main beneficiaries of the scheme are services with little exposure to competition (catering and retail, with 55% and 35% of their contributions reduced respectively), but also agri-food and metalworking (around 40% and 30%) (Cour des comptes, 2025[27]). Export sectors, which employ a higher proportion of middle- and high-skilled workers, benefit to a lesser extent from these reductions, which are significantly lower for intermediate wage levels. Labour costs at these levels remain relatively high, which can undermine cost competitiveness (Paris, 2019). However, these sectors manage to benefit from these reductions through their intermediate consumption. This strategy mirrors the one adopted in Germany following the Hartz reforms, which significantly reduced service costs and, in turn, boosted national cost competitiveness (L’Horty, Martin and Mayer, 2019[28]).
In theory and given the competitiveness objective of social security contribution policies since 2015, lower unit costs should have improved price competitiveness and market shares. However, empirical assessments reveal only modest and, in most cases, insignificant effects on export performance, be it through margin recovery, prices or the entry of new exporters (Malgouyres, 2019[29]; Guillou et al., 2016[30]; Malgouyres and Mayer, 2018[31]; Carbonnier et al., 2017[32]). These findings are consistent with the trend deterioration in the trade balance despite these measures.
Given the limited impact of tax relief on competitiveness, it would be advisable to reassess the size of these social security contribution reductions (see Chapter 1). The 2025 Finance Act has already provided for the merging of the current measures into a degressive reduction of up to 3 times the minimum wage from 2026 onwards. Lowering this threshold further by focusing on relatively low wages, for example to about 2 times the minimum wage as suggested by the Minimum wage Working Group, would focus support on low wages, where the effects on employment are most significant (Barreto et al., 2025[33]; Groupe d'Experts SMIC, 2025[34]). This shift in focus would preserve low-skilled jobs while reducing ineffective aid, as the risks to competitiveness are now more limited, given the convergence of labour costs among European partners (Figure 4.8, Panel D). This would also have a very positive impact on reducing the public deficit, as this readjustment of the amount of relief could limit its cost by EUR 7-9 billion (0.23-0.3% of GDP) depending on whether the threshold is set above or below twice the minimum wage. To avoid threshold effects and reduce the risk of low-wage traps (Bozio and Wasmer, 2024[35]), part of the savings could be used to finance training and support for salary increases. Finally, in order to limit job losses, the reform should ideally be introduced during an expansion phase of the economic cycle.
Note: Panel D: The unweighted OECD average excludes Austria, Belgium, Chile, Colombia, Costa Rica, Denmark, Finland, Iceland, Israel, Italy, Japan, Mexico, Norway, Sweden and Switzerland.
Source: OECD QuIS database - Industrial policy grants and tax expenditures; Groupe d'Experts SMIC (2025[34]) ; Cour des comptes (2025[27]); Barreto, C. et al. (2025[33]).
4.3.2. Pursuing efforts to reduce production taxes
French companies pay significantly higher production taxes than in most comparable countries, placing France second in Europe behind Sweden according to Eurostat data. Even after factoring in subsidies, net production taxes remain the highest in Europe, nearly twice as high as in Italy and four times higher than in Spain, while Germany even has negative net production taxes (Bock et al., 2025[14]) (Figure 4.9, Panel A). A gradual reduction in production taxes could partially offset the scaling-back of social security reductions and of the research tax credit (recommended below). Measures relating to tax relief, tax expenditures and production taxes should be considered in conjunction with each other in order to ensure the consistency of reforms and avoid placing an excessive burden on businesses, while at the same time generating budgetary margins.
The reduction in levies should target the most distortive taxes, such as Companies’ value-added contributions (CVAE), which hinder investment, and Companies’ social solidarity contributions (C3S), which are similar to an export tax (Urvoy, 2019[36]). Since 2019, the reduction in the CVAE, the company property tax (CFE), and the property tax on developed land (TFPB) has reduced the production tax burden significantly, but it remains high (Rexecode, 2025[37]) (Figure 4.9, Panel A). The total abolition of the CVAE (planned but not adopted in the 2026 budget) would support business activity and in particular industry, whose contribution to the CVAE and C3S is greater than its share of value added (Goutebroze and Lapierre, 2025[38]).
In order to reduce the fiscal impact, particularly on local authorities, which account for 70% of public investment, options include introducing a new inter-municipal tax to finance rising local service needs and strengthening property tax revenues by updating tax bases to current values (see Chapter 1).
4.3.3. Helping industrial companies with rising energy prices
French industrial companies benefit from relatively low electricity prices in international comparison (Figure 4.9, Panel B; (Tournier, 2025[39])). However, the reform of the electricity pricing system could weigh on industrial competitiveness, requiring some attention. In line with European guidelines, France is ending the ARENH mechanism, which offered suppliers access to nuclear electricity at EUR 42/MWh. The new system of long-term contracts for large industrial companies is designed to boost the profitability of EDF, France’s main public electricity utility, while ensuring more predictable energy costs for businesses. However, the associated increase in energy costs could weigh on the productivity and competitiveness of industries. Empirical evidence suggests that, after an initial decline in productivity in the short term, higher energy prices can encourage firms to invest more, innovate and, in the medium term, improve their productivity and competitiveness (André et al., 2023[40]; Fontagné, Martin and Orefice, 2024[41]). However, the adverse effect would be more noticeable in the short term for companies in electricity-intensive sectors, and the spillover effect on productivity in the medium term would be weaker. This highlights the need for mechanisms to preserve the competitiveness of these companies, which have less capacity to adjust. However, it is critical to make such mechanisms temporary in order to strengthen productive transformation and support competitiveness in the medium and long term (see also the next section on decarbonisation and competitiveness).
Note: Panel A: Value added data for France refer to 2024 instead of 2023.
Source: Rexecode (2025[37]) ; and Eurostat.
4.3.4. Stimulating competition in business services and simplifying administrative procedures for factory permits
Rising input costs are the main driver of increased production costs in manufacturing and services in France, although they have risen less rapidly on average than in Germany, Italy and the United States (Bock et al., 2025[14]). Reducing barriers to entry in business service sectors could curb the inflation inherent in industrial inputs and strengthen industrial competitiveness. While France generally benefits from competition-friendly regulations and relatively low barriers to entry, obstacles remain in some professional services (Figure 4.10, Panel A). Easing these regulations would better support the growth of new innovative businesses and reduce the cost and quality of intermediate inputs (OECD, 2025[42]). Reforms could focus on reviewing compulsory local examinations for foreign professionals, reducing mandatory minimum fees, and offering alternatives to mandatory membership of a professional body (for lawyers, notaries, accountants and architects), such as voluntary certifications linked to consumer information systems (Chapter 1).
Slow administrative procedures can hamper competitiveness and industrial momentum by delaying the implementation of investment projects and increasing opportunity costs. Since 2018, the authorities have reduced regulatory burdens and introduced an electronic single point of contact for business registration (OECD, 2024[43]). Nevertheless, industrial projects are still experiencing implementation delays that exceed European standards, despite the laws and measures adopted since 2020 (ASAP, renewable energies, and green industry laws) (Cour des comptes, 2024[12]). For example, the average administrative processing time for new industrial facilities in France monitored by the Directorate General for Enterprise (DGE) was supposed to fall from 13.4 months at the end of 2022 to nine months but remained unchanged at the end of 2023. Strengthening local administrative and technical capacities could speed up the permitting process for new factories. Strengthening coordination between the various administrative layers involved in the authorisation process is also essential in a context of marked administrative fragmentation (see Chapter 1). To this end, the best practice principles in licensing and permitting recently developed by the OECD could provide a useful basis for improving the efficiency of these procedures (OECD, 2025[44]).
Note: Panel A: the higher the indicator, the stricter the regulatory framework. Panel B: Concentration is measured by the share of total revenue held by the four largest groups of companies in a given market. The data are weighted according to market size and expressed as cumulative growth (percentage points).
Source: OECD Product Market Regulation (PMR) dataset; Calligaris et al., (2025[45]).
4.4.1. Redirecting tax support measures for R&D
Intangible assets and R&D are a key factor in determining companies' export intensity and their integration into the most profitable segments of global value chains (Dimas, Stamopoulos and Tsakanikas, 2022[46]). France has four main tax mechanisms that support R&D: the “CiCo” tax credit for collaborative research, accelerated depreciation of R&D expenditure, exemptions from social security contributions for young innovative start-ups (JEI: Jeunes Entreprises Innovantes) and, above all, the research tax credit (CIR: crédit d’impôt recherche) (OECD, 2026[47]). The research tax credit is the main instrument for supporting R&D in companies and results in one of the highest tax subsidy rates in the OECD (Figure 4.11, Panel A).
The 2008 reform significantly expanded the fiscal support available to firms for qualifying research expenditures (by moving from an incremental system to a volume-based system), increasing the fiscal cost of the CIR from EUR 1.8 billion in 2007 to nearly EUR 8 billion in 2024. Despite this strong support, growth in corporate R&D expenditure has remained very moderate by international standards (Figure 4.11, Panel B). Large companies benefit the most from the tax credit, which could explain the weak effect of the scheme in raising business R&D investment and low additionality of the scheme in France (Appelt et al., 2020). Several studies suggest windfall effects for these companies (Aghion, Chanut and Jaravel, 2022[48]; Bach et al., 2021[49]; Harfi and Lallement, 2021[50]). According to estimates by the Conseil d’analyse économique, EUR 1 million in aid generates about 1.2 patents in a micro-enterprise, compared with 0.5 in a large company (Aghion, Chanut and Jaravel, 2022[48]). Similarly, OECD research shows that EUR 1 of tax relief generates EUR 1.4 of R&D in small enterprises, compared with EUR 0.4 in large enterprises, with the strongest spillover effect linked to the lower volumes of R&D activity in SMEs (Appelt et al., 2023[51]). Lowering the threshold above which a reduced rate applies (currently EUR 100 million), would improve its cost-effectiveness and reinforce the additionality of the scheme, while maintaining support for the overall R&D efforts of large companies. Countries applying a binding threshold in a two-tier system include Canada and the Netherlands. The reform would also contribute to the necessary fiscal consolidation effort (see Chapter 1) and could be combined with a reallocation of expenditure towards sectors identified as having increased needs, such as support for the development of new technologies under the France 2030 initiative (funding for which was reduced in the 2026 budget) and the financing of disruptive innovations (see sections 4.4.2 and 4.4.3 below).
Note: Panel A: the tax subsidy rate is defined as 1 minus B-index, a measure of the before-tax income needed by a “representative” firm to break even on USD 1 of R&D outlay (Warda, 2001).
Source: OECD database on tax incentives for R&D; and OECD Main Science and Technology Indicators (MSTI) database.
The horizontal design of the scheme implies that it is of greater benefit to large companies and thereby specific sectors, especially manufacturing sectors, in line with the sector level distribution of R&D efforts across firms (OECD, 2024[52]). In this context, strengthening the research tax credit incentives for SMEs would also provide greater support for innovative services, which account for 66% of SME R&D and are becoming increasingly important for industrial competitiveness (see section 4.1).
One issue that is frequently highlighted is that any reform of the research tax credit could reduce France's attractiveness for R&D investment by foreign companies. At this stage, the available data do not permit an assessment of whether there is a significant risk, as there has been no recent research on this issue. However, one study suggests that the 2008 reform did not lead to a significant influx of foreign investment (Lhuillery et al., 2021[53]). Furthermore, although France attracts 44% of European industrial FDI, only 36% of this is for new greenfield projects, compared with over 70% in the United Kingdom and Germany (Cour des comptes, 2025[27]). While these factors do not rule out the possibility that a scenario without the reform would have been even less favourable, they do suggest that the reduction in R&D costs was not enough to significantly boost the country's attractiveness. This indicates that risks of reform are presumably moderate and that other levers could be activated to enhance France's appeal, particularly in relation to the intensity of corporate research and the transfer of public research to industry (Siedschlag et al., 2013[54]). Available cross-country evidence suggests that while tax incentives can influence where multinationals locate R&D, with some risk of substitution across jurisdictions, location decisions are mainly driven by scientific excellence, skilled labour and strong intellectual property protection, particularly for basic research (Appelt et al., 2016[55]).
4.4.2. Maximising the economic impact of direct support to industrial innovation
Better targeting direct aid for industrial innovation
At 1.7% of GDP, support to industrial policy is among the highest in the OECD, a significant proportion of which is allocated to supporting innovation. Just under half corresponds to direct aid, a proportion close to the average observed for the selected sample of OECD countries (Figure 4.12, Panel A). Direct support for business innovation is framed by the five-year France 2030 plan, with a budget of EUR 54 billion. The France 2030 plan marks a departure from previous Future Investment Programmes (Programmes d’Investissements d’Avenir), both in terms of its unprecedented budgetary scale and its ambition to adopt a strategic approach oriented on missions and strategic technologies, largely focused on addressing major societal challenges. The plan focuses on several sectors considered to be strategic: energy (nuclear, hydrogen, industrial decarbonisation), transport (electric vehicles, low-carbon aviation), space and marine technologies, health and biomedicine, agri-food, electronics and telecommunications, as well as culture and creative content.
The France 2030 plan was designed to limit the traditional risks of public intervention in industrial policy. The sectors were chosen in consultation with industry players, and most of the subsidies are distributed through calls for projects. This selection method, in line with best practices, does not seek to predetermine which companies will become industry leaders, promotes competition in the allocation of aid, preserves a certain technological neutrality, and is based on the assessment of independent experts (Alvarez et al., 2025[56]; Tirole, 2015[4]). However, governance could be strengthened in several areas, as described below.
A more selective approach in selecting objectives would help focus funding and optimise the impact of public aid. France currently has a low level of technological specialisation compared to the world leaders (United States, China, Japan, Korea, Figure 4.12, Panel B), which could limit its ability to become a leader in key technologies (Bellit and Charlet, 2023[13]; Bock et al., 2025[14]). Limiting the number of areas of intervention to around 20, compared with around 40 today (Cour des comptes, 2024[12]), would reduce the risk of spreading resources too thin and should focus on disruptive innovations, as incremental innovations are already supported by the research tax credit.
The programme's cost-effectiveness could also be improved by redirecting more aid towards repayable advances, which remain limited in number (around 7% of France 2030 aid) (Cour des comptes, 2025[57]), and by reserving grants for high-risk projects or those that are far from market-ready. Repayable advances, by sharing risk more effectively with the private sector and strengthening public leverage, reduce adverse selection, especially in the development phase. However, care must be taken to ensure that this shift does not favour projects that are less risky but also less innovative.
The effectiveness of policies supporting industrial innovation could be improved by consolidating the criteria for discontinuing projects. International experience shows that linking public support to clear objectives, such as export performance in Japan and Korea or technological targets in the United States, enhances the effectiveness of industrial policies (OECD, 2023[58]; Alvarez et al., 2025[56]; Cherif and Hasanov, 2025[59]). Continuous assessment based on specific indicators would make it possible to quickly terminate unpromising projects and reallocate resources (Criscuolo et al., 2022[3]; Millot and Rawdanowicz, 2024[2]). It would therefore be preferable to set clear and measurable economic performance targets.
Note: Panel B: specialisation is measured using patent families in a set of twelve selected breakthrough technologies, filed in at least two national or international patent offices. The specialisation index is defined as the ratio of a country’s global patent share in a given technology to its overall global patent share. The chart shows the four countries with the highest specialisation. For example, France is strongly specialised in only one technology, with a specialisation index of 1.1.
Source: OECD calculations based on OECD QuIS database; Bellit and Charlet (2023[13]).
Reducing coordination mismatches between public research and industry
The promotion of public research is a central pillar of industrial policy, as it allows scientific breakthroughs to be translated into commercial innovations. By facilitating the transfer of knowledge and technology from public laboratories to industry, the policy to promote public research reduces gaps in coordination and speeds up the industrialisation of innovations. Despite the excellent quality of its universities and the fact that its laboratories are among the European leaders for public patents, France still has only limited links with the private sector, particularly in terms of partnerships with start-ups, co-publications and collaborative funding.
France has introduced several schemes to remove barriers to technology transfers from public research to industry, but their overall effectiveness could be improved in several ways:
The transfer of technology appears to be insufficiently aligned with strategic priorities. France could draw inspiration from South Korea's approach, which combines clear state priorities, long-term roadmaps and jointly managed research-industry consortia.
Economic assessments of the ecosystem remain limited, with no comprehensive assessment since 2018 (Commission des Finances, 2025[60]). At the same time, the proliferation of schemes with similar objectives poses challenges in terms of coordination and the risk of duplication.
Since 2022, the “CiCo” tax credit for collaborative research has offered a rate of 50% for SMEs (40% for large companies) on expenditure related to public research contracts, up to a limit of EUR 6 million per year (OECD, 2026[47]). This scheme, which is still relatively new, has not yet been fully evaluated. Its effectiveness in terms of additionality will therefore need to be closely monitored and, if necessary, adjusted or discontinued in the current budgetary context.
4.4.3. Removing barriers to financing disruptive innovation and intangible assets
Challenges in financing industrial innovation
High-growth companies, particularly unicorns (i.e. start-ups valued at one billion euros or more), are a key driver of French industrial policy, and demonstrate France's ability to scale up and compete internationally in often strategic technologies. While there have been notable successes, the number of unicorns, around 30, remains significantly lower than in the United Kingdom (around 55) and the United States (around 708) (CB Insight, 2026[61]). The proportion of high-growth companies (i.e. start-ups with at least 10 employees whose workforce is growing by more than 10% per year) relative to companies with more than 10 employees is also 9%, compared with 12% in Germany and the United Kingdom (Eurostat). More generally, the limited growth of start-ups in France (see Figure 4.6) suggests that barriers exist. One possible explanation is linked to difficulties in accessing external financing, particularly for innovative firms, due to perceived high risk and a lack of guarantees, which hamper the productivity and growth of these companies (Haskel and Westlake, 2017[62]; OECD, 2024[16]; Demmou and Franco, 2021[63]). Public funding for innovative start-ups, as discussed in the previous sections, alleviates these constraints, but a strong capital market remains essential.
Deepening venture capital is a key challenge for improving the financing of innovative businesses. Despite recent progress, the French ecosystem remains heavily dependent on public funding and suffers from a shortage of private capital, particularly during the growth and industrialisation stages. A broader venture capital market would make it possible to attract more institutional investors, with greater investment capacity than traditional venture capitalists, to provide patient capital in the form of long-term equity financing as well as strategic support to innovative companies, and to facilitate the scaling up of radical innovations. In this regard, OECD comparative analyses show that the rise of institutional investors is a key driver of deeper equity markets: in Sweden, for example, venture capital investments by national pension funds have almost doubled over the last decade, contributing to deepening the market and increasing the financing capacity for innovative companies throughout their development cycle (OECD, 2025[64])
The venture capital market in France has been very buoyant over the last ten years, mainly owing to the efforts of Bpifrance (Banque Publique d'Investissement), with the volume of funds raised (from all sources) increasing ninefold between 2013 and 2023 (BPI, 2025[65]), placing France among the leading European countries. However, volumes remain relatively modest compared to the best-performing countries, limiting the market's financing potential for industrial innovation (Figure 4.13, Panel A) (Hafied, Rachiq and Roulleau, 2021[66]). As a result, ten years after their creation, French start-ups have raised on average 40% less capital than their US counterparts. This funding gap remains below the European average (50%), but higher than that observed in Germany, where it is less than 30% (Fratto et al., 2024[67]). The gap also exists in forward-looking sectors such as AI, where fundraising is ten times lower than in Israel and five times lower than in the United States (Figure 4.15). This results in a delay in financing in the more advanced stages, known as the “Death Valley phase”, which slows the transformation of innovative start-ups into growth companies. This can push some start-ups to relocate abroad in order to access more significant funding (OECD, 2025[64]).
Better mobilising household savings and institutional investors
The French household saving rate is high, at close to 19% of gross disposable income, and could play an important role in expanding venture capital and financing industrial start-ups. France has traditionally encouraged investment in SMEs through tax measures such as the ISF-PME scheme, which allowed wealth tax deductions, and the Madelin scheme, which provides income tax reductions. While these schemes have mobilised significant volumes of capital (Sénat, 2024[68]; Midy, 2023[69]), the ISF-PME was abolished in 2018 and the Madelin rate was reduced from 25% to 18% in 2024 due to windfall effects resulting from imperfect targeting of market failures. Conversely, the new 50% tax deduction for investments in young disruptive innovation companies (JEIR, see above) is a promising lever for directing savings towards sectors with significant market failures and deserves to be made permanent. This mechanism can support small-scale crowdfunding and strengthen venture capital, as in Germany (INVEST programme), in Israel (Yozma Fund) and the United Kingdom (Seed Enterprise Investment Scheme, SEIS) (Box 4.3). To mobilise more capital, it would be appropriate, as in these countries, to combine exit mechanisms (partial capital gains tax exemptions or partial deductions in the event of losses).
Nevertheless, a risk-sharing mechanism of this kind should be accompanied by a set of measures designed to limit the risks of moral hazard and the state’s exposure to significant contingent liabilities. First and foremost, risk-sharing should be accompanied by a cap on returns in order to limit budgetary costs. Furthermore, these guarantees would need to be accompanied by clear governance rules, particularly with respect to oversight and transparency regarding exposure to risk. They should also be temporary, targeted at clearly identified market failures, and pursue specific additionality objectives by supporting projects or firms that would not have access to sufficient financing on market terms, without replacing private financing. To this end, it would be desirable to reserve these guarantees for radical innovations, i.e. innovations based on technological breakthroughs or characterised by high uncertainty and risks that are difficult for the market alone to finance.
Innovative industrial firms could benefit from increased institutional funding, although this market remains smaller than in the United States and Scandinavia, where there are more pension and sovereign wealth funds. Launched in 2019, the Tibi initiative aims to mobilise these players. Phase 1 (2020-2022) committed EUR 6 billion to advanced growth funds, placing France at the forefront of fundraising in Europe (EUR 8.3 billion in 2023). Phase 2 is more focused on industrial projects, IPOs and attracting foreign investors. While the initiative has succeeded in mobilising significant volumes, with EUR 12.5 billion raised since 2020, investment has been mainly focused on mature companies and some projects financed under the Tibi label to exhibit limited innovation (Sénat, 2024[68]). Increased participation by public venture capital funds of funds (such as FNVI) in Tibi-approved funds would increase public-private leverage and better direct capital towards technology-intensive projects. It would also help to reduce the funding gap in the advanced stages, known as the "Death Valley phase".
Regulatory constraints, such as Solvency II, limit institutional investors' involvement in financing industrial innovation, particularly start-ups, as they require larger capital reserves to reduce the financial risk associated with these investments. To remedy this, within the existing regulatory framework, one option would be to introduce targeted public guarantees: partial coverage of losses, combined with a cap on returns, could encourage these investors to finance high-risk technologies, following the Israeli model. Similarly, tax rules are hindering the expansion of the capital market. As in many OECD countries, the corporate tax system in France continues to favour debt over equity, with interest deductibility making debt financing more attractive. A better balance is needed in the taxation on debt and equity, for example by limiting the deductibility of interest or making returns on capital partially deductible.
INVEST in Germany
In Germany, the INVEST scheme offers a 20% entry subsidy for private individuals investing in innovative start-ups in a range of EUR 10 000 to EUR 500 000, with an annual cap of EUR 100 000 per investor and EUR 3 million per company. Investors must hold their shares for at least three years. The scheme also includes an exit subsidy of 25% of the gains realised, in the form of a tax exemption or tax credit, provided that the initial investment benefited from the entry subsidy.
Seed Enterprise Investment Scheme (SEIS) in the United Kingdom
Established in 1994, the Enterprise Investment Scheme (EIS) is one of three UK venture capital schemes based on tax incentives, alongside the Venture Capital Trust (VCT) and the Seed Enterprise Investment Scheme (SEIS). The EIS helps small businesses in the growth phase to raise funds by offering tax incentives to investors who purchase new ordinary shares. Launched in 2012, the SEIS complements the EIS by targeting businesses in the very early stages of their life, offering enhanced tax incentives (50% compared to 30% for the EIS).
Yozma Fund in Israel
Launched in 1993, the Yozma Fund played a decisive role in developing venture capital in Israel. With USD 100 million in assets, this public fund of funds created ten private VC funds, combining public and private resources in equal measure, with each fund requiring a domestic VC investor, a foreign VC fund/company and an Israeli investment company/bank. An innovative mechanism allowed private investors to buy back the government's share after five years if the fund proved successful, facilitating the rapid privatisation of the sector. A reduced version, Yozma Fund 2.0, was relaunched in 2023.
Source: OECD (2025[64])
Consolidating early-stage financing to maximise its impact
Industrial start-ups require significant and patient financing from the early stages of their development, due to their specific technical needs (machinery, laboratories, prototype costs, etc.). The public funding system covering the entire early stages of innovation has been strengthened in recent years, using a hybrid venture capital structure via funds such as French Tech Seed and the Fonds National d'Amorçage. These funds mainly provide equity capital, and some also combine private co-financing, particularly with business angels, thereby promoting leverage and risk sharing. Bpifrance plays a central role, representing more than a third of the capital of partner funds specialising in seed funding between 2018 and 2021 (BPI, 2025[65]). Through its relatively long-term co-investments (7 years compared to 3-4 years for the private sector) (Cour des comptes, 2024[12]), Bpifrance mobilises long-term capital, which is essential to supporting the long profitability cycle of innovative industrial projects.
However, the volumes raised by business angels remain relatively low, leading to a fragmentation of financing and the risk that it will be insufficient to ensure the growth potential of start-ups (Figure 4.13, Panel A). For example, Germany has 36 networks that have invested more than EUR 26.2 billion in around 670 companies (Tracxn, 2025[70]). In France, nearly EUR 74 million has been invested in around 441 start-ups, illustrating a high level of dispersion (France Angel, 2024[71]). While volumes remain limited, public funding needs to be concentrated on a small number of projects in order to maximise impact, while ensuring that target selection remains the preserve of experts, particularly through funds of funds.
Source: OECD (2026[72]); EBAN Annual Statistics Compendium for 2024; BPI (2025[73]).
Strengthening the role of banks in financing the intangible assets of industrial SMEs
Investment in intangible assets is essential if industrial companies are to scale up, position themselves in higher value-added segments, and more easily integrate into complex value chains (Tsakanikas et al., 2022[74]). However, only 30% of industrial SMEs and mid-sized enterprises invest in intangible assets each year, mainly among exporting companies, with smaller companies struggling to reach the critical investment mass necessary for profits to materialise (Durand and Boulongne, 2021[75]). This points to underinvestment and untapped opportunities for productivity growth and competitiveness gains.
One possible explanation for the funding gap in intangible assets for SMEs is the difficulty in financing these through loans (Demmou and Franco, 2021[63]). While industrial SMEs are heavily dependent on access to bank credit to finance their activities, this type of loan is not well suited to financing their intangible assets, forcing them to rely heavily on self-financing, which puts additional pressure on their cash flow. Accordingly, as is the case in other OECD countries, bank credit contributes significantly to the financing of tangible assets in France (43% of financing) but significantly less to intangible assets (23% of financing) (Lé and Vinas, 2022[76]). This situation is particularly restrictive for micro-enterprises, which report a 40% rejection rate for bank loans for their investment in intangible assets, twice as high as for their investment in tangible assets (BPI, 2020[77]).
Bpifrance's innovation loan guarantees and unsecured loans (prêts sans garanties, PSG) are helping adapt the banking model to the financing of innovative industrial projects by facilitating the mobilisation of intangible assets as collateral. Loan guarantees (in the range of 50-70%) reduce banking risk, and the delegation of guarantees of up to EUR 200 000 encourages banks to learn about financing innovation (BPI, 2025[73]). However, these loans still mainly go to traditional sectors such as catering, construction and retail. Only 16% go to industry, 18% to the services sector and 6% to the information and communication sector (Figure 4.13, Panel B). They could be directed more towards forward-looking sectors, while maintaining their current support for micro-enterprises with a financing gap.
The rapid expansion of these instruments, while desirable to fill the funding gap for innovative SMEs, particularly in the industrial sector, also entails certain macroeconomic risks. As Bpifrance’s unsecured loans (PSG) can crowd out private financing, priority should be given to public-private co-financing to ensure that risk is shared, with unsecured loans being reserved primarily for high-risk projects. Furthermore, the default rate for unsecured loans remains nearly three times higher than that of other loans (Cour des comptes, 2023[78]). Their strong growth over the last decade, reaching EUR 2.5 billion at the end of 2021, highlights the need for enhanced prudential monitoring and better modelling of default risk, as the difficulty in achieving self-financing through the scheme requires regular fiscal transfers.
Adjusting the insolvency system would complement the development of bank lending for innovative industrial projects. Although it is effective in terms of provisions that favour the reallocation of resources (OECD, 2024[43]; André and Demmou, 2022[79]), the French system could do more to address the challenges of an economy based on intangible assets. Accelerated procedures for the liquidation of these assets, as well as the development of a more liquid intellectual property market, would reduce banking risk. Only 2% of French SMEs hold industrial property rights (compared to 10% in the EU) and 30% cite a lack of knowledge as an obstacle (compared to 19% in the EU), illustrating their specific difficulties. Simplified patent procedures and streamlined bankruptcy processes could improve the securitisation of intangible-intensive firms.
4.4.4. Providing a competitive environment conducive to industrial dynamism
There has been significant consolidation of industrial activity in France, like in other OECD economies, affecting market structure and competitive dynamics, and with implications for innovation and technology diffusion. Between 2011 and 2022, the product market share of the four largest industrial groups increased by 4 percentage points, compared with an average of 1.1 points in the 25 countries reviewed, to reach 60%, a level that is nevertheless close to the OECD average (Calligaris et al., 2025[45]). High concentration is not a problem in itself, but it can become one if it limits new entrants and weakens competition. This OECD study also shows marked stability in dominant positions, slightly above the European average, suggesting reduced contestability. Similarly, the number of large companies investing heavily in R&D remains dominated by incumbents (such as Sanofi, Renault and PSA/Stellantis), with very little renewal among industrial leaders (Bock et al., 2025[14]; EU R&D Scoreboard, 2024[80]). These findings highlight the need for continued vigilance regarding anti-competitive practices and risks associated with barriers to entry and mergers and acquisitions.
The rise of deep tech, fusing digital industries, artificial intelligence and disruptive innovation start-ups, raises specific issues in terms of competition and maintaining a regulatory environment conducive to innovation. OECD analyses show that productivity gains from artificial intelligence depend critically on the spread of technology and market structure, and that highly concentrated upstream AI markets can lead to technology providers capturing a disproportionate share of value, to the detriment of a wider distribution of productivity gains across the economy (Filippicci et al., 2025[81]). Frequent acquisitions of start-ups by large groups can reduce the diversity of players and consolidate dominant positions, sometimes to the detriment of innovation (predatory acquisitions) (Berger et al., 2025[82]). In France, acquisitions of this kind appear to be in the minority, estimated at 6% (Hafied, Rachiq and Roulleau, 2021[66]). The French Competition Authority nevertheless highlights the risks of high barriers to entry in the AI sector due to computing, data and skills requirements, as well as the risks of concentration and technological lock-in, requiring, among other things, increased monitoring of the risks of anti-competitive practices (Autorité de la Concurrence, 2024[83]). The OECD recommends more proactive control of mergers, including ex post analyses to verify whether innovation is continuing, lower notification thresholds and ex ante counterfactual assessments (OECD, 2020[84]). These measures require strengthening the resources of the French Competition Authority so as to protect competitive dynamics in the knowledge economy.
4.5.1. Increasing the resilience of the industrial base to shocks in value chains
France remains exposed to shocks to global value chains
France's growing integration into global value chains, combined with the increased fragmentation of production processes, have increased its dependence on foreign inputs. Since 2000, the share of these inputs in industrial production has risen from 29% to 39% (Bonneau and Nakaa, 2020[85]). While this has helped boost productivity, it has also increased the economy's exposure to climate-related, health, and geopolitical shocks, as illustrated by the COVID-19 crisis and recent trade tensions. In this context, France has refocused part of its industrial policy on strengthening resilience to international shocks, notably through France Relance and France 2030, which support the reshoring of strategic inputs. This objective is consistent with other industrial priorities, in particular promoting the electric vehicle battery sector, as its production requires critical raw materials such as lithium, cobalt and nickel, which are concentrated in a limited number of countries.
The degree of France's vulnerability to external shocks varies depending on the indicators used. Measures based on trade concentration indicate relatively limited exposure as the concentration of French imports and exports is well below the OECD average, with an average of 152 imported products concentrated among a few partners, compared with an average of 256 products in the OECD (OECD, 2025[86]). Although the volume of these products is marginal, a disruption to them could nevertheless have significant repercussions.
OECD research indicates that France's vulnerability to global shocks appears to be higher when considering indirect links arising from global value chains, in addition to direct trade relations. Based on the METRO CGE model, France (like Canada, Germany, and the United Kingdom) is more exposed due to its close vertical links with major foreign economies (Arriola, Kowalski and van Tongeren, 2024[87]) (Figure 4.14, Panel A). Manufacturing sectors appear particularly vulnerable as they are highly integrated internationally, both in terms of access to inputs and end markets. A stress test carried out by the OECD on the supply of advanced technology products exported by China, including certain critical minerals, simulates a temporary 50% drop in exports to all of its trading partners. The estimated impact on France would be around 4% of production, and once transmission mechanisms via global value chains are taken into account, it appears sizeable in the short term, at around 15% of production, although it remains smaller than in other countries, such Germany and Korea. Furthermore, the shock tends to fade in the medium term as downstream producers adapt (Berthou, Haramboure and Samek, 2024[88]).
Finding a balance between reshoring, diversification and corporate flexibility
Faced with the risks associated with dependence on global value chains, particularly strategic inputs, reducing the vulnerability of the French economy requires a response that includes partial reshoring, diversification and increasing companies' flexibility when faced with shocks.
To reduce its dependence on critical minerals, France is supporting research and industrial projects aimed at optimising the circular economy, in particular through the recovery and reuse of materials contained in end-of-life batteries. It is also financing the reshoring of strategic inputs, in particular projects for the production, refining and recycling of critical materials. This strategy, led by France 2030, is part of the European Critical Raw Materials Act (CRMA) initiative, which sets a target of meeting 25% of strategic material needs through recycling by 2030.
These initiatives must be scaled up and coordinated at the European level to strengthen France's strategic autonomy and take advantage of the economic security offered by the single market in the face of rising protectionism. On the other hand, the scope of reshoring strategies should remain limited given the risks involved. OECD simulations comparing an ‘interconnected’ trade regime with a ‘localised’ regime indicate that a reshoring policy coupled with higher tariffs, subsidies and import restrictions could reduce France's GDP by around 5 percentage points, a magnitude comparable to the OECD average (Arriola et al., 2020[89]).
France has also implemented a number of measures to promote supplier diversification, notably in the form of Bpifrance’s support for exploring new markets. However, this strategy has its limits, as alternative suppliers may themselves depend on supply chains dominated by the country that is being avoided, thereby shifting vulnerability rather than eliminating it. This is particularly true in the context of the current trend towards reshoring or relocation to partner countries (OECD, 2025[86]). OECD research shows that, in order to be fully effective, diversification policies must be accompanied by adaptation measures aimed at increasing substitutability. This requires progress in the harmonisation and certification of key intermediate inputs, which must be achieved at European level (Berthou, Haramboure and Samek, 2024[88]).
Note: Panel B: The OECD average is a simple average. The boxplots show the minimum and maximum values of the index among OECD countries.
Source: OECD (2025[86]); OECD Services Trade Restrictiveness Index database.
France is somewhat behind in adopting digital technologies, including artificial intelligence (see Section 4.1). Increasing the pace of take-up could strengthen business resilience by enabling real-time monitoring of supply chains and faster responses to shocks (OECD, 2025[86]). Reducing barriers to trade in services can also boost the digital interconnection of supply chains. According to the STRI indicator (Figure 4.14, Panel B), restrictions on IT and telecommunications services in France are below the global average but remain above those in the best-performing countries, particularly due to licensing and qualification‑related limits on professional mobility, indicating scope for improvement (OECD, 2025[90]).
4.5.2. Consolidating strengths in artificial intelligence
Artificial Intelligence (AI) is a major potential enabler of competitiveness for industry through its effects on corporate productivity, innovation, and value chains (Filippucci et al., 2025[15]). However, realising these benefits will depend on the scale and pace of adoption, as well as the ability of companies to effectively integrate AI into their structures and develop it (see Figure 4.4, Panel B). The development of AI can also lead to a dominant, “winner-takes-all” situation, where a small number of players with control over data, computing power, and basic models can regulate access to technology for other companies, posing challenges in terms of economic sovereignty. In this context, France AI, the umbrella framework for implementing France’s national AI strategy, and the France 2030 plan identify artificial intelligence as a key technology for French industrial competitiveness, with a twofold ambition: to foster wider uptake and bolster France’s leadership in innovative, competitive AI models.
According to the OECD AI Index, France ranks among the most advanced OECD countries in terms of artificial intelligence ecosystems (Figure 4.15, Panel A). Along with Finland and Sweden, it is one of the first European countries to have adopted a national strategy on artificial intelligence. It devotes significant resources to R&D, notably through the National AI Research Programme launched in 2018 (nearly EUR 700 million, or 45% of the national strategy budget) (OECD, 2024[16]). That said, US investment remains over seven times higher, and significantly higher when adjusted for the size of the economies, suggesting that France needs to specialise in certain parts of the value chain (Fonteneau et al., 2025[91]) (Panel B).
Note: Panel B: the EU average is a simple average.
Source: OECD (2026[92]), data from AI Investment, last updated on 03 November 2025, accessed on 05 January 2026, https://oecd.ai/.
Access to computing infrastructure (computing power, cloud, etc.) is a prerequisite for development and innovation in AI and is therefore a matter of economic sovereignty for Europe. With 5% of the world's 500 largest supercomputers, France ranks far behind the United States and China, but among the best-performing European countries (OECD, 2024[16]). However, computing resources and expertise are increasingly dominated by large private players, widening the gap with the public sector and limiting its ability to develop and regulate advanced AI systems. In this context, a European consortium with EUR 7 billion in funding for the period 2021-2027 (The European High Performance Computing Joint Undertaking) is pooling computing capacities between countries and, since 2023, has been opening them up to start-ups, SMEs and researchers (OECD, 2024[16]). France's competitiveness could benefit from continued investment in supercomputers and the strengthening of public-private partnerships, which would facilitate private sector access to public infrastructure and researchers' access to advanced models owned by the private sector.
The integration of AI into robotics also offers new industrial opportunities, particularly in manufacturing, logistics and healthcare. This market is experiencing strong global growth, with autonomous mobile robots expected to see their market value increase from EUR 2.3 billion in 2020 to EUR 10.5 billion in 2030 (OECD, 2024[16]). In France, the “Organic Robotics” programme (EUR 34 million over eight years, France 2030) aims to develop robots that interact with humans, while the Pendragon project is creating AI-controlled land robots for defence purposes. The projected increase in military budgets represents an opportunity to strengthen the competitiveness of this sector by leveraging civil-defence synergies and financing high-potential dual-use projects, similar to the approaches implemented in the past in the United States and Israel, particularly for GPS and the Internet (see Chapter 1). This would require an increase in defence research efforts, which amount to 0.3% of GDP (less than 2% of defence spending), compared with 12.2% of GDP in the United States (47% of defence spending), 5% of GDP in the United Kingdom, and 2.3% of GDP in Germany (about 15% of defence spending) (OECD, Annual government expenditure by function (COFOG) database). It is also important to ensure that this expenditure does not crowd out other areas of research. An in-depth analysis of the complementarities between military and civilian research would make it possible to optimise synergies, achieve economies of scale, and limit the risk of crowding out.
More generally, in order to fully leverage productivity gains and strengthen industrial competitiveness, companies must invest in ancillary assets, such as digital capabilities, digital infrastructure, and ICT and management skills (Calvino and Fontanelli, 2024[93]). These points are addressed in the following section.
4.5.3. Preserving industrial competitiveness during the green transition
The scale and stringency of climate action in France (as measured by the OECD’s Environmental Policy Stringency indicator) is above the OECD average and has been rising over the past ten years (see Chapter 3). Climate action can weigh on the competitiveness of industrial companies, which account for 18% of national emissions, increase the risk of carbon leakage (Graveda and Galle, 2025[94]), and also create incentives for earlier investment in low‑carbon technologies. An OECD study estimates that 12% of additional carbon imports can be traced back to the environmental regulatory framework (Hemmerlé, Kruse and Pisu, 2025[95]) (Figure 4.16, Panel A). The carbon border adjustment mechanism (CBAM), combined with the phase-out of free allowances for products under the CBAM, provides a partial response to this risk by introducing a carbon price at the border for imports of carbon-intensive products covered by the mechanism, in order to align carbon costs between domestic production and imports.
However, the CBAM could result in higher production costs for companies that are heavily dependent on imported carbon-intensive raw materials, thereby undermining their competitiveness in export markets outside the European Union. OECD simulations suggest that France would be moderately affected at the aggregate level in a scenario combining the CBAM, a rise in carbon prices, and the removal of free carbon allowances, particularly due to the relatively lower weight of sectors exposed to the CBAM in terms of value added (Figure 4.16, Panel B). However, within the downstream sectors not covered by the CBAM, the impact appears to be very uneven, with the equipment and motor vehicle sectors being the most affected (Dechezleprêtre et al., 2025[96]). This suggests that compensatory mechanisms may be needed, such as targeted support for decarbonisation and energy efficiency efforts. Furthermore, the potential extension of the CBAM to other industries highlights the need to step up decarbonisation and energy efficiency efforts in order to maintain industrial competitiveness, the CBAM tending to shift demand towards countries with higher carbon prices and lower emissions intensity (Dechezleprêtre et al., 2025[96]).
Note: Panel A: the country-wide leakage rate is calculated as a weighted average with sectoral imported and domestic emission bases as weights. Panel B: the Full Policy Mix Scenario includes an increase in EU ETS price, the removal of free allowance and the CBAM and is compared to the baseline scenario across CBAM industries in EU countries. The exposure to CBAM is mapped to the size of the data point and computed as the share of the CBAM-covered sector in total value added in the country. Emission intensity is computed as the total emission of sectors over their production.
Source: Hemmerlé, Kruse et Pisu (2025[95]); Dechezleprêtre et al. (2025[96]).
Recent OECD analysis shows substantial dispersion in energy efficiency among French firms at a highly disaggregated product level (8-digit), with efficiency ratios ranging from 1 to 9. Bringing the least efficient firms up to the 25th percentile of their sector could reduce energy use by around 35%, while improving their productivity. At the same time, progress in energy efficiency is associated with higher productivity (De Lyon and Dechezleprêtre, 2025[97]). This highlights significant scope for energy-efficiency policies in France, which could strengthen firms’ resilience and competitiveness by reducing exposure to energy shocks. Reducing across-the-board support to energy consumption, including fossil-fuels subsidies (see Chapter 3) and redirecting it towards support conditional on energy-efficiency improvements would strengthen long-term competitiveness and firms’ ability to adjust to shocks.
France has adopted several measures aimed at decarbonising its production and developing a green technology industry, notably through the Green Industry Act (Loi Industrie Verte). Ecological transition contracts commit the highest-emitting companies to reducing their emissions by 45% by 2030. Progress in the green industrial transition will also depend on the electrification of processes and new energy efficient technologies, supported by the France 2030 plan, which allocates half of its budget to supporting key sectors (batteries, green hydrogen, carbon capture). In addition to encouraging investment in these sectors, a tax credit for Investments in Green Industry (C3V) was introduced in 2024, as well as specific credit guarantees for SMEs, which often face high capital costs. These measures seem appropriate and should be maintained, given the difficulties in accessing finance for the greening of productive capital. Access to finance is particularly limited for integrated technologies, such as the electrification of industrial processes (i.e. technologies incorporated into production processes and requiring organisational or technical adaptations) compared to ‘add-on’ technologies, such as the installation of filters or solar panels. The former require greater investment but also offer higher returns in terms of energy and production efficiency (Costa et al., 2024[98]). It would therefore be advisable to focus aid more on this type of investment, for example in the form of a higher rate.
France has one of the most decarbonised electricity mixes in the EU, largely as a result of nuclear power, which is a major asset for industrial decarbonisation and energy security. However, the nuclear sector faces growing difficulties, including delays and cost overruns in new reactors and very high investment needs (around EUR 460 billion by 2040) (Cour des comptes, 2025[99]). These challenges are weakening EDF, France’s main public electricity utility, and its ability to compete internationally. While the France 2030 strategy seeks to address these issues, progress is limited by skills shortages, supply-chain constraints, and project management challenges. At the same time, renewable energy deployment, especially solar, remains insufficient, notably in terms of grid integration and managing variability.
Financial institutions could also play a greater role in incentivising the decarbonisation of industry. In 2023, only 3-6% of non-financial corporate securities held by banks, insurers and funds were aligned with the European Taxonomy, while 30-45% concerned potentially eligible companies, contingent on their transition efforts (Jourde, Piquard and Salakhova, 2024[100]). Banks, which are particularly exposed to the construction and real estate sectors, could provide stronger incentives to decarbonise these industries through their lending decisions and associated interest rates, following the example of some segments of the financial markets, where companies with higher CSR ratings benefit from favourable interest rate differentials (D’Arcangelo et al., 2025[101]). To strengthen the role of financial institutions in decarbonisation, the French authorities could accelerate the integration of the Green Taxonomy and impose mandatory alignment indicators.
4.6.1. Better integrating PhD graduates into the corporate sector
PhD programmes provide training in key skills for applied research and are an essential factor in the transfer of knowledge between universities and industry (Afcha, García-Quevedo and Mas-Verdú, 2023[102]). Despite this fact, the number of PhD holders has declined over the last decade in France, while it has increased among leading innovators. PhD holders represent only 1% of 25–64-year-olds, compared with an average of 1.1% in the European Union, nearly 2.1% in the United States, 1.9% in Germany and 1.7% in the United Kingdom (OECD, 2025[103]). Furthermore, their presence in the private sector is limited, with 11% of PhD holders compared to almost 57% of engineers, who typically do not hold a PhD in France (Figure 4.17, Panel A) (Pommier and Lazarus, 2024[104]). This reflects the specific structure of the French higher education system, where it is engineering graduates from highly selective engineering programmes who traditionally occupy technical and managerial positions in companies rather than PhD holders. This situation is partly due to companies' low level of R&D investment, focus on development rather than research, and limited links with universities. Furthermore, France has traditionally favoured engineers, which can be attributed to the choice of specialisations but also to a negative perception of PhD holders, despite their proven productivity in terms of patents (Riedinger and Zaiem, 2013[105]). This suggests an imbalance of information that requires public intervention (ANRT, 2023[106]).
The main lever for integrating PhD graduates remains the CIFRE mechanism of industrial agreements for training through research, which each year supports more than 2 000 PhD students working in companies, under a joint supervision with a research public centre. PhD graduates under the CIFRE scheme are 17% more likely to obtain stable employment three years after completing their thesis, with a 7% salary increase and better integration into R&D, despite lower academic career opportunities (Malgouyres, 2020[107]). The Young Doctors scheme (Dispositif Jeunes Docteurs), abolished in 2025, temporarily doubled the research tax credit base for hiring young PhD graduates on permanent contracts (allowing firms to recover up to 120% of a young PhD’s salary cost through the research tax credit). While relatively dated, existing assessments suggest that the scheme helped to partially reduce the recruitment gap with engineers (Bernela et al., 2018[108]). However, the perceived quality of the jobs under the scheme remained low, suggesting windfall effects that justified its suppression. A replacement with a different scheme would be useful to correct the information imbalance between companies and university-trained PhD graduates. This would complement other initiatives aimed at improving the integration of PhD graduates, in particular the new France-PhD agency at public employment service France Travail and the PhD-Talent platform, both of which connect companies and PhD graduates.
4.6.2. Attracting highly skilled labour
Attracting highly skilled individuals is a key lever for industrial competitiveness, particularly in emerging sectors (AI, biotechnology, green energies), where skills shortages can be an obstacle to innovation. France is one of the most attractive countries for start-up founders, ranking third among OECD countries. It has the most favourable visa and admission policies for this category and stands out for the financial support it offers and the existence of specific pathways for international start-up employees (Andersson, 2025[109]). However, it appears to have less appeal for entrepreneurs and highly skilled workers. In terms of labour market integration, France ranks below the OECD average for the employment rate of highly skilled immigrants and the proportion of temporary contracts among this group (Figure 4.17, Panel B). Since 2019, France has nevertheless made progress in attracting foreign workers and entrepreneurs, but has fallen behind in attracting international students.
A closer examination of the flows of skilled AI specialists, who are highly sought after by competing countries, reveals that France lost more AI talent than it gained in 2022, marking a decline from 2019 when the situation was reversed (OECD, 2024[16]). France's relative lack of appeal is due in particular to high visa refusal rates for skilled workers, administrative burdens linked to the continuing lack of digitised visa procedures, complicated procedures, and high minimum capital requirements for entrepreneurs. Measures to attract skilled workers and facilitate the entry process need to be strengthened, and administrative procedures need to be simplified.
Note: Panel B: The OECD Talent Attractiveness index is a composite indicator assessing how attractive countries are for four talent profiles (highly skilled workers, entrepreneurs, international students, start-up founders). It aggregates multiple indicators across seven dimensions, quality of opportunities, income and taxes, future prospects, family environment, skills environment, inclusiveness, and quality of life, plus visa and residence permit accessibility, with equal weighting applied to dimensions. The figure for the OECD is a simple average.
Source: Ministère chargé de l'Enseignement Supérieur et de la Recherche (2024), État de l'Enseignement supérieur, de la Recherche et de l'Innovation en France n°17; OECD Dashboard Talent Attractiveness: How does your country compare in each dimension? (2024).
4.6.3. Using management to improve productivity and industrial competitiveness
Sound management practices encourage innovation and the adoption of digital technologies and are therefore a key lever for industrial competitiveness. In a rapidly changing world marked by the rise of AI, performance depends on flexibility, leveraging collective knowledge, and a culture of continuous learning (Lassébie and Quintini, 2022[110]). Earlier estimates suggest that differences in managerial practices could account for up to half of the productivity gap between France and the United States (Bloom, Sadun and Van Reenen, 2016[111]). In France, 43% of employees work in firms that have adopted a learning organisational model, i.e. firms where work practices encourage employee autonomy, cooperation, and problem-solving to sustain innovation over time, a proportion close to the EU average but well below the Nordic countries (54-62%) (Benhamou and Lorenz, 2020[112]). Levels of autonomy remain low, with 40% of workers reporting a high degree of autonomy, compared to over 50% in Europe (European Commission, 2020[113]). On the corporate side, shortcomings persist in setting objectives, mobilising teams, and providing incentive systems (Viganola and Yaya Diallo, 2025[114]) (Figure 4.18).
While public authorities cannot directly transform management, they can act as a catalyst, particularly by supporting SMEs, which are often ill-equipped to structure their management practices. The development of inter-company learning networks can contribute to this. The United Kingdom, for example, funds exchanges between mentors from large companies and managers of SMEs and has launched the Small Business Leadership Programme to offer small business managers access to high-level leadership training (Demmou and Franco, 2021[63]). The French authorities could seek inspiration from these initiatives, while also promoting good practices in public administrations could also have a knock-on effect on the private sector.
Involving trade unions and employee representatives in the organisation of work fosters autonomy, teamwork, and self-assessment, which are essential for innovation and performance. The public authorities could strengthen mandatory employer-employee negotiations by better integrating management issues as a factor in performance and well-being (IGF-IGAS-IGERS, 2024[115]). These negotiations, which currently focus on professional equality, quality of life and value sharing, could also include the organisation of work.
Note: Panel A: The OECD and EU averages are simple averages.
Source: OECD calculations based on data from World Bank Enterprise Surveys (WBES); Eurostat.
Highly competitive product markets (as discussed in Section 4.3) can also stimulate managerial efficiency by pushing managers to optimise their practices in order to remain competitive, which translates into better resource allocation, increased innovation, and tighter management of internal processes (Roy and Chakraborty, 2023[116]). Following the same mechanisms, less tariff protection and easier international trade increase competitive pressure on local businesses, encouraging management to keep improving (Bloom, Sadun and Van Reenen, 2016[111]).
4.6.4. Improving adult skills to better manage industrial transitions
Better aligning training with strategic priorities
Maintaining industrial competitiveness depends on the ability to adapt technical skills in response to the growing integration of digital technologies, including artificial intelligence, and robotics, which in turn calls for a greater emphasis on lifelong learning. An OECD simulation suggests that the sectoral skills gap between adults in France and those in the three best-performing countries accounts for more than a third of the sectoral productivity gap (Figure 4.19, Panel A) (Andrews, Egert and de la Maisonneuve, 2025[117]). However, according to OECD data, only 30% of adults take part in lifelong learning, compared with an OECD average of 40% and nearly double that in the highest-performing countries (Finland, Norway, United States), with this under-participation observed at all levels of education (OECD, 2025[118]) (Panel B). However, data from DARES indicate a slightly higher rate, with 36% of adults accessing job-related vocational training, as do Eurostat figures, which include initial training. Over and above the differences in the number of participants reported across surveys, which highlight differences in scope, the key issue lies in ensuring that the training programmes on offer are aligned with labour market requirements. Vocational training programmes have been expanded and strengthened under the Skills Investment Plan (PIC) and the call for expressions of interest in ‘Skills and careers of the future’ under the France 2030 investment plan. However, their ability to meet labour market needs could be improved in several areas.
The simplification of the Personal Training Account (CPF) scheme in 2018 increased access to short courses, but they are still not sufficiently aligned with companies' skills requirements and strategic priorities, which limits the courses’ impact on competitiveness, particularly for industrial SMEs. Around half of all new applicants enter training programmes in sectors in high demand (for a variety of reasons, such as a lack of appeal due to working conditions, low pay, or a skills shortage), and 20% in sectors in very high demand (DARES, 2025[119]). Nevertheless, fewer than one in five training courses are directed at professions in high demand due to a skills shortage (DARES, 2023[120]), and training courses related to industrial professions account for only around 6% of all courses, with little contribution from companies to finance more expensive technical training (IGF-IGAS-IGERS, 2024[115]). Better aligning training schemes with skill requirements would make them more effective, as training is not necessarily the best solution when the underlying problem is the lack of appeal of certain professions. Furthermore, it is important to ensure that schemes are aligned with the priorities of France 2030. For example, in Singapore, eligibility criteria give priority to training courses that are aligned with the country's strategic priorities and the needs of businesses in growth sectors (OECD, 2025[118]).
CPF-funded training courses continue to be predominantly short (averaging 89 hours, and 66 hours for those in employment) and fewer than 2% of them are certifying (Caisse des dépôts et consignation, 2025[121]). OECD surveys suggest that, by international standards, the duration is even shorter, with a higher-than-average proportion of short courses (ranging from one day to one week) within the OECD (OECD, 2025[118]). These characteristics limit the depth of learning and, ultimately, the effective acquisition of new skills. In addition, the development of modular and cumulative qualifications, which are more harmonised and recognised across professions, would make them more relevant to labour market needs, particularly in view of the spread of AI and rapid technological developments. Although qualifications in the National Directory of Professional Certification (RNCP) are now organised by skill sets, their standardisation across professions remains limited (Cour des comptes, 2023[122]). The development of micro-credentials with cross-sector recognition and better coordination between the CPF and certification programmes would make training courses more relevant to labour market needs.
Note: Panel A: Country-level productivity gap explained by sector-level PIAAC differences, compared to top 3 PIAAC performing countries in each sector. The bar for each country is a weighted (by value added share) average of the PIAAC contribution of each sector to overall productivity. The overall average is a simple average of the contributions in each country. PIAAC scores used for the calculations are the simple averages of the PIAAC scores on literacy, numeracy and problem solving. Panel B: Adults aged 25-65; formal and non-formal job-related learning in the 12 months prior to the survey. The OECD figure is a simple average.
Source: Andrews, Egert and de la Maisonneuve (2025[117]); OECD (2025[118]).
The Act of 24 October 2025, which transposes national cross-industry agreements to promote the employment of experienced employees and concerning the development of social dialogue includes, in particular, provisions on career transitions and retraining. It has simplified the retraining system by replacing the two separate schemes for internal mobility and skills development (Pro-A) and external retraining for employees in jobs at risk (Transco) with a single mechanism: the retraining period. This makes it possible to support an employee’s professional development or retraining, either within the company or at another company. In addition, the professional transition scheme allows an employee to take temporary leave from their post via a dedicated leave scheme to undertake certified training in order to change careers. This reform aims to make professional transitions clearer and more accessible, whereas the previous schemes were considered complex and not widely used.
Strengthening support mechanisms for industrial SMEs
Participation in lifelong learning remains much lower in SMEs, particularly small industrial enterprises: 15% of employees receive training in companies with fewer than 10 employees, compared with 61% in companies with 1 000 or more employees (IGF-IGAS-IGERS, 2024[115]). This disparity reflects higher opportunity costs, linked to the difficulty of replacing staff, the risk of losing trained employees to rivals, and more limited access to external financing. SMEs receive support from national authority France Compétences and skills operators (OPCOs), which devote 8 to 13% of their funds to skills development in companies with fewer than 50 employees. However, human resource support mechanisms remain limited, with fewer than 5 000 SMEs receiving support each year, out of more than 55 000 in the 32 industrial sectors (IGF-IGAS-IGERS, 2024[115]). Targeted reinforcement of public support for SMEs would help to accompany the digital and industrial transitions. Additional measures, such as pooling schemes for temporarily replacing employees, early exit reimbursement clauses, and public loans, could also reduce barriers to training. Progress in adopting digital technologies, especially AI, is also essential, with modular digital learning formats providing the flexibility SME employees need.
Strengthening the upskilling of less qualified and older workers
A lack of basic literacy and numeracy skills limits a workforce's ability to acquire more technical skills, thereby impeding adaptation to technological change and the industry's capacity to embrace innovation and new technologies. Nearly 30% of adults in France have low literacy, numeracy, and problem-solving skills but less than 20% of this group receive training (OECD, 2025[118]). The proportion is even lower for certified training courses, which are attended by less than 4% of low-skilled adults (DARES, 2023[120]). Conversely, over 60% of individuals in the highest skill category (level 4) take part in training, a significantly greater proportion than in most OECD countries (OECD, 2025[118]). Similarly, the participation gap between tertiary graduates and those less qualified is one of the highest among OECD countries (Figure 4.19, Panel B).
The participation of older workers in training is among the lowest in the OECD, at less than 20%. This is a major challenge in France, where the OECD’s PIAAC survey observes a more marked decline of skills over the lifespan than the OECD average. This reflects the continuing inadequacy of support throughout working life, particularly in terms of helping people adapt to changes in the labour market and technological developments. The National Inter-professional Agreement (ANI) on the employment of older workers, which was enacted in October 2025, sets out several training schemes designed to enhance the employability of older workers and facilitate their career changes, notably through more in-depth career progression interviews, the introduction of a certified retraining period, and wider access to the Validation of Acquired Experience mechanism. These provisions are a welcome development. It is also important to provide training proactively, before skills erode significantly, particularly for workers approaching mid-career stages. In Singapore, the SkillsFuture scheme provides extra targeted support in mid-career (over 40 years of age) to prevent skills obsolescence and could serve as a good practice to consider (OECD, 2025[123]). Developing modular courses, as discussed above, would also help foster participation in training.
4.6.5. Improving the quality of career guidance in schools
Labour shortages in France remain high, particularly in industry, even if they have fallen significantly (Figure 4.20, Panel A), which partly reflects the low appeal of career paths in industry. According to projections by DARES and France Stratégie, around 141 000 industrial jobs could remain unfilled by 2030, mainly for technicians, engineers, and skilled workers, despite a moderate net job supply linked to deindustrialisation and retirements. However, demand for training in these professions remains limited. The proportion of students in STEM subjects at secondary level, two-year post-secondary level, and training to be technicians, remains below the European average, while it exceeds the OECD average at master's level. This lack of interest is partly explained by insufficient career guidance and difficulties in finding employment, with vocational graduates having a lower employment rate than the OECD average (OECD, 2025[103]).
In order to better meet labour market demands, particularly in the industrial sector, France introduced reforms to vocational secondary education in 2023 and apprenticeships in 2018, strengthening links between educational institutions and businesses through work-study programmes (OECD, 2024[43]). In 2023, 33% of upper secondary students were in work-study programmes, up from 28% in 2021 but still below the OECD average (45%) (OECD, 2025[103]) (Figure 4.20, Panel B). Furthermore, growth in apprenticeships has mainly benefited higher levels of educational attainment, with 44% of apprentices having completed five years of higher education, compared with 38% at secondary level (Demongeot and Lombard, 2024[124]). The 2025 reform, which refocuses financial support to young people with less than three years of higher education and priority occupations, is welcome given the cost of the scheme and its windfall effects. Making work-study programmes a standard feature of secondary schools and higher technician certificate programmes (BTS) and strengthening the quality assurance of training courses remain also essential. Other non-financial measures could make apprenticeships more attractive to employers, particularly SMEs, including support, mentoring, and educational engineering services (OECD, 2022[125]).
Note: Panel A: The synthetic tension indicator developed by Dares and Pôle Emploi reports for each profession and each geographic area recruitment difficulty anticipated by employers, the number of job vacancies relative to the number of job seekers, and the ease with which job seekers exit Pôle Emploi's lists. The tension indicator is then standardised, i.e. centred and reduced, using the 2014-2018 period and all professions as a baseline. A value of 0 corresponds to a level of tension identical to that of the reference period for all professions and a negative (positive) value indicates lower (higher) tension.
Source: Ducatel V., Villedieu P., Chartier F. et Lainé F. (2026), Les tensions sur le marché du travail en 2024, Dares Résultats, n° 5; and OECD Education at a Glance 2025 database.
The previous Economic Survey also highlighted the need to strengthen lifelong learning for teachers in STEM subjects, who appear to be less well trained than the OECD average, which limits the educational use of technology (OECD, 2024[126]). Furthermore, STEM subjects are taught in silos in France, whereas integrated approaches have been shown to boost student engagement, creativity and problem-solving skills, and could help to make science and technology courses more attractive (OECD, 2018[127]).
France is notable for the strong correlation between socio-economic background and academic performance, resulting in significant growth losses and a sub-optimal allocation of human capital (Andrews, Égert and de la Maisonneuve, 2024[23]) (see above Figure 4.7, Panel A). A potentially large pool of innovators remains untapped; at equivalent levels, a child in the top 1% income bracket is seven times more likely to become an innovator than a child whose parents are below the median income level (Feng, Jaravel and Richard, 2022[128]). The concentration of disadvantaged pupils, which is higher than the OECD average, reinforced by residential segregation and the growth of private schools under state contract, is associated with lower academic performance (OECD, 2019[129]; Claes and Moulin, 2025[130]).
The previous Economic Survey presented an in-depth analysis of the French education system and made several recommendations aimed at improving its performance and reducing inequalities (Table 4.1). In particular, it recommended increasing diversity, targeting resource allocation more effectively, improving the attractiveness of the teaching profession by raising salaries and giving greater recognition to career paths, attracting experienced teachers to priority schools, and increasing the autonomy of schools. Furthermore, the French education system continues to lag behind in the use of new cognitive activation teaching practices that enable the development of cross-curricular skills essential for lifelong learning, requiring greater teacher training in these areas (OECD, 2024[43]).
References
[102] Afcha, S., J. García-Quevedo and F. Mas-Verdú (2023), “Gaining or losing PhDs: What are the effects on firms’ linkages with universities?”, Technological Forecasting and Social Change, Vol. 186, p. 122211, https://doi.org/10.1016/j.techfore.2022.122211.
[48] Aghion, P., N. Chanut and X. Jaravel (2022), “Renforcer l’impact du Crédit d’impôt recherche”, Conseil d’analyse économique Focus 90, https://cae-eco.fr/static/pdf/cae_Focus090.pdf.
[56] Alvarez, B. et al. (2025), “Enseignements des politiques industrielles passées”, Trésor Eco Direction Générale du Trésor 358, https://www.tresor.economie.gouv.fr/Articles/c94a3af3-2a9b-4167-8f95-0105a5bf44bf/files/1effaa8d-cc75-4882-9bd4-4095af960921.
[109] Andersson, L. (2025), “Measuring and assessing talent attractiveness in OECD countries, second edition: Methodology paper”, OECD Social, Employment and Migration Working Papers, No. 318, OECD Publishing, Paris, https://doi.org/10.1787/133b6085-en.
[40] André, C. et al. (2023), “Rising energy prices and productivity: short-run pain, long-term gain?”, OECD Economics Department Working Papers, No. 1755, OECD Publishing, Paris, https://doi.org/10.1787/2ce493f0-en.
[79] André, C. and L. Demmou (2022), “Enhancing insolvency frameworks to support economic renewal”, OECD Economics Department Working Papers, No. 1738, OECD Publishing, Paris, https://doi.org/10.1787/8ef45b50-en.
[17] Andrews, D., C. Criscuolo and P. Gal (2016), The Best versus the Rest: The Global Productivity Slowdown, Divergence across Firms and the Role of Public Policy, OECD Productivity working papers, https://doi.org/10.1787/63629cc9-en.
[117] Andrews, D., B. Egert and C. de la Maisonneuve (2025), “Adult Skills and Productivity: New Evidence from PIAAC 2023”, OECD Economics Department Working Papers, Vol. 1834, https://www.oecd.org/content/dam/oecd/en/publications/reports/2025/06/adult-skills-and-productivity_5e9a03df/12ac6e8c-en.pdf.
[23] Andrews, D., B. Égert and C. de la Maisonneuve (2024), “From decline to revival: Policies to unlock human capital and productivity”, OECD Economics Department Working Papers 1827, https://doi.org/10.1787/8d0d232c-en.
[106] ANRT (2023), “Pour un grand plan national pour le doctorat”, https://www.anrt.asso.fr/sites/default/files/2024-start/ANRT_pour_un_grand_plan_national_pour_le_doctorat_oct.2023.pdf.
[51] Appelt et al. (2023), “The Impact of R&D Tax Incentives: Results from the OECD microBeRD+ Project”, OECD Science, Technology, and Industry Paper 159, https://www.oecd.org/content/dam/oecd/en/publications/reports/2023/10/the-impact-of-r-d-tax-incentives_bc42ab04/1937ac6b-en.pdf.
[55] Appelt, S. et al. (2016), “R&D Tax Incentives: Evidence on design, incidence and impacts”, OECD Science, Technology and Industry Policy Papers, No. 32, OECD Publishing, Paris, https://doi.org/10.1787/5jlr8fldqk7j-en.
[89] Arriola, C. et al. (2020), “Efficiency and risks in global value chains in the context of COVID-19”, OECD Economics Department Working Papers, No. 1637, OECD Publishing, Paris, https://doi.org/10.1787/3e4b7ecf-en.
[87] Arriola, C., P. Kowalski and F. van Tongeren (2024), “Shocks in a highly interlinked global economy”, OECD Trade Policy Papers, No. 283, OECD Publishing, Paris, https://doi.org/10.1787/4a2f0127-en.
[83] Autorité de la Concurrence (2024), “Avis relatif au fonctionnement concurrentiel du secteur de l’intelligence artificielle générative”, https://www.autoritedelaconcurrence.fr/fr/avis/relatif-au-fonctionnement-concurrentiel-du-secteur-de-lintelligence-artificielle-generative.
[49] Bach et al. (2021), “Les impacts du crédit impôt recherche sur la performance économique des entreprises”, Rapport IPP 33, https://www.ipp.eu/wp-content/uploads/2021/06/impact-credit-impot-recherche-performance-entreprises-juin-2021.pdf.
[33] Barreto et al. (2025), “Low-wage employment in France: A cross-country perspective”, OECD Social, Employment and Migration Working Paper 313, https://www.oecd.org/content/dam/oecd/en/publications/reports/2025/01/low-wage-employment-in-france_ac641c84/82539f44-en.pdf.
[13] Bellit, S. and V. Charlet (2023), “L’innovation de rupture, terrain de jeu exclusif des start-up?”, Les Notes de La Fabrique, https://www.la-fabrique.fr/wp-content/uploads/2023/12/n48-innovation-de-rupture_terrain-de-jeu-exclusif-des-start-up_web.pdf.
[22] Ben Hassine, H. and R. Fotso (2025), “Quelle dynamique des PME et des ETI en France depuis 1994 ?”, Document de Travail France stratégie, https://www.strategie-plan.gouv.fr/files/2025-06/FS-2025-DT-01-ETI_1.pdf.
[112] Benhamou, S. and E. Lorenz (2020), “Les organisations du travail apprenantes : enjeux et défis pour la France”, Document de travail France Stratégie, https://www.strategie-plan.gouv.fr/publications/organisations-travail-apprenantes-enjeux-defis-france.
[82] Berger, M. et al. (2025), “Acquisitions and their effect on start-up innovation: Stifling or scaling?”, OECD Science, Technology and Industry Working Papers, No. 2025/10, OECD Publishing, Paris, https://doi.org/10.1787/b4efd3ab-en.
[108] Bernela et al. (2018), “Une évaluation des effets du Dispositif Jeunes Docteurs sur l’accès aux emplois de R&D”, Rapport France Stratégie, https://www.strategie-plan.gouv.fr/files/files/Publications/Rapport/rapport_giret_et_al_djd_cir_version_octobre_2018%20%281%29.pdf.
[10] Berthou, A. and G. Gaulier (2021), “Quel bilan de la compétitivité prix et coût dans les exportations de la France depuis le début des années 2000 ?”, Bulletin de la Banque de France, Vol. 235/6, https://www.banque-france.fr/system/files/2023-01/bdf235-6_competitivite-prix-cout.pdf.
[88] Berthou, A., A. Haramboure and L. Samek (2024), “Mapping and testing product-level vulnerabilities in granular production networks”, OECD Science, Technology and Industry Working Papers, No. 2024/02, OECD Publishing, Paris, https://doi.org/10.1787/9bcde495-en.
[111] Bloom, N., R. Sadun and J. Van Reenen (2016), “Management as a technology?”, NBER Working Paper 22327, https://www.nber.org/system/files/working_papers/w22327/revisions/w22327.rev0.pdf.
[14] Bock, S. et al. (2025), “L’Europe sous contraintes: des coûts maitrisés, des gammes maitrisées”, OFCE Policy Brief, Vol. 148, https://www.ofce.sciences-po.fr/pdf/pbrief/2025/OFCEpbrief148.pdf.
[85] Bonneau, C. and M. Nakaa (2020), “Vulnérabilité des approvisionnements français et européens” 274.
[35] Bozio, A. and E. Wasmer (2024), “Les politiques d’exonérations de cotisations sociales : une inflexion nécessaire”, France Stratégie, https://www.strategie-plan.gouv.fr/publications/mission-bozio-wasmer-politiques-dexonerations-de-cotisations-sociales-une-inflexion.
[73] BPI (2025), “Action de BPI France en faveur des micro-entreprises, PME et ETI françaises”, https://www.bpifrance.fr/sites/default/files/2025-03/2024_Impact%20de%20Bpifrance%20en%202023.pdf.
[65] BPI (2025), “Capital risque: 10 ans d’action de BPI France”, Le Lab BPI France, https://lelab.bpifrance.fr/Etudes/capital-risque-en-10-ans-bpifrance-a-fortement-contribue-a-l-essor-du-tissu-de-financement-des-start-ups-francaises.
[77] BPI (2020), “Evaluation de l’impact économique des prêts sans garanties de BPI France”, https://www.bpifrance.fr/sites/default/files/2021-08/2020%2006%2023%20-%20Rapport%20%C3%A9valuation%20pr%C3%AAts%20sans%20garantie%20Bpifrance%20VF%20%281%29.pdf.
[19] Bunel, S. et al. (2025), “Revisiting the European performance gap vis-a-vis the United States”, Eco Notepad 391, https://www.banque-france.fr/en/publications-and-statistics/publications/revisiting-european-performance-gap-vis-vis-united-states.
[121] Caisse des dépôts et consignation (2025), Mon compte formation Rapport annuel 2024, https://politiques-sociales.caissedesdepots.fr/sites/default/files/Rapport-Annuel-MCF_2024-SA.pdf.
[45] Calligaris, S. et al. (2025), “Concentration and business dynamics in product markets”, OECD Science, Technology and Industry Working Papers, No. 2025/04, OECD Publishing, Paris, https://doi.org/10.1787/d714066e-en.
[93] Calvino, F. and L. Fontanelli (2024), Ai Users are Not All Alike: The Characteristics of French Firms Buying and Developing Ai, Elsevier BV, https://doi.org/10.2139/ssrn.5010540.
[32] Carbonnier, C. et al. (2017), “Evaluation interdisciplinaire des impacts du CICE en matière de compétitivité internationale, d’investissement, d’emploi, de résultat net des entreprises et de salaires”, Science Po Rapport, https://www.strategie-plan.gouv.fr/files/files/Publications/2017%20SP/2017-10-04%20-%20Rapport%202017%20du%20comit%C3%A9%20de%20suivi%20du%20CICE/liepp_rapport_final_25_septembre_2017_0.pdf.
[61] CB Insight (2026), The Unicorn list, https://www.cbinsights.com/research-unicorn-companies.
[59] Cherif, R. and F. Hasanov (2025), “Industrial Policy, Asian Miracle Style”, Journal of Economic Perspectives, Vol. 39/4, pp. 101-125, https://doi.org/10.1257/jep.20251448.
[130] Claes and Moulin (2025), “Private vs. Public Schooling: The Role of School Composition”, Document de travail INED, https://www.ined.fr/fichier/s_rubrique/35889/dt304.claes_moulin_compressed.en.pdf.
[60] Commission des Finances (2025), “Les dispositifs de valorisation de la recherche et leur financement”, Rapport d’information de l’Assemblée Nationale, https://www.abg.asso.fr/js/kcfinder/upload/files/Rapport_AN_Financement_dispositifs_TT.pdf.
[98] Costa, H. et al. (2024), “Making the grass greener: The role of firm’s financial and managerial capacity in paving the way for the green transition”, OECD Economics Department Working Papers, No. 1791, OECD Publishing, Paris, https://doi.org/10.1787/cdffe7eb-en.
[57] Cour des comptes (2025), “Analyse de l’exécution budgétaire 2024: Mission Investir pour la France de 2030”, Rapport, https://www.ccomptes.fr/sites/default/files/2025-04/NEB-2024-Investir-pour-France-2030.pdf.
[99] Cour des comptes (2025), “Le modèle économique d’Électricité de France”, https://www.ccomptes.fr/fr/publications/le-modele-economique-delectricite-de-france-edf.
[27] Cour des comptes (2025), “Maîtriser la dynamique des allégements généraux de cotisations sociales”, https://www.ccomptes.fr/sites/default/files/2025-05/20250526-RALFSS-2025-Maitriser-dynamique-allegements-generaux-de-cotisations-sociales.pdf.
[12] Cour des comptes (2024), “Dix ans de politiques publiques en faveur de l’industrie”, https://www.ccomptes.fr/sites/default/files/2024-11/20241128-10-ans-de-politiques-publiques-en-faveur-industrie.pdf.
[78] Cour des comptes (2023), “BPI France”, Rapport portant sur une entreprise publique, https://www.ccomptes.fr/sites/default/files/2023-10/20230612-RPEP-S2023-0494-1-BPIfrance.pdf.
[122] Cour des comptes (2023), “La formation professionelle des salariés”, https://ccomptes.fr/system/files/2023-06/20230630-Formation-professionnelle-salaries.pdf.
[3] Criscuolo, C. et al. (2022), “An industrial policy framework for OECD countries: Old debates, new perspectives”, OECD Science, Technology and Industry Policy Papers, No. 127, OECD Publishing, Paris, https://doi.org/10.1787/0002217c-en.
[101] D’Arcangelo, F. et al. (2025), “The effect of climate policies on firm financing and investment through the banking channel”, Journal of Environmental Management, Vol. 387, p. 125866, https://doi.org/10.1016/j.jenvman.2025.125866.
[26] DARES (2025), “Dépenses en faveur de l’emploi et du marché du travail en 2023”, DARES RESULTATS 14, https://dares.travail-emploi.gouv.fr/sites/default/files/6d34ddfb9e82cff1e65ef32725857f88/Dares_R%C3%A9sultats_D%C3%A9penses%20en%20faveur%20de%20l%27emploi%20et%20du%20march%C3%A9%20du%20travail%20en%202023.pdf.
[119] DARES (2025), Rapport final du comité scientifique de l’évaluation du Plan d’investissement dans les compétences, https://dares.travail-emploi.gouv.fr/sites/default/files/a18512b3397c9b05c9001f8f4d7e5e43/Rapport%20final%20du%20comit%C3%A9%20scientifique%20de%20l%27%C3%A9valuation%20du%20PIC_FINAL.pdf.
[120] DARES (2023), “Quatrième rapport du comité scientifique de l’évaluation du Plan d’investissement dans les compétences”, Ministère de travail, de l’emploi et de l’insertion, https://dares.travail-emploi.gouv.fr/sites/default/files/d632851009ae84c3cd386c33ff6d1a16/4eme%20rapport_CS_PIC.pdf.
[97] De Lyon, J. and A. Dechezleprêtre (2025), “The great dispersion in energy productivity between firms”, OECD Science, Technology and Industry Working Papers, No. 2025/24, OECD Publishing, Paris, https://doi.org/10.1787/0fa543ce-en.
[96] Dechezleprêtre, A. et al. (2025), “Carbon Border Adjustments: The potential effects of the EU CBAM along the supply chain”, OECD Science, Technology and Industry Working Papers, No. 2025/02, OECD Publishing, Paris, https://doi.org/10.1787/e8c3d060-en.
[8] Demmou, L. (2010), Le recul de l’emploi industriel en France entre 1980 et 2007 - Ampleur et principaux déterminants : un état des lieux, Economie et Statistiques, https://www.insee.fr/fr/statistiques/1377172?sommaire=1377175.
[63] Demmou, L. and G. Franco (2021), “Mind the financing gap: Enhancing the contribution of intangible assets to productivity”, OECD Economics Department Working Papers, No. 1681, OECD Publishing, Paris, https://doi.org/10.1787/7aefd0d9-en.
[124] Demongeot, A. and F. Lombard (2024), “L’apprentissage au 31 décembre 2023”, Note d’information DEPP, https://www.education.gouv.fr/l-apprentissage-au-31-decembre-2023-414743.
[46] Dimas, P., D. Stamopoulos and A. Tsakanikas (2022), “Intangibles and Participation in Global Value Chains in the EU: Evidence from the GLOBALINTO Input-Output Intangibles Database”, Economic and Business Review, Vol. 24/3, pp. 152-160, https://doi.org/10.15458/2335-4216.1303.
[75] Durand, R. and R. Boulongne (2021), “Investissement dans l’Immatériel pour l’Industrie”, Rapport BPI France à la Ministre Déléguée Chargée de l’Industrie, https://presse.bpifrance.fr/bpifrance-presente-la-premiere-etude-a-grande-echelle-sur-le-lien-entre-investissement-immateriel-et-performances-des-entreprises-francaises.
[7] Duranton, G. and D. Puga (2004), “Micro-Foundations of Urban Agglomeration Economies”, Handbook of Regional and Urban Economics: Cities and Geography, Vol. 4, https://www.sciencedirect.com/handbook/handbook-of-regional-and-urban-economics/vol/4/suppl/C.
[76] Erel, I. (ed.) (2022), “Firm Listing Status, Firm Size, and the Financing of Investment”, The Review of Corporate Finance Studies, Vol. 13/3, pp. 818-857, https://doi.org/10.1093/rcfs/cfac038.
[80] EU R&D Scoreboard (2024), “2024 EU industrial R&D investment scoreboard”, Office des publications de l’Union européenne, https://data.europa.eu/doi/10.2760/0775231.
[24] European Commission (2023), “SMEs and skills shortages France”, Flash Eurobarometer 537, https://europa.eu/eurobarometer/surveys/detail/2961.
[113] European Commission (2020), “Job autonomy and pressure at work”, Statistics, https://ec.europa.eu/eurostat/statistics-explained/index.php?title=Job_autonomy_and_pressure_at_work_-_statistics.
[128] Feng, J., X. Jaravel and E. Richard (2022), “Pour une stratégie nationale d’innovation pour tous”, Note du conseil d’analyse économique, https://cae-eco.fr/static/pdf/cae_Focus089.pdf.
[81] Filippicci, F. et al. (2025), Macroeconomic productivity gains from Artificial Intelligence in G7 countries, OECD Artificial Intelligence Papers Series, https://www.oecd.org/content/dam/oecd/en/publications/reports/2025/06/macroeconomic-productivity-gains-from-artificial-intelligence-in-g7-economies_dcf91c3e/a5319ab5-en.pdf.
[15] Filippucci, F. et al. (2025), “Macroeconomic productivity gains from Artificial Intelligence in G7 economies”, OECD Artificial Intelligence Papers, No. 41, OECD Publishing, Paris, https://doi.org/10.1787/a5319ab5-en.
[41] Fontagné, L., P. Martin and G. Orefice (2024), “The many channels of firm’s adjustment to energy shocks: evidence from France”, Economic Policy, Vol. 39/117, pp. 5-43, https://doi.org/10.1093/epolic/eiae011.
[6] Fontagné, L., P. Mohnen and G. Wolff (2014), “Pas d’industrie, pas d’avenir ?”, Les notes du conseil d’analyse économique 13, https://cae-eco.fr/Pas-d-industrie-pas-d-avenir.
[91] Fonteneau, F. et al. (2025), “Advancing the measurement of investments in artificial intelligence”, OECD Artificial Intelligence Papers, No. 47, OECD Publishing, Paris, https://doi.org/10.1787/13e0da2f-en.
[71] France Angel (2024), “Bilan de l’activité 2023 des Business Angels”, https://actubusinessangels.com/2024/03/26/bilan-de-lactivite-2023-des-business-angels/.
[67] Fratto, C. et al. (2024), “The scale-up gap”, European Investment Bank, https://www.eib.org/attachments/lucalli/20240130_the_scale_up_gap_en.pdf.
[25] Gallois, L. (2012), “Pacte pour la compétitivité de l’industrie francaise”, Rapport pour le Premier Ministre, https://www.vie-publique.fr/files/rapport/pdf/124000591.pdf.
[20] Gal, P. et al. (2026), Business investment in the face of the digital transformation: Initial evidence, OECD Economics Department Working Papers, https://www.oecd.org/content/dam/oecd/en/publications/reports/2026/03/business-investment-in-the-face-of-the-digital-transformation_e10f8cdd/4f89aa3e-en.pdf.
[38] Goutebroze, H. and A. Lapierre (2025), Compétitivité industrielle et prélèvements obligatoires, Conseil des Prélèvements Obligatoires, https://www.ccomptes.fr/sites/default/files/2025-09/20250922-Tracer-cadre-fiscal-et-social-pluriannuel%20pour-industrie-francaise-rapport-particulier-4_0.pdf.
[94] Graveda, M. and Q. Galle (2025), Garantir la compétitivité d’une industrie décarbonée par une tarification efficace du carbone, DGE Thema, https://www.entreprises.gouv.fr/files/files/Publications/2025/Th%C3%A9mas/thema-35-tarification-carbone.pdf.
[34] Groupe d’Experts SMIC (2025), Salaire minimum interprofessionnel de croissance, https://www.vie-publique.fr/rapport/301429-rapport-du-groupe-dexperts-smic-25-novembre-2025.
[30] Guillou, S. et al. (2016), “L’impact du CICE sur la marge intensive des exportateurs”, Rapport d’évaluation pour France Stratégie, https://www.ofce.sciences-po.fr/pdf-articles/actu/rapport-de-recherche-de-ofce.pdf.
[66] Hafied, F., C. Rachiq and G. Roulleau (2021), “Capital-risque et développement des start-ups”, Trésor-Eco Direction Général du Trésor, https://www.tresor.economie.gouv.fr/Articles/f830eb26-dce7-4b50-91b9-3df5607fce17/files/59b4e67d-3fb6-47fb-9e6c-c43a6c8a5a97.
[50] Harfi, M. and R. Lallement (2021), “Evaluation du crédit d’Impot recherche”, Avis de la CNEPI France Stratégie, https://www.strategie-plan.gouv.fr/files/files/Publications/2021/0601%20CNEPI/fs-2021-rapport-cnepi-cir-juin.pdf.
[62] Haskel, J. and S. Westlake (2017), Capitalism without Capital, Princeton University Press, https://doi.org/10.2307/j.ctvc77hhj.
[5] Hausmann, R., J. Hwang and D. Rodrik (2006), “What you export matters”, Journal of Economic Growth, Vol. 12/1, pp. 1-25, https://doi.org/10.1007/s10887-006-9009-4.
[95] Hemmerlé, Y., T. Kruse and M. Pisu (2025), “How different climate policies affect carbon leakage through trade: Cross-country evidence for the manufacturing sector”, OECD Economics Department Working Papers, No. 1851, OECD Publishing, Paris, https://doi.org/10.1787/5819ce91-en.
[115] IGF-IGAS-IGERS (2024), “Tensions sur les effectifs et compétences dans l’industrie et dispositifs de formation associés”, Rapport au Gouvernement, https://www.enseignementsup-recherche.gouv.fr/sites/default/files/2023-11/22-23-130a-30270.pdf.
[100] Jourde, T., T. Piquard and D. Salakhova (2024), “Les acteurs financiers français face au défi de la transition climatique”, Banque de France Bloc-notes Eco, https://www.banque-france.fr/fr/publications-et-statistiques/publications/les-acteurs-financiers-francais-face-au-defi-de-la-transition-climatique.
[131] Kalantzis, Y. and C. Thubin (2017), Les causes de la d;esindustrialisation en France, Bloc-notes Éco Banque de France, https://www.banque-france.fr/fr/publications-et-statistiques/publications/les-causes-de-la-desindustrialisation-en-france.
[9] Kalantzis, Y. and C. Thubin (2017), Les causes de la désindustrialisation en France, Bloc-notes Éco Banque de France, https://www.banque-france.fr/fr/publications-et-statistiques/publications/les-causes-de-la-desindustrialisation-en-france.
[110] Lassébie, J. and G. Quintini (2022), “What skills and abilities can automation technologies replicate and what does it mean for workers?: New evidence”, OECD Social, Employment and Migration Working Papers, No. 282, OECD Publishing, Paris, https://doi.org/10.1787/646aad77-en.
[28] L’Horty, Y., P. Martin and T. Mayer (2019), “Baisses de charges : stop ou encore ?”, Notes du conseil d’analyse économiques 19, https://shs.cairn.info/revue-notes-du-conseil-d-analyse-economique-2019-1-page-1?lang=fr#s2n7.
[53] Lhuillery, S. et al. (2021), “La R&D des groupes francais et le CIR”, Rapport CNEPI pour France Stratégie, https://www.strategie-plan.gouv.fr/files/files/Publications/2021/0601%20CNEPI/2021_rapport_final_neoma_bs_cir_et_rd_des_groupes.pdf.
[107] Malgouyres, C. (2020), “Évaluation des effets du dispositif Cifre sur les entreprises et les doctorants participants”, Rapport de l’IPP 28, https://www.ipp.eu/wp-content/uploads/2020/10/evaluation-effets-cifre-entreprises-doctorants-ipp-octobre-2020.pdf.
[29] Malgouyres, C. (2019), “Coût du travail et exportations : analyses sur données d’entreprises”, Rapport Instituts des Politiques Publiques 20, https://www.ipp.eu/publication/janv-2019-cout-du-travail-et-exportations-analyses-sur-donnees-dentreprises/.
[31] Malgouyres, C. and T. Mayer (2018), “Exports and labor costs: evidence from a French policy”, Review of World Economics, Vol. 154/3, https://sciencespo.hal.science/hal-03391928/document.
[69] Midy, P. (2023), “Soutenir l’investissement dans les startups, PME innovantes et PME de croissance”, https://mission-midy.fr/assets/rapport.pdf.
[2] Millot, V. and Ł. Rawdanowicz (2024), “The return of industrial policies: Policy considerations in the current context”, OECD Economic Policy Papers, No. 34, OECD Publishing, Paris, https://doi.org/10.1787/051ce36d-en.
[72] OECD (2026), Financing SMEs and Entrepreneurs 2026: An OECD Scoreboard, OECD Publishing, Paris, https://doi.org/10.1787/075d8058-en.
[47] OECD (2026), Tax incentives for R&D and innovation, http://France - Details | INNOTAX Portal.
[92] OECD (2026), The OECD.AI Index, OECD Publishing, Paris, https://doi.org/10.1787/32c01014-en.
[64] OECD (2025), “Benchmarking government support for venture capital: A comparative analysis”, OECD SME and Entrepreneurship Papers, No. 71, OECD Publishing, Paris, https://doi.org/10.1787/81e53985-en.
[103] OECD (2025), Education at a Glance 2025: OECD Indicators, OECD Publishing, Paris, https://doi.org/10.1787/1c0d9c79-en.
[44] OECD (2025), Licensing and Permitting: How to Manage Risks While Supporting Growth, OECD Best Practice Principles for Regulatory Policy, OECD Publishing, Paris, https://doi.org/10.1787/68fc3301-en.
[42] OECD (2025), OECD Economic Outlook, Volume 2025 Issue 2: Resilient Growth but with Increasing Fragilities, OECD Publishing, Paris, https://doi.org/10.1787/9f653ca1-en.
[123] OECD (2025), OECD Employment Outlook 2025: Can We Get Through the Demographic Crunch?, OECD Publishing, Paris, https://doi.org/10.1787/194a947b-en.
[21] OECD (2025), “OECD insights on productivity and business dynamics: France country note”, https://www.oecd.org/content/dam/oecd/en/publications/reports/2025/06/oecd-insights-on-productivity-and-business-dynamics-country-notes_2e4ed11c/france_33355104/b18cf1cf-en.pdf.
[90] OECD (2025), OECD Services Trade Restrictiveness Index: Policy Trends up to 2025, OECD Publishing, Paris, https://doi.org/10.1787/9953845b-en.
[86] OECD (2025), OECD Supply Chain Resilience Review: Navigating Risks, OECD Publishing, Paris, https://doi.org/10.1787/94e3a8ea-en.
[118] OECD (2025), Trends in Adult Learning: New Data from the 2023 Survey of Adult Skills, Getting Skills Right, OECD Publishing, Paris, https://doi.org/10.1787/ec0624a6-en.
[126] OECD (2024), “Building Competencies for Digital and Green Innovation in Higher Education”, OECD Education Spotlights, No. 12, OECD Publishing, Paris, https://doi.org/10.1787/d3869c1f-en.
[16] OECD (2024), “Digital Economic Outlook 2024: Embracing the Technology Frontier”, Vol. 1, https://doi.org/10.1787/a1689dc5-en.
[52] OECD (2024), How do governments direct support for innovation?: Lessons from recent OECD measurement and impact analysis (MABIS) work, OECD Publishing, Paris, https://doi.org/10.1787/c1d93d1c-en.
[43] OECD (2024), OECD Economic Surveys: France 2024, OECD Publishing, Paris, https://doi.org/10.1787/bd96e2ed-en.
[1] OECD (2023), “France: Quantifying Industrial Strategy”, OECD QUIS Project, https://www.oecd.org/content/dam/oecd/en/topics/policy-sub-issues/quantifying-industrial-strategies/QuIS-France.pdf.
[58] OECD (2023), OECD Reviews of Innovation Policy: Korea 2023, OECD Reviews of Innovation Policy, OECD Publishing, Paris, https://doi.org/10.1787/bdcf9685-en.
[125] OECD (2022), “Engaging employers in quality apprenticeships – making it happen locally”, https://www.oecd.org/en/publications/engaging-employers-in-quality-apprenticeships-making-it-happen-locally_6725cbbc-en.html.
[84] OECD (2020), “Start-ups, Killer Acquisitions and Merger Control,”, http://www.oecd.org/daf/competition/start-ups-killer-acquisitions-and-merger-control-2020.pdf.
[129] OECD (2019), “Le renforcement de la mixité sociale à l’école a-t-il un impact sur l’équité des résultats d’apprentissage?”, https://www.oecd.org/content/dam/oecd/fr/publications/reports/2019/06/does-greater-social-diversity-in-schools-have-an-impact-on-equity-in-learning-outcomes_fa8e55b8/1ce0ae6a-fr.pdf.
[127] OECD (2018), “A Brave New World: Technology & Education”, Spotlight 15, https://www.oecd.org/content/dam/oecd/en/publications/reports/2018/06/a-brave-new-world_91aeb9d2/9b181d3c-en.pdf.
[104] Pommier, S. and X. Lazarus (2024), “Recommandations pour la reconnaissance du doctorat dans les entreprises et la société”, Rapport Ministere de l’Enseignement Supérieur, de la Recherche et de l’Espace, https://www.enseignementsup-recherche.gouv.fr/fr/vers-une-meilleure-valorisation-du-doctorat-dans-le-monde-de-l-entreprise-et-de-l-industrie-le-97855.
[11] Rexecode (2025), “Etude Rexecode-SKEMA de la compétitivité des exportations françaises auprès des importateurs étrangers”, Document de Travail Rexecode 96, https://www.rexecode.fr/media/documents/document-de-travail/2025/etude-competitivite-skema-rexecode-document-de-travail-n.96-juin-2025.
[37] Rexecode (2025), “La compétitivité française en 2024”, Document de travail 94, https://www.rexecode.fr/media/documents/document-de-travail/2025/la-competitivite-francaise-en-2024-document-de-travail-fevrier-2025.
[105] Riedinger, N. and M. Zaiem (2013), “Y a-t-il un problème d’insertion des titulaires de doctorat dans les centres de R&D des entreprises”, Économie & prévision, Vol. n° 197-198/1, pp. 177-184, https://doi.org/10.3917/ecop.197.0177.
[116] Roy, U. and I. Chakraborty (2023), “Market concentration, agency cost and firm performance: a case study on Indian corporate firms”, Economic Change and Restructuring, Vol. 56/4, pp. 2645-2693, https://doi.org/10.1007/s10644-023-09529-1.
[68] Sénat (2024), “Financer l’entreprise de demain”, Rapport d’information 70, https://www.senat.fr/rap/r24-070/r24-0701.pdf.
[54] Siedschlag, I. et al. (2013), “What determines the location choice of R&D activities by multinational firms?”, Research Policy, Vol. 42/8, pp. 1420-1430, https://doi.org/10.1016/j.respol.2013.06.003.
[4] Tirole, J. (2015), “Market Failures and Public Policy”, American Economic Review, Vol. 105/6, pp. 1665-1682, https://doi.org/10.1257/aer.15000024.
[39] Tournier, L. (2025), L’électrification de l’industrie : un enjeu de décarbonation et de compétitivité, DGE Thema, https://www.entreprises.gouv.fr/files/files/Publications/2025/Th%C3%A9mas/thema-36-electrification.pdf.
[70] Tracxn (2025), “Angel network in Germany”, https://tracxn.com/d/investor-lists/angel-networks-in-germany/__j3f3F_tOJu_ILQI-CzVdnZyQOFYIihXVKKBDn30JdIs.
[74] Tsakanikas, A. et al. (2022), “Intangibles, innovation, and sector specialization in global value chains: A case study on the EU’s and the UK’s manufacturing industries”, Technological Forecasting and Social Change, Vol. 177, p. 121488, https://doi.org/10.1016/j.techfore.2022.121488.
[36] Urvoy, C. (2019), “Impôts sur la production quel impact sur la compétitivité ? Analyse de trois taxes sur données d’entreprises”, Conseil d’analyse économique Focus 35, https://cae-eco.fr/static/pdf/cae-focus035v2.pdf.
[114] Viganola, D. and M. Yaya Diallo (2025), “How Management Practices Differ in the EU-27”, Management Practices Enterprise Note 42, https://espanol.enterprisesurveys.org/content/dam/enterprisesurveys/documents/research-1/IDU1b17303ed12bae141a219c5215d5795945e21.pdf.
[18] WIPO (2025), World Intangible Investment Highlights 2025, https://www.wipo.int/web-publications/world-intangible-investment-highlights-2025/en/world-intangible-investment-highlights-2025.html.