This chapter discusses the importance of broadening the benefits from and participation in innovation across different social groups, regions and industries. It discusses how the particular challenges facing science, technology and innovation (STI) policymakers in 2025 – accelerating frontier technology development, building resilience and improving sustainability – interact with the channels through which the benefits of innovation are distributed. It concludes by identifying key implications for STI policymakers.
OECD Science, Technology and Innovation Outlook 2025
Driving Change in a Shifting Landscape
Abstract
Long-standing challenges, changing contexts. Many of the participation and inclusion challenges facing science, technology and innovation (STI) policy in 2025 – such as underrepresentation, concentration of opportunity and the importance of diffusion – echo patterns observed throughout the history of technological change. While the participation imperative is not new, the evolving context of rapid technological shifts, geopolitical competition and urgent societal challenges makes it even more vital for policymakers to address barriers that have persisted across past transformations.
Innovation is inherently concentrated, but broader benefits require deliberate diffusion efforts. Innovation activities naturally cluster among leading firms, sectors and regions due to economies of scale and knowledge spillovers. However, translating these concentrated innovations into economy-wide productivity gains and societal benefits requires dedicated policies and investments in diffusion mechanisms.
Broadening participation is a key lever for expanding innovation benefits. Beyond the diffusion of existing technologies, widening who participates in creating and shaping innovation directions can enhance both the quality and societal relevance of technological development while ensuring benefits reach more diverse populations.
Regional and industrial concentration shapes inclusion opportunities. Innovation activities are highly concentrated among leading firms, sectors and regions, creating uneven opportunities for participation and benefit. Workers in highly innovative environments have greater access to high-value jobs and career advancement while those in lagging areas face limited prospects.
STI policies may face trade-offs between excellence and inclusion. Policies directly promoting participation and those supporting centres of excellence for reasons of expediency play important roles for STI policies. What is key for the STI policy mix is supporting frontier advancement while facilitating broad-based access to benefits and participation.
Co-ordination across policy domains is essential. Persistent participation gaps despite targeted STI interventions demonstrate that achieving inclusion requires co-ordinated action across education, labour market, social and regional development policies. STI policies alone cannot address deep-rooted structural barriers and need to be co-ordinated with other policy domains.
Technological innovation and diffusion significantly improve people’s lives across and beyond the OECD. Whether through improvements in labour productivity, the quality of and access to medical and educational services, or in ensuring stable and affordable access to energy, technological innovation and diffusion can be significant forces for positive economic, societal and environmental progress. As policymakers increasingly direct STI investments toward strategic objectives – competitiveness, resilience and sustainability – this chapter examines why and how ensuring their broadly distributed benefits has become essential for policy success.
Broadening the participation in and widening the benefits from innovation have been recurring themes in international STI policy discussions, including in the 2018 and 2021 editions of the OECD Science, Technology and Innovation Outlook (OECD, 2018[1]; 2021[2]; Paunov and Planes-Satorra, 2021[3]). These discussions underscore the point that wider participation in technological innovation and diffusion not only enhances economic outcomes but can also improve the quality and societal relevance of innovation itself. While these insights remain valid for 2025 and beyond, policymakers face new challenges arising from emerging trends in STI.
The first challenge is the post-COVID increased focus in directed STI policies addressing strategic imperatives such as technological resilience and competitiveness (OECD, 2024[4]; Paunov and McGuire, 2022[5]). While this strategic focus is essential for national competitiveness, the risk is that the urgency and scale of these investments may inadvertently reinforce existing patterns of concentration. When finite public resources are rapidly deployed to achieve technological breakthroughs, they often flow to established centres of excellence – leading research institutions, flagship companies and innovation hubs – that can deliver results quickly, potentially widening gaps with other regions and actors unless accompanied by parallel investments in diffusion and capacity building.
The second consideration is that the vast majority of employment across the OECD is in low or medium innovation-intensive sectors, including those such as healthcare and education, where technological diffusion could significantly improve productivity and social outcomes. It is therefore important to ensure that efforts to push the technological frontier do not impede efforts to accelerate the diffusion of technological innovation where uplifts from productivity could be both economically and socially significant.
The third consideration is that innovation-intensive manufacturing industries face the dual challenge of adopting digital and low-carbon technologies simultaneously, creating a temporal challenge for firms and workers in those industries to contribute to and benefit from technological innovation.
The fourth consideration is that market concentration dynamics in some innovation-intensive sectors may affect the distribution of innovation opportunities and benefits. Understanding these dynamics and their implications for STI policy design requires policy attention and potential co-ordination with other policy domains (OECD, 2024[6]; 2023[7]).
These dynamics suggest several areas where STI policymakers can take action:
Strengthening diffusion policies to broaden participation in and benefits from technological innovation, revisiting best practices from technological, regional and sectoral diffusion initiatives for the contemporary context.
Embedding strategic reflections in the policy design process on how the strategic agenda‑setting for STI may impact participation in and benefits from publicly funded STI, and how policymakers in STI and other policy areas could address the economic and social implications of these impacts.
Better understanding of the relationship between competition and technological innovation and diffusion, both within and across economies.
This chapter begins with a brief overview of the differentiated and at times lag effects of technological innovation and its diffusion through history. It then proceeds with a discussion of a simplified framework to clarify how participation in innovation affects the direction of technological progress and the distribution of its outcomes. The chapter then discusses how participation in technological innovation and its diffusion interact with the imperatives facing STI policymakers today, such as competitiveness and resilience. It then outlines some of the key channels through which the framework set out could inform STI policy in 2025 and the years ahead. The chapter concludes with some overarching considerations for policymakers as they consider the importance of broadening the participation in and benefits from technological innovation as they design and implement STI policies in the years ahead.
The transformative changes currently underway, including the digital and green transitions, share critical characteristics with transformative technological changes throughout history, particularly in how they create concentrated benefits during development phases and uneven diffusion patterns that can limit the ability of all sectors of the population to benefit. Historical analysis of major technological transitions, notably the First Industrial Revolution (1760-1840), reveals persistent patterns where the benefits of breakthrough innovations initially concentrate among those with access to capital and technological infrastructure while broader societal gains emerge only through deliberate diffusion mechanisms and often significant social and political intervention (Acemoglu and Johnson, 2024[8]; 2023[9]; Hobsbawm, 1962[10]).
The mechanisation of British textile production, for instance, demonstrates how technological advancement can simultaneously drive productivity growth while displacing entire categories of skilled workers – handloom weavers saw their real wages more than halve between 1806 and 1820 as power looms replaced traditional craftsmanship, illustrating the stark short-term distributional consequences of technological change (Feinstein, 1998[11]; Voth, 2003[12]).
What distinguishes contemporary digital and green transitions from earlier technological revolutions is not just their speed and scope, but the recognition that broad-based and widely shared benefits from innovation are neither automatic nor guaranteed. Unlike previous eras where distributional consequences were often treated as inevitable byproducts of progress, today’s transitions occur within policy frameworks that explicitly acknowledge the need for participatory approaches to technology development and diffusion.
The evidence from historical transitions suggests that without deliberate intervention, technological progress can create persistent inequalities, even after decades of economic growth. Improvements in living standards for displaced workers during the Industrial Revolution were minimal and required sustained social and political action to materialise (see Feinstein (1998[11]); Allen (2007[13]); or Acemoglu and Johnson (2024[8])). This historical perspective underscores why current policy debates around frontier technologies like artificial intelligence (AI), quantum computing and clean energy systems must simultaneously address the concentration of development capabilities and the mechanisms for ensuring broad-based access to the benefits of technological progress.
The framework that follows in this chapter builds on these historical insights to distinguish between development dynamics, which determine who participates in innovation processes and where innovation capabilities concentrate, and diffusion dynamics, which shape how innovations spread across different populations, regions and economic sectors over time. Understanding this distinction is crucial for contemporary STI policy, as the evidence suggests that addressing inclusion challenges requires different approaches at the development stage (where the focus is on broadening participation in frontier research and innovation) versus the diffusion stage (where the emphasis shifts to ensuring widespread adoption and benefit-sharing across diverse communities and regions).
How does technological innovation impact society, and how does broader participation in innovation affect the direction of technological innovation?
There are two key dimensions for understanding how technological innovation interacts with socio‑economic outcomes. The first is to consider how the development and diffusion of new technologies affects the outcomes of different groups, and the skills and capacities they may require to benefit from these technologies. The second is to consider how participation in the development of technological innovation affects the direction and quality of that innovation. These two dimensions interact in ways that affect the inclusiveness of the gains from innovation (e.g. better health outcomes, better education access, higher productivity) and the alignment of innovation with social needs.
These outcomes are how different groups in society benefit from new technologies and innovations in the following ways (Figure 3.1):
Direct impacts [quadrant A]: Different socio-economic groups may benefit differently from new technologies and innovations. Several conditions determine access to the benefits of innovation, including price, geographic availability, required infrastructure, user capabilities (such as digital or technical skills) as well as the very purpose of the innovation, which may cater to the needs of specific groups of the population. Social innovations are a specific sub-set of innovations aimed at addressing unmet social needs – particularly those affecting marginalised or underserved groups.
Indirect impacts [quadrant B]: Wider economic and social impacts accrue indirectly by affecting returns to labour and capital impacted by innovation and technology. Changes in production processes driven by innovation, such as automation and the wider application of AI in production, can alter demands for skills or capacities and affect demands for different types of assets (capital, land, etc.). Moreover, technological progress can change the relative returns to labour and capital (see Autor et al. (2017[14]) and Guellec and Paunov (2017[15]) for a discussion).
Participation refers to the opportunities that different groups in society have to shape technological progress in the following ways (Figure 3.1):
Direct participation [quadrant C]: Individuals can engage directly in the design and development of technology and innovation by being part of the research and innovation workforce. Such engagement requires having specialised skills and capacities, which vary depending on the types of activities undertaken. Beyond professional research and innovation roles, citizens more generally can engage in a myriad of other ways, ranging from weak engagements, such as providing user feedback that influences product development, to more substantial involvement, like contributing to open-source and citizen science projects (OECD, 2025[16]).
Indirect participation [quadrant D]: Individuals may also engage indirectly in shaping innovation and technology development by participating in industry or policy decision-making processes (e.g. taking key investment decisions, shaping institutional choices over adoption); engaging in public consultations, participatory technology assessment or participatory research agenda-setting exercises influencing the directions of public or private choices on innovation and technology; and, more passively, shaping demand.
The relationship between innovation and inclusion can, therefore, be explored through four dimensions, as illustrated in Figure 3.1.
Building on the historical context and framework from the previous section, this section examines how participation dynamics interact with today’s STI policy imperatives. Key priorities policymakers face in 2025 include accelerating the development of strategic technologies, navigating green and digital transitions while ensuring competitiveness, enabling broad-based diffusion to meet productivity needs, building consensus around STI directions, and operating with agility amid fiscal constraints. Where possible, evidence from gender, regional and industrial participation is provided to illustrate these dynamics.
Navigating excellence and inclusion in an era of strategic competition
This subsection addresses how the framework’s distinction between outcomes and participation – across both direct and indirect channels – plays out at the frontier of STI policy. Frontier-oriented investments and “directional” STI policies are intended to deliver significant economic and technological gains, but they risk concentrating both the direct benefits (A: who reaps the returns from new STI products) and the indirect structural effects (B: who benefits from changing skill and capital demands) among established actors. Frontier-oriented STI investments thus face a resource allocation challenge: directing finite budgets toward established centres of excellence – which may be entirely rational for achieving rapid breakthroughs – can inadvertently concentrate both innovation returns and structural benefits among already capable actors. This creates tension between the immediate imperative to develop strategic technologies and the broader goal of ensuring widespread access to innovation benefits, particularly when existing diffusion mechanisms are already strained.
The rapid pace of digital and green transitions, combined with intensifying geopolitical competition around frontier technologies, has fundamentally altered the policy landscape for STI systems. Major economies have launched significant directed investments – such as the European Union’s European Chips Act – that aim to build capabilities in AI, quantum computing, semiconductors and clean energy technologies. These policies reflect a new era where STI policy is increasingly “directional”, pursuing specific mission-oriented objectives while navigating complex trade-offs between technological leadership and inclusive participation (Arnold et al., 2023[17]; Larrue, 2021[18]; Mazzucato, 2018[19]; OECD, 2024[4]).
A variety of metrics show that technological innovation is already highly concentrated at the firm, sectoral and regional levels. For evidence at the firm level, data on top research and development (R&D) investors show that the top 100 companies (in terms of R&D investments) account for around a striking 50% of global R&D in 2023 (Figure 3.2) There is also significant concentration within that group of top R&D investors: on average, the top 10 companies invested more than double the amount than the top 50 (including the top 10), and the top 50 invested on average over 50% more than the top 100. Top R&D investors also account for an important part of national R&D.
Source: European Commission: Joint Research Centre, NINDL, E., NAPOLITANO, L., CONFRARIA, H., RENTOCCHINI, F., FAKO, P., GAVIGAN, J. and TUEBKE, A., The 2024 EU Industrial R-D Investment Scoreboard, Publications Office of the European Union, Luxembourg, 2024, https://data.europa.eu/doi/10.2760/9892018, JRC140129.
Innovation is also geographically concentrated in a limited set of “innovation leader” regions across OECD countries, which collectively generate the bulk of national R&D and scientific output. While such a concentration can encourage economies of scale and knowledge spillovers, persistent innovation leadership and a decline in entry and job reallocation rates signal rising barriers to new entrants. However, this concentration pattern may be evolving in frontier technology areas: in fields like AI and quantum computing, smaller firms such as OpenAI, Anthropic and specialised quantum start-ups are driving significant breakthroughs alongside established tech giants, suggesting that technological disruption can still create opportunities for new entrants to challenge incumbent advantages.
In the context of public support for frontier technology development, the question for policymakers is to what extent the pursuit of research excellence and support for pre-existing clusters with strong capacities reinforces these concentration dynamics, and what are the implications of that concentration? When governments allocate finite resources to leading firms, research institutions and regions that already possess established capabilities, they may achieve technological breakthroughs more efficiently but risk widening gaps with lagging actors (quadrants B and C in Figure 3.1). For instance, while progress in AI and automation may benefit those with complementary skills and capital assets, it can simultaneously reduce returns to routine labour and concentrate benefits in technologically advanced regions. The challenge for policymakers is determining when such concentration is a necessary short-term cost for long‑term competitiveness and when it creates structural barriers that ultimately undermine innovation system resilience.
Concentrating resources in frontier technology development can be a highly effective strategy, especially when leading firms, industries and research institutions – often clustered in specific regions – already possess strengths that policies can leverage. In a globally competitive environment, building on these existing capabilities may offer the best chance for success. Spreading resources too broadly can dilute their impact, resulting in well-intentioned but ultimately ineffective outcomes. Furthermore, technological diffusion may occur over time through market forces, collaboration and talent mobility. They also depend on whether the effects of concentration are temporary or long-lasting, i.e. whether frontier developments will eventually lead to improved outcomes that spread more broadly across society over time.
The downside, however, may be a widening gap with those already lagging behind, challenging also the wider diffusion of frontier technologies, which in turn can reduce possibilities for their further development as demand and user experiences play important roles in shaping innovation pathways. The allocation of limited public resources to firms and institutions that already have well-established capabilities may further reduce possibilities for others to participate in frontier technology development and benefit from its diffusion.
Recent OECD work has widely discussed tension between concentration (which may promote excellence and rapid technological development) and inclusion. The OECD’s Industrial Policy Framework demonstrated that effective industrial strategies must balance productivity growth with addressing societal challenges, including inclusion, through co-ordinated approaches that address complementarities between different policy instruments (Criscuolo, Gonne and Lalanne, 2022[20]). Recent OECD analysis shows that industrial policies increasingly target societal goals, with green transition objectives comprising 18.6% of national STI strategies, followed by social and regional inclusion at 9.9% (Paunov and Einhoff, 2025[21]).
The potential unintended negative impact of these STI policies (as well as the above-mentioned frontier technology policies) on broader participation does not imply that they should not be pursued, as they address other important policy priorities – such as economic competitiveness, resilience and national security. Strategies that support frontier technology development are an example of the many rationales justifying STI policy. From a Schumpeterian perspective, technological change inherently involves disruption and renewal, and OECD countries have well-established social policy frameworks that can help mitigate the social and regional impacts of such transitions. The key point is to recognise that technology policy choices are not neutral, and that they have an impact on the distribution of wealth, income and opportunities across firms, industries and places (and ultimately individuals) well into the future. The other key insight is that there is a role for STI policy to complement other efforts in shaping the nature of transitions.
From development to diffusion and adoption: The need for policy differentiation
The relationship between knowledge creation (quadrants C and D in Figure 3.1, who participates in and shapes STI development and governance) to widespread technology adoption and benefit-sharing (quadrants A and B in Figure 3.1, who directly and indirectly gains from STI products and changing economic returns) reveals distinct yet interdependent policy challenges. Effective STI systems require not only excellence at the frontier, but also robust, inclusive mechanisms for diffusion, ensuring diverse actors have the capacity and opportunity to participate in innovation and be able to enjoy its benefits.
Frontier-oriented STI policies now face a double mandate: continue pushing the technological boundary and consider how diffusion and adoption policies can be integrated into these policies. Recent OECD evidence on AI adoption shows why this matters: uptake is still twice as high in large firms and advanced regions as in smaller firms and peripheral areas, owing to scale-dependent fixed costs, data requirements and superior absorptive capacity (OECD, 2024[22]).
Effective adoption of innovation is essential for ensuring that the benefits of new technologies reach a broad swathe of society and the economy but is hindered by highly uneven spatial and sectoral innovation activity. Workers in “leader” firms, sectors and regions consistently achieve higher wages and revenues, while those in “laggard” environments capture fewer gains. For example, patenting is overwhelmingly concentrated in large urban areas, with 90% of patent applications coming from urban inventors and large urban regions having significantly higher patenting rates than medium-sized or smaller areas (OECD, 2024[23]).
This reflects both the structural advantages of co-location and the cascading disadvantages faced by less connected or less innovative regions. Moreover, increasing entrenchment among market leaders, as seen in the declining rates of firm entry and job mobility, limits the spread of new technologies to lagging places and firms. Addressing these challenges requires targeted policy interventions – such as improving infrastructure and connectivity, advancing skills development in underperforming areas, and supporting technology adoption in small and medium-sized enterprises (SMEs) – to facilitate broader and more equitable diffusion and participation throughout the innovation system.
The push to align economic competitiveness with sustainability presents its own particular diffusion challenges. The evolution of labour and skill demands in emerging sectors, such as the move to electric vehicle production, requires new competencies not always possessed by workers in sunset industries (Curtis, O’Kane and Park, 2023[24]). If skills development and retraining lag behind technological change, this transition may end up benefiting only a narrow pool of workers in specific regions or firms. STI policy must, therefore, link innovation support to strategic workforce development and ensure that knowledge, infrastructure and market opportunities reach participants beyond traditional innovation leaders, including those in less advantaged sectors and places.
Policy design has to distinguish between two separate bottlenecks. Knowledge diffusion determines who can join the inventive process while technology adoption decides who can turn new ideas into social and economic value. Because the obstacles differ, skills and research networks on the one hand, data readiness, finance and managerial know-how on the other, each stage needs its own toolkit. Capacity building must simultaneously broaden participation in technology creation and equip diverse regions, sectors and communities to adopt innovations at scale.
Diffusion policies, whether embedded in technology frontier programmes or deployed in parallel, are the main lever for widening the benefits of and participation in STI; Table 3.1 shows examples of relevant diffusion policies. They only work when actors possess sufficient absorptive capacity, i.e. the ability to spot, absorb and apply external knowledge. Investing in those capacities is thus essential to share the gains from innovation more equitably, including across borders. For developing economies, building such capabilities is a prerequisite for meaningful engagement in global STI systems.
Translating this into practice poses a governance challenge: responsibilities for funding and decision making must be split judiciously between national and subnational authorities so that national strategic aims align with regional strengths and opportunities. Regional policymakers, in particular, need diagnostic tools to see where their ecosystems can credibly participate in strategic, frontier-technology domains.
Survey data from the G7 and Brazil (OECD, 2025[25]) reveal that diffusion frictions – acute skill shortages, low data maturity and uncertainty over returns – now trump simple awareness gaps. Firms rate three policy responses the highest:
1. modernised qualification frameworks plus hands-on, sector-specific training
2. higher quality, easily accessible public data
3. streamlined collaboration with universities and dedicated diffusion agencies.
Although dedicated technology diffusion agencies are well regarded, they currently serve only a minority of firms, underscoring the need for scalable sign-posting services, SME-oriented vendor-selection guidelines and clear accountability frameworks for safe AI use. Comparable international surveys and rigorous evaluation of these agencies are critical to identify and share what works.
Sources: European Commission (2025[26]), APC (2025[27]), NGen (2025[28]), Federal Ministry for Economic Affairs and Energy (2025[29]), Japan Science and Technology Agency (2025[30]) , EC-OECD (2025[31]).
Broadening participation in and the benefits from research and innovation
If policymakers want the benefits of the development and diffusion of technologies to be truly broad-based and shared, STI policies need to dramatically accelerate efforts to address participation imbalances. While this subsection focuses on gender as a salient lens, the structural and systemic challenges described here apply – often with added complexity – to other groups that face barriers, such as people from lower income backgrounds, minority groups and regions that have undergone significant deindustrialisation.
Over the past decade, women’s participation in STI has notably increased, although gaps persist. Globally, the share of 25-34-year-olds with tertiary education rose from 23% to 27.5% between 2013 and 2021, and in OECD countries from 45.6% to 53.7% (OECD, 2024[32]), outpacing men in both cases. Yet this progress masks persistent disparities in key fields. In 2021, just one-third (32.5%) of graduates in science, technology, engineering and mathematics (STEM) were women, up only marginally from 31% in 2013. Representation varies sharply by discipline: women make up a slight majority in natural sciences, mathematics and statistics (53.6% of graduates), but only 27.8% in engineering and 22.7% in information technology.
These educational gaps carry through into research and innovation careers. Despite a modest increase – from 34.7% to 35.6% between 2013 and 2021 – on average women still account for just over a third of R&D personnel in OECD countries. National shares vary considerably, with some countries approaching gender parity (Iceland, Latvia and Lithuania) while others continue to record comparatively low shares despite progress in recent years (Czechia, Korea and Japan at ~22%). In patenting, the share of women inventors fell from 13.4% in 2013 to 11.3% in 2019 and remains below 7% in some innovation‐intensive economies such as Austria, Germany and New Zealand.
Improving women’s participation in STI has been a policy focus for a long time, with policymakers implementing targeted financing schemes, such as scholarships and research grants, for women to engage in STI training and activities. Table 3.2 provides examples of policy initiatives in this area.
Sources: STEM Learning (2025[33]); Korea Foundation for Women in Science, Engineering and Technology (2025[34]); MINTvernetzt (2025[35]); Innovative Initiatives Digital Luxembourg (2025[36]); The Research Council of Norway (2019[37]).
Broadening participation in the governance and steering of science, technology and innovation
This subsection foregrounds quadrant D of the framework illustrated in Figure 3.1 – diversity in decision making and leadership in STI – by highlighting that participatory governance is increasingly critical as technological transformations accelerate. As technological change profoundly reshapes society, the capacity for a range of groups – not just technical experts – to influence the direction, priorities and norms of innovation becomes central to the legitimacy, equity and societal alignment of STI policy. Ensuring effective and democratic participation in STI governance elevates indirect forms of participation to a policy priority and brings the interplay between who shapes innovation and who ultimately benefits from it into sharper focus across all four quadrants.
The scale and societal implications of transformative technological developments require participatory approaches to STI policy design (quadrant D of Figure 3.1). As in previous technological revolutions, today’s frontier technologies will fundamentally reshape work, governance and social relations in ways that technical experts alone cannot fully anticipate or evaluate; the direction of technology development and diffusion is also much more actively shaped by policy than in the past.
The 2024 OECD Framework for Anticipatory Governance of Emerging Technologies emphasises early societal engagement to surface concerns, inform design choices and guide innovation toward more equitable outcomes (OECD, 2024[38]). This requires investing in the capacity building, consultation processes and institutional arrangements that enable diverse groups to contribute meaningfully to shaping technological directions.
When thinking about participation in the governance and agenda-setting of STI going forward, it is important to note that policymakers do so in a context where historical participation gaps remain unresolved. Women, for example, remain under-represented across entrepreneurial and leadership roles in STI. An analysis of a dataset of start-ups listed on Crunchbase covering firms founded between 2000 and 2017 in OECD and BRICS countries found that women-only founding teams accounted for less than 6% of all start-ups, while those with at least one female co-founder made up approximately 15% (Lassébie et al., 2019[39]). Evidence on European venture capital (VC)-funded start-ups based on data from private sources (Pitchbook and the European Data Cooperative) covering 39 000 investors and 85 000 entrepreneurs that were active in Europe between 2011 and 2021 (accounting for 80% of total VC firms and 52% of start-ups) indicates that, between 2011 and 2021, women comprised only 10% of founders and chief executive officers (CEOs) in those start-ups, while start-ups led solely by women captured a mere 2% of total VC funding. Among investors, only about one in seven senior-level VC investors was female, 90% of whom worked in predominantly male teams (European Investment Fund, 2023[40]).
These participation imbalances extend into senior roles across sectors shaping the future of STI. In 2020, fewer than 5% of Silicon Valley 150 companies had female CEOs, and women held only 26-34% of senior posts at Google, Apple, Facebook, Amazon and Microsoft (European Centre for Women and Technology, 2024[41]). In pharmaceuticals, just 17% of board seats are held by women (ISPE, 2021[42]), and in deep‑tech start-ups under one-quarter of founding teams include a woman, with overall female founder share at 14% (Davila et al., 2024[43]). Academia and public administration show similar patterns – women lead 23.6% of higher education institutions and occupy 31.1% of board roles in the European Union (European Commission, 2021[44]).
Note: Organisations covered are central administrations, also referred to as ministries and/or departments of a national government led by a minister. Senior administrators are the sum of level 1 and level 2 administrators. Level 1 administrators include all administrative (non-political) positions from the head of the ministry down to the level of head of directorate or similar, where a directorate is a major section within the ministry. Level 2 administrators include all positions below the head of directorate down to the level of head of division/department, where a division/department is the first level of organisation below the directorate (i.e. the second level of functional organisation). Data refer to the OECD-EU countries plus Iceland, Norway the United Kingdom and Türkiye.
Source: OECD (2023[45]) based on European Institute for Gender Equality (EIGE) Gender Statistics (database); women and men in decision-making (WMID) authorities.
Similarly, women are also under-represented in public decision-making bodies. In August 2023, they accounted for 33% of members of parliament in national parliaments across the European Union. Only six national parliaments had more than 40% women members, while seven had less than 25%. In the case of senior management positions in national administrations, gender disparities tend to be lower and decreasing over time, although significant differences remain across countries (Figure 3.3) (EIGE, 2024[46]). When looking at research-funding organisations in the European Union, the gender composition of presidents and members of the highest decision-making body tends to be predominantly male, with some notable exceptions (Figure 3.4).
Percentage of total
Source: EIGE, https://eige.europa.eu/gender-statistics/dgs/indicator/wmidm_educ__wmid_resfund, accessed on 18 August 2025.
Industrial and regional innovation concentration: The competition policy dimension
This subsection discusses how the concentration of innovation activities – across firms, sectors and regions – shapes both the distribution of outcomes (quadrants A and B) and the opportunities for participation (quadrants C and D) within STI systems. A high and rising concentration in R&D investment and innovation capacity can lead to substantial gains in technological progress but also risks reinforcing disparities in who benefits (quadrants A and B) and who is able to participate meaningfully in the innovation process and decision making (quadrants C and D). The framework clarifies that such market dynamics are not neutral: they can reinforce exclusionary patterns unless actively addressed. Recognising these risks, recent OECD work highlights the critical, complementary role of competition policy in maintaining contestable markets, safeguarding access for new entrants, and ensuring that both the outcomes and opportunities created by innovation are widely shared, not just captured by a handful of leading actors or regions.
While concentration can drive technological advancement through economies of scale and scope in R&D – particularly critical for frontier technologies like AI, quantum computing and advanced semiconductors that require massive capital investments – it also creates significant barriers to participation for smaller firms, emerging regions and new entrants.
The OECD’s extensive work on competition policy in digital markets demonstrates that these concentration effects are not merely incidental but reflect underlying market power dynamics that can become self‑reinforcing, as leading platforms leverage their positions to acquire talent, emerging competitors and complementary technologies (OECD, 2025[47]; 2024[6]). Addressing these inclusion challenges effectively requires co-ordinated intervention from competition policymakers, who possess the analytical tools and enforcement mechanisms to assess market contestability, prevent anti-competitive consolidation and ensure that innovation ecosystems remain accessible to diverse participants. The OECD’s Competition and Innovation Framework emphasises that competition policy has a key role in facilitating innovation diffusion and allowing innovations to spread across markets, complementing STI policies’ efforts to broaden participation while maintaining the competitive dynamics essential for continued technological progress (OECD, 2023[7]).
International competition also plays a decisive role in shaping the opportunities – and limitations – in low‑carbon and green technologies. For example, while the People’s Republic of China’s strategic dominance in green technology manufacturing has profoundly altered the global competitive landscape, it has also led to a rapid deployment of what are critical technologies for economic decarbonisation (ITIF, 2020[48]). The consolidation of production in China degraded European manufacturing capacity but may also have curtailed potential technological pathways where European firms might have excelled, such as advanced thin-film technologies, perovskite-silicon tandems and other alternative photovoltaic approaches. The implication is, therefore, that the international dimension of technological competition can have a profound impact on the ability of local innovation and industrial ecosystems to participate in the opportunities that can emerge in transitions.
Co-ordinated policy action
This subsection draws together the full four-quadrant framework by emphasising that realising both broad participation in, and broad benefits from, innovation requires policy coherence beyond the STI domain itself. While STI policy can directly influence who participates in developing and using new technologies (quadrants C and D in Figure 3.1) as well as who reaps their benefits (quadrants A and B in Figure 3.1), persistent barriers to inclusion are often rooted outside STI’s traditional mandate.
The evidence demonstrates that STI policies alone cannot, therefore, address all dimensions relating to the participation in STI and the benefits from its outcomes. STI governance must, therefore, co-ordinate with social, education and economic policies to tackle structural barriers that limit participation and benefit; this, of course, is not new, but the speed and implications of frontier technological innovation raises the importance of action.
While technological innovation can bring substantial benefits for OECD Member countries and their citizens, the pursuit of technological leadership and the uneven diffusion of technological innovation can reinforce or deepen existing divides in who participates in and benefits from the STI system. The three ‑quadrant framework that was introduced – linking direct and indirect participation with direct and indirect outcomes – provides one approach for policymakers to assess and address the multifaceted impacts of technological progress. Four key takeaways emerge for STI policymakers in 2025.
First, STI policies operate across multiple dimensions of participation and inclusion simultaneously. The illustrative framework shows that policies designed to advance one quadrant – such as excellence-based initiatives targeting frontier development – can have unintended consequences across the others. For instance, while such policies may accelerate innovation, they can also concentrate benefits in already-advantaged regions and actors. Effective STI policy requires explicit consideration of cross-quadrant effects and complementary measures to address potential exclusionary impacts.
Second, addressing global challenges demands both excellence and inclusion. The scale and speed required for digital and green transitions cannot be achieved through concentration in innovation alone. Evidence from the gender participation case study illustrates that under-representation across the innovation pipeline – from education through leadership – represents a fundamental constraint on innovation capacity. Broader participation unlocks diverse perspectives, accelerates diffusion and generates the societal legitimacy essential for successful transitions.
Third, successful inclusion policies require co-ordination beyond STI. The persistent nature of participation gaps – despite decades of targeted interventions – underscores that STI policies alone are insufficient. Structural barriers require co-ordinated action across education, labour market, social and regional development policies. The framework’s emphasis on both direct and indirect pathways highlights the need for this multi-policy approach.
References
[8] Acemoglu, D. and S. Johnson (2024), “Learning from Ricardo and Thompson: Machinery and labor in the Early Industrial Revolution and in the age of artificial intelligence”, Annual Review of Economics, Vol. 16, pp. 597-621, https://doi.org/10.1146/annurev-economics-091823-025129.
[9] Acemoglu, D. and S. Johnson (2023), Power and Progress: Our Thousand-year Struggle Over Technology and Prosperity, PublicAffairs.
[13] Allen, R. (2007), “Pessimism preserved: Real wages in the British Industrial Revolution”, Oxford Economics Series Working Papers, No. 314, Oxford Univeristy, Department of Economics, https://ideas.repec.org/p/oxf/wpaper/314.html.
[27] APC (2025), “Advanced Propulsion Centre UK”, https://www.apcuk.co.uk/.
[17] Arnold, E. et al. (2023), “Navigating green and digital transitions: Five imperatives for effective STI policy”, OECD Science, Technology and Industry Policy Papers, No. 162, OECD Publishing, Paris, https://doi.org/10.1787/dffb0747-en.
[14] Autor, D. et al. (2017), “Concentrating on the fall of the labor share”, American Economic Review, Vol. 1087/5, pp. 180-185, https://doi.org/10.1257/aer.p20171102.
[20] Criscuolo, C., N. Gonne and G. Lalanne (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.
[24] Curtis, E., L. O’Kane and R. Park (2023), Workers and the Green-Energy Transition: Evidence from 300 Million Job Transitions, National Bureau of Economic Research, Cambridge, MA, https://doi.org/10.3386/w31539.
[43] Davila, A. et al. (2024), Women Founders in European Deep Tech Startups, European Investment Bank et al., https://supernovas.eitcommunity.eu/wp-content/uploads/2024/11/Executive_Summary_Supernovas_Digital_FV.pdf?utm_source=chatgpt.com.
[31] EC-OECD (2025), “STIP Compass”, https://stip.oecd.org/stip/.
[46] EIGE (2024), Gender Equality in the European and National Parliaments Ahead of the 2024 Election Year, European Institute for Gender Equality, https://eige.europa.eu/sites/default/files/documents/gender-equality-in-the-european-parliament-and-in-national-parliaments-in-the-european-union-ahead-of-the-2024-election-year.pdf.
[41] European Centre for Women and Technology (2024), “Women in tech statistics: Gender diversity in tech [2024]”, news, https://ecwt.eu/news/gender-diversity-in-tech-2024.
[26] European Commission (2025), “About S3”, https://ec.europa.eu/regional_policy/policy/communities-and-networks/s3-community-of-practice/about_en.
[44] European Commission (2021), She Figures 2021: Gender in Research and Innovation – Statistics and Indicators, Publications Office of the European Union, https://op.europa.eu/en/web/eu-law-and-publications/publication-detail/-/publication/67d5a207-4da1-11ec-91ac-01aa75ed71a1.
[40] European Investment Fund (2023), The VC Factor: Gender Lens Edition, European Investment Fund, https://www.eif.org/news_centre/publications/vc-factor-gender-lens-edition.pdf.
[29] Federal Ministry for Economic Affairs and Energy (2025), “Clusters4Future”, https://www.clusterplattform.de/CLUSTER/Navigation/EN/Cluster-Policy/National-Level/Clusters4Future/Clusters4Future.html.
[11] Feinstein, C. (1998), “Pessimism perpetuated: Real wages and the standard of living in Britain during and after the Industrial Revolution”, The Journal of Economic History, Vol. 58/3, pp. 625-658, https://doi.org/10.1017/s0022050700021100.
[15] Guellec, D. and C. Paunov (2017), “Digital innovation and the distribution of income”, NBER Working Paper Series, No. 23987, National Bureau of Economic Research, Cambridge, MA, https://doi.org/10.3386/w23987.
[10] Hobsbawm, E. (1962), The Age of Revolution: 1789-1848, Vintage.
[36] Innovative Initiatives Digital Luxembourg (2025), “Women in tech”, web page, https://innovative-initiatives.public.lu/initiatives/women-tech.
[42] ISPE (2021), Women in Pharma: Empowering Women as Industry Leaders, International Society for Pharmaceutical Engineering, https://ispe.org/sites/default/files/attachments/public/PE_JanFeb21_CompleteIssue_v5_LR.pdf.
[48] ITIF (2020), The Impact of China’s Production Surge on Innovation in the Global Solar Photovoltaics Industry, Information Technology & Innovation Foundation, Washington, DC, https://itif.org/publications/2020/10/05/impact-chinas-production-surge-innovation-global-solar-photovoltaics.
[30] Japan Science and Technology Agency (2025), “Science and Technology Research Partnership for Sustainable Development (SATREPS)”, https://www.jst.go.jp/global/english/index.html.
[34] Korea Foundation for Women in Science, Engineering and Technology (2025), WISET, https://wiset.or.kr/eng/sub01_01_01.do.
[18] Larrue, P. (2021), “The design and implementation of mission-oriented innovation policies: A new systemic policy approach to address societal challenges”, OECD Science, Technology and Industry Policy Papers, No. 100, OECD Publishing, Paris, https://doi.org/10.1787/3f6c76a4-en.
[39] Lassébie, J. et al. (2019), “Levelling the playing field: Dissecting the gender gap in the funding of start-ups”, OECD Science, Technology and Industry Policy Papers, No. 73, OECD Publishing, Paris, https://doi.org/10.1787/7ddddd07-en.
[19] Mazzucato, M. (2018), “Mission-oriented innovation policies: Challenges and opportunities”, Industrial and Corporate Change, Vol. 27/5, pp. 803-815, https://doi.org/10.1093/icc/dty034.
[35] MINT Vernetzt (2025), MINT Vernetzt, https://www.mint-vernetzt.de/.
[28] NGen (2025), “Next Generation Manufacturing”, https://www.ngen.ca/.
[47] OECD (2025), Competition and Digital Economy, OECD Publishing, Paris, https://www.oecd.org/en/topics/sub-issues/competition-and-digital-economy.html.
[16] OECD (2025), “Embedding citizen science into research policy”, OECD Science, Technology and Industry Policy Papers, No. 175, OECD Publishing, Paris, https://doi.org/10.1787/a1cfb1a8-en.
[25] OECD (2025), The Adoption of Artificial Intelligence in Firms: New Evidence for Policymaking, OECD Publishing, Paris, https://doi.org/10.1787/f9ef33c3-en.
[6] OECD (2024), Competition Policy in Digital Markets: The Combined Effect of Ex Ante and Ex Post Instruments in G7 Jurisdictions, OECD, Paris, https://www.oecd.org/content/dam/oecd/en/publications/reports/2024/10/competition-policy-in-digital-markets_554eb7d5/80552a33-en.pdf.
[22] OECD (2024), “Fostering an inclusive digital transformation as AI spreads among firms”, OECD Policy Briefs, OECD Publishing, Paris, https://www.oecd.org/en/publications/fostering-an-inclusive-digital-transformation-as-ai-spreads-among-firms_5876200c-en.html.
[38] OECD (2024), “Framework for Anticipatory Governance of Emerging Technologies”, OECD Science, Technology and Industry Policy Papers, No. 165, OECD Publishing, Paris, https://doi.org/10.1787/0248ead5-en.
[4] OECD (2024), “OECD Agenda for Transformative Science, Technology and Innovation Policies”, OECD Science, Technology and Industry Policy Papers, No. 164, OECD Publishing, Paris, https://doi.org/10.1787/ba2aaf7b-en.
[23] OECD (2024), OECD Regions and Cities at a Glance 2024, OECD Publishing, Paris, https://doi.org/10.1787/f42db3bf-en.
[32] OECD (2024), Population with tertiary education [data set], https://www.oecd.org/en/data/indicators/population-with-tertiary-education.html.
[7] OECD (2023), Competition and Innovation: A Theoretical Perspective: OECD Competition Policy Roundtable Background Note, OECD, Paris, https://www.oecd.org/content/dam/oecd/en/publications/reports/2023/05/competition-and-innovation-a-theoretical-perspective_6e330b82/4632227c-en.pdf.
[45] OECD (2023), Government at a Glance 2023, OECD Publishing, Paris, https://doi.org/10.1787/3d5c5d31-en.
[2] OECD (2021), OECD Science, Technology and Innovation Outlook 2021: Times of Crisis and Opportunity, OECD Publishing, Paris, https://doi.org/10.1787/75f79015-en.
[1] OECD (2018), OECD Science, Technology and Innovation Outlook 2018: Adapting to Technological and Societal Disruption, OECD Publishing, Paris, https://doi.org/10.1787/sti_in_outlook-2018-en.
[21] Paunov, C. and J. Einhoff (2025), “Innovation policy transformed? Unveiling a new paradigm through natural language processing”, OECD Science, Technology and Industry Working Papers, No. 2025/21, OECD Publishing, Paris, https://doi.org/10.1787/5ee60cb5-en.
[5] Paunov, C. and H. McGuire (2022), “Towards a new vision of innovation through COVID-19? A comparative reading of 11 countries’ strategies”, OECD Science, Technology and Industry Policy Papers, No. 136, OECD Publishing, Paris, https://doi.org/10.1787/15475840-en.
[3] Paunov, C. and S. Planes-Satorra (2021), “What future for science, technology and innovation after COVID-19?”, OECD Science, Technology and Industry Policy Papers, No. 107, OECD Publishing, Paris, https://doi.org/10.1787/de9eb127-en.
[33] STEM Learning (2025), “About STEM ambassadors”, web page, https://www.stem.org.uk/stem-ambassadors.
[37] The Research Council of Norway (2019), Policy for Gender Balance and Gender Perspectives in Research and Innovation, The Research Council of Norway, Lysaker, Norway, https://www.forskningsradet.no/contentassets/19527ed7d0b149d6b9b310f8bb354ce9/nfr_gender_policy_orig-1.pdf.
[12] Voth, H. (2003), “Living standards during the Industrial Revolution: An economist’s guide”, American Economic Review, Vol. 93/2, pp. 221-226, https://doi.org/10.1257/000282803321947083.