This chapter reviews the policy instruments through which governments across OECD Member countries and partner economies support data spaces (including functionally comparable architectures). It examines regulatory frameworks that require or encourage multi-party data sharing, the design of public investment on both the supply and the demand side of the bootstrapping problem, standardisation and cross-border interoperability, and certification and public trust frameworks. It concludes with the role of government as data space operator and anchor participant, including inter-governmental data spaces.
Data Spaces between Co‑ordination Challenges, Financial Viability and Public Policy
Abstract
Governments across OECD Member countries and Key Partner economies are deploying data spaces and functionally comparable architectures through a widening range of policy instruments. Drawing on examples from all regions, this chapter organises selected policy initiatives by the type of policy intervention, which are often combined. Section 4.1 examines regulatory frameworks that require or encourage multi-party data sharing. Section 4.2 analyses public investment and institutional design. Section 4.3 explores standards and cross-border interoperability, while Section 4.4 looks at certification and public trust frameworks. Section 4.5 examines inter-governmental (G2G) data spaces and government deployment as operator and anchor participant.
Regulatory instruments across jurisdictions are driving data space adoption through two distinct channels:
Some explicitly mandate or promote data space architectures, as with the European Health Data Space (EHDS) regulation (European Union, 2025[1]) or India’s creation of regulated consent managers under Data Empowerment and Protection Architecture (DEPA) (NITI Aayog, 2020[2]).
Others impose requirements that can most readily be met through federated, governed data exchange without specifically targeting data spaces, as with the EU Battery Regulation (European Union, 2023[3]), or the Trusted Exchange Framework and Common Agreement (TEFCA) in the United States (HHS, 2026[4]).
In the European Union (EU), several instruments create requirements that align structurally with data space building blocks. The EU Data Governance Act (DGA) (European Union, 2022[5]) requires separation between data intermediation and other commercial activities. The EU AI Act (European Union, 2024[6]) imposes provenance and traceability obligations for high-risk AI systems. The EU Data Act (European Union, 2023[7]) establishes fair access to data from connected products and interoperability requirements. The EU Battery Regulation requires carbon footprint declarations and digital battery passports (DPPs) involving multi-stakeholder data exchange across supply chains. Digital product passports more broadly are an emerging regulatory instrument requiring machine-readable lifecycle data to accompany products; their traceability, provenance and multi-stakeholder access requirements map onto data space infrastructure.
The EU DGA’s neutrality requirement (Article 12) deserves particular attention, not least because it is now subject to significant revision. As adopted in 2022, the EU DGA requires data intermediation services to maintain structural (legal) separation between intermediation and other commercial activities. This is a requirement incompatible with the integrated service models of major cloud platforms, which derive revenue from bundling data intermediation with compute, analytics and artificial intelligence (AI) services. For organisations subject to these requirements, data spaces offer a structurally aligned pathway that proprietary platforms do not.
The EU Digital Omnibus proposal (European Commission, 2025[8]), published in November 2025 alongside the EU Data Union Strategy (European Commission, 2025[9]), would substantially alter this landscape. The proposal repeals the EU DGA as a stand-alone regulation and consolidates its provisions, together with those of the EU Open Data Directive (European Union, 2019[10]) and the EU Free Flow of Non-Personal Data Regulation (European Union, 2018[11]), into the EU Data Act as a single instrument. In the process, two key changes affect data intermediation: registration shifts from mandatory to voluntary (with only registered providers permitted to use a recognition label); and the separation requirement is relaxed from structural (legal) to functional separation. The neutrality protections are retained in principle. Providers, for example, may not exploit data beyond what is needed to operate the intermediation service. However, the shift from structural to functional separation lowers the barrier for integrated platform operators.
Whether these changes strengthen or weaken the regulatory case for data spaces depends on perspective. On the one hand, a lighter data intermediation regime may encourage more providers to enter the market. On the other, it may also erode the distinctive regulatory advantage of data spaces over commercial platforms under the EU DGA’s stricter separation requirement. Bisière et al. (2025[12]) argue that neutrality requirements constrain the cross-subsidisation strategies available to data space operators (the “see-saw effect”). However, they also point out these same requirements prevent commercial platforms from serving the data intermediation function.
The regulatory effect that encourages adoption of data spaces extends beyond the EU. Third-country suppliers and manufacturers must comply with EU product regulations when exporting to the European market. The EU Battery Regulation and the EU Ecodesign for Sustainable Products Regulation (European Union, 2024[13]) may require these suppliers and manufacturers to adopt or join data spaces. This extraterritorial reach transforms regulatory alignment from a European policy concern into a global market access requirement.
While discussions on data spaces tend to focus on EU developments, they have global relevance. The most striking evidence comes from Japan, where the Automotive and Battery Traceability Centre (ABtC) was driven explicitly by the EU Battery Regulation. Thus, Japanese manufacturers adopted data space principles (federation, provider control, common data models) not because of any domestic policy mandate but rather because compliance with EU regulations required precisely the multi-stakeholder traceability infrastructure that data spaces provide (Box 4.1). The initiative is built on the Open Data Spaces Reference Architecture Model (ODS-RAM) and is less than two years old; its long-term viability remains untested. Project Hyperion, a Dutch initiative at the implementation stage, applies the data space model to DPPs in aviation maintenance. It shares aircraft component data across civil and military supply chains through a trusted data space (iSHARE/FIWARE stack). The EU Ecodesign for Sustainable Products Regulation extends DPP requirements into broader product categories.
In the health sector, regulation is the primary catalyst rather than a secondary consideration. The Flemish Health Data Space (implementation stage) is explicitly driven by the EHDS regulation. Besides the EHDS, it lists the EU Data Act, the EU DGA and the EU AI Act as its motivating legal frameworks. The data space exists because the regulatory environment demands a compliant infrastructure for cross-institutional health data sharing. Indeed, the OECD has identified that such an infrastructure requires dedicated governance frameworks that promote data availability, while ensuring privacy protection as called for in the OECD Recommendation on Health Data Governance (OECD, 2017[14]).
Non-European jurisdictions have arrived at functionally comparable regulatory frameworks through independent institutional pathways. In the United States, TEFCA operates as a federated “network of networks” for health data exchange. Developed by the Assistant Secretary for Technology Policy within the Department of Health and Human Services, and co‑ordinated by the Sequoia Project, the network features 11 designated qualified health information networks. Together, they interconnect 14 214 organisations representing over 79 000 unique connections. This entails the exchange of nearly 500 million health records as of February 2026, up from approximately 10 million in January 2025 (HHS, 2026[4]). Six defined exchange purposes (treatment, payment, healthcare operations, public health, government benefits determination and individual access services) provide purpose-bound, contractually enforceable constraints on data use. TEFCA addresses the same multi-party co‑ordination problem that European health data spaces target. However, it does so through a regulatory and contractual framework rather than through a data space architectural stack.
The Automotive and Battery Traceability Centre (ABtC) was established in February 2024 by 14 Japanese automotive manufacturers and two industry associations (the Battery Association for Supply Chain and the Japanese Auto Parts Industries Association). To address the growing global demand for disclosure of information regarding CO2 emissions and due diligence across the entire supply chain, ABtC operates a traceability service. This enables members to compile total CO2 emissions (CFP) throughout the battery lifecycle. To that end, they share and link data securely among members within the automotive and battery supply chains, while safeguarding their trade secrets and other sensitive information.
The Centre has four features: (i) ensuring data interoperability between the different applications adopted by its users; (ii) storing data in a federated manner and protecting it using blockchain technology; (iii) data transfer mechanisms using smart contract based on prior agreements; and (iv) the development of common rules for data sharing with ABtC acting as the main co-ordinator (e.g. for defining data models and identifiers).
The value proposition of ABtC is based on enhancing the ability to share data securely and efficiently (at lower costs) across value chains; providing cost-sharing arrangements to reduce compliance burdens, including for small and medium-sized enterprises; and optimising global supply chain management. The revenue model of ABtC is based on a combination of member (basic) fees and transaction (data use) fees (Annex A). It has been accepted to some funding programmes run by the Japanese government for its proof-of-concept experiments, which could eventually be a new service offered in its data space.
ABtC has also obtained accreditation as a “public interest digital platform operator” under a certification framework administered by the Information-technology Promotion Agency (Section 4.4). A notable feature is the sequenced expansion strategy. It begins with carbon footprint traceability for batteries (a “collaborative service” addressing a shared regulatory challenge). Subsequently, it expands to other vehicle components and full vehicles. Eventually, it moves towards “competitive services” such as battery passports and supply chain resilience applications (Annex A). ABtC’s operator has adopted the Open Data Spaces Reference Architectural Model and its protocols for technical standards and data-sharing rules. In so doing, it avoids the need to build its own infrastructure from scratch.
Section 3.4 of Chapter 3 established that data spaces face a bootstrapping problem operating at two levels: the collective investment in shared infrastructure, and the individual participant’s onboarding investment. Public policy has responded primarily through two channels: (i) time-limited project funding directed at infrastructure and consortium formation; and (ii) permanent institutional anchoring, in which governments establish or designate organisations with a long-term mandate to co‑ordinate data space ecosystems beyond any single funding cycle.
The distinction between the two channels matters. The bootstrapping problem requires sustained co‑ordination over time horizons that typically exceed project grant cycles of two to four years (Section 3.4 of Chapter 3). The time-limited funding channel itself operates on two sides of the bootstrapping problem: supply-side investment in shared infrastructure (Section 4.2.1) and demand-side support for participant adoption (Section 4.2.2). The balance between the two has significant implications for whether funded data spaces reach operational viability and scale.
Several jurisdictions have dedicated permanent institutions to provide the long-term co‑ordination that project funding cannot. Finland’s approach, led by Sitra (the Finnish Innovation Fund), encompasses over 30 data space projects across 15 sectors involving more than 100 organisations (Sitra, 2024[15]). Sitra has launched the Data Spaces Alliance Finland as a national co‑ordination network. It also co‑ordinates Towards the European Health Data Space, an EU joint action on the EHDS. In the Netherlands, the Centre of Excellence for Data Sharing and Cloud, co‑ordinated by the Netherlands Organisation for Applied Scientific Research, has identified 47 active data-sharing initiatives supported by over 80 service providers; nearly 80% of these initiatives depend on government funding (CoE-DSC, 2024[16]). In Germany, the cross-ministerial National Data Institute is designed as a hub for co‑ordinating data ecosystems across public, private and academic sectors. These institutional arrangements share a common design logic: they separate the co‑ordination function from the project-funding cycle.
At the EU level, for instance, the European Data Spaces Support Centre (DSSC) is a consortium of 12 organisations. It is co‑ordinated by the Fraunhofer Institute and Research Units and funded under the EU Digital Europe Programme (October 2022 to March 2026). The DSSC provides cross-sectoral co‑ordination through the Blueprint, the Community of Practice, and guidance on participant adoption and sustainability (DSSC, 2024[17]; 2025[18]). However, the DSSC itself a time-limited project (now continued through a follow-on DSSC 2 phase). This illustrates the tension between the co‑ordination function it serves and the project-funding model through which it is financed.1
4.2.1. Supply-side: Funding data space infrastructure
The dominant policy response to date has been supply-side funding directed at data space infrastructure and consortium formation. As of November 2025, the EU had invested approximately EUR 336 million in Common European Data Spaces across 14 sectors and domains through the Digital Europe Programme. Ongoing EU investment of around EUR 100 million supports their further roll-out (European Commission, 2025[9]). The EU Data Union Strategy, published in November 2025 alongside the EU Digital Omnibus simplification package (Section 4.1), marks a shift towards implementation and deployment (European Commission, 2025[9]). The strategy identifies three priority areas: scaling up access to data for AI; streamlining data rules for legal certainty and reduced compliance costs; and safeguarding the EU’s position on international data flows. In particular, the strategy introduces “data labs” to link data spaces to AI ecosystems. In this way, it repositions data spaces from a governance instrument to AI supply chain infrastructure.
At the national level, approaches vary significantly in scale and design. Germany committed EUR 200 million to the Gaia-X Funding Competition (Förderwettbewerb Gaia-X), administered by the Bundesnetzagentur. The competition led to 11 selected projects and approximately EUR 117.4 million awarded by end 2024, spanning health, mobility, automotive, energy and manufacturing (Bundesnetzagentur, 2022[19]). The funding competition phase concluded in March 2025, with the Gaia-X Hub Germany shifting to an onboarding and community co‑ordination role. The German approach channels substantial public resources through competitive selection, reflecting the pre-competitive research and development (R&D) transfer model that also produced the IDSA Reference Architecture (Chapter 2).
France has integrated data space funding into its broader France 2030 cloud strategy. To that end, Bpifrance launched a dedicated call for projects with significant impact on structuring French industrial sectors. However, the specific data spaces allocation within France 2030’s broader EUR 2.5 billion AI and digital financing envelope has not been publicly broken out (Bpifrance, 2023[20]).
Canada has launched the first dedicated national data spaces programme outside Europe and Asia. The National Data Spaces Pilot Funding Program, managed by the Digital Research Alliance of Canada, runs from January 2026 to March 2028 with a budget of up to CAD 3.15 million (Digital Research Alliance of Canada, 2025[21]).
The People’s Republic of China (hereafter “China”) has adopted the most explicitly state-directed approach to data spaces. The National Data Administration published a Trusted Data Space Development Action Plan (2024-2028) in November 2024 (National Data Administration, 2024[22]). It designated 63 pilot projects across enterprise, industry and city levels in 2025, and targets the creation of more than 100 trusted data spaces by 2028.
These supply-side instruments share a structural limitation that becomes clearer when two distinct purposes of public funding are distinguished, as participants at the OECD Expert Workshop emphasised (Annex A). The first purpose is the development of schemas, protocols and interoperability frameworks (i.e. the technical foundation layer). Both public and private entities can host these technical components, and public investment in this layer resembles pre-competitive R&D with broad spillover benefits. The second purpose is the creation of actual data spaces or ecosystems. Here, the case for public funding is more conditional. It depends on whether a specific collective-action problem has been identified that no market actor has the incentive or legitimacy to solve.
EU funding programmes were not explicitly designed around this distinction or the bootstrapping rationale identified in Section 3.4 of Chapter 3. The European Commission’s stated framing centres on competitiveness, digital sovereignty and the single market for data, not on correcting a formally identified collective-action failure. The bootstrapping justification is an analytical overlay by economists (Bisière et al., 2025[12]; Otto and Tardieu, 2025[23]), not the stated rationale of the policymakers who designed the funding instruments. If the instruments were designed for the first purpose (pre-competitive R&D and technology transfer) rather than the second (solving the specific co‑ordination problem faced by data spaces), the fact that only one in nine entries has reached operational status becomes easier to understand. The instruments were simply not designed for the problem they are now expected, or may need, to solve.
The IDSA Radar data are consistent with this diagnosis. EU funding accounts for 46% of implementation-stage entries. However, it drops to 20% at operational stage, where bilateral or multilateral contracts replace grant agreements as the dominant governance model (Figure 4.1) (Section 3.3 of Chapter 3).
Note: The public funding rate is the share of entries at each development stage that self-report the EU or its members as a funding source. Funding source is reported by 99 of 235 entries (42%); the remaining entries either did not report or do not have funding. Entries that do not report a funding source are included in the denominator (all entries at that stage), which means the rates shown are lower bounds. EU funding sources include Horizon 2020, Horizon Europe, the Digital Europe Programme, EU4Health and Connecting Europe Facility.
Source: Based on IDSA (n.d.), IDSA Data Spaces Radar, www.dataspaces-radar.org.
Yet only 15% of European data-sharing initiatives, including data spaces, have a clear revenue model (Moonen et al., 2025[24]). The gap between funding prevalence and revenue readiness points to an incentive misalignment in subsidy design (discussed as “moral hazard” in the economics literature).2 Where funding instruments reward observable compliance activities (consortium formation, architectural alignment, dissemination) rather than harder-to-measure outcomes (operational viability, participant willingness to pay), rational actors optimise for the former (Holmström and Milgrom, 1991[25]). Lerner (2009[26]) documents this dynamic in public venture capital and entrepreneurship programmes. The mechanism – that subsidy design shapes what recipients optimise for – applies by analogy to consortium funding.
The collective-action problem that public funding aims to solve is thus compounded by a second-order problem in which the funding regime itself may generate misaligned incentives. Workshop participants cautioned that long-term public funding may also create dependency effects that prevent data space stakeholders from being efficient and innovative (Annex A). This concern is distinct from the incentive distortion at the margin discussed above; it points to the possibility that the availability of continued public support removes the pressure to find viable revenue models altogether.
The diagnosis is increasingly recognised. Recent work by both practitioner and research communities has placed business value, use cases and sustainable revenue models at the centre of the data spaces agenda (Mertens et al., 2024[27]; DSSC, 2025[18]; Moonen et al., 2025[24]). The DSSC, for instance, notes the focus is shifting “from technical proof of concepts or governance definitions to the practical realities of delivering real value through use cases, attracting, onboarding and retaining participants and reaching a working business model with long-term viability” (DSSC, 2025[18]). The OECD Expert Workshop held in March 2025 confirmed this finding across practitioners, policymakers and industry experts from Europe, Japan and North America (Annex A).
Prioritising use-case value over infrastructure development was consistently associated with stronger prospects for financial viability. Workshop participants further noted that not all data spaces can be considered public goods. A data space may solve a collective-action problem that no single market actor has the incentive or legitimacy to address (e.g. supply chain traceability across competitors). This solution may create a case for treating the shared infrastructure layer as a public good. However, the dependence of many data spaces on public funding does not in itself demonstrate that they are public infrastructure. It may instead indicate that viable business models have not yet been found.
4.2.2. Demand-side: Funding participant adoption
Far less common is public funding that directly addresses the second level of the bootstrapping problem identified in Section 3.4 of Chapter 3: the onboarding costs that individual participants face when joining a data space. The dominant instruments fund infrastructure (connectors, semantic models, governance frameworks) but not the integration of individual firms into that infrastructure. This gap is significant because even where shared infrastructure exists, participation costs (adapting internal systems, implementing connectors, mapping data to shared semantic models) can deter adoption, particularly by small and medium-sized enterprises (SMEs) (DSSC, 2025[18]).
Spain’s Data Spaces Kit (Kit de Espacios de Datos) is one of the principal exceptions (Red.es, 2025[28]). Administered by Red.es, Spain’s public entity for digitalisation, the programme subsidises the integration of public and private entities into sectoral data spaces on a non-competitive basis. Drawing on EUR 60 million in NextGenerationEU funds, the Data Spaces Kit provides grants of up to EUR 30 000 for private entities, specifically data providers, and EUR 50 000 for public administrations. The non-competitive, low-threshold design targets breadth of adoption rather than depth of capability. This strategy is suited to building participant density in sectors where the co‑ordination problem requires many parties to join simultaneously. The design directly addresses the level-2 bootstrapping problem by reducing the private onboarding cost for each participant, lowering the expected-return threshold that each must clear.
Japan’s approach through the Information-technology Promotion Agency (IPA) illustrates a different demand-side instrument: public investment in knowledge infrastructure and capacity building. IPA has developed a structured educational programme (the “Data Spaces Academy”) comprising data space guides; a Chief Data Officer handbook for executives on data-driven management; and a data maturity framework targeting C-suite executives, business departments and IT personnel (IPA, 2024[29]; IPA, 2025[30]). The rationale, as articulated at the OECD Expert Workshop, is that businesses first need to understand what data spaces are before they can participate; low awareness, limited skills and low data maturity are significant barriers to adoption, particularly among SMEs (Annex A). This capacity-building approach addresses the demand-side bootstrapping problem through a different channel than direct subsidy. Rather than reducing participant onboarding costs financially (as Spain’s programme does), it reduces the informational and organisational barriers that prevent potential participants from recognising the value of joining in the first place.
India’s DEPA offers a structural alternative to demand-side subsidy (Section 4.1). Rather than funding participant onboarding, the regulatory design creates a category of regulated intermediaries (consent managers) that can charge fees for facilitating data exchange (NITI Aayog, 2020[2]). This could make the intermediary role itself commercially viable. By presenting a standardised interface to both data providers and data users, consent managers may reduce participant onboarding costs by design rather than by subsidy. India’s AA framework has achieved substantial adoption without equivalent public subsidy to the European model. That said, it benefitted from regulatory backing by the Reserve Bank of India and co‑ordination by Sahamati, a non-profit industry alliance. The underlying design insight is that creating a commercially viable intermediary category may address the bootstrapping problem more durably than subsidising infrastructure. As such, the design has broader applicability even where the specific consent-based architecture does not transfer directly.
The near-absence of systematic demand-side funding across jurisdictions represents a significant gap in policy design. The three approaches identified above (direct subsidy for onboarding costs, public investment in capacity building and regulatory design that creates commercially viable intermediary categories) each address a different dimension of the level-2 bootstrapping problem. The two-level bootstrapping analysis in Section 3.4 of Chapter 3 suggests that addressing only the supply side (shared infrastructure) while leaving the demand side (participant adoption) unfunded may help explain why data spaces that receive substantial infrastructure investment nonetheless fail to attract sufficient participation.
Standardisation provides the technical substrate for both domestic and cross-border interoperability. As detailed in Box 2.1 of Chapter 2 (Section 2.2 of Chapter 2), activity is proceeding across multiple bodies: the Eclipse Dataspace Protocol (DSP) is in ISO standardisation as ISO/IEC DIS 26450; CEN/CENELEC and ETSI are developing standards for data space components; and the Institute of Electrical and Electronics Engineers has engaged in related standardisation efforts.
The three major reference architectures (Section 2.2 of Chapter 2) each reflect a different model of public-private collaboration. The IDSA Reference Architecture Model (IDSA-RAM) originated from publicly funded research (approximately EUR 10 million from the German Federal Ministry of Education and Research) carried out by the Fraunhofer Institutes. The resulting model was then transferred to IDSA. ODS-RAM was developed by IPA, a government agency under the Ministry of Economy, Trade and Industry (METI), and published jointly by METI and IPA in February 2025; the Data Society Alliance, a private sector association created to implement government data strategy, co‑ordinates ecosystem adoption. The DSSC Blueprint is produced by a consortium of 12 organisations co‑ordinated by the Fraunhofer Institutes and Research Units. It is funded by the European Commission under the EU Digital Europe Programme, with the Commission governing at arm’s length through the grant framework (see Section 4.2).
Cross-border interoperability between data space ecosystems has advanced furthest between Japan and Europe. Developed under METI with IPA’s Digital Architecture Design Center, version 1 of the ODS-RAM was published in February 2025 (METI/IPA, 2025[31]). In October 2025, IPA, the Data Society Alliance, the Robot Revolution & Industrial IoT Initiative and the University of Tokyo designated “Open Data Spaces” as the common concept for Japan’s approach. In April 2026, IPA published the ODS-RAM Version 2 (IPA, 2026[32]) alongside a companion design philosophy document (Tsuda, 2026[33]) that elaborates the architectural paradigm (Section 3.3 of Chapter 3). Interoperability with European architecture has been demonstrated in practice: IPA and Catena-X completed a cross-border proof of concept in March 2025 for product carbon footprint data exchange for batteries, driven by EU Battery Regulation compliance requirements (Section 4.1) (METI, 2025[34]). METI established a certification programme for the Ouranos Ecosystem in May 2025 (METI, 2025[34]). (See Section 4.4).
Other jurisdictions are building architectures that address the same co‑ordination problem, in some cases through independent technical choices. A Green Paper on structuring data spaces in Brazil, prepared by NEO-UFRGS under the Brazilian-German Digital Dialogue, published in July 2024 (NEO-UFRGS, 2024[35]), outlines eight strategies for developing data spaces in Brazil. It has since established direct links to the European standardisation ecosystem: an IDSA Hub at the Eldorado Research Institute (March 2024, the first in the Americas) and IDSA Competence Center recognition (December 2025). These steps do not yet constitute demonstrated interoperability. However, they align Brazil’s emerging data space architecture with the IDSA standards track, reducing the technical distance that future cross-border data exchange would need to bridge.
In the United States, TEFCA achieves federated health data exchange through a common contractual framework rather than through a data space architectural stack (Section 4.1). China’s Trusted Data Space Development Action Plan (2024-2028) emphasises blockchain-based traceability, digital contracts with automatic compliance execution and real-time monitoring (National Data Administration, 2024[22]).
These parallel developments show architectural convergence on common functions (federation, traceability, usage control) without institutional co‑ordination. This is a precondition for future cross-border interoperability but not yet evidence of it.
Where cross-border interoperability has been demonstrated, regulatory requirements have created the concrete demand. The Japan-Catena-X proof of concept for battery traceability, driven by EU Battery Regulation compliance, is the clearest case. Standardisation activity provides the technical substrate. So far, however, organisations mainly invest in cross-border interoperability for regulatory reasons rather than market-related rationales (Section 4.1).
Workshop participants drew an analogy with the Internet to argue that standardisation discussions should focus on the protocol layer rather than on how protocols are implemented or used in business (Annex A). The implication is a “coopetition” model in which co‑operation on protocols coexists with competition on the services built on top. Identifying what to standardise and keeping it minimal are important design choices. Applied to data spaces, this framing reinforces the analytical distinction between the technical foundation layer and the ecosystem layer identified in Section 4.2.1. It suggests that supply-side public investment may be most productively concentrated at the protocol level. Conversely, demand-side support (Section 4.2.2) remains important for promoting adoption of the services and use cases built on top.
Standardisation approaches across jurisdictions range from the EU’s multi-body process through Japan’s demonstrated cross-border interoperability to China’s state-directed architecture. This diversity constitutes a natural experiment in technical governance design. Its full value depends on systematic measurement, a requirement that current data collection infrastructure does not meet (see Box 1.1 of Chapter 1). The IDSA Radar captures primarily European initiatives; DEPA, TEFCA, ODS and China’s trusted data spaces are absent or underrepresented. Extending comparable data collection to non-European federated data-sharing initiatives is a precondition for evidence-based policy guidance that is genuinely international in scope.
Alongside regulation, public investment and standardisation, certification is emerging as a distinct policy instrument for establishing trust in data space operators and distinguishing those that serve a public interest from those that are purely commercial. Two contrasting approaches are in operation: a registration-based regime and certification.
The EU DGA establishes a registration-based regime for data intermediation services. As adopted in 2022, the regime requires mandatory registration with national competent authorities and structural (legal) separation between intermediation and other commercial activities. The EU Digital Omnibus proposal (Section 4.1) would shift registration from mandatory to voluntary and relax the separation requirement from structural to functional. Under both variants, the mechanism is procedural: operators declare compliance with specified requirements; the competent authority registers (or, under the Omnibus, recognises) them. The regime does not assess the substantive quality or public benefit of the intermediation service, nor does it distinguish between operators serving different purposes or sectors.
Conversely, Japan’s “public interest digital platform operator” certification, administered by IPA, certifies operators against substantive criteria for reliability and public benefit. This includes the protection of trade secrets, security, interoperability and data provider control (METI, 2025[34]). The accreditation of ABtC under this framework (Box 4.1) illustrates how certification can provide institutional legitimacy beyond what a private sector consortium structure alone confers. At the same time, certification also signals to potential participants that the operator has been independently assessed against defined standards. The OECD Expert Workshop discussed this certification approach as a potential model for other jurisdictions seeking to distinguish between data spaces that serve a public interest and those that are purely commercial (Annex A).
The two approaches reflect different policy logics. Registration lowers entry barriers and avoids the administrative cost of substantive assessment but provides limited information to potential participants about operator quality or public benefit. Certification imposes higher entry costs and requires institutional capacity to assess operators but generates a trust signal. This signal may help address the bootstrapping problem (Section 3.4 of Chapter 3) by reducing participants’ uncertainty about operator reliability. Whether certification-based or registration-based frameworks prove more effective at establishing durable trust in data intermediaries remains an open empirical question. However, the two approaches offer complementary design insights for jurisdictions developing their own policy frameworks.
Government-backed trust frameworks for digital identity can complement these instruments. For example, the European Digital Identity Wallet – established under the EU's revised eIDAS Regulation (European Union, 2024[36]) – provides the identity layer on which data space participation and usage control depend (Section 3.3.3 of Chapter 3). This underlines the role that governments can play as an anchor participant in data spaces discussed in Section 4.5.
A further policy channel, distinct from regulation, funding, standardisation and certification, is government deployment of data spaces for public sector operations. These configurations span a spectrum. At one end are ecosystems bounded entirely within the public sector to optimise state operations (inter-governmental, G2G). These include cross-ministry planning, public health monitoring or integrated social services. At the other end are arrangements that extend outward to businesses and citizens (G2G2B). In the middle are mixed ecosystems in which government anchors participation among private actors (see Box 4.2). The examples in this section cover this full range.
Rather than subsidising private actors to overcome collective-action barriers, government as operator creates the infrastructure directly and generates the initial data flows and network effects that make subsequent participation attractive. This allows governments to lead by example. By operating data spaces for their own functions, governments demonstrate that the architecture works at scale, legitimise the underlying standards and lower the perceived adoption risk for private actors considering participation. In so doing, government can also address the bootstrapping problem (Section 3.4 of Chapter 3) through a different mechanism than public funding. In some cases, this can lower onboarding costs for subsequent participants.
The choice between funding private data spaces and operating public ones is not binary. The most effective configurations may involve government anchoring initial participation while enabling private extension. The Finnish Maritime Data Space (Section 3.4 of Chapter 3), anchored by the state-owned enterprise Fintraffic, illustrates this logic. In this sector, government operates alongside private participants rather than as sole operator. Several examples across sectors illustrate how this channel operates in practice.
Estonia’s X-Road, operational since 2001 and now governed by the Nordic Institute for Interoperability Solutions (comprising Estonia, Finland and Iceland), provides a longer-running example. Serving approximately 52 000 organisations with nearly 3 000 e‑services and processing more than 1 billion data transactions per year, X-Road is implemented in over 20 countries (e-Estonia, n.d.[37]). It functions as a data space under the broad definition proposed in Section 2.3 of Chapter 2: a governance framework, supported by enabling technical and organisational services, that facilitates trusted and interoperable data sharing among participants. Its evolution is now converging with the narrow definition: the X-Road 8.0 “Spaceship” initiative is aligning its trust framework with data space protocol stacks, including Gaia-X compatibility (NIIS, 2025[38]). X-Road illustrates how long-running government interoperability infrastructure can provide the institutional foundation on which data space capabilities are incrementally layered, a trajectory that inverts the more common pattern of building data space architecture from scratch. It also shows how integration with legacy government systems can proceed incrementally rather than through wholesale replacement.
The EHDS regulation, which entered into force in March 2025, takes a different approach by legislatively mandating government bodies as data space operators. Member States must designate national digital health authorities (for primary use in healthcare delivery) and health data access bodies (for secondary use in research), with the European Commission developing and maintaining central cross-border infrastructure. This is the first EU regulation that explicitly requires establishment of government-operated data space components. It creates a distinct governance model from the consortium-led approach used for most other EU sectoral data spaces.
The EU Public Procurement Data Space (PPDS), launched in September 2024, is among the most documented cases of government as direct data space operator. Managed by the European Commission (DG GROW), the PPDS connects national procurement databases with EU-level procurement data from Tenders Electronic Daily through a federated data layer built on the eProcurement Ontology as common semantic framework. Over 250 000 public authorities engage in public procurement across the EU, spending an estimated EUR 2.5 trillion annually. This represents around 15% of gross domestic product, making it one of the largest potential data spaces by transaction volume (European Commission, n.d.[39]). EU countries can connect their national databases on a voluntary basis; an OECD-implemented project funded by the EU Technical Support Instrument is supporting a cohort of Member States in connecting to the PPDS (OECD, n.d.[40]). The PPDS is notable not only as a government-operated data space; the government’s own procurement activity also generates the data that populate the infrastructure. Consequently, the government is simultaneously operator, data provider and primary user.
Inter-governmental data-sharing frameworks are increasingly being developed to enable co‑ordinated use of data across government entities. These aim to overcome government data silos and promote more effective data use within the government. Such approaches support interoperability, accountability and data stewardship, while reducing unnecessary centralisation of sensitive information. In so doing, they enable trusted and scalable data sharing across public authorities.
From a conceptual perspective, whether such arrangements meet the narrow definition of data spaces (Section 2.3 of Chapter 2) depends on their governance design. Public authorities are typically legally distinct organisations with separate mandates and accountability structures, and separate data controllers. An inter-governmental framework in which each authority retains custody and control over its data under a shared co‑ordination framework can therefore satisfy the federated governance feature of the narrow definition. Conversely, only the broad definition applies where a single lead agency concentrates governance authority and operates a central repository, or where usage conditions are fixed centrally rather than defined by the providing authorities. In either configuration, such frameworks support scalable, co‑ordinated and trusted data sharing and use in the public sector in an effective manner.
Israel’s inter-ministerial framework in the construction sector illustrates the federated configuration: it aggregates and shares planning and infrastructure data across government entities while maintaining data management within the respective authorities. This approach combines domain-level data control with cross-government co‑ordination to improve efficiency and planning outcomes (Government of Israel, 2021[41]; Government of Israel, 2025[42]).
Collective and infrastructural value can also characterise private or mixed data spaces. However, purely public sector data-sharing frameworks differ in that such value cannot be captured through market-based mechanisms. As a result, their sustainability does not rely on revenue generation or cost recovery through user participation, but rather on public funding justified by the public value they generate. In these contexts, data-sharing initiatives are assessed primarily in terms of their contribution to public value, such as improved administrative efficiency, policy coherence and enhanced service delivery. For their part, considerations of financial viability remain relevant insofar as they inform the effective and proportionate use of public resources (OECD, 2019[43]).
The funding implication is that purely public sector data infrastructure, including maintenance of the data themselves, is sustained as a permanent utility through recurrent public budgets, in the same way as other standing government infrastructure. This contrasts with time-limited project funding whose co‑ordination limits Section 4.2 discusses. Estonia’s X-Road, continuously operated and publicly funded for more than two decades, illustrates the durability of this model.
In the United States, the Genesis Mission – established by Executive Order in November 2025 (The White House, 2025[44]) – illustrates a government-operator approach focussed on AI-relevant proprietary data rather than public sector transactions. The US Department of Energy is building a federated platform connecting the 17 national laboratories. In so doing, it aims to combine federal scientific datasets with industry and academic contributions through public-private partnerships to train AI for priority innovation sectors. As its distinguishing feature, the programme explicitly raises the question of intellectual property attribution for contributed data, a challenge that has constrained comparable efforts worldwide (Section 3.3 of Chapter 3). Implementation is at an early stage, but the programme provides another model in which government acts as anchor operator for AI-oriented data sharing without adopting the data spaces label.
Government procurement can also serve as a demand-side driver for data space adoption, complementing the supply-side instruments discussed in Section 4.2.1. The Interoperable Europe Act (European Union, 2024[45]), which entered into force in April 2024, mandates interoperability assessments for EU and national digital public services, effectively requiring the infrastructure that underpins data spaces without explicitly naming them. The eFTI Regulation on electronic freight transport information similarly requires Member State authorities to accept information shared electronically by operators via certified platforms by July 2027, creating de facto demand for data-space type infrastructure in logistics. Where governments require interoperable, standards-based data exchange in their own procurement and operations, they create market demand for data space components that private sector providers can then extend to other contexts.
More generally, existing data-sharing initiatives offer infrastructure and participation that data spaces can build on. Some legislation or sectoral arrangements already require or organise the publication or exchange of data. By making the associated infrastructure compatible with data space protocols, this approach could add usage control, cross-border federation and mechanisms for value creation, without requiring new legislation. The evolution of Estonia's X-Road towards data space compatibility, discussed above, illustrates this trajectory for government interoperability infrastructure.
References
[12] Bisière, C. et al. (2025), “The economics of data spaces”, Policy Paper, SE Digital Center and Toulouse School of Economics, France, https://www.tse-fr.eu/sites/default/files/TSE/documents/DigitalCenter/policy_paper/the_economics_of_data_spaces_july_2025_policy-paper.pdf.
[20] Bpifrance (2023), “Appel au projets: Espace de données”, https://www.bpifrance.fr/nos-appels-a-projets-concours/appel-a-projets-espace-de-donnees (accessed on 14 April 2026).
[19] Bundesnetzagentur (2022), “Funding competition: Innovative and practical applications and data spaces in the Gaia-X digital ecosystem”, https://www.bundesnetzagentur.de/EN/Areas/Digitalisation/Data/GaiaX/start.html (accessed on 14 April 2026).
[16] CoE-DSC (2024), Landscape Scan: Dutch Data Sharing Initiatives & Service Providers, initial trial version, https://coe-dsc.nl/wp-content/uploads/2025/01/Dutch-Data-Sharing-Landscape-Scan-2024-report.pdf.
[21] Digital Research Alliance of Canada (2025), “National data spaces pilot funding program: Call for proposals”, https://www.alliancecan.ca/en/opportunities/national-data-spaces (accessed on 16 April 2026).
[18] DSSC (2025), “Participant adoption and sustainability in data spaces”, https://dssc.eu/publications (accessed on 16 April 2026).
[17] DSSC (2024), “Generative AI and data spaces”, https://dssc.eu/space/Generative+AI+and+Data+Spaces (accessed on 14 April 2026).
[37] e-Estonia (n.d.), “X-Road”, https://e-estonia.com/solutions/interoperability-services/x-road/ (accessed on 16 April 2026).
[9] European Commission (2025), “Data union strategy: Unlocking data for AI”, https://digital-strategy.ec.europa.eu/en/library/data-union-strategy-unlocking-data-ai (accessed on 6 March 2026).
[8] European Commission (2025), Proposal for a Regulation amending Regulations (EU) 2022/868, (EU) 2023/2854 and Directive (EU) 2019/1024 (Digital Legislation Omnibus), European Commission, Brussels.
[47] European Commission (n.d.), “Simpl: Cloud-to-edge federations empowering EU data spaces”, https://digital-strategy.ec.europa.eu/en/policies/simpl (accessed on 14 July 2026).
[39] European Commission (n.d.), “The Public Procurement Data Space (PPDS) – Internal market, industry, entrepreneurship and SMEs”, https://single-market-economy.ec.europa.eu/single-market/public-procurement/digital-procurement/public-procurement-data-space-ppds_en (accessed on 31 March 2026).
[1] European Union (2025), Regulation (EU) 2025/327 on the European Health Data Space, European Union, Brussels.
[6] European Union (2024), Artificial Intelligence Act European Parliament legislative resolution of 13 March 2024 on the proposal for a regulation of the European Parliament and of the Council on laying down harmonised rules on Artificial Intelligence (Artificial Intelligence Act) and amending certain Union Legislative Acts (COM(2021)0206 – C9-0146/2021 – 2021/0106(COD)), European Union, Brussels, https://www.europarl.europa.eu/doceo/document/TA-9-2024-0138_EN.pdf.
[36] European Union (2024), Regulation (EU) 2024/1183 amending Regulation (EU) No 910/2014 as regards establishing the European Digital Identity Framework, European Union, Brussels, https://eur-lex.europa.eu/eli/reg/2024/1183/oj.
[13] European Union (2024), Regulation (EU) 2024/1781 establishing a framework for setting ecodesign requirements for sustainable products (Ecodesign for Sustainable Products Regulation), European Union, Brussels.
[45] European Union (2024), Regulation (EU) 2024/903 laying down measures for a high level of public sector interoperability across the Union (Interoperable Europe Act), European Union, Brussels.
[3] European Union (2023), Regulation (EU) 2023/1542 concerning batteries and waste batteries, European Union, Brussels.
[7] European Union (2023), Regulation (EU) 2023/2854 on harmonised rules on fair access to and use of data (Data Act), European Union, Brussels.
[5] European Union (2022), Regulation (EU) 2022/868 of the European Parliament and of the Council of 30 May 2022 on European data governance and amending Regulation (EU) 2018/1724 (Data Governance Act), European Union, Brussels, https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=CELEX%3A32022R0868.
[10] European Union (2019), Directive (EU) 2019/1024 of the European Parliament and of the Council of 20 June 2019 on open data and the re-use of public sector information, European Union, Brussels, https://eur-lex.europa.eu/legal-content/EN/TXT/?qid=1561563110433&uri=CELEX:32019L1024.
[11] European Union (2018), Regulation (EU) 2018/1807 on a framework for the free flow of non-personal data in the European Union, European Union, Brussels.
[42] Government of Israel (2025), Infrastructure for Growth 2026, Government of Israel, https://www.gov.il/BlobFolder/generalpage/plan-infrastructure-2025/he/files_Infrastructure240326.pdf.
[41] Government of Israel (2021), “Aggregation and sharing of information for planning, implementation, and improvement of regulation in the construction sector”, Government Decision No. 854 (1 March 2021), Government of Israel, https://www.gov.il/he/pages/dec854_2021.
[4] HHS (2026), “TEFCA – America’s national interoperability network reaches nearly 500 million health records exchanged”, https://www.hhs.gov/press-room/tefca-americas-national-interoperability-network-reaches-nearly-500-million-health-records-exchanged.html (accessed on 7 April 2026).
[46] Holmström, B. (1979), “Moral hazard and observability”, The Bell Journal of Economics, Vol. 10/1, p. 74, https://doi.org/10.2307/3003320.
[25] Holmström, B. and P. Milgrom (1991), “Multitask principal-agent analyses: Incentive contracts, asset ownership”, Journal of Law, Economics, & Organization, Vol. 7/special issue, pp. 24-52, https://www.sfu.ca/~allen/HolmstromMilgrom.pdf.
[32] IPA (2026), “Open Data Spaces Reference Architecture Model (ODS-RAM)”, https://open-dataspaces.gitbook.io/ods-docs/ods-ram/v2 (accessed on 1 April 2026).
[30] IPA (2025), “Agreement to jointly promote the dataspace technical concept as ’open data spaces’”, https://www.ipa.go.jp/en/pressrelease/press20251015.html (accessed on 9 April 2026).
[29] IPA (2024), “Data Spaces Academy”, https://www.ipa.go.jp/en/digital/data/data-spaces-academy.html (accessed on 28 April 2026).
[26] Lerner, J. (2009), Boulevard of Broken Dreams: Why Public Efforts to Boost Entrepreneurship and Venture Capital Have Failed — and What to Do About It, Princeton University Press.
[27] Mertens, C. et al. (2024), “Data spaces business models”, Position Paper, Zenodo (CERN European Organization for Nuclear Research), https://doi.org/10.5281/zenodo.14101303.
[34] METI (2025), “Outstanding initiatives to promote industrial data sharing selected under the project certification of Ouranos ecosystem”, https://www.meti.go.jp/english/press/2025/0509_002.html (accessed on 22 April 2026).
[31] METI/IPA (2025), Open Data Spaces Reference Architecture Model (ODS-RAM), Ministry of Economy, Trade and Industry (METI) and the Information-Technology Promotion Agency, Government of Japan.
[24] Moonen, N. et al. (2025), Sustainable Revenue Models for Data Sharing Initiatives, Jheronimus Academy of Data, Tilburg University, The Netherlands, https://www.government.nl/site/binaries/site-content/collections/documents/2025/04/22/sustainable-revenue-models-for-data-sharing-initiatives/Report_Sustainable_Revenue_Models_for_Data_Sharing_Initiatives.pdf.
[22] National Data Administration (2024), Trusted Data Space Development Action Plan (2024-2028), National Data Administration, People’s Republic of China.
[35] NEO-UFRGS (2024), Condições estruturantes para o desenvolvimento de data spaces no Brasil: Uma visão de especialistas [Green Paper: Structuring data spaces in Brazil], Deutsche Gesellschaft für Internationale Zusammenarbeit (GIZ) GmbH on behalf of the German Federal Ministry for Digital and Transport (BMDV), https://digital-dialogues.net/en/news-details/publication-green-paper-on-structuring-data-spaces.
[38] NIIS (2025), “X-Road 8 Spaceship”, https://x-road.global/spaceship (accessed on 22 April 2026).
[2] NITI Aayog (2020), Data Empowerment and Protection Architecture: Draft for Discussion, NITI Aayog, Government of India, https://www.niti.gov.in/sites/default/files/2023-03/Data-Empowerment-and-Protection-Architecture-A-Secure-Consent-Based.pdf.
[43] OECD (2019), Public Value in Public Service Transformation: Working with Change, OECD Publishing, Paris, https://doi.org/10.1787/47c17892-en.
[14] OECD (2017), Recommendation of the Council on Health Data Governance, OECD/LEGAL/0433, https://legalinstruments.oecd.org/en/instruments/OECD-LEGAL-0433 (accessed on 10 March 2026).
[40] OECD (n.d.), “Leveraging the EU Public Procurement Data Space (PPDS)”, https://www.oecd.org/en/about/programmes/sg-reform/leveraging-the-eu-public-procurement-data-space-ppds.html (accessed on 10 April 2026).
[23] Otto, B. and H. Tardieu (2025), Data Sharing: Closing the Gap in the Draghi Report’s Call for AI Competitiveness in Europe, Le Grand Continent.
[28] Red.es (2025), “Kit Espacios de Datos”, (Kit Data Spaces), https://www.red.es/es/iniciativas/proyectos/kit-espacios-de-datos (accessed on 25 April 2026).
[15] Sitra (2024), “Gaia-X Finland builds one of Europe’s leading data sharing business networks in Finland”, https://www.sitra.fi/en/articles/gaia-x-finland-builds-one-of-europes-leading-data-sharing-business-networks-in-finland/ (accessed on 15 April 2026).
[44] The White House (2025), “Launching the Genesis mission”, https://www.whitehouse.gov/presidential-actions/2025/11/launching-the-genesis-mission/ (accessed on 28 April 2026).
[33] Tsuda, M. (2026), Why Open Dataspaces: Design Philosophy and Architectural Paradigm, Information-technology Promotion Agency, Tokyo, https://www.ipa.go.jp/en/digital/architecture-guidelines/individual-link/p1o1lf000001xv4n-att/WhyOpenDataspaces_en.pdf.
← 1. Another example of EU funding includes Simpl, the open-source smart middleware procured by the European Commission to provide common infrastructure for the common European data spaces and cloud-to-edge federations (European Commission, n.d.[47]).
← 2. “Moral hazard” is used here in its principal-agent sense (Holmström, 1979[46]): The funder (principal) cannot directly observe whether recipients (agents) are allocating effort towards operational viability or towards satisfying formal milestones. Holmström and Milgrom (1991[25]) formalise this as a multi-task principal-agent problem: when some tasks are more measurable than others, incentives tied to the measurable task crowd out effort on the unmeasurable one. In this context, compliance is measurable; progress towards operational viability is not.