This chapter examines the geography of greenfield foreign direct investment (FDI) across EU regions. It reviews recent trends in inflows and stocks to place the regional analysis in context. It analyses the distribution of FDI across regions and the extent of disparities, distinguishing between within- and between-country differences. It considers the persistence of regional divides over time, including patterns of stability and mobility across different levels of FDI activity. Finally, it highlights how the clean and digital transitions are reshaping the sectoral composition of FDI.
Connecting FDI and SMEs for Productivity and Innovation in Europe
Abstract
Foreign direct investment (FDI) is a cornerstone of Europe’s economy, but its contribution has become increasingly unpredictable. For decades, it has created jobs, fuelled innovation, and supported regional development, yet its role has shifted markedly in recent years. Over the past decade alone, inflows have swung sharply – from a peak of USD 7.39 billion in 2015 to just USD 0.02 billion in 2022 (a change of – 99.7%) – reflecting successive global and regional shocks. The COVID-19 pandemic, Russia's war of aggression against Ukraine, and growing pressures for economic self-sufficiency have all disrupted investment patterns, triggering sharp declines, corporate restructuring, and only a partial recovery across Europe. Despite this volatility, the stock of inward FDI has more than tripled since 2005, reaching USD 12.4 trillion in 2024 – about one-third of the OECD total and one-quarter of the global total. This underlines the deep integration of multinational enterprises in the EU economy and the significant scope for linkages with domestic firms.
The geography of this investment, however, is highly uneven, with disparities far exceeding those observed in other macroeconomic variables, such as GDP. Over the past two decades, a handful of hubs – such as Eastern and Midland in Ireland, Noord-Holland, and Île-de-France – have captured the bulk of inflows, with volumes often exceeding USD 65 billion, while many regions have attracted virtually none. Unequal access remains stark even after adjusting for population: in 2022, the regional Gini coefficient for FDI per capita reached 0.65, more than triple that of GDP per capita (0.19) and well above typical levels of income disparities. About two-thirds of this disparity reflects differences within Member States rather than between them – for instance, between globally connected capitals and lagging peripheral regions – fuelling pronounced winner-takes-most dynamics not only across Europe, but also within national borders.
Regional FDI divides are also stubbornly persistent across EU regions. Since 2003, around 85% of regions have remained in the same position – whether among the biggest winners or those receiving the least investment – with only limited upward mobility even during periods of rapid expansion. The global and regional shocks of recent years have altered investment volumes but barely touched the underlying geography of FDI. At the same time, its sectoral profile is being reshaped by the twin transition: digital activities, now the largest contributor, account for over one-third of inflows, while clean energy has almost doubled its share to 20% in 2023–2024. Meanwhile, conventional manufacturing corridors have lost ground but continue to play a pivotal role, with strong clusters persisting across parts of France and central and eastern Europe.
Overall, while FDI is central to Europe’s economic and social future in green and digital transitions, its opportunities remain concentrated in a small number of long-standing beneficiary regions, risking deeper divides and threatening cohesion and shared prosperity. Turning this trend around will require targeted action to broaden the geography of investment and help lagging regions and their SMEs integrate into global value chains.
Address entrenched regional divides. Around 85% of EU regions have remained in the same FDI tier since 2003, with leading hubs entrenched at the top and lagging regions stuck at the bottom. Policies should prioritise mobility and convergence to avoid long-term exclusion.
Focus on within-country FDI disparities. Nearly two-thirds of FDI disparities arise inside Member States, underscoring the importance of national and regional strategies to complement EU-wide frameworks.
Seize opportunities from sectoral shifts in FDI. Digital now accounts for more than one-third of inflows and clean energy for almost 20% in 2023-24. Regions able to align their ecosystems with these growth sectors are better placed to catch up.
1.2.1. Inward FDI is volatile, shaped by successive crises and uneven recoveries
Over the past two decades, the EU-27 has attracted substantial volumes of FDI, with flows peaking in 2015, and recent years marked by high volatility. Earlier cycles of expansion in the mid-2000s and contraction during the global financial crisis set the stage for a relatively stable period in the early 2010s, when inflows hovered around USD 300-400 billion annually (Figure 1.1, Panel A). Since 2015, however, swings have become sharper. Inflows peaked at a record USD 1 trillion in 2015, before falling back and then collapsing during the COVID-19 pandemic to just USD 72 billion in 2020, an 80% drop compared to the previous year. A rebound followed in 2021 (USD 378 billion), but momentum quickly faltered, with inflows plunging to only USD 3.6 billion in 2022- the lowest in two decades. A modest recovery has since taken place, with inflows rising to USD 81 billion in 2023 and USD 271 billion in 2024. This pattern broadly mirrors global FDI cycles, which also peaked in 2015 but by 2023-24 had settled at around half that level. The EU’s sharper fluctuations partly reflect the outsized role of intra-company transactions and the concentration of flows in financial hubs such as Ireland, Luxembourg, and the Netherlands (OECD, 2025[1]).
This volatility contrasts with more stable economic indicators such as GDP or employment and reinforces the concentration of investment in established hubs. Multinationals tend to retreat to major financial and corporate centres during downturns and expand selectively during recoveries, leaving peripheral or less established regions with fewer opportunities to attract projects (Borga, Ibarlucea Flores and Sztajerowska, 2020[2]; Nguyen, Pham and Sala, 2022[3]). For SMEs, these swings shape the scope for integration: expansion phases can open new supply-chain channels, while downturns limit opportunities to only the most competitive or proximate firms.
Meanwhile, the stock of inward FDI in the European Union has accumulated steadily over the past two decades and remains consistently high. From USD 3.5 trillion in 2005, it more than doubled by 2013 (USD 8.5 trillion) and climbed further to nearly USD 12.0 trillion in 2019 (+40.6%). It continued to expand, reaching USD 13.1 trillion in 2023 (+9.8% from 2019), before easing slightly to USD 12.4 trillion in 2024 (-5.2% from 2023) (Figure 1.1, Panel A). Even so, the European Union accounted for around one-third of total OECD inward FDI stock and nearly one-quarter of the global total in 2024, making clear the embedded role of foreign-owned firms in its economy, even as annual flows have weakened.
Viewed relative to GDP, FDI has historically played a larger role in the EU-27 than in the OECD or globally but has been more susceptible to fluctuations. The ratio of inflows to GDP peaked at 7.4% in 2015, well above the OECD (3.7%) and world average (3.4%) in the same year (Figure 1.1, Panel B). Since then, the trend has been downward, with the European Union falling below both comparators in 2020 (0.5%) and again in 2022, when inflows relative to GDP were virtually zero (0.02%). More recently, however, the EU ratio has begun to recover, reaching 1.4% of GDP in 2024, though still below its historical highs.
Official inflows (stocks and flows)
Note: Stocks (Panel A, bars) are in trillion USD; flows (Panel A, line) are in billion USD.
OECD and global comparison (Panel B) shows official FDI inflows as a proportion of GDP.
Source: OECD International Direct Investment Statistics (IDIS).
A set of contemporary factors, including pressures for supply-chain reconfiguration, reshoring, and strategic self-sufficiency, are reshaping the nature of FDI in the 2020s. These trends are particularly visible in energy, semiconductors, and critical raw materials (European Commission, 2024[4]). The war in Ukraine and the ensuing European energy crisis have heightened investor uncertainty and accelerated the push for resilience in energy and critical technologies. At the same time, policy frameworks are shifting: a common EU screening mechanism and more active Member State scrutiny mean that inflows into sensitive sectors such as ICT, defence, and advanced manufacturing are increasingly regulated (European Commission, 2024[4]). Investor origins are also diversifying: while the US and UK remain dominant, the share of Chinese investment has declined, replaced by offshore financial centres and emerging players such as India and South Africa.
Official FDI statistics provide a robust picture of national FDI inflows, but their coverage does not always capture the subnational and sectoral detail needed to understand how FDI interacts with local SME ecosystems. To examine these dimensions, the analysis draws on the REGIO-FDI Panel, which provides the empirical basis for the rest of this chapter and chapter 2. This harmonised dataset, which spans 2003 to 2024, links project-level investment records to socio-economic indicators and enterprise data at the regional level. Below summarises its construction and scope (Box 1.1), with further technical details in Annex 1.A.
The REGIO-FDI Panel captures greenfield foreign direct investment (FDI) projects in EU regions over 2003-2024. Project-level data from the Financial Times’ fDi Markets database are mapped to NUTS-2 regions using an OECD correspondence table, ensuring high geographic precision.
Geographic classification. NUTS (Nomenclature of Territorial Units for Statistics) is Eurostat’s standard classification of regions: NUTS-1 refers to major socio-economic regions (3-7 million inhabitants), NUTS-2 to basic regions for the application of regional policies (800 000-3 million inhabitants), and NUTS-3 to small regions for detailed diagnosis (150 000-800 000 inhabitants). This chapter focuses on NUTS-2, which offers a balance between cross-country comparability and policy relevance.
Data sources. fDi Markets is a commercial database that records new and expansion greenfield FDI projects worldwide. Greenfield FDI is the focus of this analysis as it reflects regional attractiveness and development opportunities. In contrast, mergers and acquisitions are harder to allocate regionally and often lie outside the mandate of investment promotion agencies (OECD, 2018[5]). A limitation is the “headquarters effect,” where projects are sometimes attributed to the corporate head office rather than the site of operations, though this bias is systematic across countries.
Coverage. To contextualise FDI, the panel integrates Eurostat indicators on population, GDP per capita, productivity, employment, and sectoral enterprise activity from Structural Business Statistics. Projects are reclassified from fDi Markets’ proprietary taxonomy into the NACE Rev. 2 framework, enabling consistency with regional enterprise data. The resulting unbalanced panel captures total and sectoral inward FDI alongside socio-economic and enterprise indicators.
Limitations. The panel is based on commercial records of announced greenfield projects, which - while extensive - may not capture all investment flows. Disaggregated SME data at NUTS-2 are unavailable; instead, enterprise units are used as a proxy, reflecting SME ecosystems given that SMEs account for 99% of EU firms (European Commission, 2024[6]). The analysis focuses on linkage potential - the conditions enabling FDI-SME connections - rather than realised impacts, and findings, while informative, should not be interpreted as causal.
1.3.1. FDI concentration creates sharp regional divides across the European Union
Investment is heavily concentrated in a handful of metropolitan and industrial hubs, leaving many EU regions with minimal exposure. Eastern and Midland in Ireland and Île-de-France in France each attracted more than USD 65 billion in greenfield FDI between 2003 and 2024, while several regions received less than USD 3 billion over the same period (Figure 1.2, Panel A). The contrast between the top and bottom regions is striking (see Box 1.2). This uneven geography limits access to global value chains, finance, innovation, and international networks (OECD, 2022[7]). Such divergence risks entrenching territorial and social divides (European Commission, 2024[6]).
These gaps remain pronounced even after adjusting for population size. On a per-capita basis, large regions such as Lombardia and Sicilia in Italy fall sharply in the ranking, while smaller territories including Zeeland and Flevoland in the Netherlands or the province of Luxembourg in Belgium rise considerably despite modest absolute inflows (Figure 1.2, Panel B). In smaller regions, where foreign projects constitute a greater share of economic activity, multinationals tend to be more visible, increasing the likelihood of interactions and knowledge transfer with domestic firms (Brookings Institution, 2015[8]; OECD, 2023[9]) Conversely, in larger regions with significant inflows but sizeable domestic economies, FDI projects may represent a smaller share of overall activity, reducing their relative visibility and potential for spillovers. Yet, visibility alone does not translate into impact - effective linkages depend on broader local conditions, including the presence of capable SMEs, supportive institutions, and absorptive capacity that enable FDI to drive productivity and job creation (OECD, 2023[9]; Davies, Ghodsi and Guadagno, 2025[10]).
Note: Data represent cumulative greenfield capital investment (2003-2024) and corresponding per capita averages at the NUTS-2 level.
Source: OECD calculations based on fDi Markets data.
Within the overall distribution, a few standout regions capture a disproportionate share of inward FDI. Leading recipients include Eastern and Midland (Ireland), Île-de-France (France), Cataluña (Spain), and Noord-Holland (Netherlands), each of which attracted more than USD 60 billion in greenfield investment between 2003 and 2024 (Figure 1.3). Even within this top tier, disparities persist: Eastern and Midland received 74% more than Noord-Holland, with total inflows of USD 106 billion and USD 61 billion respectively. At the other end of the spectrum, regions such as Molise (Italy), Warmińsko-Mazurskie (Poland), and several Greek island territories received less than USD 1 billion over the same period. In some cases, cumulative inflows were as low as USD 100-300 million.
Cumulative greenfield FDI between 2003-24 (in USD billions)
Source: OECD based on Financial Times fDi Markets.
1.3.2. FDI disparities have widened, outpacing income inequality
Regional gaps in FDI remain larger and more volatile than income disparities. Per capita FDI has consistently shown greater dispersion than GDP per capita, with regional Gini index averaging around 0.55 between 2003 and 2023 - more than twice the level of GDP per capita - and peaking at 0.65 in 2022 before reducing slightly in 2023 (Figure 1.4, Panel A). These gaps intensified during periods of crisis and recovery, including the global financial crisis and, more recently, the COVID-19 pandemic, likely because investment decisions are more sensitive to cyclical shocks and investor confidence, amplifying fluctuations during downturns and recoveries (Kimura and Flood, 2025[11]; IMF, 2012[12]). By contrast, GDP per capita inequality has followed a gradual downward trend, falling from 0.24 in 2003 to 0.19 in 2023. This suggests that that while regional income disparities have narrowed in many parts of Europe, investment gaps have remained wider and more volatile.
Country-level trends reveal sharp divergences, shaped by national structures and policy environments. In Germany, France, Italy, Spain, Sweden, and Hungary, FDI disparities in 2022-2023 stood well above long-term averages (2003-2021), signaling a marked widening driven by uneven regional capacity to attract and retain foreign projects (Figure 1.4, Panel B). Disparities changed little in Poland and the Netherlands and rose only moderately in Ireland. A few countries - such as Portugal, Slovakia, Lithuania, and Croatia - even recorded gaps below their long-term averages, although in these cases low or declining investment volumes limit broader interpretation. The scale of FDI disparities relative to income inequality also varies widely across Member States, reflecting differences in economic structures, institutional settings, and historical trajectories. Yet, in most countries, FDI gaps remain deeper than income gaps, underscoring the higher volatility of regional investment compared to income inequality - in part because income redistribution and cohesion policies feature more prominently on national policy agendas than subnational investment equalisation.
Gini coefficients of FDI per capita and GDP per capita, long-run trends and recent shifts
Note: Disparities are measured using the Gini coefficient (0-1). Panel A shows the evolution of disparities over time across all EU NUTS-2 regions. Panel B compares long-term averages (2003-2021) with recent estimates (2022-2023), with vertical lines showing the change. Results are omitted for countries where NUTS-2 coverage prevents meaningful estimates (e.g. SI, MT, LU, HR, EE, CY). For countries with only two NUTS-2 regions (e.g. LT, HR), results should be interpreted with caution as values are mechanically close to zero.
Source: OECD calculations based on fDi Markets and Eurostat.
1.3.3. FDI disparities are driven more by within-country than between country divides
A decomposition of EU-wide disparities shows that most FDI differences arises within countries rather than between them. On average over 2003-2023, around two-thirds of disparities in FDI per capita reflected gaps between regions of the same Member State (within-country), rather than across Member States (between-country), a share that has consistently remained above 50% (Figure 1.5, Panel A). The within-country component has also shown notable variation over time, spiking during the global financial crisis and again in 2021 before easing slightly in recent years. Structural change and recent shocks have reinforced this shift. The long-run dismantling of trade and investment barriers has reduced cross-country differences while exposing persistent divides within Member States (Kimura and Flood, 2025[11]; European Commission, 2017[13]). GDP per capita disparities, meanwhile, were shaped more by cross-country differences in the early 2000s, but the within-country share has steadily increased, reaching similar levels to FDI by 2023.
These findings are consistent with earlier OECD evidence showing that regional FDI disparities across the OECD are likewise driven more by within-country than between-country divides. Across OECD regions, within-country differences accounted for around two-thirds of total FDI per capita disparities over the past two decades, peaking at nearly 70% during the COVID-19 pandemic (OECD, 2022[7]). This broader pattern reinforces that regional rather than national factors - such as agglomeration, infrastructure, and institutional quality - are decisive in shaping investment geography.
Even within countries, most FDI disparities are local rather than macro-regional. Distinguishing between broad NUTS-1 macro-regions and NUTS-2 regions within them - the EU’s standard units for policy implementation (see Box 1.1) - reveals that local gaps consistently account for the majority of FDI disparities, easing only slightly from 64% in 2003-2021 to 61% in 2022-2023 (Figure 1.5, Panel B). This distinction is important, as it shows whether inequality reflects a few high-performing localities inside otherwise balanced macro-regions, or instead broad divides between larger national territories. Cross-country experiences vary: In Belgium, Bulgaria, Hungary, Sweden, and Spain, disparities are mainly localised, while in Germany, France, Portugal, and Poland, macro-regional divides dominate. Over time, Austria and Romania moved towards stronger local disparities, while the Netherlands and Greece shifted towards macro-regional gaps, and Italy saw a modest decline in its local share. Overall, this suggests that while local clustering drives most FDI inequality in Europe, in several large economies macro-regional divides remain a persistent concern, which influences whether investment gaps take the form of concentrated enclaves or wider territorial fractures.
Theil index decomposition of FDI per capita disparities, 2003-2023 (higher values = more unequal)
Note: Panel A shows the evolution of FDI per capita disparities across EU NUTS-2 regions, 2003-2023, benchmarked against GDP per capita disparities. Disparities are measured using the Theil index, decomposed into between-country (across Member States) and within-country (across NUTS-2 regions) components. Panel B shows the composition of disparities within countries, using NUTS-1 as the grouping level. The Theil index is decomposed into within-region (blue) and between-region (orange) components. Each country has two stacked bars: 2003-2021 (left) and 2022-2023 (right). Lighter bars indicate a rising within-region share; darker bars indicate a rising between-region share. Countries are ordered by the direction of change. Estimates are restricted to countries with both NUTS-1 and NUTS-2 breakdowns (AT, PT, RO, PL, BG, BE, IT, EL, NL, HU, SE, ES, FR, DE).
Source: OECD calculations based on fDi Markets and Eurostat.
1.3.4. Most regions remain in the same FDI tier over time, even in the face of recent shocks
FDI disparities are highly persistent, with most regions remaining in the same tier over time. Between 2003-2013 and 2014-2022, about 84% of regions stayed in in the same position, whether among the biggest winners or those receiving the least investment (Table 1.1, Panel A).1 Stability was strongest at the extremes: two-thirds of leading regions kept their top position and none dropped to the bottom, while just over half of lagging regions (52%) remained stuck at the lowest level. Among the large middle, nearly 90% stayed in place, with movements more frequently downwards. Recent shocks have not altered this picture. Comparing 2014-2022 with 2023-2024 shows a nearly identical pattern, with 82% of regions in the same tier (Table 1.1, Panel B). Energy transition pressures, supply-chain realignments, and new policy incentives have triggered only modest reshuffling: a handful of regions such as Berlin, Piemonte, and Rhône-Alpes advanced, while others like Brabant Wallon, Bourgogne, and Lubelskie slipped. At the bottom, half of lagging regions moved into the middle tier, particularly in Southern Europe. Yet these shifts were incremental, leaving the divide between high-exposure centres and low-exposure peripheries largely intact.
Transition matrices of NUTS-2 regions across FDI exposure tiers, by period
Note: FDI exposure tiers are defined by decile rankings of cumulative capital investment, grouped as Top 10%, Middle 80%, and Bottom 10%. Figures show counts of NUTS-2 regions and row percentages.
Source: OECD calculations based on fDi Markets and Eurostat.
Beyond how much FDI regions receive, its sectoral mix influences how investment contributes to local economies. Whether projects are concentrated in manufacturing, services, or emerging sectors such as digital and clean energy influences the opportunities and risks for local firms. Manufacturing projects tend to integrate into supply chains and transfer technology, while digital and clean energy projects can foster new clusters and innovation ecosystems but often demand specialised infrastructure and skills (Mercer-Blackman, Xiang and Khan, 2021[14]; Amendolagine et al., 2023[15]). In Austria, for instance, high-tech manufacturing industries such as motor vehicles, electronics, and pharmaceuticals have benefited from sustained greenfield FDI, helping to strengthen value-chain integration and drive productivity growth (OECD, 2023[16]). The sectoral composition of FDI is not only crucial for regional growth and how evenly benefits are distributed, but also signals opportunities to channel foreign capital into productive and sustainable sectors. This calls for upskilling workers, strengthening innovation and supplier ecosystems, and aligning infrastructure and regulatory frameworks with evolving sectoral demands (Kimura and Flood, 2025[11]). Against this backdrop, recent shifts in sectoral composition are altering Europe’s FDI landscape, with implications for both the geography of inflows and their potential impact.
1.4.1. Shifting sectoral patterns in greenfield FDI
Greenfield FDI is shifting from traditional industries towards digital, and more recently, clean energy. Between 2003 and 2013, manufacturing and services together absorbed more than half of inflows, with manufacturing alone accounting for over one-third (Figure 1.6). Digital was already significant at around 16%, while clean energy represented about one-tenth. Between 2014 and 2022, rapid growth in data centres, cloud services, and other ICT activities doubled digital’s share to more than one-third, overtaking manufacturing for the first time. Services lost ground in the same period, while clean energy remained broadly stable. More recently, in 2023-2024, digital climbed further, manufacturing continued its gradual decline, and clean energy doubled its share to close to one-fifth, overtaking services as the third-largest recipient. Together, these shifts mark a structural transformation in FDI: digital has become the main growth engine, clean energy is expanding rapidly where enabling conditions exist, and traditional industries continue to play an important but declining role across many regions. Recent international evidence highlights that growth in clean-tech and digital investment is strongest where complementary policies support infrastructure upgrades, workforce skills, and sustainable investment incentives (OECD/UNCTAD, forthcoming[17]).
Note: Investment is classified using the NACE Rev. 2 framework, with fDi Markets subsectors mapped to NACE codes and grouped into seven exhaustive categories: primary, construction, utilities (non-clean), services, manufacturing, clean energy and environment, and digital. Clean energy is identified using granular subsector data from fDi Markets (renewable and low-carbon power generation), while digital follows the OECD typology (digital services, ICT goods, electronic components, and telecom). These categories are mutually exclusive and collectively exhaustive, ensuring no double counting across sectors).
Source: OECD calculations based on fDi Markets.
As a result of these aggregate shifts, Europe’s FDI geography is shaped by distinct hubs of specialisation that are gradually evolving towards digital and clean energy. Between 2003 and 2022, two manufacturing corridors were especially prominent: one running across Portugal, Spain, and southern France, and another spanning southern Germany, Austria, Czechia, and Slovakia, with extensions into parts of Poland and Hungary (Figure 1.7, Panel A). These hubs specialised in a broad range of high-value manufacturing activities, which are deeply integrated into European and global value chains, while other clusters were more fragmented. In Czechia, for example, sustained inflows into advanced manufacturing - particularly automotive and electrical equipment - have reinforced industrial clusters with growing links to clean-energy technologies and component production (OECD, 2024[18]). Since 2023, new hubs have begun to emerge, particularly in digital and clean energy, concentrated in regions such as Spain, Portugal, France, and parts of Eastern Europe where renewable resources align with policy support and infrastructure upgrades (Figure 1.7, Panel B) (OECD/UNCTAD, forthcoming[17]). Traditional manufacturing corridors continue to attract investment, but they now coexist with newer clusters, reflecting a gradual diversification of Europe’s investment landscape and alignment with the European Commission’s priorities on digitalisation, clean technology, and industrial resilience (European Commission, 2023[19]). Other forms of specialisation, such as in services, utilities, or primary industries, remain less prominent and more localised, typically linked to capital regions or resource-specific projects.
Regional distribution of dominant sectors across two time periods
Source: OECD calculations based on fDi Markets data.
These contrasts highlight why national averages can be misleading and underscore the importance of place-based strategies. The coexistence of established manufacturing strongholds with emerging digital and clean-energy hubs reveals the sharp territorial differences that shape Europe’s investment landscape. Some regions are well positioned to capture new opportunities, while others risk being left behind if structural barriers persist. Effective policy must therefore build on local strengths - such as resource endowments, industrial ecosystems, and innovation capacities - while tackling constraints in infrastructure, skills, and connectivity that limit regions’ ability to attract and benefit from new investment. Ensuring that lagging regions can seize these opportunities will require targeted policy support to strengthen local absorptive capacity, foster entrepreneurship, and link domestic firms to new investment flows (OECD, 2023[9]). These dynamics have profound implications for local businesses, influencing their potential to scale, diversify, and participate in emerging value chains.
These geographic and sectoral patterns do not reveal how inward investment connects with local economies. Chapter 2 examines the conditions for such linkages, including regional readiness, conditional employment effects, SME ecosystems, and sectoral overlap.
References
[15] Amendolagine, V. et al. (2023), “Do green foreign direct investments increase the innovative capability of MNE subsidiaries?”, World Development, Vol. 170, p. 106342, https://doi.org/10.1016/j.worlddev.2023.106342.
[2] Borga, M., P. Ibarlucea Flores and M. Sztajerowska (2020), “Drivers of divestment decisions of multinational enterprises - A cross-country firm-level perspective”, OECD Working Papers on International Investment, No. 2019/03, OECD Publishing, Paris, https://doi.org/10.1787/5a376df4-en.
[8] Brookings Institution (2015), Understanding FDI spillover mechanisms, https://www.brookings.edu/articles/understanding-fdi-spillover-mechanisms/.
[10] Davies, R., M. Ghodsi and F. Guadagno (2025), “Innovation interactions: Multinational spillovers and local absorptive capacity”, Working Paper 265, https://wiiw.ac.at/innovation-interactions-multinational-spillovers-and-local-absorptive-capacity-dlp-7344.pdf.
[6] European Commission (2024), Micro & small businesses make up 99% of enterprises in the EU, https://ec.europa.eu/eurostat/fr/web/products-eurostat-news/w/ddn-20241025-1.
[4] European Commission (2024), Revision of the EU Foreign Direct.
[19] European Commission (2023), A Green Deal Industrial Plan for the Net-Zero Age.
[13] European Commission (2017), My Region, My Europe, Our Future: Seventh report on economic, social and territorial cohesion, https://ec.europa.eu/regional_policy/sources/reports/cohesion7/7cr.pdf.
[12] IMF (2012), “Research Summaries: Foreign Direct Investment and the Crisis: Is This Time Different?”, IMF Research Bulletin, Vol. 13/1, p. 1, https://doi.org/10.5089/9781475502183.026.
[11] Kimura, Y. and M. Flood (2025), “Attracting FDI in regions: Green and strategic opportunities”, OECD Regional Development Papers, No. 122, OECD Publishing, Paris, https://doi.org/10.1787/085c06a1-en.
[14] Mercer-Blackman, V., W. Xiang and F. Khan (2021), Understanding FDI Spillovers in the Presence of GVCs, World Bank, Washington, DC, https://doi.org/10.1596/1813-9450-9645.
[3] Nguyen, T., B. Pham and H. Sala (2022), “Being an emerging economy: To what extent do geopolitical risks hamper technology and FDI inflows?”, Economic Analysis and Policy, Vol. 74, pp. 728-746, https://doi.org/10.1016/j.eap.2022.04.005.
[1] OECD (2025), Modest growth in global FDI flows in 2024 masks weaker.
[18] OECD (2024), Strengthening FDI and SME Linkages in Czechia, OECD Publishing, Paris, https://doi.org/10.1787/4c97d104-en.
[16] OECD (2023), FDI Qualities Review of Austria: Closing Gender Gaps and Empowering Women, OECD Publishing, Paris, https://doi.org/10.1787/fc785be5-en.
[9] OECD (2023), Policy Toolkit for Strengthening FDI and SME Linkages, OECD Publishing, Paris, https://doi.org/10.1787/688bde9a-en.
[7] OECD (2022), “The geography of foreign investment in OECD member countries: How investment promotion agencies support regional development”, OECD Business and Finance Policy Papers, No. 20, OECD Publishing, Paris, https://doi.org/10.1787/1f293a25-en.
[5] OECD (2018), Mapping of investment promotion agencies in OECD countries, OECD Publishing, Paris, https://doi.org/10.1787/098e4f0e-en.
[17] OECD/UNCTAD (forthcoming), Toolbox for harnessing investment policies for clean energy, digital transformation, and public health, G20.
The regional-level dataset used throughout this report – referred to as REGIO-FDI – was developed to provide a harmonised and spatially precise view of greenfield foreign direct investment across EU regions over the period 2003-2024. The dataset integrates multiple data sources and applies consistent definitions across geography, time, and sectors.
The greenfield FDI data were sourced from fDi Markets, which provides project-level records with detailed location information at the city or district level. These geographic markers were spatially linked to EU NUTS-2 regions using an in-house correspondence table developed by the OECD. This table mapped thousands of localities to their corresponding NUTS-2 codes, achieving high geographic precision since city-level granularity is typically finer than NUTS-2. Regional classifications follow the 2024 NUTS revision to ensure consistency with the most up-to-date territorial boundaries.
To contextualise FDI inflows, the project integrates two additional sources of regional data from Eurostat:
1. Regional socio-economic indicators – including population, GDP per capita, labour productivity, and employment – drawn from Eurostat’s regional databases.
2. Structural Business Statistics (SBS) – used to capture the sectoral distribution of enterprise activity, including employment and turnover by sector and region.
All indicators were harmonised into a regional-year panel spanning 2003-2024. Where appropriate, variables such as GDP per capita and labour productivity were converted into constant price terms to ensure comparability over time.
To assess the alignment between FDI and regional economic structures, the report relies on a classification system based on the NACE Rev. 2 framework. Sectoral data on enterprise activity are based on NACE 2-digit codes.
The original fDi Markets database uses a proprietary set of subsectors. To enable cross-source comparability, a custom correspondence table was developed to map these subsectors into the NACE Rev. 2 structure. As fDi Markets classifies investments at a more granular level than NACE-2, this process enabled a relatively robust sectoral alignment across datasets.
Due to classification changes in Eurostat’s enterprise statistics prior to 2008, sectoral analysis of SME activity is limited to the period 2008-2022. Moreover, SME activity in this report refers to overall enterprise activity, as disaggregated data by firm size were not consistently available at the NUTS-2 level across countries.
The final REGIO-FDI dataset is structured as a balanced panel of EU NUTS-2 regions by year, capturing:
Total and sectoral greenfield FDI inflow
Population and economic indicators (e.g. GDP per capita, labour productivity)
Enterprise activity by sector (number of units, employment, wages)
This harmonised structure allows for systematic analysis of spatial disparities, sectoral alignment, and structural correlates of FDI attraction and spillovers at the regional level.
← 1. The 84% persistence rate corresponds to the share of regions that remained in the same tier between the two periods. It is calculated as the sum of regions on the diagonal of the transition matrix in Table 1.1, Panel A: 15 (Top) + 161 (Middle) + 12 (Bottom) = 188 divided by the total number of regions (224), giving 84%. Similar calculations yield 82% for Panel B.