This chapter analyses the enabling conditions that attract FDI and influence its diffusion through SME linkages across EU regions. It examines structural drivers such as skills, R&D, and infrastructure, and introduces an FDI Readiness Index, a composite benchmark of regional capacity to both attract investment and translate it into local benefits. The chapter also examines domestic SME ecosystem capacities and their presence in sectors where foreign investors are active, highlighting the conditions under which spillovers are more likely to occur. It classifies regions into four groups - leaders, untapped, emerging, and lagging - to assess mismatches between capacity and realised FDI flows. Finally, it explores how FDI influences regional labour markets and employment, and reveals distinct regional profiles of top, middle, and bottom FDI recipients, pointing to opportunities for convergence through targeted improvements in human capital, innovation, and SME upgrading.
Connecting FDI and SMEs for Productivity and Innovation in Europe
Abstract
Whether foreign direct investment (FDI) becomes a driver of long-term regional growth in the European Union depends not only on how much arrives but on the conditions it meets. Some regions manage to convert foreign projects into productive jobs and thriving small and medium-sized enterprises (SMEs), while others see investment remain isolated, with few local linkages or even competitive pressures that displace domestic firms. These differences reflect a web of interconnected economic, social, legal and policy factors that determine both how attractive a place is to investors and how deeply new activities take root. Among these, a skilled workforce, strong innovation capacity and reliable transport networks are key features repeatedly associated with greater ability to attract high-value investment and turn it into knowledge transfer, supplier relationships and better jobs for local firms.
Regional FDI patterns across 2003-2024 reveal that economic strength alone does not guarantee better investment outcomes. The analysis shows that while GDP per capita is strongly associated with the volume of FDI a region attracts, income levels do not fully explain why some places convert investment into stronger outcomes. Education, R&D spending and transport connectivity each matter individually, but regions that combine strength in all three attract and embed significantly more FDI. To capture this combined effect, a composite FDI Readiness Index was developed. It aggregates tertiary education attainment, R&D expenditure and transport infrastructure into a single measure of a region’s overall capacity to attract and embed investment. Regions in the top quartile of the index account for the majority of FDI inflows, while those in the bottom quartile score much lower and attract only a small share. Several Central and Eastern European regions have strengthened all three dimensions and, in parallel, recorded faster FDI growth than the EU average, suggesting that upgrading readiness may support better investment outcomes.
Once investment arrives, the structure and strength of local SMEs are closely linked to whether foreign projects integrate and generate wider benefits (see also Chapter 3). SMEs make up 99% of firms across the European Union, but their capacity to connect with multinationals varies widely. Enterprise density ranges from fewer than 20 to more than 200 firms per 1 000 inhabitants, and productivity spans an 18-fold gap - from EUR 15 700 to EUR 279 000 value added per worker. Regions where SMEs are both more productive and more concentrated in medium/high-tech manufacturing or knowledge-intensive services - about 21% of firms in the strongest ecosystems compared with 15% in the weakest - are more often associated with stronger linkages to foreign investors and a greater potential to absorb new technologies.
How closely foreign and domestic firms match by sector can influence how well investment connects to the local economy. Regions where foreign investors enter industries already active locally attract more than half of all greenfield FDI, while those with little overlap receive only about one-tenth. Such alignment can open opportunities for suppliers, learning and skilled labour mobility, but it does not guarantee positive outcomes: strong competition can still displace weaker firms, and valuable spillovers can also arise when sectoral overlap is low, but supply-chain links involving different sectors are strong.
Taken together, these patterns point to a clear divide in the potential for FDI to generate local linkages. Regions that capture the bulk of foreign investment typically combine strong human capital, vibrant innovation activity and well-developed transport networks with SME bases that are both productive and concentrated in technology- and knowledge-intensive sectors. By contrast, regions attracting only a small share of investment often display weaker skills, lower R&D spending, limited connectivity and SME structures focused on lower-technology activities. As a result, the scope for spillovers - through supplier linkages, technology transfer and skill upgrading - remains highly uneven, challenging efforts to close regional gaps and promote more balanced development.
Build the core enablers of investment. Regions that combine skills, innovation and transport connectivity attract and embed more high-value FDI. Strengthening these pillars together - rather than in isolation - should be a priority for places seeking to move up the readiness ladder.
Boost SME productivity and technological depth. SMEs drive linkages but often lack the scale or capabilities to partner with multinationals. Support for upgrading, digital adoption, quality certification and management skills can help them integrate into global value chains.
Align local industry with high-spillover sectors. FDI delivers stronger knowledge transfer where local firms are present in medium/high-tech manufacturing and knowledge-intensive services. Industrial diversification and cluster strategies can increase this match.
Close the gap where foreign and local sectors diverge. Weaker connections between foreign investors and domestic industries signal miss opportunities for collaboration. Supplier-matching, cluster initiatives and targeted promotion can strengthen these linkages and boost knowledge transfer.
Balance attraction with resilience. High overlap can also bring crowding-out risks. Complementing investment promotion with SME support, innovation policies and workforce development helps local firms adapt and share the gains from FDI.
Whereas Chapter 1 showed that FDI in the European Union is both volatile and spatially concentrated, this chapter focuses more closely on FDI–SME linkages. It examines the enabling conditions that attract investment and influence how its benefits diffuse through SME interactions. The analysis is structured around three pillars: 1) structural drivers (skills, R&D, infrastructure), 2) SME ecosystem strength, and 3) sectoral overlap. It then explores how these profiles differ across regions with varying levels of FDI.
FDI performance depends on strong foundations that both attract investment and shape how it is utilised. Regions need the right conditions to draw in foreign projects, but also to ensure these projects generate meaningful local benefits. The extent to which foreign companies hire local talent, integrate into regional value chains, or collaborate with institutions is influenced by structural enablers such as infrastructure, skills, and innovation capacity. These factors play a double role: they make regions more appealing to investors, but they also determine how effectively investment is harnessed once it arrives. For policymakers, the challenge is therefore twofold – to strengthen the foundations that draw investment in, and to ensure those same foundations support the transition from attraction to impact.
2.2.1. Multiple enabling factors influence regional FDI performance
Evidence from OECD and academic research consistently points to a recurring set of conditions associated with stronger FDI outcomes. A number of factors can support stronger FDI performance, including economic scale, productivity, skills, and innovation capacity (OECD, 2022[1]; Crescenzi, Pietrobelli and Rabellotti, 2013[2]; Chen et al., 2010[3]; OECD, 2022[4]; Islam and Beloucif, 2023[5]).1 Transport infrastructure – including air, rail, and road connectivity – can enhance market access and mobility, though its impact depends on how well its configuration aligns with regional economic and spatial profiles (OECD, 2023[6]; Fageda, 2016[7]; Halaszovich and Kinra, 2020[8]; Gaus, Daniele and Lembcke, 2025[9]). Human capital and skills development can attract higher-value activities and help local firms absorb knowledge from foreign partners (OECD, 2022[1]; Jones, 2017[10]). Innovation capacity can strengthen local learning networks and increase the likelihood that investment translates into productivity gains (OECD, 2022[1]). Less tangible conditions also shape investor confidence: governance quality, coherent development strategies and environmental performance are increasingly relevant, even if harder to capture in standard indicators (OECD, 2022[1]; OECD, 2022[11]; Mistura and Roulet, 2019[12]; OECD, 2025[13]). Together, these factors can help regions integrate into global value chains, provide the right mix of labour and services, and foster ecosystems suited to high-value, knowledge-intensive investment (Jones, 2017[10]). The following analysis examines how these drivers – individually and in combination – relate to patterns of FDI attraction and spillover potential across EU regions.
2.2.2. Skills, innovation, and infrastructure jointly strengthen FDI attraction
More affluent regions tend to attract more foreign investment, but income alone does not fully explain where it lands. Across EU regions, investment generally increases with higher income levels and rises especially sharply once a certain level of development is reached (Figure 2.1). Yet, wealth on its own cannot fully account for the geography of FDI, as many regions with similar income levels show very different outcomes. More concretely, the association between FDI and GDP shows wide variation around the average trend, meaning that some middle-income regions attract much more investment than expected, while others lag behind despite comparable income.
EU NUTS-2 regions, 2003-2024 (average)
Source: OECD calculations based on fDi Markets and Eurostat.
Enabling factors shed light on why regions with similar income levels attract very different volumes of foreign investment. Education, which captures core elements of human capital, is one of the clearest dividing lines in this regard. In regions with high levels of tertiary education, the link between income and investment is strong and positive (Figure 2.2, Panel A). In regions with medium levels of education, the relationship is much weaker, and in regions with low attainment it even turns negative. At first glance, such distinct trends seem counterintuitive; however, it reflects the fact that as regions develop, rising wages and costs can erode their earlier price advantage. Without parallel improvements in skills, such regions may struggle to compete for higher-value projects, leaving them stuck in a “middle ground” with neither a cost edge nor a strong talent base. By contrast, regions with a well-educated workforce signal to investors that skilled labour is available, boosting the benefits of higher income and drawing in more knowledge-intensive projects. A similar story emerges for other enabling factors, including innovative capacity and infrastructural development. Regions that invest more in research and development strengthen the link between income and investment, while those with low innovation spending risk falling behind (Figure 2.2, Panel B). Transport infrastructure also makes a difference: regions with denser road and rail networks are better able to translate rising incomes into higher investment, while those with weak connectivity attract relatively little despite similar income levels (Figure 2.2, Panel C).2
EU NUTS-2 regions, 2003-2024 (average)
Note: Each panel groups region into high, medium, and low categories based on terciles of the respective indicator – tertiary educational attainment, R&D spending per capita, or transport infrastructure index. Each point represents a NUTS-2 region, and fitted lines illustrate the association between FDI per capita and the corresponding regional characteristic.
Source: OECD calculations based on fDi Markets and Eurostat.
Complementarities between enabling factors matter more than individual strengths. Panel regressions using annual data from 2003 to 2024 confirm that GDP per capita is a positive and consistently significant predictor of FDI intensity (measured as average FDI per capita). By contrast, the effects of education, R&D, and infrastructure are weaker or mixed when considered individually, with R&D in particular often having a negative link with FDI.3 Their combined influence, however, is much stronger. Education and R&D reinforce each other: innovation systems deliver greater returns when backed by a skilled workforce. Infrastructure quality can also amplify the benefits of education, as better connectivity strengthens the impact of skills on investment attraction. But infrastructure cannot compensate for weak innovation systems. In regions with low R&D capacity, new transport links do not make the area more attractive for high-value projects; instead, they can make it easier for firms and workers to shift activity to nearby hubs with stronger innovation ecosystems, underlining the local innovation gap. The evidence therefore points to a clear conclusion: regions that build skills, innovation, and infrastructure together in mutually reinforcing ways are far better placed to attract and retain foreign investment than those relying on a single strength in isolation.4
A composite index of regional FDI readiness
Building on the preceding analysis, a composite FDI Readiness Index provides a benchmark of regional capacity to attract and benefit from investment. The index is constructed for all EU NUTS-2 regions and brings together three consistently significant drivers of FDI performance: tertiary education attainment, R&D expenditure per capita, and transport infrastructure density. Taken together, these indicators give a summary view of both a region’s ability to draw in foreign investment and its potential to translate inflows into wider economic benefits (see Box 2.1 for details on the methodology).
The FDI Readiness Index is composed of three equally weighted components, each identified in the empirical analysis as a consistently significant driver of regional investment performance:
Tertiary education attainment: share of the population aged 25-64 with tertiary qualifications, reflecting the availability of advanced skills.
R&D expenditure per capita: measured in euros, serving as a proxy for innovation capacity and technological readiness.
Transport infrastructure density: combined density of motorway and rail infrastructure (km/km²), capturing physical connectivity to markets.
Each component is standardised by year (z-scores) to control for EU-wide shifts over time and then averaged to produce a balanced composite score. To reflect long-term structural capacity rather than short-term fluctuations, scores are aggregated over the 2003-2024 period.
The index is expressed in z-scores (standard deviations from the EU average, typically ranging from -3 to +3) to allow for comparability across years and regions.
Wide disparities in FDI readiness mirror Europe’s broader regional divide. Northern and Western regions dominate the top quartile, supported by strong skills bases, dense transport networks, and robust innovation systems - for example, Île-de-France in France and Baden-Württemberg in Germany. By contrast, many Southern and Eastern European regions remain in the lower quartiles, with weaknesses often concentrated in one or two dimensions, such as limited R&D expenditure in North-East Romania or weaker transport connectivity in parts of Southern Italy (Figure 2.3). Addressing these asymmetries is critical for achieving a more balanced investment landscape, as regions with weaker enabling conditions face greater challenges in attracting high-value projects and reaping the benefits of globalisation.
Composite index reflecting tertiary education attainment, R&D expenditure per capita, and transport infrastructure density, 2003-2024 (average)
Note: The index is expressed in z-scores (standard deviations from the EU average). Regions are grouped into quartiles for presentation.
Source: OECD calculations based on fDi Markets data.
Some regions, however, are beginning to narrow the gap by strengthening their structural fundamentals. Since 2003, the largest improvements in readiness have been recorded mainly in Central and Eastern Europe, including regions in Poland (e.g. Mazowieckie, Dolnośląskie, Lubelskie, Śląskie), Czechia (Jihozápad, Moravskoslezsko), and Romania (București-Ilfov, Centru, Nord-Vest). Other notable improvers include parts of Bulgaria, Hungary, Spain, and Portugal. Shared characteristics among these regions include rapid gains in tertiary education attainment, rising R&D spending from relatively low starting points, and targeted upgrades to transport connectivity. These advances have enabled them to move closer to the EU average and close part of the long-standing gap with more advanced Northern and Western regions. Importantly, they have also experienced faster growth in FDI per capita than the EU average, suggesting that improvements in readiness are beginning to translate into stronger investment attraction (see Box 2.2).
While overall disparities in readiness remain large, a group of regions have made substantial progress since 2003. The 20 regions with the largest gains in the Readiness Index are predominantly concentrated in Central and Eastern Europe, including several in Poland (e.g. Mazowieckie, Dolnośląskie, Lubelskie, Śląskie), the Czech Republic (e.g. Jihozápad, Moravskoslezsko), and Romania (e.g. București-Ilfov, Centru, Nord-Vest). Other notable improvers include regions in Bulgaria, Hungary, Spain, and Portugal.
These regions have recorded the strongest gains in tertiary education attainment, R&D expenditure, and transport connectivity, enabling them to move closer to the EU average. Their progress highlights a clear process of catch-up, narrowing part of the long-standing gap with more advanced regions.
Despite these advances, most of the top 20 improvers still remain below the highest-scoring Northern and Western regions in absolute terms. Nevertheless, their trajectory underscores the potential for convergence. Importantly, these regions have also experienced faster growth in FDI per capita than the EU average, suggesting that improvements in readiness have begun to translate into stronger investment attraction (Figure 2.4).
Top 20 regions with the largest gains in readiness since 2003, compared with the EU-27 average
Note: The dashed blue line shows the average Readiness Index of the top 20 improvers (left axis), while the solid blue line shows their average FDI per capita (two-year average, right axis). Grey lines represent the corresponding EU-27 averages: dotted for the Readiness Index and solid for FDI per capita. Readiness Index is expressed as a z-score; FDI per capita is smoothed using a two-year average.
Source: OECD calculations based on fDi Markets and Eurostat.
Classifying regions by capacity and performance
To explore the relationship between structural readiness and realised investment, the FDI Readiness Index is plotted against cumulative FDI per capita (Figure 2.5). This yields four groups (see Annex Table 2.B.1. for the full list of regions):
1. Leaders (green): combine strong absorptive capacity with high realised inflows. These regions - mainly in Northern and Western Europe - not only attract investment but also possess the structural conditions required to embed it effectively. Example regions include Île-de-France (FR10), Oberbayern (DE21), and Eastern and Midland (IE06).
2. Untapped (orange): display strong fundamentals but relatively weak inflows, suggesting potential barriers linked to investment promotion, governance, or market access. Examples include Salzburg (AT32), Rhône-Alpes (FRK2), and Friesland (NL12), which combine strong skills, infrastructure, and innovation capacity but attract less investment than their readiness would suggest.
3. Emerging (blue): attract substantial inflows despite weaker fundamentals, often reflecting sector-specific or cost-based advantages. While promising, such positions may not be sustainable without further upgrading. Examples include Moravskoslezsko (CZ08) in Czechia, Śląskie (PL22) in Poland, and Centru (RO12) in Romania.
4. Lagging (red): face dual challenges of low absorptive capacity and low inflows, underscoring structural and institutional weaknesses that limit both attraction and spillover potential. Examples include Severozapaden (BG31) in Bulgaria, Calabria (ITF6) in Italy, and Oberpfalz in Germany.
This typology illustrates that investment attraction and diffusion are interdependent, yet distinct, dimensions of regional performance. Regions with strong fundamentals but low inflows reveal untapped potential, suggesting that structural capacity alone is not sufficient without effective promotion and connectivity to global investment networks. Conversely, regions that attract investment despite weaker foundations often rely on short-term or cost-based advantages that may prove unsustainable without upgrading their skills, infrastructure, and innovation systems. The most resilient regions are those that combine both – high readiness and strong inflows – reinforcing the importance of integrated policies that strengthen SME ecosystems, enhance innovation capacity, and align investment promotion with long-term regional development strategies.
NUTS-2 regions in the European Union, 2003-2024
Note: The figure plots regions by their average FDI inflows per capita (asinh-transformed) and their FDI attraction/ diffusion capacity (Readiness Index, expressed in z-scores based on education, R&D, and infrastructure). The typology is based on EU-wide median splits, classifying regions as Leaders, Untapped, Emerging, or Lagging. The asinh transformation is used instead of logs because it behaves like a logarithm for large values while remaining defined at zero, ensuring regions with no or very small inflows are retained in the analysis.
Source: OECD calculations based on fDi Markets data.
2.2.3. FDI can support labour markets under specific conditions
Having contrasted capacity and realised inflows, the analysis now examines direct employment effects – the most immediate channel through which FDI projects influence regional labour markets. FDI can have positive labour market effects when projects reach critical mass and connect to regional strengths; effects weaken when investment is dispersed or concentrated in already saturated hubs. Across EU regions (2003-2024), more foreign projects are clearly associated with more jobs: a 1% rise in project numbers corresponds to a 1–2% increase in employment (see Box 2.3). Controlling for underlying regional characteristics strengthens this association, suggesting that initial estimates likely understate FDI’s job-creating potential (see Annex Table 2.C.1. ). This aligns with recent evidence that regional FDI inflows raise employment in a non-linear manner, with stronger effects where sectoral specialisation and spatial spillovers are present (Petreski and Olczyk, 2025[14]; Javorcik, 2014[15]; Jude and Silaghi, 2016[16]; Crescenzi, Ganau and Storper, 2021[17]).
The employment effect is not uniform. Returns diminish once regions already host high levels of investment, meaning that the largest employment gains are often observed in places catching up from lower starting points. The structure of inflows is also important: regions that build critical mass within specific industries benefit most, as concentration strengthens supply chains, supports workforce development, and fosters productivity spillovers. By contrast, more diversified inflows are generally associated with weaker job effects, although this pattern is not consistent across all models.
Labour market characteristics further shape outcomes. Interestingly, higher tertiary education does not always translate into stronger job effects. Many projects generate mid-skill opportunities in manufacturing, logistics, or construction, which may not fully align with highly educated workforces. This mismatch underscores that while skills are vital for attracting high-value investment, not all projects create jobs that directly correspond to advanced qualifications.
Overall, the findings suggest that attracting FDI is only the first step. Regions maximise employment benefits when inflows align with their sectoral strengths and workforce profiles. This calls for policies that not only expand investment but also strengthen local labour markets, build clusters, and ensure that skills strategies correspond to the types of jobs most likely to be created. Yet employment is only one channel through which FDI affects regional economies. Longer-term gains depend on whether foreign projects integrate with domestic business structures, enabling the diffusion of knowledge, technology, and market access.
Econometric analysis confirms that inward FDI contributes positively to job creation in EU regions. Using panel regressions for 2003-2024:
A 1% increase in projects is associated with a 1.0-1.9% rise in jobs.
Diminishing returns are observed once regions already attract high volumes of FDI.
Specialisation strengthens employment effects, as regions concentrating inflows in particular industries generate stronger gains than those with more dispersed inflows.
Education does not always amplify job creation, as many FDI projects create mid-skill opportunities that may not match highly educated workforces.
Source: OECD calculations based on fDi Markets and Eurostat; Annex Table 2.C.1. for regression results.
Beyond immediate employment gains, the lasting value of FDI depends on how deeply it embeds within local economies. SMEs are both a driver of investment attraction and a channel for its diffusion. Strong SME and entrepreneurial ecosystems make regions more attractive to foreign investors by providing suppliers, partners, and innovation capabilities (OECD, 2022[1]; Gyan, Bright and Samuel, 2023[18]). At the same time, they determine whether spillovers take root - through technology transfer, supply-chain integration, and skill upgrading - or remain limited (OECD, 2023[6]). Recognising this dual role is essential for policies that not only strengthen local firms but also enhance regional attractiveness to high-value investment.
Yet, foreign investment alone does not guarantee spillovers or sustained regional upgrading. While it can generate jobs and stimulate demand, its broader benefits depend on whether foreign projects connect with capable local firms that can absorb, adapt, and diffuse new knowledge. The extent to which these benefits materialise hinges on the presence of SMEs with sufficient scale, productivity, and sectoral alignment. Evidence shows that spillovers are strongest in technology- and innovation-intensive sectors, where foreign firms are more likely to transfer advanced processes, demand high-quality inputs, and engage in collaborative R&D with local partners (Davies, Ghodsi and Guadagno, 2025[19]; Chetia et al., 2025[20]; Newman et al., 2015[21]; Jude, 2015[22]). Regions whose SMEs are concentrated in such sectors are therefore better placed to integrate into high-value segments of global value chains and capture knowledge from foreign investors.
SMEs dominate Europe’s business landscape but differ greatly in their capacity to connect with foreign investors. They account for around 99% of all firms across EU regions, making enterprise data a useful proxy for SME ecosystems (European Commission, 2024[23]). Yet sharp contrasts remain. Enterprise density ranges from fewer than 18 firms per 1 000 inhabitants in Nord-Est (Romania) to more than 200 in Prague (Czechia), compared with an EU average of 58. Productivity spans from EUR 15 700 per worker in Yuzhen Tsentralen (Bulgaria) to EUR 279 000 in Southern Ireland – an 18-fold gap, with a median of EUR 91 100 (2019, constant prices). Sectoral composition also varies widely, from under 7% of enterprises in knowledge-intensive services and high-tech manufacturing in the weakest regions to over 34% in the strongest, with a median of around 18%. These differences shape how effectively foreign projects can source locally, engage in technology partnerships, and contribute to upgrading skills.
Stronger SME ecosystems increase the likelihood that foreign investment will embed locally, but scale alone is not enough. Cumulative greenfield FDI per capita (2003-2024) is positively associated with enterprise density, though with wide variation, suggesting that density by itself does not guarantee deeper connections with foreign projects – particularly where firms are concentrated in lower-technology activities (Figure 2.6, Panel A). The link with productivity is stronger and more consistent, indicating that more productive firms are better equipped to meet the standards of high-value, technology-driven projects (Figure 2.6, Panel A). Sectoral composition also plays a decisive role: regions with a larger share of enterprises in knowledge-intensive services and medium- to high-technology manufacturing are better aligned with investor demand in high-spillover industries, raising the likelihood of supplier linkages and collaborative innovation (Figure 2.6, Panel C).
EU NUTS-2 regions, 2008-2022
Note: FDI per capita is shown using the inverse hyperbolic sine (asinh) transformation. This transformation behaves like a natural logarithm for large values but is defined at zero, allowing regions with no or very small inflows to be retained while reducing skewness in the distribution.
Source: OECD calculations based on fDi Markets and Eurostat.
Regions with strong SME ecosystems are best placed to embed foreign investment and generate lasting benefits. The clearest pattern is that regions performing well across multiple SME dimensions - density, productivity, and technological alignment - are most likely to capture spillovers. Weakness in any one dimension can erode this potential: dense but low-productivity networks may be overlooked for supplier roles, while productive firms in low-technology sectors may lack the capabilities foreign investors require for deeper integration. These findings highlight that SMEs are not only beneficiaries of FDI spillovers but also key determinants of where investment locates and how firmly it embeds within regional economies.
2.3.1. Sectoral overlap between FDI and enterprise units
Building on the role of SME ecosystem strength, the extent to which linkages materialise also partly depends on whether foreign projects enter the same industries where domestic firms are concentrated. Spillovers occur when knowledge, technology, or practices introduced by foreign investors spread into the wider economy; they can be positive - raising productivity, upgrading skills, and strengthening supply chains - or negative if stronger foreign firms crowd out domestic competitors or draw talent away. The literature suggests that when foreign and domestic firms operate in the same sectors, horizontal spillovers are more likely, transmitted through competition and imitation (OECD, 2023[6]; Jordaan, Douw and Qiang, 2020[24]; Blomström and Kokko, 1998[25]). At the same time, several studies suggest that vertical linkages, which are investigated closely in Chapter 3, often represent the more important transmission channel for FDI spillovers (Jordaan, Douw and Qiang, 2020[24]; Crespo and Fontoura, 2007[26]; Javorcik, 2004[27]). Ultimately, the scale and direction of spillovers depend not only on sectoral alignment but also on firm capabilities, absorptive capacity, and supportive policies.
Evidence shows that the degree of sectoral overlap between foreign investment and domestic enterprise activity varies widely across Europe. It is measured using an enterprise-FDI horizontal linkage proximity (HLP) score on a scale from 0-1, with higher values indicating stronger overlap between foreign projects and the local enterprise base; that is, the extent to which foreign investment enters the same sectors where domestic firms are concentrated. The strongest overlaps are found in regions such as Île-de-France, Cataluña, Oberbayern, Norte, and Mazowieckie, as well as parts of the Netherlands, Portugal, and Central Europe, where foreign investment patterns mirror the local enterprise base most closely (Figure 2.7). By contrast, the lowest values are concentrated in Greek and Southern Italian regions, parts of Portugal and Spain, and several peripheral regions in Eastern and Northern Europe. These contrasts also translate into scale: regions in the top third of alignment accounted for 54% of all greenfield FDI between 2016 and 2022, compared with just 10% for those in the bottom third.
Regression analysis confirms that FDI tends to follow the footprint of the domestic enterprise base, though only to some extent. A 1% increase in the SME share of firms corresponds to a 0.16% increase in the FDI share, while the employment-based measure yields a slightly stronger elasticity of 0.25 (Annex Table 2.D.1). However, the strength of this relationship varies across sectors, indicating that the potential for alignment between foreign projects and domestic enterprise structures is not uniform across the economy (Annex Figure 2.D.1).
NUTS-2 regions, 2016-2022
Note: The Enterprise-FDI Horizontal Linkage Proximity (HLP) index measures the similarity between the sectoral structure of foreign investment and that of local enterprises. It is calculated as one minus the weighted mean absolute difference between sectoral shares of FDI and SMEs, yielding higher values when the two distributions are more closely aligned. Higher scores indicate greater potential for horizontal linkages, which may facilitate knowledge spillovers but can also create crowding-out pressures. Results should be interpreted with caution, as the index does not capture vertical linkages that occur through supply chains.
Source: OECD calculations based on fDi Markets and Eurostat.
High overlap does not automatically translate into positive outcomes, but it can signal untapped potential. Regions with strong overlaps may benefit from closer interaction between foreign and domestic firms, yet they are also more exposed to risks of crowding and intensified competition. Some regions with low FDI inflows nevertheless score high on HLP, reflecting robust SME structures but barriers that still limit investment - suggesting opportunities where improvements in connectivity, skills, or investment promotion could unlock benefits. Conversely, some high-FDI regions register low scores, as projects cluster in specialised niches with few SME counterparts, limiting wider diffusion.
The HLP index is only one lens, but it underlines the importance of enabling conditions. It cannot capture project quality, SME capabilities, or the strength of supply-chain linkages, yet it provides valuable insight into where FDI is more likely to connect with local enterprises. Whether such connections yield lasting benefits depends on broader policies - ranging from infrastructure and innovation systems to skills and market integration - that determine if potential linkages are realised as sustainable growth.
The geography of FDI in the European Union is highly unequal and structurally persistent, with clear implications for SME linkages. Regions in the top decile attract on average USD 1.37 billion in capital investment, compared with USD 456 million in the middle tier and just USD 60 million in the bottom tier (Table 2.1). On a per-capita basis, project density is more than 200 times higher at the top than at the bottom.
These divides are reinforced by broader development differences. GDP per capita is nearly 50% higher in leading regions, and their markets are larger, with regional GDP (PPS) of around EUR 54 billion compared with EUR 32 billion in the bottom tier. Employment rates are stronger (69% versus 59%), unemployment is lower (6.9% versus 12.1%), and poverty is markedly reduced (13% versus 21%). Structural readiness also diverges sharply: top-tier regions combine dense transport networks, tertiary education attainment of 35% (compared with 22% at the bottom), and double the per capita R&D spending.
Note: Definitions of all indicators used are provided in Annex Table 2.E.1.
1. Transport infrastructure index is based on the product of motorway and rail density. For readability, the table reports the geometric mean (millions), consistent with the log-transformed specification used in Section 2.2.
2. The FDI Readiness Index reflects regional capacity to attract and embed investment, combining tertiary education attainment, R&D expenditure per capita, and transport infrastructure density (Box 2.1 for methodological details).
Source: OECD calculations based on fDi Markets and Eurostat.
SME ecosystems further amplify these divides. Enterprise productivity in top-tier regions is nearly twice as high as in lagging regions, while the share of SMEs in high-spillover sectors such as medium/high-tech manufacturing and knowledge-intensive services is 20.5% compared with 15.3% at the bottom. SME density varies less, suggesting that productivity and technological orientation matter more than scale alone for spillover potential (Table 2.2).
FDI sectoral orientation mirrors these patterns. Nearly 70% of inflows to leading regions are in digital and clean-tech, compared with just 26% in the bottom tier. Although lagging regions register a higher share of projects in these sectors (59% versus 43% in the top tier), the very small scale of investment limits their ability to generate meaningful spillovers. Alignment with local SMEs, measured by the Horizontal Linkage Proximity (HLP) index, peaks at 0.75 in top-tier regions but falls to 0.51 in the bottom tier, determining whether foreign projects embed locally or remain disconnected from domestic economies.
Note: Definitions of all indicators used in Annex Table 2.E.2.
1. The SME-FDI Horizontal Linkage Proximity (HLP) index measures how closely the sectoral structure of foreign investment aligns with that of local SMEs, averaged over 2016-2022 (see Section 2.2 for methodological details).
Source: OECD calculations based on fDi Markets and Eurosta).
Top-tier hubs combine high inflows, advanced readiness, and productive SME ecosystems, but this also raises risks of crowding-out. Middle-tier regions show mixed outcomes, with some underperforming relative to their fundamentals, while bottom-tier regions face structural disadvantages that constrain their ability to capture benefits even when high-potential projects arrive. These findings remain descriptive and diagnostic, but they underscore that the potential for FDI-SME linkages is highly uneven across Europe, encompassing both opportunities for knowledge diffusion and risks of exclusion.
Signs of convergence nevertheless point to the potential for catch-up. Several regions in Central and Eastern Europe - including parts of Poland, the Czech Republic, and Romania - have recorded some of the largest gains in readiness since 2003, driven by improvements in education, R&D, and connectivity. These advances have already begun to translate into faster growth in FDI per capita than the EU average, showing that targeted investment in structural fundamentals can narrow long-standing divides.
References
[25] Blomström, M. and A. Kokko (1998), “Multinational Corporations and Spillovers”, Journal of Economic Surveys, Vol. 12/3, pp. 247-277, https://doi.org/10.1111/1467-6419.00056.
[3] Chen, H. et al. (2010), “Buy local? The geography of venture capital”, Journal of Urban Economics, Vol. 67/1, pp. 90-102, https://doi.org/10.1016/j.jue.2009.09.013.
[20] Chetia, P. et al. (2025), “FDI spillovers, innovation and the role of industrial clusters: Evidence from innovative Indian manufacturing firms”, Economic Modelling, Vol. 152, p. 107240, https://doi.org/10.1016/j.econmod.2025.107240.
[17] Crescenzi, R., ,. Ganau and M. Storper (2021), “Does foreign investment hurt job creation at home? The geography of outward FDI and employment in the USA”, Journal of Economic Geography, Vol. 22/1, pp. 53-79, https://doi.org/10.1093/jeg/lbab016.
[2] Crescenzi, R., C. Pietrobelli and R. Rabellotti (2013), “Innovation drivers, value chains and the geography of multinational corporations in Europe”, Journal of Economic Geography, Vol. 14/6, pp. 1053-1086, https://doi.org/10.1093/jeg/lbt018.
[26] Crespo, N. and M. Fontoura (2007), “Determinant Factors of FDI Spillovers – What Do We Really Know?”, World Development, Vol. 35/3, pp. 410-425, https://doi.org/10.1016/j.worlddev.2006.04.001.
[19] 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.
[23] 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.
[7] Fageda, X. (2016), “International Air Travel and fDi Flows: Evidence from Barcelona”, Journal of Regional Science, Vol. 57/5, pp. 858-883, https://doi.org/10.1111/jors.12325.
[9] Gaus, D., F. Daniele and A. Lembcke (2025), “Transport accessibility and foreign investment: An empirical analysis of European regions”, OECD Regional Development Papers, No. 158, OECD Publishing, Paris, https://doi.org/10.1787/db7e1fde-en.
[18] Gyan, S., J. Bright and A. Samuel (2023), The Dual Nature of FDI: Boosting Local Startups and SMEs While Posing Challenges.
[8] Halaszovich, T. and A. Kinra (2020), “The impact of distance, national transportation systems and logistics performance on FDI and international trade patterns: Results from Asian global value chains”, Transport Policy, Vol. 98, pp. 35-47, https://doi.org/10.1016/j.tranpol.2018.09.003.
[5] Islam, M. and A. Beloucif (2023), “Determinants of Foreign Direct Investment: A Systematic Review of the Empirical Studies”, Foreign Trade Review, Vol. 59/2, pp. 309-337, https://doi.org/10.1177/00157325231158846.
[15] Javorcik, B. (2014), “Does FDI Bring Good Jobs to Host Countries?”, The World Bank Research Observer, Vol. 30/1, pp. 74-94, https://doi.org/10.1093/wbro/lku010.
[27] Javorcik, B. (2004), “Does Foreign Direct Investment Increase the Productivity of Domestic Firms? In Search of Spillovers Through Backward Linkages”, American Economic Review, Vol. 94/3, pp. 605-627, https://doi.org/10.1257/0002828041464605.
[10] Jones, J. (2017), “Agglomeration economies and the location of foreign direct investment: A meta‐analysis”, Journal of Regional Science, Vol. 57/5, pp. 731-757, https://doi.org/10.1111/jors.12335.
[24] Jordaan, J., W. Douw and C. Qiang (2020), Foreign Direct Investment, Backward Linkages, and Productivity Spillovers: What Governments Can Do to Strengthen Linkages and Their Impact.
[22] Jude, C. (2015), “Technology Spillovers from FDI. Evidence on the Intensity of Different Spillover Channels”, The World Economy, Vol. 39/12, pp. 1947-1973, https://doi.org/10.1111/twec.12335.
[16] Jude, C. and M. Silaghi (2016), “Employment effects of foreign direct investment: New evidence from Central and Eastern European countries”, International Economics, Vol. 145, pp. 32-49, https://doi.org/10.1016/j.inteco.2015.02.003.
[12] Mistura, F. and C. Roulet (2019), “The determinants of Foreign Direct Investment: Do statutory restrictions matter?”, OECD Working Papers on International Investment, No. 2019/01, OECD Publishing, Paris, https://doi.org/10.1787/641507ce-en.
[21] Newman, C. et al. (2015), “Technology transfers, foreign investment and productivity spillovers”, European Economic Review, Vol. 76, pp. 168-187, https://doi.org/10.1016/j.euroecorev.2015.02.005.
[13] OECD (2025), “OECD FDI Regulatory Restrictiveness Index 2024: Key findings and trends”, OECD Business and Finance Policy Papers, No. 93, OECD Publishing, Paris, https://doi.org/10.1787/88e61362-en.
[6] OECD (2023), Policy Toolkit for Strengthening FDI and SME Linkages, OECD Publishing, Paris, https://doi.org/10.1787/688bde9a-en.
[1] OECD (2022), FDI Qualities Policy Toolkit, OECD Publishing, Paris, https://doi.org/10.1787/7ba74100-en.
[11] OECD (2022), Framework for Screening Foreign Direct Investment into the EU: Assessing effectiveness and efficiency, OECD Publishing, Paris, https://doi.org/10.1787/f75ec890-en.
[4] 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.
[14] Petreski, M. and M. Olczyk (2025), Foreign Direct Investment and Job Creation in EU Regions, https://doi.org/10.48550/arXiv.2503.23999.
Note: ln(Tertiary Educational Attainment) refers to the natural logarithm of the share of the population aged 25-64 with tertiary education (ISCED levels 5-8), measured at the NUTS-2 level. ln(Infrastructure) is the natural logarithm of road and rail network density (combined), calculated as total kilometres of road and rail per square kilometre of land area at the NUTS-2 level. ln(GERD/PPS) denotes the natural logarithm of Gross Domestic Expenditure on R&D (GERD), adjusted for purchasing power standards (PPS), expressed as a share of regional GDP. ln(GDP per capita) is the natural logarithm of real GDP per capita in PPS. Standard errors are reported in parentheses. All models include year fixed effects, and random-effects GLS regression is applied
* p < 0.10, ** p < 0.05, *** p < 0.01.
Note: Regions are classified into four groups by plotting their cumulative greenfield FDI inflows per capita against their attractiveness (inflows) and diffusion capacity (Readiness Index, expressed in z-scores). Thresholds are set at the EU medians of the two distributions: “Leaders” score above the median on both, “Untapped” above median on diffusion capacity but below median on inflows, “Emerging” above median on inflows but below median on diffusion capacity, and “Lagging” below median on both.
Note: The variable ln(FDI projects) refers to the natural logarithm of the number of greenfield foreign direct investment projects announced in each NUTS-2 region. ln(FDI concentration) represents the Herfindahl-Hirschman Index (HHI) of FDI project shares across sectors within a region, expressed in logarithmic form. ln(GDP per capita) denotes the natural logarithm of regional GDP per capita in purchasing power standards (PPS). Tertiary education (%) captures the share of the population aged 25-64 with tertiary education (ISCED levels 5-8). Tech employment share measures the proportion of total employment in high- and medium-tech manufacturing and knowledge-intensive services. The variable ln(FDI projects)² is included to capture non-linearities in the FDI-employment relationship. Lag dep. variable refers to the first lag of the dependent variable (ln jobs created), included in the dynamic panel specifications to account for persistence over time. Standard errors are reported in parentheses. All specifications exploit the panel structure of the data (NUTS-2 regions observed annually) and include region and year fixed effects. Estimation is based on fixed-effects GLS regression.
* p < 0.10, ** p < 0.05, *** p < 0.01.
Note: Standard errors in parentheses. Robust standard errors clustered at the region level. Year fixed effects included. * p < 0.10, ** p < 0.05, *** p < 0.01. The number of observations differs slightly across specifications due to variation in data availability for enterprise units versus employment.
Linear trendlines of FDI capital share (%) and enterprise share (%) by sector, 2008-2022
Note: Trendlines represent fitted linear relationships between the share of FDI capital and the share of enterprise units across sectors and regions. A steeper slope indicates stronger alignment between inward FDI and the local enterprise base. Solid lines show sector-specific trends, while the dotted line shows the overall relationship across sectors.
Source: OECD calculations based on fDi Markets and Eurostat.
← 1. The OECD FDI-SME Policy Toolkit also highlights other enabling factors, including the embeddedness of foreign affiliates, productivity gaps between MNEs and SMEs, SME absorptive capacity, diffusion channels such as value chain linkages and labour mobility, as well as broader framework conditions like access to finance, digital readiness, and policy support for linkages (OECD, 2023[6]).
← 2. Across all three mediators - skills (Panel A), R&D (Panel B), and infrastructure (Panel C) - weaker or even negative associations in some higher-income regions may reflect a “missing middle” dynamic. As regions develop, rising costs reduce their earlier price advantage, but without strong enabling conditions they may not yet offer the capacities needed to attract higher-value projects, leaving them relatively less appealing to foreign investors.
← 3. Full regressions are provided in Table A A.1.
← 4. Regression results including all interaction terms confirm that the education-R&D complementarity remains the strongest and most robust driver of FDI. The transport-education link is positive but less consistent, while the transport-R&D interaction is weak or negative, suggesting that infrastructure alone cannot offset gaps in innovation capacity.