Volume 25 Issue 3 Thematic Issue: Internationalization and Foreign Direct Divestment Flows in Central and Eastern European Economies Article 4 September 2023 The Impact of FDI and Financial Depth on EU Regional Growth: The Impact of FDI and Financial Depth on EU Regional Growth: Income and Spatial Heterogeneity Income and Spatial Heterogeneity Marialena Petrakou European Investment Bank, Luxembourg, Luxembourg Randolph Luca Bruno University College London, School of Slavonic and East European Studies, London, UK and Catholic University of the Sacred Heart, Department of Economic Policy, Milan, Italy Nick Phelps University of Melbourne, Melbourne School of Design, Melbourne, Australia Follow this and additional works at: https://www.ebrjournal.net/home Part of the International Business Commons, International Economics Commons, and the Regional Economics Commons Recommended Citation Recommended Citation Petrakou, M., Bruno, R., & Phelps, N. (2023). The Impact of FDI and Financial Depth on EU Regional Growth: Income and Spatial Heterogeneity. Economic and Business Review, 25(3), 164-181. https://doi.org/10.15458/2335-4216.1325 This Original Article is brought to you for free and open access by Economic and Business Review. It has been accepted for inclusion in Economic and Business Review by an authorized editor of Economic and Business Review. ORIGINAL ARTICLE The Impact of FDI and Financial Depth on EU Regional Growth: Income and Spatial Heterogeneity MarialenaPetrakou a, * ,RandolphLucaBruno b,c ,NickPhelps d a European Investment Bank, Luxembourg, Luxembourg b University College London, School of Slavonic and East European Studies, London, UK c Catholic University of the Sacred Heart, Department of Economic Policy, Milan, Italy d University of Melbourne, Melbourne School of Design, Melbourne, Australia Abstract Background and objective: The paper explores the impact of foreign direct investment and nancial development on regional growth at the EU regional level for 2005–2017. Both FDI and nancial development are important determinants of the regions’ growth, but not for all EU regions homogeneously. Some EU regions seem to benet more than others, depending on certain characteristics, which implies that FDI attraction policies need to bear in mind not only country specicities, but also regional specicities, hence conrming the need for developing FDI attraction policies at the subnational level: nancial development, capacity building, and Investment Promotion Agencies are key, for example. Methods: The methodology used in the paper relies on a beta-convergence model and on xed effects estimation. In addition, a GMM difference model accounts for endogeneity. Results: Our empirical ndings indicate that, in less wealthy (and more peripheral) regions compared to wealthy regions, FDI productivity spillovers are more signicant. In other words, in less wealthy regions, the imitation effect prevails over the competition effect. Conclusions: FDI and nancial development are important determinants of regional growth, especially for less developed and peripheral regions. Contribution/value: Financial development is shown to be a crucial determinant for economic growth at the regional level, especially for peripheral regions, which raises essential policy implications, especially for the sake of economic disparities in the EU NUTS 2 regions. In other words, local access to nance, especially to bank credit, plays a crucial role for regional growth, despite the continuous integration of nancial markets. Also, there is an income and geographic heterogeneity when it comes to estimating FDI spillovers; therefore, the impact of FDI on growth is not always homo- geneous across territories, which challenges the idea of simple “bright” or “dark” sides to the effects of FDI. Keywords: FDI spillovers, Financial depth, Regional growth, Spatial heterogeneity, EU JEL classication: F23, R11, F30 Introduction T he role of Multinational Enterprises (MNEs) in shaping economic, nancial, and institutional paths in the territorial areas they decide to locate has been well scrutinized from both academic and policy perspectives. Academic and policy orthodoxy suggests that MNEs and foreign direct investment (FDI) are channels of scarce capital and potentially positive local externalities for the host economy, such as growth, employment, and productivity spillovers (Blomström & Kokko, 1998; Girma & Wakelin, 2001; Orji et al., 2021). However, despite the extensive em- pirical literature on FDI spillovers, there is no real con- sensus on the unconditional benets deriving from foreign investment for host economies characterized by uneven economic development (Iammarino & Mc- Cann, 2013) or “dark-side” outcomes (Phelps et al., Received 3 August 2022; accepted 22 May 2023. Available online 5 September 2023 * Corresponding author. E-mail addresses: maria-eleni.petrakou@eib.org (M. Petrakou), randolph.bruno@ucl.ac.uk (R. L. Bruno), nicholas.phelps@unimelb.edu.au (N. Phelps). https://doi.org/10.15458/2335-4216.1325 2335-4216/© 2023 School of Economics and Business University of Ljubljana. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/ licenses/by-nc-nd/4.0/). ECONOMIC AND BUSINESS REVIEW 2023;25:164–181 165 2018). The literature has investigated various factors conditioning such spillovers, for example, education, institutional governance, informal institutions, and social capital (Borensztein et al., 1998; Casi & Resmini, 2017; Ogbuabor et al., 2020), primarily focusing on country-level analyses. However, the conditional na- ture of the impact of FDI is often the most apparent at the subnational level, where regional specicities such as income levels and geographical location can potentially inuence the impact of FDI on growth. Hence, examining FDI spillovers at the regional level should account for such heterogeneities. This paper makes an important contribution to the literature on the geographical impact of FDI. In general, the role of FDI as a determinant of eco- nomic growth at the national level has been well established in the academic literature (Blomström & Kokko, 1998; Borensztein et al., 1998). Nevertheless, the exploration of FDI spillovers in a subnational context has been less widespread, mainly due to a lack of data availability. Despite that, recent research has focused on examining the regional dimension of FDI spillovers in the European Union (EU) and highlighted that FDI-induced spillovers can indeed be localized. When Krisztin and Piribauer (2023) estimated the impact of inward FDI on economic growth at the EU NUTS 2 level using a Bayesian spa- tial autoregressive model and hence accounting for space–time dynamics, it was shown that there is an interdependent spatial relationship between FDI and economic output which indeed proves the mutual ef- fect of FDI on growth. Other studies have revealed that informal institutions, i.e., cultural proximity, so- cial capital, and generalized trust, tend to enhance a region’s capacity to absorb FDI spillovers and that the benets of foreign rms’ presence depend on foreign afliates’ country of origin (Casi & Resmini, 2017; 1 Monastiriotis, 2016). Therefore, regional specicities as well as MNEs’ ownership features can determine FDI spillover effects, which implies that FDI positive externalities are not always universally benecial. Despite the value added, previous research has ne- glected the varied features of the EU regions, explor- ing the impact of FDI on regional growth without dis- tinguishing distinct differences in income levels and geographical positioning of regions within EU mar- kets. The novelty of the paper is that it pays attention to regions’ specicities and addresses the question: which regions benet the most from FDI? Richer or poorer? Central or peripheral ones? Identifying a pat- tern in which less developed and peripheral regions seem to have gained a lot from foreign presence leads to a specic policy formulation where FDI needs to continue being encouraged at the regional level. In addition, this paper lls a gap in the literature by addressing the role of regional nance in eco- nomic growth at the regional level, signifying the importance of local access to nance for growth. So far, nancial development has been proven to ex- ert a positive impact on growth at the national level (Alfaro et al., 2004; Durusu-Ciftci et al., 2017; Rajan & Zingales, 1998), while its effect on growth at the subnational level has not been sufciently explored, which renders this paper quite timely in terms of pol- icy relevance. A large stream of work in economic geography and regional studies is focused on the phenomenon of “nancialization” (e.g., French et al., 2011; Pike & Pollard, 2010), but the role of nancial institutions in regional growth remains poorly under- stood. Recently, there has been a shift in paradigm whereby the need for a spatial monetary policy be- comes highly relevant for ensuring equal distribution of economic development, especially after the recent Covid crisis. Studies have argued that “nancialized economies” (national and regional) can be exposed to external shocks—hence, any spatial dimensions to monetary policy should not be independent from re- gional development policy (Sokol & Pataccini, 2022). In order to contribute to this stream of the lit- erature, this paper examines the effect of FDI and nancial depth on growth at the pan-European re- gional level for a time span of 13 years (2005–2017) by testing the moderating role of income level and centrality or peripherality of the region. Using an originally constructed dataset comprising multiple databases (Orbis, Bankscope, Eurostat), it explores the spatial concentration of FDI in the EU regions and the potentially diverse effect of foreign investment on European regions’ economic-growth patterns. It also provides one of the rst systematic explorations of the role of nancial-sector development in determining regional growth patterns. The paper is organized as follows. Section 1 pro- vides a summary of the theories underpinning the relationship between FDI, nance, and economic growth and an empirical review of the impact of FDI and nance on growth at the regional level. Section 2 presents the geographical distribution of FDI and nance within the EU NUTS 2 (Nomenclature of Ter- ritorial Units for Statistics) regions. Section 3 describes the empirical strategy, and Section 4 presents the re- sults. Finally, Section 5 concludes with a summary 1 The paper dissects the impact of FDI on economic growth using a large European cross-regional sample for 2005–2007 and shows that “FDI-induced spillovers do exist and enhance the economic growth of local economies. Foreign presence, however, is not universally benecial. Positive spillovers, in fact, are associated with EU-originating foreign rms and FDI in services.” (Casi & Resmini, 2017, p. 1500). 166 ECONOMIC AND BUSINESS REVIEW 2023;25:164–181 of the key ndings, suggestions for future research, and implications for policy related to the role of - nancial development and FDI in regional economic performance. 1 Nexus of FDI, nancial development, and economic growth In the next few sections, the paper develops a structured literature review on the intertwined re- lationship between FDI, nancial development, and economic growth. In particular, we focus on the the- oretical framework on the joint and separate impact of FDI and nancial development on growth in ag- gregate terms, then we discuss the impact of FDI and nancial development on growth particularly at the regional level, where two subsections dissect the lit- erature on FDI spillovers and growth vs the literature about the impact of nancial development on eco- nomic growth. Finally, Section 1.3 summarizes. 1.1 Theoretical framework on the impact of FDI and nancial development on economic growth FDI is expected to improve the host economy’s technology and increase its total factor productiv- ity, based on the transfer of new technologies and know-how and more efcient production processes based on imitation and greater competition (Casi & Resmini, 2017). Solow’s (1956, 1957) seminal work on the theory of economic growth, which was pop- ularized by Sala-i-Martin (1996) for the regional level, portrays FDI as a purely exogenous factor. How- ever, FDI can enhance economic growth via several channels that reach beyond increases in local nan- cial and physical capital endowments. In neoclassical growth models à la Solow (1956), this implies that foreign investment contributes to factor accumula- tion, which complements local endowments, and technological progress incorporated in the so-called “Solow residual.” Romer (1986, 1994) proposed the theory of technological change growth, which con- siders FDI as part of the key endogenous process of human-capital accumulation (attraction of talent by MNE subsidiaries), innovation spillovers (technology transfer from headquarters), and knowledge transfer (MNE which possess superior managerial skills). FDI-induced spillovers can benet the host econ- omy through many channels. For example, increased competition forces domestic rms to improve their production efciency, resulting in productivity gains for the whole region, as suggested by the litera- ture on rm heterogeneity (Barrios et al., 2005). In addition, imitation of new products and processes brought by foreign afliates allows domestic rms to enhance their productivity and improve their tech- nology through reverse engineering (Glass & Saggi, 2002), and spillovers can be enabled by human capital mobility. Foreign rms tend to require relatively high- skilled labour and usually invest in staff training. As a result, labour mobility from foreign to domes- tic rms facilitates the transmission of know-how and technologies from the foreign to the domestic rms (Fosfuri et al., 2001). Foreign rms also increase demand for local inputs, which allows direct trans- fer of technology and know-how to domestic rms (Markusen & Venables, 1999; Rodriguez-Clare, 1996). Numerous works document how foreign sub- sidiaries transfer knowledge and technology to host economies through sophisticated and cutting-edge training of employees, contributing to human capital accumulation, technological upgrading, and acqui- sition of managerial skills (Borensztein et al., 1998). Training can take the form of formal training and in- duction sessions or on-the-job training in day-to-day activities such as product design or product devel- opment (Blomström & Kokko, 1998; Padilla-Pérez, 2008). Training can potentially expand the human capital level in the host economy due to the increased availability of skilled labour for local rms and other institutions such as public bodies, knowledge cen- tres, and consultancies. Highly qualied labour might move to a local research centre or academic establish- ments to conduct research and share the knowledge acquired from the MNE with local public institutions. Even without this mechanism, everyday human and business interactions among individuals working in similar industries are likely to promote knowledge diffusion (Alfaro et al., 2004). In addition, local research communities and pub- lic organizations may interact with foreign afliates, enabling exchange of knowledge and information via collaboration or aftercare (Phelps & Fuller, 2001; Young et al., 1994). These collaborative projects may take the form of knowledge exchange programmes, such as MNE managers providing input to universi- ties’ taught materials (Padilla-Pérez, 2008), or devel- opment of links to facilitate FDI knowledge spillovers (D’Este & Patel, 2007). Positive spillovers may be generated, also, by the independent entrepreneurial endeavours (e.g., spin-offs) of former MNE employ- ees (Padilla-Pérez, 2008). Lastly, FDI productivity spillovers can occur via horizontal–vertical integration between foreign afli- ates and domestic rms. Vertical integration refers to a supplier–buyer relationship between a domestic rm and a foreign afliate (backward–forward linkage), and investment in higher product quality and tech- nology upgrading results in productivity spillovers. ECONOMIC AND BUSINESS REVIEW 2023;25:164–181 167 Knowledge transfer can occur through forward link- ages between domestic customers and foreign rms acting as suppliers (Stojˇ ci´ c & Orli´ c, 2020; 2 Zanfei, 2012). In other words, productivity spillovers can oc- cur through knowledge transfer, higher product qual- ity requirements, and technology upgrading (Mariotti et al., 2015), while forward linkages between domestic customers and foreign suppliers may enable pro- ductivity improvements. On the one hand, products, processes, and technologies sold to downstream do- mestic rms may embody superior knowledge (Orlic et al., 2018), and, on the other, increased competi- tion in upstream sectors may force all input suppliers to increase efciency, leading to higher quality and cheaper inputs for downstream rms (Markusen & Venables, 1999). Regarding the role of nancial development in economic growth, a strong theoretical background supports the positive link between nancial markets and national economic growth. The connection be- tween the nancial system and economic growth can be traced back to Schumpeter in 1911. Schumpeter considered the banking system to be a key factor of growth based on its ability to encourage allocation of savings to productive investment and encouragement of innovation (King & Levine, 1993a). Other studies show that advanced nancial markets provide rms with access to liquidity and allow them to diversify their portfolios and be entrepreneurial and inno- vative, stimulating economic growth (Durusu-Ciftci et al., 2017). The nancial system can act as a “lu- bricant” by providing access to capital for rms and encouraging easier ows of funds (Orji et al., 2022), although in some cases a threshold effect has been detected in the nance–growth relationship, whereby beyond a certain threshold, nancial development can even be harmful to growth (Law & Singh, 2014). A well-functioning/well-developed nancial sys- tem prevents productive enterprises from using their xed assets as physical collateral when requesting nancing from a nancial institution. If rms are forced to use physical assets, for example, buildings or machinery, as collateral, this generates a “distor- tion,” which forces the bank to seize these tangible assets in the case of default. In well-developed nan- cial systems, rms can use their intangible assets to guarantee their nancial transactions, protecting their physical assets, which can be put to better use to en- able R&D or invest in high-margin projects (Rajan & Zingales, 2001). Therefore, it can be argued that the more developed the nancial system and the institu- tions around it, the easier the access to capital for rms will be. This effect is stronger if the host economy is dominated by Small and Medium-sized Enterprises (SMEs), which tend to be more credit-constrained and, thus, more in need of external nance (Palacín- Sánchez & Di Pietro, 2016). 1.2 The impact of FDI and nancial development on economic growth at the regional level 1.2.1 FDI spillovers at the regional level in the empirical literature Despite the theoretical framework provided above, we observe that the focus of analysis has recently shifted progressively to the sub-national level, due to data availability for FDI and the increasing im- portance of geographical proximity as a channel of positive FDI externalities (Bournakis, 2021). Knowl- edge transfer and demonstration effects are facilitated by spatial proximity and close social interactions, es- pecially in the presence of non-mobile labour and high transportation costs. In such cases, foreign afl- iates might prefer to source their domestic supplies locally and promote MNE embeddedness by form- ing backward linkages between foreign subsidiaries and domestic rms (Phelps et al., 2003). Subsequently, they can improve knowledge transmission through frequent business interactions, making co-location a necessary but insufcient condition for FDI spillovers (Resmini, 2019; Stojˇ ci´ c & Orli´ c, 2020). The availability of reliable and detailed information on FDI inows at the sub-national level is quite recent. FDI-induced effects depend on several factors, oper- ating at different levels, for example, domestic and foreign rms, local institutions and the rms’ envi- ronment, and the potential interactions among them. Pecuniary externalities and technological spillovers have different spatial dimensions. The market me- diates pecuniary externalities and does not require proximity between the MNE and the domestic rms; however, technological spillovers, which are limited in space and decrease with distance, require proxim- ity (Audretsch, 1998; Keller, 2002). The transmission channels mentioned above have a relevant spatial dimension and are more effective if they involve ag- glomerated activities (Drifeld, 2006; Ottaviano & Thisse, 2004). This implies that co-location is a nec- essary, but not sufcient condition for the occurrence of FDI-induced effects. Foreign rms prefer to source 2 Stojˇ ci´ c and Orli´ c (2020) state: “The improvements in productivity of domestic rms may also come through supplier–buyer vertical interactions with foreign counterparts. Input quality presents competitive advantage for foreign rms for which they may be willing to share technical and managerial knowledge, product design, quality procedures, and nancial management experience with domestic suppliers through backward linkages” (p. 1058). 168 ECONOMIC AND BUSINESS REVIEW 2023;25:164–181 inputs locally to save on transportation costs (Javorcik & Spatareanu, 2011; Rodriguez-Clare, 1996). Crespo et al. (2009) show that the concept of spillovers is inherently spatial and refers to FDI externalities that likely disseminate initially to neigh- bouring regions; they also show that the magnitude and direction of FDI externalities depend heavily on regional characteristics. In other words, agglomera- tion and proximity are essential in estimating FDI spillovers because, as spin-off activity increases, the local employment opportunities for former MNE workers also increase. Moreover, the foreign afli- ate is likely to develop linkages with neighbouring industries and local suppliers or distribution compa- nies to minimize transaction costs and foster bilateral communication with the domestic industry and insti- tutional stakeholders such as local chambers of com- merce (Drifeld, 2006; Monastiriotis, 2016). Therefore, reduced distance and geographical proximity might encourage disseminating FDI benets to the local economy, especially if the latter has good absorptive capacity. In this paper we further contribute and add value to the literature by providing an analysis of the impact of FDI on EU regional growth based on a beta-convergence model. A beta-convergence model is considered a convenient framework and is de- rived directly from the theoretical foundations of the Solow model (Barro & Sala-i-Martin, 1992). It has been used extensively to investigate regional growth in the EU (Monfort, 2008). Previous studies show that human capital formation, agglomeration economies (e.g., urbanization economies), geographical location, infrastructure, and economic structure are important determinants of EU regional economic performance (Boschma et al., 2012; Crescenzi et al., 2016; Crescenzi & Rodríguez-Pose, 2012; Petrakos et al., 2011). 1.2.2 The impact of nancial development on economic growth at the regional level in the empirical literature Several studies nd a positive association between nancial depth and growth at the country level (Demirgüç-Kunt & Maksimovic, 1998; Durusu-Ciftci et al., 2017; King & Levine, 1993b; Rajan & Zingales, 1998), but few examine the inuence of nancial depth on regional growth. A recent study investigated the impact of regional nancial efciency (allocation efciency of nancial resources) on economic growth in China, and it was shown that nancial efciency is quite heterogenous across the Chinese regions and that there is a threshold effect on the impact of nan- cial efciency on economic growth (Hu et al., 2019). In a similar vein, Hasan et al. (2009) investigated 147 EU NUTS 2 regions during 1996–2004 and found that higher bank prot efciency and nancial quality improve regional economic growth. Lucchetti et al. (2001) tested the efciency of the banking sector in the Italian regions during the period 1982–1994. They found that the more efcient the regions’ nancial institutions (e.g., the greater their ability to allocate credit to the most productive rms), the greater was the impact on regions’ economic growth. While some argue that the deepening of nancial markets and the enforcement of creditors’ rights are important elements for securing economic growth (Durusu-Ciftci et al., 2017) at the national level, the regions’ nancialization and regional nance have been increasingly gaining attention as potentially im- portant parameters for shaping the development of uneven spatial trajectories (Sokol, 2017). From a pol- icy perspective, this implies that in times of crises (e.g., the recent Covid crisis and 2008–2009 nancial crisis), regions can be left exposed, without a strong banking/nancial system to boost growth and sus- tain private businesses. Therefore, analysts have re- cently focused attention on the role of policy-making institutions such as central banks in involvement in regional development policies and potential inter- vention to safeguard territorial cohesion and avoid exacerbating regional disparities (Sokol & Pataccini, 2022). According to the World Bank Global Financial De- velopment Database (2022), 3 introduced by ˇ Cihák et al. (2012) in “Benchmarking nancial systems around the world,” there are several indicators of - nancial development. Financial development can be proxied by nancial access, depth, efciency, and sta- bility. Each of these indicators is measured differently. For instance, nancial access can be measured as number of bank branches per 100,000 adults in a given country or region; nancial depth can be measured as the ratio of bank deposits divided by GDP; nancial efciency can be measured as the bank’s return on eq- uity or assets; and nancial stability can be measured as the ratio of the bank’s non-performing loans to its gross loans. 1.3 Theoretical framework on the impact of FDI and nancial development on economic growth: The value added of the paper In the previous sections we have argued that the macro literature (1.1) and the regional literature (1.2) on FDI, nancial development (FD), and growth 3 https://thedocs.worldbank.org/en/doc/5882f2b2117b882d58a78f9c64ea3613-0050062022/original/20220909-global-nancial-development-database.xlsx ECONOMIC AND BUSINESS REVIEW 2023;25:164–181 169 should be more connected in order to grasp the com- plexities of the impact of the former (FDI and FD) on the latter (growth). The overarching theoretical framework can be traced back to the Solow model and the Romer “extension” to endogenous growth theory, whereas an important part of the literature has furthermore tried to assess the existence (ab- sence) or (possibly) irrelevance of spillover effects (horizontally and vertically) from FDI. The literature on FD has developed much earlier, though, focusing on the key element of credit provision and possible bottlenecks. In order to assess such phenomena of the impact of FDI and FD on growth, scholars have looked at different countries, regions, periods, and channels of transmission. Compared to the above- mentioned studies, our study provides an additional insight on the role of nance in EU regional growth for the noteworthy period of 2005–2017, emphasizing a geographical heterogeneity related to the differential impact of nance on growth. This paper tests whether the period 2005–2017 shows regional convergence/divergence patterns for the 252 EU NUTS 2 regions under study. The EU- 24 4 regions exhibit conditional b-convergence when their GDP per capita growth rates are negatively re- lated to their initial income, in other words, when the poorer regions tend to catch up with the wealthier ones in the sample (Barro & Sala-i-Martin, 1992; Sala- i-Martin, 1996). Finally, this paper focuses on regional nancial depth 5 as a measure of nancial development and explores its inuence as a driver of regional-level growth. In fact, we clearly dene nancial depth as the ratio of bank deposits to GDP in each EU NUTS 2 region (see Table A1) and test it as a determinant of growth for 250 EU NUTS 2 regions during 2005–2017. 2 Geographical distribution of FDI and nancial depth across EU NUTS 2 regions How are FDI and nancial depth distributed across EU NUTS 2 regions? Using ne-grained maps, we portray the regional distribution of FDI in the EU NUTS 2 regions, by unveiling patterns of foreign- owned rms’ and nancial-depth concentration. The map of FDI presence is illustrated in Fig. 1 below. Some regions in the EU tend to exhibit a much higher presence of foreign enterprises than others. For instance, West Midlands in the UK 6 has a foreign presence of .4, which means that 40% of that region’s turnover stems from foreign rms, based on its major automotive manufacturing cluster (Bryson & Taylor, 2006). Regions in North and South Holland also show high FDI ratios: North Holland includes the capital city, Amsterdam, and South Holland includes indus- trial city of Rotterdam, which in turn includes the port of Rotterdam, which operates as a trade hub for north- west Europe and has a high concentration of foreign rms. FDI presence shows a clear metropolitan/ periphery pattern—metropolitan regions continue to attract a high volume of foreign-rm activities. In addition, some border regions, especially along the old East–West frontier, seem to attract signicant FDI activity. Generally, foreign presence dominates in Central and Eastern Europe (CEE) and the UK and Ireland, implying that FDI activity follows an East–Northwest geographical distribution. Previous studies conrm that some CEE countries, such as Poland, the Slovak Republic, and Czech Republic, have attracted signicant amounts of FDI since the 1990s, especially from countries in western Europe, particularly Germany. CEE countries became a popular FDI destination due to geographical proximity to the investor countries, a skilled and inexpensive labour force, political stability, and prospective EU membership (Pavlínek, 2004). This is conrmed by the fact that after the fth EU enlargement in 2004 and 2007, FDI seemed to preferentially locate on the border regions of the CEE countries (“West–East border effect”), which were characterized by lower production costs and could benet from abolished border checks, due to the EU integration (Serwicka et al., 2022). Although countries in south and western Europe (especially Greece, Italy, and Portugal) are not sig- nicant recipients of foreign activity compared to the CEE 7 countries, certain non-metropolitan regions have a strong FDI presence (e.g., Sterea Ellada in Greece, Aragon and Asturias in Spain). This some- what “inated” indicator might be attributable to the method used to measure FDI presence: foreign rms’ turnover over total turnover in the same year. This 4 Denmark, Croatia, and Cyprus are excluded from the analysis because we have insufcient information on Danish regions. Croatia joined the EU in 2013, and Cyprus FDI data could be “over-inated” due to Russian investment in shell companies. 5 Unfortunately, Bankscope does not provide sufcient data to allow the construction of other nancial development variables. 6 As of February 2020, UK has no longer been a member of the European Union. 7 CEE seems to have attracted higher amounts of FDI vis-à-vis other regions during the 2010 to 2017 time period. This might be related to the reverse capital ows that occurred after the 2008 nancial crisis, which induced many foreign rms, including banks, to repatriate their capital, i.e., outward capital ows (Mihaljek, 2010). It is likely, also, that the repatriation of foreign banks’ assets caused a credit contraction in the CEE countries and affected MNCs’ subsequent location decisions to decrease foreign afliates’ turnover. Also, the high dependence of CEE countries on inward FDI from western Europe and, especially, Germany, might have exposed CEE countries to macroeconomic shocks occurring in western Europe and might be responsible for the evident decrease in FDI in CEE after the 2008–2009 nancial crisis, which was the result of a contraction in production in western Europe. 170 ECONOMIC AND BUSINESS REVIEW 2023;25:164–181 Fig. 1. FDI presence (EU NUTS 2) for 2005–2017. Source: Authors’ calculation. Fig. 2. EU NUTS 2 FDI presence by GEO group 8 (2005–2017). Source: Authors’ calculation. ratio may be large if the NUTS 2 region has relatively low levels of total economic activity (denominator) and, simultaneously, attracts only a few foreign rms with high turnover (numerator). Fig. 2 illustrates the FDI patterns developed through time for the three different geographical groups of NUTS 2 regions. Overall, it is evident that after 2012–2013 all regions experience a reversal of downward FDI trend, attributed to the economic cri- sis. FDI presence seems to be the lowest in the South EU regions and has the lowest uctuations over time. A slight rise of FDI is observed between 2014–2015 8 FDI presence by group of regions: 1 (Northwest), 2 (South), 3 (CEE). ECONOMIC AND BUSINESS REVIEW 2023;25:164–181 171 Fig. 3. Financial depth (EU NUTS 2) for 2005–2017. Source: Authors’ calculation. and afterwards it is rather stable (until 2017). The CEE group of regions seems to attract the highest FDI among the three groups of regions and it also seems to face the highest uctuations of FDI presence, es- pecially from 2013 onwards. This might be related to the reverse capital ows that occurred after the crisis and induced many foreign rms (including banks) to repatriate their capital and cause outward capital ows (Mihaljek, 2010). Moreover, the high depen- dence of the CEE countries on inward FDI from West- ern Europe and especially from Germany might have rendered the CEE countries vulnerable to macroe- conomic shocks occurring in Western Europe and hence might be responsible for the evident decrease of FDI occurring in CEE after the 2008–2009 nancial crisis, as an aftermath of the contracted production in Western Europe. For instance, the FDI stock in the au- tomotive industry in the CEE countries experienced a temporary decrease during the years of the economic crisis, reaching its lowest point in 2011 but slowly recovering in 2012, exhibiting large uctuations stemming from the changes in automotive FDI stock in Hungary and from the declining investment from Western European companies (Pavlínek, 2015). Finally, the Northwest group of regions exhibits a uctuation of FDI similar to the CEE regions, due to the resurgence of economic activity after the nancial crisis. The mapping of nancial depth in Fig. 3 9 shows that, except for a large part of Germany, Austria, Ire- land, and some parts of the UK, few regions have high levels of nancial depth. The high levels re- ported for the UK and Ireland can be explained by the fact that the former is a world-level nancial cen- tre and the latter is a country with a high presence of FDI net inows. The next ranked areas for nan- cial depth are regions in Germany and the Benelux area, which represents the heart of EU economic ac- tivity with signicant industrial and nancial centres considered safety nets for deposits in Europe. Except for some peripheral regions in the south of the EU, metropolitan regions appear to have higher levels of nancial depth. This might be due to unrecorded ac- tivity (shadow economy). 3 Methodology The present paper explores two specic phenom- ena which have received scant attention so far as determinants of regional economic performance: for- eign rms’ presence (FDI) and nancial depth of the region. These factors are included as economic- growth explanatory variables in an extended beta- convergence model. Relying on a beta-convergence framework at the regional level, where regional GDP per capita growth rate is regressed on the initial 9 Calculated as bank deposits divided by GDP for each NUTS 2 region (see Table A1 for a denition). 172 ECONOMIC AND BUSINESS REVIEW 2023;25:164–181 regional GDP per capita level, we employ a speci- cation completely in line with the seminal papers by Crescenzi and Rodríguez-Pose (2012) and Crescenzi et al. (2016) and estimate the following equation: 1logGDPpc it D aCb 1 LogGDPpc it 1 Cb 2 LogFDI it 1 Cb 3 LogFIN-DEPTH it 1 Cb 4 Log(INTERACTIONS) it 1 Cb 5 LogCONTROLS it 1 C RegionFEC TimeFEC u it (1) where i takes values between 1 and 252 for the EU regions, t takes values from 2005 to 2017, and a is the constant term. In other words, the dependent variable 1logGDPpc it D logGDPpc t – logGDPpc t 1 is the annual change of the logarithm of GDP per capita (at constant 2010 prices) in region i and LogGDPpc it 1 measures the logarithm of lagged GDP per capita, which can be interpreted as the “convergence” explanatory vari- able. When the sign of the coefcient b 1 is negative, this signies that the countries with a relatively lower Gross Domestic Product per capita grow faster. In fact, the annual difference of the log of GDP is an excellent proxy of the growth rate of the level of devel- opment we wish to capture, and in the results section we show that this is indeed the case. Finally,b 2 andb 3 are respectively the coefcients of the lagged main ex- planatory variables (log of FDI and FIN-DEPTH), and b 4 , b 5 are the coefcients of the interaction terms (ex- ploring moderation channels) and a vector of control variables, respectively. All the models include region and time xed effects, and u it is the idiosyncratic er- ror term. We rely on such specication to explicitly capture the cross-country variation levels of FDI and FIN-DEPTH on the growth of level of development. Had we used the deltas of the independent variable, we would have departed from the convergence model specication. A delayed effect and endogeneity might character- ize the relationship between GDP per capita growth and all the explanatory variables. Following the ap- proach of previous studies on beta-convergence using dynamic panel data models (Badinger et al., 2004; Crescenzi et al., 2016; Elhorst, 2010), we employ the GMM estimator in rst differences as was introduced by Arellano and Bond (1991). Therefore, as a robust- ness check, we constructed a xed effects (FE) model with region dummies and time-lagged independent variables (t 1) and a General Methods of Moment (GMM) difference model (Arellano & Bond, 1991; Blundell & Bond, 1998). The rst differenced depen- dent variable is the growth variable (GDPpc). FDI is calculated as: Foreign Firms 0 Turnover Total Turnover at the NUTS 2 aggregation level. 10 EU NUTS 2 data on FDI were constructed from rm-level data aggregated at the regional level using the Bureau van Dijk Amadeus database. 11 After aggregating the rm-level data from Amadeus at the regional level (NUTS 2), the panel dataset contains observations on foreign afliate pres- ence for 252 EU NUTS 2 regions for 13 years during 2005–2017. FIN-DEPTH measures nancial depth and is calculated as: Banking Deposits GDP nominal at the NUTS 2 aggrega- tion level. 12 The vector CONTROLS includes four explanatory variables, which are analysed below. To take account of the level of centrality or peripherality of the EU NUTS 2 regions, we use the gravity index (GRAV). The gravity index is dened as the inverse of the sum of distances among the centroids of each pair of regions weighted by the region’s population. It is measured as: P j i ( PiP j Di j ), where P is the population of region i and all other European regions j, and D is the distances between them (Petrakos et al., 2011). The gravity index takes values greater than (or equal to) 0. The higher the value of the gravity index, the more central the region’s position in the EU space, the better its accessibility, and the greater its market potential. R&D expenditure is measured as public R&D ex- penses per inhabitant (see denition in Table A1), and it has been fully recognized as a key determi- nant of long-run growth (Frenken et al., 2005; Fu, 2008). The capital–labour ratio measures gross xed capital formation divided by labour (see denition in Table A1) and builds on Solow’s growth theory (Solow, 1956, 1957; see also Boschma et al., 2012). Population density measures the number of inhab- itants per square metre (see denition in Table A1) and is included as a standard control variable to ac- count for rapid patterns of urbanization (Fujita & Thisse, 1996). The literature has extensively used the 10 An alternative measure would be the number of foreign rms in the total number of rms. However, this would greatly underestimate FDI presence: the average size of foreign rms (based on turnover or employment) is much higher than the average size of domestic rms. 11 We use Amadeus data which include micro and small enterprises. Operating revenue (turnover) is used to compute the ratio. To derive the FDI variable, we rst distinguish between foreign and domestic rms: we consider a rm to be foreign if its country of residence is different from the country of the global ultimate owner and if the percentage of foreign ownership exceeds 10% of the rm’s total shares. Then we divide foreign-rm turnover by total regional turnover. 12 To construct the variable FIN-DEPTH, we use Bankscope data. We employed unconsolidated Bankscope data for bank customer deposits at the EU city level and aggregated them to the EU NUTS 2 level. The World Bank Global Financial Database has dened several other indicators for measuring nancial depth such as a) Liquid liabilities to GDP (%), b) Central bank assets to GDP (%), c) Stock market capitalization to GDP (%), and d) Deposit money banks’ assets to GDP (%), but Bankscope did not provide sufcient data for these proxies and therefore we could not use them. ECONOMIC AND BUSINESS REVIEW 2023;25:164–181 173 Table 1. Baseline regressions (OLS with region and time dummies). (1) (2) (3) (4) (5) (6) (7) (8) (9) Full Full Full Full Less More Full Less More Developed Developed Developed Developed GDPpc ( 1) 0.222 0.208 0.220 0.221 0.230 0.365 0.220 0.229 0.364 (0.021) (0.020) (0.021) (0.021) (0.027) (0.032) (0.021) (0.027) (0.033) FDI ( 1) 0.038 0.043 0.116 0.213 0.035 0.042 0.060 0.025 (0.017) (0.017) (0.060) (0.059) (0.066) (0.017) (0.027) (0.018) FIN-DEPTH ( 1) 0.028 0.029 0.029 0.070 0.002 0.077 0.042 0.022 (0.009) (0.009) (0.009) (0.013) (0.007) (0.021) (0.050) (0.022) FDI ( 1) # Gravity ( 1) 0.025 0.055 0.020 (0.018) (0.019) (0.019) FIN-DEPTH ( 1) 0.015 0.010 0.006 # Gravity ( 1) (0.005) (0.017) (0.005) Gravity ( 1) 0.351 0.374 0.355 0.353 0.321 1.036 0.348* 0.324 1.012 (0.205) (0.214) (0.205) (0.204) (0.200) (0.268) (0.204) (0.202) (0.271) R&D ( 1) 0.006 0.013 0.010 0.009 0.009 0.008 0.010 0.010 0.007 (0.006) (0.006) (0.006) (0.006) (0.008) (0.009) (0.006) (0.008) (0.009) Capital/Labour ( 1) 0.023 0.021 0.024 0.024 0.017 0.037 0.024 0.016 0.036 (0.008) (0.008) (0.008) (0.008) (0.010) (0.018) (0.008) (0.010) (0.018) Pop. Density ( 1) 0.116 0.186 0.126 0.132 0.038 0.723 0.130 0.027 0.717 (0.195) (0.204) (0.195) (0.194) (0.194) (0.242) (0.194) (0.196) (0.240) Constant 3.012 2.591 2.975 2.941 3.184 3.526 2.933 3.240 3.470 (0.502) (0.507) (0.498) (0.497) (0.587) (0.648) (0.499) (0.587) (0.651) Observations 2,613 2,623 2,552 2,552 1,266 1,286 2,552 1,266 1,286 Number of NUTS 2 252 249 248 248 152 151 248 152 151 Adjusted R-squared 0.479 0.473 0.487 0.487 0.565 0.491 0.488 0.564 0.491 Regions FE YES YES YES YES YES YES YES YES YES Years FE YES YES YES YES YES YES YES YES YES Robust standard errors in parentheses. p< 0:01, p< 0:05, p< 0:1. variables mentioned above to estimate long-run or short-term convergence trends in the EU regions (Bar- rios et al., 2005; Fujita & Thisse, 1996; Ottaviano & Puga, 1998; Petrakos et al., 2011). Finally, the vector INTERACTIONS includes FDI*GRAVITY and FIN-DEPTH*GRAVITY and is aimed at measuring the respective conditional impacts of FDI and FIN-DEPTH on economic growth. 13 4 Empirical results Table 1 presents the estimation results of the base- line xed effects (FE) model with region and time dummies. The rst three columns explore the im- pact of FDI and FIN-DEPTH on regional economic growth. Columns 4 to 6 focus on the role of gravity as a potential moderating factor in the FDI–growth nexus, for the overall sample (column 4), the sample of less developed regions (column 5), and the sample of more advanced regions (column 6). Columns 7 to 9 present the result for the role of gravity as a po- tential moderating factor in the FIN-DEPTH–growth nexus in the overall sample (column 7), the sample of less developed regions (column 8), and the sam- ple of more developed regions (column 9). 14 Based on the interaction between FDI*GRAVITY and FIN- DEPTH*GRAVITY for the different samples, we can estimate the impact of income and spatial hetero- geneity of FDI and FIN-DEPTH. The interactions of FDI*GRAVITY and FIN-DEPTH*GRAVITY provide information on whether the centrality/peripherality of each region magnies the respective impacts of FDI and nancial depth on regional growth. First, all the columns in Table 1 show that the coefcient of GDPCAP t 1 is statistically signicant and negative, indicating that, during 2005–2017, the poorer EU NUTS 2 regions were converging towards 13 All regressions include region- and year-xed effects (FE), and the GMM-DIFF modelling strategy (Arellano & Bond, 1991; Blundell & Bond, 1998) tests for endogeneity and omitted variables bias. Tables A2 and A3 present the summary statistics and correlations. 14 Less (more) advanced regions are dened as regions with lower (higher) average per capita GDP (sample average GDP per capita € 25,473). The Appendix, Table A4, presents the list of less and more developed regions. 174 ECONOMIC AND BUSINESS REVIEW 2023;25:164–181 the wealthier ones (conrming the beta-convergence hypothesis). The value of the coefcient of lagged GDP per capita is in the range 0.20–0.36 log points, meaning that the speed of convergence is weak, that is, poorer EU regions are converging only slowly to- wards the wealthier ones. FDI and FIN-DEPTH have a positive impact on GDP per capita growth (columns 1, 2, and 3); there- fore, regional FDI presence and nancial depth con- tribute positively to regional growth. But how do income heterogeneity and gravity moderate this pos- itive impact? The opportunity to explore the level of the regions’ development as a possible conditional factor in FDI spillovers stems from the fact that our analysis includes a very diverse sample of regions in terms of per capita GDP . Also, the geography of the EU shows the presence of very heterogeneous re- gions: for example, large centrally located markets (in terms of potential measured by population) and small markets (scarcely populated) in more peripheral areas. In the regression that includes only less developed regions in the EU (column 5), the impact of FDI is positive, but is moderated negatively by gravity: this means that more peripheral and less wealthy re- gions benet relatively more from higher levels of FDI (negative coefcient of the interaction). How- ever, we can appreciate that it is not the case for the sample of more developed regions (column 6). Col- umn 7 also shows that the impact of nancial depth is positive for the full sample but again is moder- ated negatively by gravity, 15 meaning that peripheral regions—regardless of the level of development— benet relatively more from higher levels of FIN- DEPTH. In other words, peripherality in the EU is not a barrier to the benets deriving from foreign invest- ment and nancial markets alike. This is an important result in the literature on the role of nancial develop- ment in Europe from a regional perspective. We have run a robustness check for the possibil- ity that both FDI and FIN-DEPTH are affected by endogeneity. We have employed a GMM difference one-step robust model 16 that suits regional level data (see Crescenzi & Rodríguez-Pose, 2012). The GMM- DIFF regression results are presented in Table 2. The results of the baseline model are mostly conrmed, with the exception of the lack of signicance of the FIN-DEPTH variable and its interaction with gravity (column 7). We interpret this as further conrma- tion that FDI, compared to internal nance, boosts regional economic growth more and might be chan- nelled through the same positive spillovers. 17 For each regression we report the p-value of the Sargan test, the Hansen test, and the cross-sectional depen- dence test developed by Friedman (De Hoyos & Saradis, 2006) to rule out endogeneity and cross- sectional dependence, respectively. 18 Our empirical ndings indicate that, in less wealthy (and more peripheral) regions, compared to wealthy regions, FDI productivity spillovers and knowledge externalities, arising from proximity to foreign in- vestors, are more signicant. In other words, in less wealthy regions, the imitation effect prevails over the competition effect, and perhaps the larger the technological gap between local and foreign rms, the more local rms can learn to increase their pro- ductivity (Meyer & Sinani, 2009). The coefcient of FIN-DEPTH, which estimates the effect of local nan- cial depth on regional growth, is generally positive and statistically signicant, indicating that better- developed regional nancial markets boost economic performance. This nding is important as it illustrates that even though capital markets are currently glob- alized, nancial intermediation at the local level and access to local capital still matters, especially for less advanced regions (Cavallaro & Villani, 2022). Concerning the control variables, the region’s po- sition on the EU map and its distance from markets (expressed by the gravity index) seem to have a neg- ative direct effect on growth due, perhaps, to the fact that peripheral regions grew more than centrally lo- cated regions in the time period considered (Bruno & Cipollina, 2018). 19 Furthermore, a non-linear effect drives the negative result for R&D expenditure. 20 The capital–labour ratio, used to proxy for capital inten- sity, seems to be another indicator of convergence: the lower the initial level of the capital-to-labour ratio, the higher the growth potential (Boschma et al., 2012; Frenken et al., 2005). Finally, increased population density—linked indirectly to economies of scale— seems also to contribute to regional growth (Fujita & Thisse, 1996; Ottaviano & Puga, 1998). To sum up, the above ndings suggest that FDI and FIN-DEPTH are important determinants of regional 15 The lack of signicance of FIN-DEPTH (results in columns 8 and 9) is driven also by limited statistical power; the last two regressions include only half the total number of observations. 16 We have adopted “orthogonal” transformation of the missing value to achieve more efcient use of statistical power. 17 The (very marginally) signicant result for the interaction between FIN-DEPTH and gravity (column 8) does not change the overall pattern of results and, again, may be driven by the reduced number of observations. 18 We would like to thank one referee for this suggestion. The use of the Friedman test is appropriate in this context due to the relatively unbalanced nature of the panel, where the Peasaran and Frees tests are less powerful (De Hoyos & Saradis, 2006). 19 This is depicted in the Appendix, Fig. A1. 20 The table is available upon request. ECONOMIC AND BUSINESS REVIEW 2023;25:164–181 175 Table 2. GMM-DIFF (one-step robust orthogonal). (1) (2) (3) (4) (5) (6) (7) (8) (9) Full Full Full Full Less More Full Less More Developed Developed Developed Developed GDPpc ( 1) 0.308 0.321 0.315 0.314 0.337 0.407 0.316 0.344 0.405 (0.031) (0.035) (0.034) (0.034) (0.039) (0.054) (0.034) (0.038) (0.055) FDI ( 1) 0.074 0.109 0.299 0.245 0.097 0.108 0.051 0.007 (0.035) (0.037) (0.125) (0.115) (0.153) (0.035) (0.038) (0.050) FIN-DEPTH ( 1) 0.110 0.113 0.111 0.114 0.005 0.122 0.023 0.017 (0.025) (0.026) (0.027) (0.019) (0.013) (0.095) (0.046) (0.037) FDI ( 1) # Gravity ( 1) 0.065 0.070 0.030 (0.039) (0.038) (0.040) FIN-DEPTH ( 1) 0.003 0.035 0.003 # Gravity ( 1) (0.035) (0.021) (0.009) Gravity ( 1) 2.818 3.232 2.937 2.899 2.863 1.880 2.925 2.915 1.883 (0.560) (0.659) (0.624) (0.633) (0.659) (0.675) (0.592) (0.637) (0.677) R&D ( 1) 0.017 0.034 0.026 0.027 0.023 0.013 0.026 0.021 0.013 (0.011) (0.012) (0.011) (0.011) (0.010) (0.015) (0.010) (0.010) (0.015) Capital/Labour ( 1) 0.004 0.018 0.008 0.008 0.010 0.002 0.008 0.014 0.003 (0.014) (0.016) (0.015) (0.015) (0.016) (0.021) (0.015) (0.017) (0.021) Pop. Density ( 1) 2.624 3.059 2.748 2.726 2.592 1.764 2.738 2.646 1.753 (0.571) (0.664) (0.630) (0.637) (0.700) (0.684) (0.607) (0.672) (0.681) Observations 2,013 2,033 1,971 1,971 989 982 1,971 989 982 Number of NUTS 2 245 243 242 242 145 142 242 145 142 ar1p 0 0 0 0 4.34 10 9 1.04e-09 0 5.52e-09 9.51e-10 ar2p 0.000126 0.000498 0.000715 0.00100 1.68e-05 0.00988 0.000715 1.14e-05 0.00779 Sargan p-v 0 0 0 0 0 0 0 0 0 Hansen p-v 0.979 0.953 0.975 0.980 1.000 1.000 0.973 1.000 1.000 CSD(Friedman) p-value 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 N 2013 2033 1971 1971 989 982 1971 989 982 j 300 287 296 296 266 289 296 266 289 Robust standard errors in parentheses. p< 0:01, p< 0:05, p< 0:1. FDI and Finance considered as endogenous, IV second lag. All other variables considered as not-strictly exogenous and instrumented with internal IV (from rst lag) and external instruments. growth, especially for less developed and peripheral regions. The greater growth impact of FDI and FIN- DEPTH on regions with low and modest levels of development can be attributed to the fact that less developed EU regions, having less capital in general, and less foreign capital in particular, have more mar- gin to learn new things and absorb new technologies from foreign investors, whereas advanced regions have already established their production patterns and might not have much to gain from a foreign investor. For instance, for less wealthy regions, a for- eign investor could contribute more added value to their existing economic structure than in economi- cally and technologically mature economies. 5 Discussion and policy implications This paper conrms the existence of a relevant income and spatial heterogeneity when estimating the impact of foreign direct investment and nan- cial depth on growth at the EU regional level. The main value added of the paper is the exploitation of a purpose-built detailed regional dataset, which has enabled us to estimate FDI spillovers at the re- gional level, thereby lling a gap, given that the literature has mostly focused on identifying the FDI and nancial-depth impact at the national, sectoral, or rm level (Bruno & Cipollina, 2018; Nicolini & Resmini, 2010). The paper has shed light on foreign afliates’ role in determining regional economies’ growth trajectories in an EU context and has explored whether FDI spillovers occur in a homogenous or het- erogeneous way across geographical territories. Our analysis shows that those regions that seem to absorb the most externalities stemming from the presence of foreign afliates are the less wealthy (and often peripheral) EU regions. This nding could challenge the argument that spatial concentration of FDI could exacerbate regional disparities or cause crowding-out effects (Lee et al., 2022). Furthermore, the heterogene- ity identied challenges relating to the idea of simple bright or dark sides to the effects of FDI and high- lights the essentially empirical nature of the issue and the necessity for contextualized policy interven- tions (Phelps et al., 2018). Therefore, in the context of FDI spillovers, the intra-national heterogeneity of 176 ECONOMIC AND BUSINESS REVIEW 2023;25:164–181 EU regions and the core–periphery patterns seem to benet the less advanced EU areas disproportion- ately more, implying that these regions have more to gain from foreign investment. This result is in line with previous research on the spatial heterogene- ity of within-country or cross-country FDI spillovers (Crespo et al., 2009; Monastiriotis, 2016; Monastiri- otis & Jordaan, 2010; Xu & Sheng, 2012); however, it provides a unique insight into the intra-national heterogeneity of FDI spillovers in the EU regions. Policy-wise, this evidence sheds new light on the design of regional development policy by conrm- ing that sub-national Investment Promotion Agencies (IPAs) have indeed an important task to full, by potentially prioritizing less advanced and peripheral regions in terms of FDI attraction, where informa- tion asymmetries and institutional weaknesses can be stronger (Crescenzi et al., 2021). Another key contribution of the paper is the ex- ploration of nancial development as an important determinant of regional growth and the potential heterogeneity impact across EU regions. Regional - nance is shown to positively affect regional growth; in particular, peripheral EU regions seem to benet more from a developed regional nancial system. These re- gions generally include mostly small rms (or SMEs) that do not have easy access to global/national capital markets and have to rely more on the local banking system to access nance. In other words, the paper offers a unique insight to the territorial dimension of nancial development as a growth determinant, despite the EU’s continuous nancial integration and globalization, and highlights that access to nance also matters for regional economies. This calls for considering that among the physical and commercial infrastructures that are known to be determinants of both the location of FDI and its spillovers, much more information is required on the role of nancial in- stitutional development above and beyond physical and commercial infrastructures as a potential future, explicit policy focus. In other words, while the pos- sibilities for regional-level institutions to affect the embeddedness of and spillovers from MNEs have been explored (Crescenzi et al., 2021; Phelps et al., 2003; Young et al., 1994), regional nancial-sector de- velopment policies have been overlooked so far. For instance, the institutional and monitoring role of re- gional nancial institutions should be enhanced, as part of sub-national or national development pol- icy, sustaining their capacity to act as “lubricants” of the local economy. To go one step further, hav- ing seen the importance of regional nance for the peripheral regions of the EU, one could argue that “regional nancialization” (on the institutional and capacity-building side) could progressively become a pillar/objective of governments’ development policy through potential government subsidies. In parallel, current EU nancial instruments such as guaran- tees, subsidized loans, and equity nancing, targeting mostly SMEs, could be further reinforced in order to address market failures and information asymmetries in the EU periphery. Lastly, one could argue that a spatial monetary policy could be deployed (via cen- tral banks) to ensure easier access to nance at the regional level. Consequently, the paper re-captures the essential role of regional development policy, both in helping less developed regions to attract foreign direct investment and addressing market distortions arising from lack of access to nance. We would like to conclude with some suggestions for future research. Future scholars could ground some of the aggregate regional-level ndings in this paper in rm-level quantitative and qualitative anal- yses, drawing on insights from the nancial, FDI policy, MNE, and SME communities. More detailed and context-specic knowledge about how rms ac- cess nance, from which nancial institutions, via which business - service intermediaries, and with what effect would help to inform policy interven- tions designed to promote regional growth. More generally, any “new” industrial or regional strate- gies in Europe should be framed as both vertically and horizontally integrated, place-sensitive develop- ment policies. In other words, “a coherent industrial strategy at various levels of governance, whether regional and/or national” (Crescenzi & Iammarino, 2017; Iammarino, 2018) should tackle individual and social isolation across geographical space (the pe- riphery is still the underdog). Identication of the subnational dimensions to these structural transfor- mations has been advanced signicantly by academic research at the intersection of international business studies and economic geography. However, a rethink- ing of regional development from this perspective continues to present challenges in terms of policy de- sign, and beyond a sole focus on maximization of FDI spillovers and over-reliance on FDI strategies that dis- regard the level of MNEs’ embeddedness (Zoltán & Gábor, 2022). Place-sensitive industrial connectivity policies require territorial differentiation within both the core and peripheral regions, and across and within regions in the same country. Acknowledgments This research received funding from the European Union Horizon 2020 Research and Innovation Ac- tion (Grant Agreement no. 822781-GROWINPRO). We thank Alessandro De Trane for excellent research assistance. ECONOMIC AND BUSINESS REVIEW 2023;25:164–181 177 References Alfaro, L., Chanda, A., Kalemli-Ozcan, S., & Sayek, S. (2004). FDI and economic growth: The role of local nancial markets. Jour- nal of International Economics, 64(1), 89–112. Arellano, M., & Bond, S. (1991). Some tests of specication for panel data: Monte Carlo evidence and an application to employment equations. The Review of Economic Studies, 58(2), 277–297. Audretsch, B. (1998). Agglomeration and the location of innovative activity. Oxford Review of Economic Policy, 14(2), 18–29. Badinger, H., Müller, W., & Tondl, G. (2004). Regional convergence in the European Union, 1985–1999: A spatial dynamic panel analysis. Regional Studies, 38(3), 241–253. Barrios, S., Görg, H., & Strobl, E. (2005). Foreign direct investment, competition and industrial development in the host country. European Economic Review, 49(7), 1761–1784. Barro, R. J., & Sala-i-Martin, X. (1992). Convergence. Journal of Po- litical Economy, 100(2), 223–251. Blomström, M., & Kokko, A. (1998). Multinational corporations and spillovers. Journal of Economic Surveys, 12(3), 247–277. Blundell, R., & Bond, S. (1998). Initial conditions and moment re- strictions in dynamic panel data models. Journal of Econometrics, 87(1), 115–143. Borensztein, E., De Gregorio, J., & Lee, J. W. (1998). How does foreign direct investment affect economic growth? Journal of International Economics, 45(1), 115–135. Boschma, R., Minondo, A., & Navarro, M. (2012). Related variety and regional growth in Spain. Papers in Regional Science, 91(2), 241–256. Bournakis, I. (2021). Spillovers and productivity: Revisiting the puzzle with EU rm level data. Economics Letters, 201, Article 109804. Bruno, R. L., & Cipollina, M. (2018). A meta-analysis of the indirect impact of foreign direct investment in old and new EU member states: Understanding productivity spillovers. The World Econ- omy, 41(5), 1342–1377. Bryson, J., & Taylor, M. (2006). The functioning economic geography of the West Midlands. West Midlands Regional Observatory. Casi, L., & Resmini, L. (2017). Foreign direct investment and growth: Can different regional identities shape the returns to foreign capital investments? Environment and Planning C: Politics and Space, 35(8), 1483–1508. Cavallaro, E., & Villani, I. (2022). Beyond nancial deepening: Re- thinking the nance–growth relationship in an uneven world. Economic Modelling, 116, Article 106009. Crescenzi, R., Di Cataldo, M., & Giua, M., (2021). FDI inows in Eu- rope: Does investment promotion work? Journal of International Economics, 132, Article 103497. Crescenzi, R., Di Cataldo, M., & Rodríguez-Pose, A. (2016). Govern- ment quality and the economic returns of transport infrastruc- ture investment in European regions. Journal of Regional Science, 56(4), 555–582. Crescenzi, R., & Iammarino, S. (2017). Global investments and regional development trajectories: The missing links. Regional Studies, 51(1), 97–115. Crescenzi, R., & Rodríguez-Pose, A. (2012). Infrastructure and re- gional growth in the European Union. Papers in Regional Science, 91(3), 487–513. Crespo, N., Fontoura, M. P ., & Proença, I. (2009). FDI spillovers at regional level: Evidence from Portugal. Papers in Regional Science, 88(3), 591–607. ˇ Cihák, M., Demirgüç-Kunt, A., Feyen, E., & Levine, R. (2012). Benchmarking nancial systems around the world . (Policy Research Working Paper 6175). World Bank Group. De Hoyos, R. E., & Saradis, V . (2006). Testing for cross-sectional dependence in panel-data models. The Stata Journal, 6(4), 482– 496. Demirgüç-Kunt, A., & Maksimovic, V . (1998). Law, nance, and rm growth. The Journal of Finance, 53(6), 2107–2137. D’Este, P ., & Patel, P . (2007). University–industry linkages in the UK: What are the factors underlying the variety of interactions with industry? Research Policy, 36(9), 1295–1313. Drifeld, N., (2006). On the search for spillovers from foreign di- rect investment (FDI) with spatial dependency. Regional Studies, 40(1), 107–119. Durusu-Ciftci, D., Ispir, M. S., & Yetkiner, H. (2017). Financial devel- opment and economic growth: Some theory and more evidence. Journal of Policy Modeling, 39(2), 290–306. Elhorst, J. P . (2010). Applied spatial econometrics: Raising the bar. Spatial Economic Analysis, 5(1), 9–28. Fosfuri, A., Motta, M., & Rønde, T. (2001). Foreign direct investment and spillovers through workers’ mobility. Journal of International Economics, 53(1), 205–222. French, S., Leyshon, A., & Wainwright, T. (2011). Financializing space, spacing nancialization. Progress in Human Geography, 35(6), 798–819. Frenken, K., van Oort, F. G., Verburg, T., & Boschma, R. A. (2005). Variety and regional economic growth in the Netherlands. SSRN. Fu, X. (2008). Foreign direct investment, absorptive capacity and regional innovation capabilities: Evidence from China. Oxford Development Studies, 36(1), 89–110. Fujita, M., & Thisse, J. F. (1996). Economics of agglomeration. Jour- nal of the Japanese and International Economies, 10(4), 339–378. Girma, S., & Wakelin, K. (2001). Regional underdevelopment: Is FDI the solution? A semiparametric analysis (CEPR Discussion Paper 2995). Centre for Economic Policy Research. Glass, A. J., & Saggi, K. (2002). Multinational rms and technology transfer. The Scandinavian Journal of Economics, 104(4), 495–513. Hasan, I., Koetter, M., & Wedow, M. (2009). Regional growth and nance in Europe: Is there a quality effect of bank efciency? Journal of Banking & Finance, 33(8), 1446–1453. Hu, M., Zhang, J., & Chao, C. (2019). Regional nancial efciency and its non-linear effects on economic growth in China. Interna- tional Review of Economics & Finance, 59, 193–206. Iammarino, S. (2018). FDI and regional development policy. Journal of International Business Policy, 1(3–4), 157–183. Iammarino, S., & McCann, P . (2013). Multinationals and economic geography: Location, technology and innovation. Edward Elgar Publishing. Javorcik, B. S., & Spatareanu, M. (2011). Does it matter where you come from? Vertical spillovers from foreign direct investment and the origin of investors. Journal of Development Economics, 96(1), 126–138. Keller, W. (2002). Geographic localization of international technol- ogy diffusion. American Economic Review, 92(1), 120–142. King, R. G., & Levine, R. (1993a). Finance and growth: Schumpeter might be right. The Quarterly Journal of Economics, 108(3), 717– 737. King, R. G., & Levine, R. (1993b). Finance, entrepreneurship and growth. Journal of Monetary Economics, 32(3), 513–542. Krisztin, T., & Piribauer, P . (2023). A joint spatial econometric model for regional FDI and output growth. Papers in Regional Science, 102(1), 87–106. Law, S. H., & Singh, N., (2014). Does too much nance harm eco- nomic growth? Journal of Banking & Finance, 41, 36–44. Lee, C. C., Lee, C. C., & Cheng, C. Y., (2022). The impact of FDI on income inequality: Evidence from the perspective of nancial development. International Journal of Finance & Economics, 27(1), 137–157. Lucchetti, R., Papi, L., & Zazzaro, A. (2001). Banks’ inefciency and economic growth: A micro–macro approach. Scottish Journal of Political Economy, 48(4), 400–424. Mariotti, S., Mutinelli, M., Nicolini, M., & Piscitello, L. (2015). Productivity spillovers from foreign multinational enterprises to domestic manufacturing rms: To what extent does spatial proximity matter? Regional Studies, 49(10), 1639–1653. Markusen, J. R., & Venables, A. J. (1999). Foreign direct investment as a catalyst for industrial development. European Economic Re- view, 43(2), 335–356. Meyer, K. E., & Sinani, E. (2009). When and where does foreign direct investment generate positive spillovers? A meta-analysis. Journal of International Business Studies, 40(7), 1075–1094. Mihaljek, D. (2010). The spread of the nancial crisis to central and eastern Europe: Evidence from the BIS data. In R. Matousek 178 ECONOMIC AND BUSINESS REVIEW 2023;25:164–181 (Ed.), Money, banking and nancial markets in central and eastern Europe (pp. 5–31). Palgrave Macmillan. Monastiriotis, V . (2016). Institutional proximity and the size and geography of foreign direct investment spillovers: Do European rms generate more favourable productivity spillovers in the European Union neighbourhood? Environment and Planning C: Government and Policy, 34(4), 676–697. Monastiriotis, V ., & Jordaan, J. A. (2010). Does FDI promote regional development? Evidence from local and regional productivity spillovers in Greece. Eastern Journal of European Studies, 1(2), 174–192. Monfort, P ., 2008. Convergence of EU regions: Measures and evolution (European Union Regional Policy Working Paper 01/2008). Eu- ropean Commission. Nicolini, M., & Resmini, L. (2010). FDI spillovers in new EU mem- ber states. Economics of Transition, 18(3), 487–511. Ogbuabor, J. E., Orji, A., Eigbiremolen, G. O., Manasseh, C. O., & Onuigbo, F. N. (2020). The role of institutions in the FDI–Growth relationship in a developing economy: A new evidence from Nigeria. Studia commercialia Bratislavensia, 13(4), 348–363. Orji, A., Nwagu, G. U, Ogbuabor, J. E, Nwosu, E. O., & Anthony- Orji, O. I. (2021). Foreign direct investment and growth nexus: Further evidence from Africa’s largest economy. Journal of In- frastructure Development, 13(1), 65–78. Orji, A., Ogbuabor, J. E, Okolomike, F. C., & Anthony-Orji, O. (2022). Capital inows, nancial development and growth in ECOWAS countries: A new empirical insight. Applied Economet- rics and International Development, 22(1), 61–80. Orlic, E., Hashi, I., & Hisarciklilar, M. (2018). Cross sectoral FDI spillovers and their impact on manufacturing productivity. In- ternational Business Review, 27(4), 777–796. Ottaviano, G. I. P ., & Puga, D. (1998). Agglomeration in the global economy: A survey of the “new economic geography,” The World Economy, 21(6), 707–731. Ottaviano, G., & Thisse, J. F. (2004). Agglomeration and economic geography. In J. V . Henderson & J. Thisse (Eds.), Handbook of regional and urban economics (Vol. 4, pp. 2563–2608). Elsevier. Padilla-Pérez, R. (2008). A regional approach to study technology transfer through foreign direct investment: The electronics in- dustry in two Mexican regions. Research Policy, 37(5), 849–860. Palacín-Sánchez, M. J., & Di Pietro, F. (2016). The role of the regional nancial sector in the capital structure of small and medium- sized enterprises (SMEs). Regional Studies, 50(7), 1232–1247. Pavlínek, P . (2004). Regional development implications of foreign direct investment in Central Europe. European Urban and Re- gional Studies, 11(1), 47–70. Pavlínek, P . (2015). Foreign direct investment and the development of the automotive industry in Central and Eastern Europe. In B. Galgóczi, J. Drahokoupil & M. Bernaciak (Eds.), Foreign invest- ment in Eastern and Southern Europe (pp. 209–256). ETUI. Petrakos, G. C. (1996). The regional dimension of transition in Central and East European countries: An assessment. Eastern European Economics, 34(5), 5–38. Petrakos, G., Kallioras, D., & Anagnostou, A. (2011). Regional con- vergence and growth in Europe: Understanding patterns and determinants. European Urban and Regional Studies, 18(4), 375– 391. Phelps, N. A., Atienza, M., & Arias, M. (2018). An invitation to the dark side of economic geography, Environment and Planning A: Economy and Space, 50(1), 236–244. Phelps, N. A., & Fuller, C. (2001). Taking care of business: Aftercare and the state–multinational enterprise nexus in Wales. Environ- ment and Planning C: Government and Policy, 19(6), 817–832. Phelps, N. A., Mackinnon, D., Stone, I., & Braidford, P . (2003). Em- bedding the multinationals? Institutions and the development of overseas manufacturing afliates in Wales and North East England. Regional Studies, 37(1), 27–40. Pike, A., & Pollard, J. (2010). Economic geographies of nancializa- tion. Economic Geography, 86(1), 29–51. Rajan, R. G., & Zingales, L. (1998). Financial depth and growth. American Economic Review, 17(4), 467–482. Rajan, R. G., & Zingales, L. (2001). Financial systems, industrial structure, and growth. Oxford Review of Economic Policy, 17(4), 467–482. Resmini, L. (2019). Foreign direct investments, global value chains and regional development. In R. Capello & P . Nijkamp (Eds.), Handbook of regional growth and development theories. (2nd ed., pp. 218–238). Edward Elgar. Rodriguez-Clare, A. (1996). Multinationals, linkages, and economic development. The American Economic Review, 86(4), 852–873. Romer, P . M. (1986). Increasing returns and long-run growth. Jour- nal of Political Economy, 94(5), 1002–1037. Romer, P . M. (1994). The origins of endogenous growth. The Journal of Economic Perspectives, 8(1), 3–22. Sala-i-Martin, X. X., (1996). Regional cohesion: Evidence and the- ories of regional growth and convergence. European Economic Review, 40(6), 1325–1352. Serwicka, I. E., Jones, J., & Wren, C. (2022). Economic integration and FDI location: Is there a border effect within the enlarged EU? The Annals of Regional Science. Advance online publication. Sokol, M. (2017). Financialisation, nancial chains and uneven geo- graphical development: Towards a research agenda. Research in International Business and Finance, 39, 678–685. Sokol, M., & Pataccini, L. (2022). Financialisation, regional eco- nomic development and the coronavirus crisis: A time for spatial monetary policy? Cambridge Journal of Regions, Economy and Society, 15(1), 75–92. Solow, R. (1956). A contribution to the theory of economic growth. Quarterly Journal of Economics, 70(1), 65–94. Solow, R. (1957). Technical change and aggregate production func- tion. The Review of Economics and Statistics, 39(3), 312–320. Stojˇ ci´ c, N., & Orli´ c, E. (2020). Spatial dependence, foreign invest- ment and productivity spillovers in new EU member states. Regional Studies, 54(8), 1057–1068. Xu, X., & Sheng, Y. (2012). Are FDI spillovers regional? Firm-level evidence from China. Journal of Asian Economics, 23(3), 244–258. Young, S., Hood, N., & Peters, E. (1994). Multinational enterprises and regional economic development. Regional Studies, 28(7), 657–677. Zanfei, A. (2012). Effects, not externalities. The European Journal of Development Research, 24(1), 8–14. Zoltán, G., & Gábor, L., 2022. FDI-based regional development in Central and Eastern Europe: A review and an agenda. Tér és Társadalom, 36(3), 68–98. ECONOMIC AND BUSINESS REVIEW 2023;25:164–181 179 Appendix Table A1. Description of variables. Name Indicator Type Denition Source Reference Regional GDP growth rate % Regional Economic Development Dependent GDPCAPt GDPCAPt 1 GDPCAPt 1 NUTS 2 Eurostat Regional statistics by NUTS classication World Bank Initial GDP/capita Initial level of economic growth Independent GDPCAP t 1 (NUTS 2) Eurostat Regional statistics by NUTS classication Barro and Sala-i-Martin (1992); Solow (1956) Foreign afliates’ presence (NUTS 2) FDI Independent Foreign Firms 0 Turnover Total Turnover NUTS 2 Amadeus Meyer and Sinani (2009); Monastiriotis and Jordaan (2010) Financial Depth Financial depth Independent Banking Deposits GDP nominal NUTS 2 Bankscope & Eurostat ˇ Cihák et al. (2012) Gravity Centrality & Accessibility Independent P j i ( pipj di j )NUTS 2 Eurostat; GISCO Petrakos (1996) Population Density Agglomeration economies Independent Inhabitants per square meter (NUTS 2) Eurostat Regional statistics by NUTS classication Petrakos (1996) Capital Labour Ratio Capital Intensity Independent Gross Fixed Capital Formation Employment NUTS 2 Eurostat Regional statistics by NUTS classication Solow (1956) Table A2. Summary statistics. Variable Observations Mean Std. Dev Min Max GDP per capita growth 2,552 0.021046 0.06111 0.25256 0.292501 GDP per capita 2,552 25473.06 11865.36 2400 65400 Foreign Presence 2,552 0.237451 0.160246 0 1 Financial Depth 2,552 0.342477 0.585716 4.76E 5 12.02749 “Gravity” 2,552 26.36502 27.59608 1.644867 249.7496 Research and Development 2,552 468.9388 522.3606 0 3884.3 Capital Labour Ratio 2,552 147082.9 91613.28 13601.4 964864.9 Population Density 2,552 397.529 937.6592 3.3 11357.1 Table A3. Correlation table (SE in parenthesis, *** 1% signicance). GDP per capita growth GDP per capita Foreign Presence Financial Depth “Gravity” Research and Development Capital Labour Ratio GDP per capita 0.0491*** (0.0064) 1 Foreign Presence 0.0873*** (0.000) 0.1218*** (0.000) 1 Financial Depth 0.0011 (0.9511) 0.2647*** (0.000) 0.0604*** (0.0007) 1 “Gravity” 0.0072 (0.6901) 0.2647*** (0.000) 0.1574*** (0.000) 0.0834*** (0.000) 1 Research and Development 0.0054 (0.7737) 0.7055*** (0.000) 0.1076*** (0.000) 0.2113*** (0.000) 0.1819*** (0.000) 1 Capital Labour Ratio 0.0693*** (0.0003) 0.4999*** (0.000) 0.0124 (0.4962) 0.1243*** (0.000) 0.1983*** (0.000) 0.3264*** (0.000) 1 Population Density 0.0061 0.736 0.0539*** (0.002) -0.0265 (0.1309) -0.0222 (0.2057) 0.4081*** (0.000) 0.1277*** (0.000) 0.0877*** (0.000) 180 ECONOMIC AND BUSINESS REVIEW 2023;25:164–181 Table A4. Less developed and developed regions. Regions with lower-than-average GDP per capita Regions with higher-than- average GDP per capita Abruzzo Alsace Alentejo Aquitaine Algarve Arnsberg Anatoliki Makedonia Thraki Auvergne Andalucía Basse-Normandie Aragón Bedfordshire Attiki Buckinghamshire Basilicata Berlin Brandenburg Bourgogne Bucuresti Bratislavský kraj Budapest Braunschweig Burgenland Bremen Calabria Bretagne Campania Cataluña Cantabria Centre Castilla Leon Champagne-Ardenne Castilla-la Mancha Cheshire Centro (PT) Comunidad Foral de Navarra Centru Comunidad de Madrid Chemnitz Corse Ciudad Autónoma de Ceuta (ES) Cumbria Ciudad Autónoma de Melilla (ES) Darmstadt Comunidad Valenciana Derbyshire Cornwall Detmold Devon Dorset Dolnoslaskie Drenthe Dresden Düsseldorf Dytiki Ellada East Anglia Dytiki Makedonia East Wales Dél-Alföld Yorkshire Dél-Dunántúl Eastern Scotland Eesti Eastern and Midland Extremadura Emilia-Romagna Franche-Comté Essex Galicia Etelä-Suomi Illes Baleares Flevoland Ionia Nisia Freiburg Ipeiros Friesland Jihovýchod Friuli-Venezia Giulia Jihozápad Gelderland Kentriki Makedonia Gießen Kriti Wiltshire Kujawsko-Pomorskie Greater Manchester Közép-Dunántúl Groningen La Rioja Hamburg Languedoc-Roussillon Hampshire Latvija Hannover Limousin Haute-Normandie Lincolnshire Helsinki-Uusimaa Lorraine Worcestershire Lubelskie Highlands Lubuskie Inner London East Lódzkie Inner London West Lüneburg Karlsruhe Malopolskie Kassel Malta Kent Mecklenburg-Vorpommern Koblenz Molise Kärnten Moravskoslezsko Köln Nord-Est Lancashire Nord-Vest Lazio Table A4. continued. Regions with lower-than-average GDP per capita Regions with higher-than- average GDP per capita Norte Rutland Notio Aigaio Leipzig Nyugat-Dunántúl Liguria Opolskie Limburg Outer London Lombardia Peloponnisos Luxembourg Pest Länsi-Suomi Picardie Marche Podkarpackie Mellersta Podlaskie Merseyside Pomorskie Midi-Pyrénées Principado de Asturias Mittelfranken Prov. Hainaut Münster Prov. Liège Niederbayern Prov. Luxembourg Niederösterreich Prov. Namur Noord-Brabant Puglia Noord-Holland Região Autónoma da Madeira (PT) Nord-Pas-de-Calais Região Autónoma dos Açores (PT) Norra Región de Murcia North Eastern Scotland Sachsen-Anhalt North Yorkshire Sardegna Northern Ireland Severen Northern and Western Severoiztochen Northumberland Severovýchod Oberbayern Severozapaden Oberfranken Severozápad Oberpfalz Shropshire Oberösterreich Sicilia Outer London South Slaskie Outer London West Sostines regionas Overijssel South Yorkshire Pays-de-la-Loire Southern Scotland País Vasco Sterea Ellada Piemonte Stredné Slovensko Pohjois- ja Itä-Suomi Strední ˇ Cechy Poitou-Charentes Strední Morava Praha Sud Prov. Antwerpen Sud-Est Prov. Brabant wallon Sud-Vest Prov. Limburg (BE) Swietokrzyskie Prov. Oost-Vlaanderen Tees Prov. Vlaams-Brabant Thessalia Prov. West-Vlaanderen Thüringen Provence-Alpes-Côte d’Azur Umbria Provincia Autonoma di Bolzano/Bozen Vest Provincia Autonoma di Trento Vidurio Rheinhessen-Pfalz Voreio Aigaio Rhône-Alpes Vzhodna Slovenija Région de Bruxelles-Capitale Východné Slovensko Saarland Warminsko-Mazurskie Salzburg Warszawski stołeczny Schleswig-Holstein West Wales and the Valleys Schwaben Wielkopolskie Småland Yugoiztochen Steiermark Yugozapaden Stockholm Yuzhen Stuttgart (continued on next page) ECONOMIC AND BUSINESS REVIEW 2023;25:164–181 181 Table A4. continued. Regions with lower-than-average GDP per capita Regions with higher-than- average GDP per capita Zachodniopomorskie Surrey, East and West Sussex Zahodna Slovenija Sydsverige Západné Slovensko Tirol Área Metropolitana de Lisboa Toscana Észak-Alföld Trier Észak-Magyarország Tübingen Unterfranken Utrecht Valle d’Aosta/Vallée Veneto Vorarlberg Västsverige Weser-Ems West Central Scotland West Midlands West Yorkshire Wien Zeeland Zuid-Holland Åland Île de France Östra Mellansverige Övre Norrland Source: Authors’ calculation. Fig. A1. Map of average GDP per capita yearly growth.