Balázs Páger (p), Zoltán Gál

Bennour Mohamed Hsin

This study aims to clarify the relationship between the presence of foreign-owned companies and the entry rates of new firms (i.e., firm births) in Hungarian sub-regional districts, situating our findings within the existing literature. Previous studies have explored the significance of foreign-owned companies in the Hungarian economy, including their role in introducing new knowledge and enhancing regional research and development (R&D), their impact on differing growth trends among Hungarian counties, their contribution to entrepreneurial activities at the national level, and their influence on the industrial structure of Hungarian districts.
The literature on the relationship between foreign-owned firms and entrepreneurship presents conflicting results. Some studies have highlighted the positive effects of foreign-owned companies on domestic firm entry, such as increasing demand in local resource markets, raising the technological level of the host country or region, enhancing the productivity of domestic firms, and contributing to overall economic performance. Conversely, other empirical findings suggest a more mixed relationship, indicating that foreign-owned firms can displace local firms in the market or attract scarce local resources, such as qualified workers, away from domestic firms.
Our study's primary assumption is that foreign-owned companies may facilitate the entry of new firms, thereby boosting the local economy in these districts. Our analysis utilizes a panel dataset spanning from 2010 to 2019, comprising data from the Hungarian Central Statistical Office and the Databank of the HUN-REN Centre for Economic and Regional Studies. We use the firm entry rate in districts (number of entering companies related to the number of operating companies), and the change in employment within foreign-owned companies is our model’s primary explanatory variable. Additionally, we include several control variables to describe the socioeconomic conditions of the districts, ensuring a comprehensive analysis.
Initial results suggest a positive relationship between the entry of new firms and the presence of foreign-owned companies; however, this positive effect appears to manifest with a delay of a few years.

Gabor Lux (p)

Zoltan Gal

This paper discusses ongoing reindustrialisation processes in Hungary as an example of the evolving geographic complexity of industrial development in Central and Eastern Europe (CEE). Foreign Direct Investment (FDI) has played a key role in reintegrating CEE regions into global value chains as part of an “integrated periphery”, but this growth model has also come with trade-offs, and shown exhaustion recently in the form of an emerging regional development trap. Simultaneously, the spatial patterns of industry have become more complex as deepening integration, industrial upgrading, and geographic expansion have taken place in different regional contexts. This paper introduces the concept of supplemental reindustrialisation to explain this shift from the emergence of integrated peripheries in CEE to an era where the FDI-driven growth model undergoes internal transformation while facing increasing growth constraints resulting from the exhaustion or “inversion” of previous competitive advantages. The paper thus explores the regional development contradictions of the FDI-based development model, highlighting the emergence of the regional development trap, and the exhaustion of the current growth path. These phenomena are then examined in the context of ongoing reindustrialisation in Hungary, using county-level data and further evidence to capture the complex geographies of supplemental reindustrialisation, simultaneously characterised by geographic broadening, deepening FDI embeddedness, but also evident growth limits in the successful areas of post-socialist industrial development. The paper untangles the emerging spatial patterns and driving forces of this process, while examining the policy implications of supplemental reindustrialisation within the context of new European and CEE industrial policies.

Zoltan Gal (p), Bennour Med Hsin

Gabor Lux

This study explores the complex role of Foreign Direct Investment (FDI) in regional development across the Visegrad countries (V4) and Austria as a non DME (Dependent Market Economies) benchmark, with a focus on productivity, specialization, Global Value Chain (GVC) positioning, and knowledge complexity. While FDI has been a catalyst for economic integration, productivity gains, and sectoral specialization, it has also reinforced structural dependencies that hinder regional upgrading.
Our study provides novel insights into how regional roles (supplier, assembler, outsider and controller) influence FDI’s effects on productivity and specialization. While FDI boosts productivity in supplier and assembler regions, it negatively affects regional GVC positioning (upstreamness) and knowledge diversity. However, its influence on GVC positioning is predominantly negative, with only marginal benefits for outsider regions. Furthermore, FDI is negatively associated with knowledge diversity, indicating that it encourages sectoral concentration rather than fostering broad-based technological upgrading. The study also finds that knowledge complexity remains largely unaffected by FDI, suggesting that technological advancement requires complementary policies beyond foreign capital inflows.
These results contribute to ongoing debates on on FDI-led growth models by demonstrating that while FDI stimulates regional economic activity, it can also lock regions into lower-value-added roles within GVCs. This research underscores the need for targeted policies to enhance local absorptive capacities, encourage technological upgrading, and reduce excessive reliance on foreign investment. By integrating regional-level FDI patterns with regional specialization dynamics, our study offers critical insights for policymakers striving to balance FDI attraction with sustainable economic development in the V4.

Zsuzsanna Zsibók (p), Zoltán Gál, Ildikó Egyed

Balázs Páger

The economic development model of Central and Eastern European countries has faced significant challenges, particularly after the COVID-19 crisis and subsequent energy and supply chain disruptions of the early 2020s. However, early warning signs of stagnating labour productivity in specific regions were already evident in the preceding decade. These challenges vary not only between countries but also among sub-national territorial units, strongly influenced by sectoral specialization.
Through an empirical analysis of sub-national labour productivity growth and its sectoral decomposition, including shift-share analysis, this study finds that increasing reliance on foreign direct investment alone cannot resolve efficiency issues in non-metropolitan areas—or, ultimately, in these economies as a whole. While economic restructuring played a key role in driving labour productivity growth in the early 2010s, its broader positive effects have diminished over time. Our findings highlight a fundamental trade-off faced by non-metropolitan regions: improving labour productivity often comes at the cost of reduced domestic value-added content. Addressing this challenge requires targeted investment policies and human capital strategies that foster both productivity growth and higher local economic integration.

Bennour Mohamed Hsin (p)

Zsuzsanna Zsibók

The complexity literature offers an elegant implementation of the principle of relatedness. However, it faces limitations due to having more categories than regional anchors, given its reliance on co-occurrence, which can obscure actual relationships. This prompts the need to capture these relationships using non-linear methods. In this paper, we leverage Random Forest, a tree-based machine learning algorithm, to capture complex, non-linear relationships among patent classes. We first construct the Revealed Comparative Advantage (RCA) matrix at the NUTS2 regional level for each year from 1978 to 2018, then train a Random Forest model for each of the 641 patent classes, using a 4-year lag structure to predict future regional expertise. Our results effectively capture relationships between technologies and successfully predicted the structural break in patenting activities observed in 2011, particularly for highly complex technologies, using data from 2007. Additionally, we address class imbalance in our data and utilize these predictions to propose a comparative framework assessing regions' efficiency in leveraging their technological landscape advantages. This framework also opens opportunities for identifying optimal regional diversification strategies through network-based and Bayesian approaches.