Regional digital economy development lowers the cost of providing banking services across geographical distance, but its effect on local banking concentration is not unidirectional. Using nearly 820,000 commercial bank branch records and a provincial digital economy index for 31 Chinese provinces from 2013 to 2023, this paper applies a spatial Durbin error model to distinguish the absorption of local digital resources, exposure to neighboring digital capacity, and unobserved regional shocks. Local digital economy development is significantly associated with greater banking concentration, suggesting that digital resources may initially reinforce the competitive advantages of incumbent institutions. By contrast, digital economy development in neighboring regions is associated with lower local concentration, reflecting the external competitive pressure created by greater cross-regional service capacity. A three-regime model and bank-type decomposition further show that digital competition follows distinct financial-geographical pathways: local digitalization is more closely associated with the expansion of city commercial banks, whereas neighboring digitalization more often facilitates the cross-regional entry of nationwide joint-stock banks, potentially generating greater openness across bank categories alongside increased concentration within categories. The digital economy therefore reshapes not merely the degree of banking competition, but also the spatial direction from which competition enters and the organizational carriers through which it operates. Provincial banking concentration should consequently be interpreted in the context of cross-regional digital linkages.
Although the transition to the circular economy (CE) is slowly underway, business activities that contribute to the CE struggle to find sufficient space. Land is scarce and contested by multiple societal needs, and not just any place is suitable; CE firms are highly diverse, each with their distinct location requirements. To develop effective policies that secure and create space specific to the CE’s needs, in-depth knowledge about these requirements is necessary. This paper contributes to this debate by analyzing the locational preferences of four classes of CE firms in the Netherlands, namely those that contribute towards: the reduction of raw material usage, the substitution of raw materials, product life extension, and high-grade processing. Negative binomial and Poisson models are estimated for a sample of 14,129 neighborhoods, based on business microdata and spatial data on infrastructure, zoning, and agglomeration economies. The models show that infrastructure plays a smaller role than expected in CE firm location in the Netherlands, while the presence of business parks and the concentration of economic sectors in the area, such as the industry and energy sector, are decisive. The analysis confirms that location patterns of CE firms in reduction and extension are quite distinct from substitution and processing, highlighting the need to plan for a diverse network of business parks to stimulate circularity. Some combinations of location qualities are quite scarce, such as business parks in centrally located urban locations, suggesting that conservation and more intensive use of these location types is recommended.
This paper presents an innovative examination of the spread of COVID-19 infection rates across England over a seven-week period commencing prior to the initial lockdown and continuing until the relaxation of some COVID-19 lockdown restrictions. We examine the evolution in infection rates across socioeconomic characteristics using topological data analysis to map the joint distribution of socioeconomic characteristics and identify where and when infection rates grew differentially. Although the over 65s and those with chronic health conditions were socioeconomic groups most at risk of experiencing morbidity and mortality from COVID-19, these socioeconomic correlates reflect consequences of the spread rather than the carriers of the virus and the reasons for the spread. Our empirical results present new findings that International Territorial Level 3 areas with higher proportions of younger inhabitants who worked longer hours for higher pay experienced earlier and faster rates of increase in infections, and that commuter towns and rural areas lagged their urban metropolises in terms of infection rates. These results underscore who was infected, where labour market characteristics associated with infection rates interact across space, and the need to improve understanding of when and why areas experienced higher infection rates. This is the first paper to find that the spread of the COVID-19 virus in the UK was greater where there was a relative abundance of skilled workers who have greater professional geographical mobility.
Regional analyses are important to address spatial disparities in economic structures, but regionalization is often constrained by limited data availability. This study applies three established approaches to regionalize input-output tables (IOTs), Cross Hauling Adjusted Regionalization Method (CHARM) and two location quotient (LQ)-based approaches to analyze the economy of Lower Saxony, Germany. Three research questions are addressed: (1) How does the construction process for regional IOTs differ across the various approaches? (2) How do CHARM- and LQ-based approaches vary regarding the interpretation of regional economies? (3) What insights can be derived for Lower Saxony? The different conceptions and data requirements of the approaches affect both the construction process and the analysis. CHARM allows for the complete regional IOT construction whereas the LQ approaches do not enable detailed final demand representation. Using linkage analysis, multiplier analysis, and the hypothetical extraction method, the findings indicate that CHARM emphasizes supply-based trade flows and consumption-based effects, while the LQ-based approaches provide deeper insights into intermediate production structures. Differences in linkages and multipliers are more pronounced in import-dependent sectors. Regarding Lower Saxony’s economy, the results indicate a high degree of regional self-sufficiency. Manufacturing plays a central role in Lower Saxony, with its hypothetical extraction reducing total output by around 40%. High gross value added multipliers are observed in education and real estate. Energy supply dominates both greenhouse gas emissions and water use multipliers, reflecting the current relevance of fossil fuels. Overall, both regionalization methods provide valuable insights and should be seen as complementary.
This paper examines regional development in the European Union from a multidimensional and dynamic perspective, addressing the limited understanding of how economic, social, institutional, and environmental dimensions interact over time. While existing research often treats these dimensions separately or relies on static indicators, less attention has been paid to their temporal interdependencies and adjustment dynamics. Using a panel vector autoregressive model with common correlated effects (PVAR-CCE) applied to 233 NUTS-2 regions over the period 2000–2022, the analysis identifies the dynamic interactions linking these dimensions while accounting for cross-sectional dependence and regional heterogeneity. The results provide evidence of a structured system of interdependencies governing regional development. They reveal a dual adjustment structure characterised by a fast-moving economic–social cycle linking employment, income, and youth exclusion, and a slower structural dynamic driven by institutional quality, human capital, and environmental conditions. These dynamics vary across regions and income levels, highlighting persistent structural constraints in lagging regions and more integrated development trajectories in advanced regions. Environmental sustainability emerges as a constitutive component of regional development rather than a peripheral outcome, extending place-based approaches beyond income-based frameworks. Overall, the findings contribute to a more integrated understanding of regional development and provide an empirical basis for assessing the feasibility of EU policy objectives related to cohesion, sustainability, and balanced territorial development.
Population aging has intensified concerns regarding regional disparities in elderly longevity, even under universal healthcare systems. This study examines how socioeconomic conditions, healthcare capacity, demographic structure, and spatial interdependence influence life expectancy at age 65 (LE65) across Taiwan’s 20 counties during 2010–2023. Using a county-level panel dataset and the Generalized Spatial Two-Stage Least Squares (GS2SLS) estimator, the analysis reveals that higher household disposable income and educational attainment are associated with longer life expectancy, whereas poverty and household healthcare-related financial burdens reduce longevity outcomes. Significant positive spillover effects are identified for educational attainment and public healthcare expenditure, indicating that regional longevity is shaped by both local and neighboring conditions. Elderly cohort composition also significantly influences longevity, highlighting the importance of distinguishing cohort effects from age-related effects. The results remain robust after excluding the COVID-19 period. The findings suggest that policies promoting healthy aging should prioritize poverty reduction, educational improvement, alleviation of healthcare-related financial burdens, and stronger interregional coordination in public health investment.
This study investigates from a theoretical point of view the spatial distribution of households by income within metropolitan areas, focusing on three key determinants: the polycentric structure of employment, the spatial distribution and accessibility of public amenities, and the presence of spatial interdependencies. Utilizing a novel and comprehensive dataset, we endogenously identify employment centers within the Barcelona Metropolitan Area (BMA). After incorporating this polycentric employment structure and the availability of public amenities—such as educational institutions, healthcare facilities, childcare centers, and transportation infrastructure—we extend conventional empirical models to account for spatial effects. These effects capture the externalities inherent in residential location choices, particularly the tendency of households to cluster with others of similar income levels. Employing spatial regression techniques, we empirically assess the influence of these factors on household location patterns. The results reveal that employment subcenters exert a significant effect on residential location decisions, that certain public amenities play a statistically significant role, and that neighborhood income level functions as endogenous amenities, further shaping the spatial distribution of income.
Depopulation and territorial decline increasingly reflect everyday constraints that undermine the liveability of disadvantaged areas. Commuting can enable residents to maintain attachment to place while accessing employment and services, but it may also generate time pressure and reduce wellbeing, especially in less accessible contexts. Using data from the Italian Time Use Survey, we analyse experienced wellbeing during commuting alongside broader evaluative outcomes, including perceived time pressure and satisfaction with work–life balance, leisure time, and life overall. Taken together, our findings suggest that commuting may be one of the mechanisms through which territorial disadvantage is reflected in subjective wellbeing. Although on average residents of peripheral areas neither report lower wellbeing during commuting nor lower life satisfaction, longer daily commutes are associated with lower satisfaction with work–life balance and with leisure time. Women living in peripheral areas, moreover, report less positive commuting experiences than men and appear to be more exposed to time pressure in disadvantaged territorial contexts. These results highlight how relatively high overall life evaluations can coexist with commuting-related time constraints in peripheral contexts, suggesting that improving accessibility and reducing commuting-related time burdens may be crucial for the long-term sustainability of disadvantaged areas.
Productivity and resource allocation could be influenced by the occurrence of spillover effects due to geographical structure. This paper examines the spatial dimension of misallocation and productivity across Italian provinces, using firm-level microdata for manufacturing sector. Methodologically, we build province-level measures of misallocation and total factor productivity, and then we apply a spatial filtering approach to decompose them into two parts: a spatial component, capturing inter-regional spillovers and geographic patterns, and an idiosyncratic component, reflecting province-specific characteristics. By disentangling these two dimensions, the paper provides a clearer view of how geography and local structural conditions jointly shape the distribution of productive resources. This distinction is crucial to identify whether regional productivity disparities are primarily due to local peculiarities or spatially correlated distortions. We find significant spatial autocorrelation in both productivity and misallocation measures, where the geographical component is predominant. Eliminating distortions would raise output by over 20 percent, with roughly two-third of these gains associated with the spatial component.
In this paper, we examine how climatic conditions affect regional economic convergence in Colombia. The analysis focuses on how climatic variability interacts with initial income levels and spatial interdependence to shape regional convergence dynamics in a heterogeneous manner. Colombia provides a suitable empirical setting due to its pronounced regional inequality, strong spatial linkages, and high exposure to climatic fluctuations. Using panel data for Colombian departments over the period 2008–2019, we estimate a set of spatial convergence models that incorporate non-linear climate effects and interactions between climatic variables and initial income. The results reveal strong spatial dependence in regional growth processes, confirming the relevance of accounting for interregional spillovers in convergence analysis. Temperature deviations emerge as a key conditioning factor of convergence: temperature shocks exhibit non-linear effects on economic growth and disproportionately affect higher-income regions, thereby reinforcing relative convergence. These impacts operate through both direct local effects and indirect spatial spillovers, reshaping the geography of regional adjustment rather than acting as a uniform growth shock.
While remittances are known to reduce poverty at the household-level, their broader distributional impacts at the subnational level remain underexplored. Using panel data from 109 districts in Pakistan, drawn from three household surveys combined with remote sensing data, we employ an instrumental variable strategy to examine the role of migrant remittances in shaping regional inequalities in income, assets, education, and healthcare use. We estimate the effects of remittance incidence, income share, and per capita inflows on various economic inequality measures by using three instruments including distance to the nearest international airport alternatively interacted with weighted transaction cost, global oil prices, and weighted GDP of destination countries. Our findings are four-fold. First, remittances increase regional income inequality. A 1 percent increase in the share of recipient households increases income Gini by 0.55–0.74 and MLD by 0.44–0.69 percentage points. The effect plateaus beyond the 24% remittance income threshold. Second, remittances widen inequality in home ownership but reduce those in non-farmland assets. No significant effects are observed for farmland or commercial property ownership. These divergent patterns reflect differential market access. Third, remittances boost private schooling enrolment both at primary and secondary levels. A 1 percent rise in remittance incidence or income share increases private school enrolment by 3.31 to 5.52 percentage points. The gender gap in education also widens. Fourth, remittances increase healthcare inequality, but do not affect gender gap in healthcare use. These results are robust to spatial autocorrelation correction, and the use of alternative instruments and remittance indicators.
This paper contributes to the resource curse and regional development literature by proposing a labour-market framework that contrasts positive spillover effects with crowding-out mechanisms and by examining how these processes shape employment and income trajectories across Australian regions. Using panel data for 566 local government areas between 2011 and 2021, combined with information on the location and status of mines across the country, we investigate the relationship between mining activity and regional development outcomes. Descriptive evidence indicates that, despite national growth in mining employment, many mining-hosting regions experienced declines in the number of resident miners, suggesting a growing spatial decoupling between mining production and local labour-market benefits. To evaluate the proposed framework, we estimate reducedform growth models linking changes in mining employment to changes in non-mining employment and family income. Results indicate modest but positive average effects, suggesting that positive spillover mechanisms generally dominate, although crowding-out effects are more prevalent in some regions. Together, the conceptual framework and empirical findings help explain how labour-market dynamics shape subnational resource curse processes and underscore the need for place-based, evidence-informed development strategies.
The green transition is reshaping Europe's development geography, yet progress remains uneven across regions. Effective policy requires systematic tools to monitor how territories perform, adapt, and benefit from this transformation. This paper proposes a practical and multidimensional definition of the green transition aligned with the European Commission's strategic directions and introduces a new Regional Green Transition Performance Index covering 232 NUTS-2 regions from 2009 to 2022. The index measures progress across seven pillars-environment, energy, circular economy, sustainable development, just transition, innovation and policy, and mobility-offering the first harmonized and longitudinal framework to track regional transition performance. Beyond performance, we introduce the Regional Green Transition Opportunity Index, capturing regional assets and capacities that enable adaptation and growth, and combine it with the existing Regional Green Transition Vulnerability Index. This yields an opportunity-vulnerability typology distinguishing resilient frontrunners, missed potentials, stable but stagnant, and at-risk regions. Methodologically, we conduct extensive robustness tests in the construction of the composite indices and estimate a battery of econometric convergence models to explore spatial dynamics. Results reveal significant spatial dependence and convergence in green transition performance, highlighting territorial interdependencies across regional systems. By integrating multidimensional performance with structural vulnerabilities and opportunities, the paper provides policymakers with an evidence-based, region-level monitoring tool to design balanced, place-sensitive strategies that both mitigate risks and mobilize opportunities for a just and cohesive European green transition.
Human capital is considered one of the main drivers of long-term economic development, justifying large public investments in higher education. However, the existence of externalities in the form of agglomeration economies can induce high-skilled individuals to migrate into a reduced number of dynamic urban areas hosting the lion's share of knowledge-intensive activities. Using rich administrative datasets on labor contracts and business demographic statistics, this work builds a gravity model explaining inter-regional migration by skill level between Spanish regions during the last 22 years (2001-2022). The results indicate that the agglomeration of knowledge-intensive activities, specifically measured as regional specialization in knowledge-intensive business services, is a significant and strong determinant of high-skilled migration. This result poses an important challenge for policymakers, as it implies that efforts to increase educational opportunities in peripheral regions could be seriously undermined by initial differences in economic structures.
By introducing the concepts of regressive deconcentration and polarized residual spatial diffusion, this study advances the debate on regional development strategies in emerging economies and highlights the intertwined role of deindustrialization and Knowledge-Intensive Business Services (KIBS) in Brazil’s uneven territorial transformation. The analysis focuses on the spatial diffusion of KIBS across 558 Brazilian microregions between 2007 and 2021, applying spatial econometric techniques and dynamic panel models. The results reveal a process of regressive deconcentration, manufacturing declines in core regions without generating compensatory gains in peripheral areas. KIBS expansion follows a distinct trajectory. Rather than spreading evenly across the territory, it exhibits polarized residual spatial diffusion, remaining concentrated in regions with skilled labor, industrial activity, and infrastructure, without offsetting the losses caused by deindustrialization. If reinforced by the diffusion of digital technologies, this pattern of KIBS expansion could become a new driver of regional disparities, in contrast to the decline of manufacturing. Taken together, these dynamics underscore the ambivalent role of KIBS in structural transformation. While they foster productivity and innovation in advanced regions, they simultaneously reinforce spatial inequalities. Taken together, these dynamics suggest that inclusive development will only be achieved through policies that simultaneously strengthen human capital, foster knowledge-intensive manufacturing, and stimulate the diffusion of KIBS across a broader territorial base.
The existing literature has yet to examine whether and how the top-down urban agglomeration policy influences the spillover effect for environmental benefits. Furthermore, it remains unexplored how a region’s spatial structure shapes this policy-induced environmental spillover, creating a critical gap in understanding the role of spatial structure in sustainability-oriented regional planning. Based on a sample of 195 cities in China from 2004 to 2019, this paper examines whether the urban agglomeration policy affects the spillover effect of carbon emissions and how the spatial structure of urban agglomeration affects this relationship. Using a high-speed rail weighted Spatial Durbin Difference-in-Differences method, our results show that urban agglomeration policy enhances the “borrowed-size” effect in carbon reduction, exhibiting a carbon reduction spillover effect from leading cities to peripheral cities. Furthermore, we demonstrate that spatial structure plays an important role in shaping the patterns of the spillover effect in carbon emissions by showing that the “borrowed-size” effect is weakened in monocentric urban agglomerations. This study extends the “borrowed-size” framework into the environmental domain and establishes spatial structure as a key determinant of sustainability-oriented regional planning.
This paper studies the spatial distribution of economic activities across urban hierarchy from an agglomeration perspective by constructing a new economic geography model with manufacturing and service sectors, two types of labor with perfect geographical mobility, and two regions. Holding the size of the sectors constant, the theoretical model shows that the larger city has a disproportionately larger concentration of the manufacturing sector compared to the smaller city. We test this prediction by estimating a spatial panel model with data from 285 Chinese cities over the period 2009-2019. We find a positive and statistically significant relationship between city size and manufacturing-to-service ratio. The findings shed new light on the importance of the manufacturing sector relative to the service sector for urban economic development. Despite the nationwide domination of services, larger cities are relatively specialized in manufacturing due to agglomeration economies.
