|
on Economic Geography |
|
Issue of 2025–07–14
seven papers chosen by Andreas Koch, Institut für Angewandte Wirtschaftsforschung |
| By: | Michael Balzer |
| Abstract: | Researchers in urban and regional studies increasingly deal with spatial data that reflects geographic location and spatial relationships. As a framework for dealing with the unique nature of spatial data, various spatial regression models have been introduced. In this article, a novel model-based gradient boosting algorithm for spatial regression models with autoregressive disturbances is proposed. Due to the modular nature, the approach provides an alternative estimation procedure which is feasible even in high-dimensional settings where established quasi-maximum likelihood or generalized method of moments estimators do not yield unique solutions. The approach additionally enables data-driven variable and model selection in low- as well as high-dimensional settings. Since the bias-variance trade-off is also controlled in the algorithm, implicit regularization is imposed which improves prediction accuracy on out-of-sample spatial data. Detailed simulation studies regarding the performance of estimation, prediction and variable selection in low- and high-dimensional settings confirm proper functionality of the proposed methodology. To illustrative the functionality of the model-based gradient boosting algorithm, a case study is presented where the life expectancy in German districts is modeled incorporating a potential spatial dependence structure. |
| Date: | 2025–06 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2506.13682 |
| By: | Batabyal, Amitrajeet; Beladi, Hamid |
| Abstract: | In this paper, we analyze a dynamic model in which two stylized regions A and B use an artificial intelligence (AI)-based technology α(t) to produce a knowledge good Q(t). Even though the initial value of the AI-based technology α(0) is identical in both regions, region A saves and hence invests more than region B to make the existing AI-based technology more powerful. We show that this differential investment means that the ratio of the output of the knowledge good in region A to region B or Q_A⁄Q_B is continually rising. In other words, without targeted policy, region A will become a “leading region” that experiences economic growth and innovation ahead of region B which will become a “lagging region” that innovates less and hence tends to grow more slowly. |
| Keywords: | Artificial Intelligence, Dynamics, Economic Growth, Region, Technology |
| JEL: | O33 R11 |
| Date: | 2025–01–09 |
| URL: | https://d.repec.org/n?u=RePEc:pra:mprapa:124730 |
| By: | Shafiqullah Yousafzai; Hisahiro Naito |
| Abstract: | This study examines the effect of exports on subnational income and regional inequality between urban (trade hub) and rural (non–trade hub) areas, using nighttime luminosity as a proxy for economic activity. We construct a country-period panel dataset covering 104 countries, based on five-year average data from 1997 to 2020. Trade hub areas are defined as the union of areas within a 30 km or 50 km radius of each of the three largest ports and three international airports in a country, while all remaining areas are classified as non–trade hub areas. To address endogeneity, we employ a two-stage least squares (2SLS) approach, using predicted trade as an instrumental variable. Predicted trade is derived from a dynamic gravity equation in which time dummies are interacted with sea and air transport distances. This instrument captures variation in transportation costs driven by technological advances that have shifted trade from sea to air, thereby influencing trade volumes. Our results show that a 1\% increase in exports raises nighttime luminosity by 0.3% in trade hub areas and by 0.06\% in non–trade hub areas. Export growth also leads to population increases in trade hub areas, but not in non–trade hub areas. Furthermore, we find that a 1% increase in exports raises nighttime luminosity per capita by 0.18% in trade hub areas and by 0.06% in non–trade hub areas. These findings suggest that while exports stimulate economic activity in trade hubs, population inflows partially offset per capita gains. Nonetheless, exports significantly exacerbate regional inequality. |
| Date: | 2025–01 |
| URL: | https://d.repec.org/n?u=RePEc:tsu:tewpjp:2025-001 |
| By: | Hisaki KONO; Hoang-Minh LE; Manabu NOSE; Yasuyuki SAWADA |
| Abstract: | This paper examines the local economic impacts of industrial zones (IZs) in Vietnam, focusing on how their sectoral orientation within production networks shapes effectiveness. Using panel data on registered firms and a newly compiled dataset on IZ locations and sectoral compositions, we estimate the dynamic effects of IZ establishment on firm entry and employment through staggered difference-in-differences and synthetic control methods. We find that IZs lead to sustained increases in both firm and worker density over a 6–10 year horizon, indicating substantial local economic gains. These effects are particularly pronounced in zones oriented toward downstream industries—those that create demand for upstream suppliers—while upstream orientation does not predict stronger outcomes. We further show that backward production linkages mediate these gains, suggesting that demand-side constraints, rather than input frictions, may be more binding in developing country contexts. The results highlight not only the overall effectiveness of IZs but also the importance of aligning industrial policy design with the structure of production networks to maximize spatial development benefits. |
| Keywords: | Industrial zones, production linkage |
| JEL: | O12 O14 R11 |
| Date: | 2025–06 |
| URL: | https://d.repec.org/n?u=RePEc:kue:epaper:e-25-005 |
| By: | Thierry Warin |
| Abstract: | In the aftermath of renewed trade tensions and geopolitical realignments—exemplified by the 2025 trade war under President Trump 2.0—the dominant policy discourse posits that globalization went “too far, ” sacrificing resilience and national security at the altar of cost efficiency. This paper challenges that narrative by unpacking the implicit assumptions that undergird it, notably the belief that global trade and value chains were ever efficient in the first place. Drawing on international business literature, economic geography, and trade theory, we argue that global supply chains, far from representing optimal configurations, were largely shaped by bounded rationality, cognitive biases, and incomplete information—what we term the streetlight post bias. Contrary to the Heckscher-Ohlin-Samuelson model’s idealized vision, firm-level decisions rarely reflect first-best equilibria; instead, trade patterns have followed the more constrained logic of the gravity model and regional familiarity. The paper contends that neither globalization nor its retrenchment (via reshoring, nearshoring, or friend-shoring) guarantees a move toward a more resilient or efficient trade architecture. Instead, both may reflect alternative second-best equilibria. We propose a forward-looking framework in which big data analytics and machine learning—grounded in an economic geography perspective— can help firms and policymakers identify robust, diversified, and efficient global value chain configurations. By addressing information asymmetries and reducing decision-making bias, such tools offer a path toward a closer approximation of the first-best equilibrium. We conclude with implications for trade policy, calling for evidence-based interventions that move beyond reactive deglobalization toward intelligent, data-driven integration. À la suite du regain des tensions commerciales et des réajustements géopolitiques — illustrés par la guerre commerciale de 2025 sous la présidence de Trump 2.0 — le discours dominant en matière de politique commerciale soutient que la mondialisation est allée « trop loin », sacrifiant la résilience et la sécurité nationale sur l’autel de l’efficacité économique. Cet article remet en question ce récit en déconstruisant les hypothèses implicites qui le sous-tendent, notamment la croyance selon laquelle le commerce mondial et les chaînes de valeur sont efficients. En mobilisant les littératures en affaires internationales, géographie économique et théorie du commerce international, nous soutenons que les chaînes d’approvisionnement mondiales, loin d’incarner des configurations optimales, ont été largement façonnées par la rationalité limitée, les biais cognitifs et l’information incomplète — ce que nous appelons le biais du réverbère. Contrairement à la vision idéalisée du modèle d’Heckscher-Ohlin-Samuelson, les décisions prises au niveau des entreprises reflètent rarement un équilibre de premier rang ; au contraire, les flux commerciaux suivent plutôt la logique contrainte du modèle gravitaire et d’une familiarité régionale. L’article soutient que ni la mondialisation ni sa remise en cause (via la relocalisation, la régionalisation ou le « friend-shoring ») ne garantissent une architecture commerciale plus résiliente ou plus efficiente. Ces dynamiques pourraient au contraire représenter des équilibres alternatifs de second rang. Nous proposons un cadre prospectif dans lequel l’analyse des mégadonnées (big data) et l’apprentissage automatique (machine learning) — ancrés dans une perspective de géographie économique — peuvent aider les entreprises et les décideurs à concevoir des chaînes de valeur mondiales à la fois robustes, diversifiées et efficientes. En réduisant les asymétries d'information et les biais décisionnels, ces outils ouvrent la voie à une approximation plus précise de l’équilibre de premier rang. Nous concluons par des implications en matière de politique commerciale, plaidant pour des interventions fondées sur les données probantes, allant au-delà d’une démondialisation réactive vers une intégration intelligente et pilotée par les données. |
| Keywords: | Global Value Chains, Trade Efficiency, Streetlight Effect, Machine Learning, Supply Chain Resilience, Chaînes de valeur mondiales, Efficacité du commerce, Effet du réverbère, Apprentissage automatique, Résilience des chaînes d’approvisionnement |
| Date: | 2025–07–09 |
| URL: | https://d.repec.org/n?u=RePEc:cir:cirwor:2025s-17 |
| By: | Marc Deschamps (Université Marie et Louis Pasteur, CRESE, UR3190, F-25000 Besançon, France); Julie Le Gallo (CESAER UMR1041, INRAE, Institut Agro Dijon, France); Catherine Refait-Alexandre (Université Marie et Louis Pasteur, CRESE, UR3190, F-25000 Besançon, France) |
| Abstract: | Drawing on data from three specialized crowdfunding platforms and on semi-structured interviews conducted with project leaders located in the French Jura Arc region, we examine the role of this financing mechanism in the transition of the agri-food system. We draw in particular on proximity theory (Torre, 2018) to show that while so-called “social” proximity (ties of family, friendship, or networks) plays a decisive role in the success of crowdfunding campaigns, geographical proximity does not appear to be a determining factor in the decision to resort to crowdfunding or in the success of these campaigns. Our exploratory study also suggests that other forms of proximity (institutional or organizational) may influence the actors’ approaches. Finally, family and friendly support remains central, whereas the effect of any “geographical neighborhood” appears much more limited. |
| Keywords: | crowdfunding, transition, agri-food, proximity theory, social and geographical proximity. |
| JEL: | G29 Q14 Q20 R11 R12 |
| Date: | 2025–04 |
| URL: | https://d.repec.org/n?u=RePEc:crb:wpaper:2025-03-eng |
| By: | Belchior José , Manuel (Universidade Católica de Moçambique e Universidade Zambeze); André Luis Squarize , Chagas (Departamento de Economia, Universidade de São Paulo) |
| Abstract: | This paper investigates the spatial dynamics and convergence of multidimensional poverty across 107 districts in northern and central Mozambique between 2007 and 2017. Using census microdata and the Alkire-Foster method, we construct three indicators – headcount ratio (H), intensity of deprivation (A), and the multidimensional poverty index (MPI) – to assess the evolution of poverty across space and time. The analysis combines exploratory spatial data techniques and spatial econometric models (SLX, SLM, SDM) to test for spatial autocorrelation, absolute and conditional convergence, and the role of demographic and socioeconomic drivers in poverty propagation. Results reveal robust evidence of spatial dependence and convergence in all three indicators. While overall poverty incidence declined, intensity increased among the most deprived, particularly in rural districts. Spatial spillovers are present in both outcome variables and structural covariates, especially illiteracy, population density, and the masculinity index. These findings underscore the need for territorially integrated policy responses, as poverty reduction in one district can generate positive externalities in neighboring areas. The study highlights the relevance of spatially coordinated strategies for addressing persistent territorial inequalities and advancing multidimensional poverty reduction in structurally vulnerable regions. |
| Keywords: | Multidimensional poverty; Spatial econometrics; Regional convergence; Mozambique; Spatial spillovers |
| JEL: | C21 I32 R11 R23 |
| Date: | 2025–06–30 |
| URL: | https://d.repec.org/n?u=RePEc:ris:nereus:2025_005 |