From the coast to the interior: global economic evolution patterns and mechanisms

Coastal areas host dense populations and concentrated economic activity and have historically benefited from maritime trade and access to large markets. Yet, despite ongoing globalization and large increases in global trade since 2000, there is debate over whether economic activity will continue clustering along coasts or diffuse into interior regions. The study addresses key questions: How much GDP lies within 100 km of ice-free coasts or sea-navigable rivers? What spatial patterns of economic activity emerge under globalization? Is GDP continuing to agglomerate near coasts, diffusing inland, or both, and what mechanisms drive these dynamics? Motivated by data gaps in global accounting of GDP within/beyond 100 km of coasts, the authors adopt nighttime light (NTL) data to quantify economic concentration in globally consistent “near” (≤100 km from coasts or navigable rivers) and “far” regions. Grounded in Hirschman’s regional uneven growth theory, the study aims to accurately measure the concentration of economic factors, identify spatio-temporal evolution patterns, and explain mechanisms related to maritime transport advantages and industrialization stages. This work informs sustainable development and planning for coordinated land–sea development.

The authors note a broader literature (detailed in supplemental materials) covering: (1) definitions and standard ranges used to delineate coastal areas; (2) evidence of agglomeration of economic and social factors in coastal zones; and (3) the use of nighttime light data as a proxy for economic activity. Prior studies consistently report strong correlations between NTL and GDP, enabling spatially explicit economic estimation where conventional statistics are sparse or inconsistent.

The study proceeds in three steps: (1) validate the correlation between nighttime light (NTL) brightness and GDP; (2) estimate the economic aggregates of global “near regions” (≤100 km from ice-free coasts or sea-navigable rivers) using a Random Forest (RF) approach; and (3) analyze spatial evolution patterns via changes in regional economic proportions over time. Correlation analysis: Using SPSS 19, the authors test NTL versus GDP (constant 2010 USD) for countries with clear administrative divisions in global coastal areas (excluding those without clear statistics or small island countries). A normality check using the W test (N=2840 > 2000) yields sig.=0.000<0.005, indicating non-normality. A Spearman correlation yields ρ=0.921 with sig.=0.000<0.005 (significant at 0.01), confirming a strong association between NTL and GDP. Random Forest estimation: For each coastal economy, the study builds an RF regression model linking annual NTL data to official GDP to characterize the NTL–GDP relationship specific to that economy. Bootstrap sampling draws multiple samples; two-thirds are used for training and one-third for Out-of-Bag (OOB) validation. Multiple decision trees are fit per bootstrap sample and aggregated. Implementation uses scikit-learn in Python. For regression, outputs are averaged across trees of equal importance. Spatial gridding partitions the study area into F grid cells; each grid has dependent (e.g., NTL) and explanatory variables, forming F samples. Trained economy-specific models are then applied to NTL within “near regions” to estimate regional GDP. Regional economic proportion: The share of a region i is computed as e_i = E_i / ΣE_i × 100%, where E_i is the economic aggregate of region i and ΣE_i is the total across regions. Temporal changes in e_i reflect the direction and degree of movement of economic factors between near and far regions.

Global patterns (2000–2018): - Near-region GDP rose from USD 33,124.2 billion (2000) to USD 51,523.6 billion (2018); far-region GDP rose from USD 16,134.8 billion to USD 30,238.0 billion. - Near-region share declined from 67.25% (2000) to 63.02% (2018); far-region share increased from 32.75% to 36.98%. - GDP density disparity (near vs far) decreased from 9.3× to 7.7×, indicating narrowing spatial imbalance while maintaining strong core-periphery structure. - Overall, globalization coincided with diffusion of economic activity from coastal near regions toward interior far regions, forming a global “coastal remoteness” pattern. Intercontinental patterns (near-share → trend; density ratios approximate): - Europe: Near share fell 69.74%→68.41%; GDP density near ≈2.3× far; evolution shows mild coastal remoteness with dominant near-region economy. - North America: Near share fell 55.28%→50.87%; density ratio declined 4.7×→3.9×; coastal remoteness driven largely by the U.S. shift toward services and flexible spatial layouts. - Asia: Near share fell 79.58%→69.25% (still above global average); density ratio declined 21.9×→12.7×; strong coastal remoteness as China, India, Thailand, Bangladesh and others see inland growth; logistics initiatives (e.g., Belt and Road, rail corridors) support interior expansion. - South America: Near share fell 58.75%→52.43%; density ratio declined 5.7×→4.4×; coastal remoteness as Brazil and Argentina move toward service-driven structures. - Oceania: Near share fell 91.30%→84.24%; density ratio declined 35.6×→18.2×; coastal remoteness amid high coastal costs and external shocks prompting more interior focus (especially in Australia). - Africa: Near share rose 47.55%→48.88%; near-region density ≈10× far while comprising <10% of land; overall “coastal proximity” pattern persists, tempered by early-stage industrialization and limited capacity to sustain higher trade costs inland. Mechanisms identified: declining inland and multimodal transport costs; rising shares of services and high-tech reducing dependence on maritime transport; and diseconomies of scale plus spillover effects in over-agglomerated coastal regions.

Findings affirm that near regions retain strong agglomeration advantages consistent with location theory (lower freight costs near ports encourage clustering). However, over 2000–2018, continued globalization did not intensify coastal concentration; instead, economic activity diffused inland, yielding a global “coastal remoteness” evolution. Proposed mechanisms include: (1) rapid improvements in inland and multimodal transport lowering logistics costs and enabling production farther from ports; (2) structural shifts toward services and high-tech sectors that are less dependent on maritime freight and more spatially flexible; and (3) excessive coastal agglomeration leading to diseconomies (congestion, land costs, pollution) and spillovers that push activity to lower-density interior areas. Intercontinental differences align with industrialization stage and transport dependence: Europe and North America (post-industrial) show modest diffusion consistent with high-level equilibrium; Asia, South America, and Oceania (mid-to-late industrialization) show stronger diffusion; Africa (early stage) maintains coastal proximity. Policy implications include recognizing the enduring economic centrality of near regions while increasingly planning for interior growth, balancing investments to support coordinated land–sea development and mitigate overconcentration costs.

The study quantifies global and continental economic distributions between coastal “near” and interior “far” regions using nighttime lights and Random Forest models. Key conclusions: (1) Near regions, covering only 18.43% of land, generate about 65% of global GDP and have roughly eightfold GDP density relative to far regions, evidencing a persistent core-periphery structure consistent with low-freight coastal advantages. (2) From 2000 to 2018, near-region GDP share fell (67.25%→63.02%) while far-region share rose (32.75%→36.98%), indicating a widespread “coastal remoteness” diffusion of economic activity inland and a narrowing of density disparities. (3) Continental patterns vary with industrialization and maritime dependence: Africa shows “coastal proximity,” whereas Europe, North America, Asia, South America, and Oceania show “coastal remoteness,” with differing amplitudes. Methodologically, the study innovates by combining NTL and RF to estimate large-scale regional GDP, offering dynamic, spatially explicit characterization of near/far economic aggregates and revealing generalized coastal remoteness versus coastal proximity under globalization. Policy should anticipate continued inland growth while managing coastal pressures, tailoring strategies to regional stages of development and transport contexts.

The analysis relies primarily on nighttime light data combined with official statistics, lacking integration of broader multi-source datasets (e.g., additional remote sensing, sensor networks, social data). Causal mechanism testing is limited; the study offers theoretical explanations (maritime transport advantages, industrialization stages) without detailed econometric driver tracing. Space constraints precluded deeper analysis of why movements are relatively small in regions such as Europe and Africa. Future work should integrate multi-source data and employ metrological/statistical analyses for more robust causal inference and finer-resolution policy guidance.