2005年-世界发展银行全球_Spatial_Dimensions_of_Trade_Liberalization_and_Economic_Convergence___Mexico_1985-2002_34页_954kb

Summary of "Spatial Dimensions of Trade Liberalization and Economic Convergence: Mexico 1985-2002"

Core Content

This article investigates the spatial patterns of income and economic growth in Mexico from 1985 to 2002, focusing on how trade liberalization and the North American Free-Trade Agreement (NAFTA) have influenced regional divergence and convergence. It employs spatial economics techniques to analyze how economic activity has become geographically concentrated or dispersed.

Main Points

• Trade Liberalization and Divergence: Mexico's unilateral trade liberalization in 1985 and its entry into NAFTA led to increased income divergence across states. While there was a decrease in income dispersion from 1970 to 1985, a sharp increase in inequality followed.

• Income Clusters: The divergence has taken the form of several income clusters that do not fully align with traditional geographic regions. A "south" region is clearly identified, but the "north" is limited to U.S. border states, and a distinct "center" region does not emerge.

• Spatial Dynamics: The principal dynamics of economic divergence and convergence are not centered on the border but are driven by the underperformance of southern states and the superior performance of a convergence club in the north-central region.

• Theoretical Ambiguity: The relationship between trade liberalization and spatial patterns is not clear-cut. While some theories suggest that trade liberalization could lead to increased polarization, others argue that natural advantages and historical factors may anchor industries to certain regions.

• Transport Costs and Growth Poles: The role of transport costs and geographic proximity to the U.S. border is important but not deterministic. Some non-border states, such as Jalisco and Yucatán, benefit from natural endowments and access to water transport, suggesting that growth could occur outside the border.

• Empirical Evidence: There is little evidence that proximity to the border correlates with higher growth after 1985. Studies such as those by Rodriguez-Pose and Sanchez-Reaza (2005) and Esquivel (2000) show no significant relationship between distance and income or growth levels.

• Methodology: The paper uses sigma and beta convergence measures, as well as spatial statistics like the global and local Moran's I, to analyze regional income patterns and their evolution over time. Markov transition matrices are also used to assess the probability of states moving between income quintiles.

Key Information

• Data Sources: The analysis is based on GDP and population data from the Mexican National Institute of Statistics, Geography, and Informatics (INEGI), covering 32 states from 1970 to 2002.

• Data Adjustments: Oil production data was excluded due to inconsistencies in its allocation, and population data for Chiapas and Oaxaca was corrected for 1975–1988. Mexico State was merged with the Federal District of Mexico City due to strong labor market links.

• Results: The analysis reveals a high persistence in income distribution, with states maintaining relatively stable positions over time. The kernel density plots show that the distribution became more multimodal after 1985, indicating the formation of convergence clubs. The poorest states showed some convergence toward the mean, while the center of the distribution became depopulated.

• Conclusion: The geographic patterns of economic growth following trade liberalization are more complex than previously thought. While the north has seen some convergence toward U.S. income levels, the south has remained underperforming. The role of natural endowments and transport infrastructure is crucial in shaping these patterns, and the effects of trade liberalization may not be uniformly distributed across regions.

Authors and Funding

• Authors: Patricio Aroca (Universidad Católica del Norte, Chile), Mariano Bosch (PhD student at LSE), and William F. Maloney (World Bank).
• Funding: The research was supported by the Regional Studies Program at the World Bank.

References

• The paper references several studies including Barro and Sala-i-Martin (1995), Juan Ramon and Rivera-Batiz (1996), Esquivel (1999), Chiquiar (2005), and others.
• The authors also acknowledge helpful discussions and research assistance from various scholars and researchers.