2005年-世界发展银行全球_Spatial_Dimensions_of_Trade_Liberalization_and_Economic_Convergence__Mexico_1985-2002_37页_1mb

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

Core Content

This paper explores the spatial dimension of economic growth and convergence in Mexico from 1985 to 2002, focusing on the effects of trade liberalization and the North American Free Trade Agreement (NAFTA). It investigates whether the economic performance of Mexican states is influenced by their geographical location, particularly proximity to the U.S. border, and whether new spatial patterns of growth have emerged as a result of these policy changes.

Main Views

• Regional Economic Growth and Inequality:

  • From 1970 to 1985, there was a decrease in regional income dispersion (sigma convergence), but after 1985, a sharp increase in inequality occurred, coinciding with trade liberalization.
  • The traditional "rich North/poor South" dichotomy appears to dominate the spatial pattern of economic growth, but the paper suggests that this pattern is not as straightforward as initially expected.

• Trade Liberalization and Spatial Impact:

  • Trade liberalization, especially the reduction in tariffs and the implementation of NAFTA, is expected to shift economic activity towards the U.S. border due to lower transportation costs and access to foreign markets.
  • However, the paper finds that the correlation between growth and proximity to the border is not as strong as anticipated. Instead, the spatial patterns of growth appear to be influenced by other factors such as natural endowments and historical industrial clusters.

• Convergence Clubs and Spatial Clustering:

  • The analysis suggests that convergence clubs (groups of states that converge in income levels) have formed, but these clubs do not necessarily correspond to geographical regions.
  • The kernel density plots and transition matrices show that while some states have remained in similar income intervals, the overall pattern of spatial dependence is not as clear as the economic divergence.

• Spatial Econometric Tools:

  • The paper uses spatial econometric tools, including transition matrices and Moran's I statistic, to assess the spatial dimensions of growth and convergence.
  • These tools reveal that while there is evidence of spatial dependence, it is not as pronounced or structured as in other countries, suggesting a more complex and less predictable spatial pattern in Mexico.

Key Information

• Trade Liberalization Timeline:

  • 1985: Unilateral trade liberalization began with a significant reduction in import licensing and tariffs.
  • 1986: Mexico joined the General Agreement on Tariffs and Trade (GATT).
  • 1994: NAFTA was signed, leading to further reductions in tariffs and increased access to U.S. markets.

• Data and Methodology:

  • The study uses data from the Mexican National Institute of Statistics, Geography and Information (INEGI) for 32 states, with some corrections for oil revenue allocation and population data.
  • The data is analyzed using stochastic kernels, transition matrices, and Moran's I statistic to capture spatial patterns and convergence.

• Findings:

  • The kernel density plots show a shift from a single-peaked distribution to a multi-peaked one, indicating the formation of convergence clubs.
  • The spatially conditioned kernel suggests that while some regions (border and southern) show clustering of income levels, this is not a universal pattern across the country.
  • The Moran's I statistic confirms the presence of spatial dependence, particularly in the "bands" measure, but the degree of clustering is less pronounced than expected.

• Geographical Patterns:

  • The "rich North/poor South" pattern persists, but the growth of southern states relative to their neighbors suggests that they may not be as disadvantaged as previously thought.
  • Some southern states, like Chiapas and Oaxaca, have higher income levels relative to their neighbors, indicating that regional factors such as natural resources and infrastructure may play a significant role.

• Conclusion:

  • The spatial dimension of economic convergence in Mexico is complex and not solely determined by proximity to the U.S. border.
  • While there is evidence of spatial dependence and convergence clubs, the results suggest that the spatial patterns of growth are more nuanced and influenced by a combination of factors, including natural endowments and historical industrial locations.

Structure of the Paper

I. Introduction

• Overview of trade liberalization and its impact on regional economic growth in Mexico.
• Introduction of the concept of spatial convergence and divergence.
• Presentation of the research question: whether the spatial patterns of growth are influenced by proximity to the U.S. border.

II. Data

• Description of the data sources and corrections made to the data.
• Mention of the aggregation of the state of Mexico with Mexico DF (Federal District) to better reflect economic realities.

III. Identifying Regions – Covariance of Income Levels

• Use of kernel density plots and transition matrices to analyze income distribution over time.
• Identification of convergence clubs and their spatial implications.
• Discussion of the "twin peaks" phenomenon and its relevance to Mexico.

IV. The Spatial Dimension

• Analysis of spatially conditioned kernel density plots using different definitions of neighborhood.
• Discussion of the role of natural endowments and historical industrial locations in shaping regional economic performance.

V. Parametric Measures of Spatial Dependence

• Application of Moran's I statistic to assess the statistical significance of spatial patterns.
• Comparison of different spatial definitions (contiguity, bands, distance) in terms of their ability to capture spatial dependence.
• Findings suggest that spatial dependence is present but not as strong or structured as in other countries.

Conclusion

• The paper concludes that while trade liberalization has had a significant impact on Mexico's economy, the spatial patterns of growth and convergence are more complex than initially expected.
• The traditional "rich North/poor South" dichotomy is still present, but the growth of southern states relative to their neighbors suggests that other factors are at play.
• The use of spatial econometric tools provides a more nuanced understanding of the spatial dimension of economic convergence in Mexico.