Why Mexico's regional income convergence broke down

https://doi.org/10.1016/j.jdeveco.2004.03.009Get rights and content

Abstract

According to previous research, Mexico's trade reforms led to a divergent pattern in regional per-capita output levels. This research did not fully include Mexico's regional growth performance after 1994, when NAFTA started operating. In this paper, I extend the sample to include Mexico's regional growth patterns after this treaty was enacted and I explore what factors may account for this result. The findings suggest that the divergent pattern observed after 1985 was not reversed with NAFTA. The results also suggest that the winners from the reforms were those states initially endowed with, or able to attract, higher levels of human and physical capital and better infrastructure.

Introduction

According to previous research, the effects of Mexico's trade liberalization, which started in the mid-eighties when the country signed on to the General Agreement of Tariffs and Trade (GATT), were not uniform across its regions. The initially richer northern states that have a border with the U.S. seem to have reaped most of the benefits from the reforms, while the central and southern states hardly exhibited an improvement in their growth performance. As a consequence, the convergent pattern in regional per-capita output levels that was observed before the reforms apparently broke down after 1985. These studies, however, were mostly based on samples including only a short period of time after the reforms and, in particular, did not fully include Mexico's regional growth performance after the North American Free Trade Agreement (NAFTA) started operating in 1994. Moreover, most previous research has not explored in detail the forces that may have induced the loss of convergence across Mexico's regions.

In this paper, I extend the sample used by previous authors to include the post-NAFTA experience and explore what kind of factors may account for the loss of convergence in Mexico's regional per-capita GDP. The empirical evidence I present supports the hypothesis that, after 1985, β-convergence across Mexican states' per-capita outputs was lost, even if we control for differences in steady state per-capita output levels. Moreover, according to the results, the divergent pattern observed after that year was not reversed with the enactment of NAFTA. The findings also suggest that Mexico's reforms induced a structural change in its growth patterns, as new sources of growth related to trade and investment flows became available. Not many states, however, were prepared to take advantage of these new sources of growth. In particular, the winners from trade liberalization were apparently states initially endowed with, or able to attract, higher levels of human and physical capital and better infrastructure. This was especially true for the border region, which benefited additionally from its proximity to the U.S. market. In contrast, southern states, whose labor force is more concentrated in traditional agriculture and have the greatest deficiencies in human capital and basic infrastructure, seem to be the losers from trade liberalization.

The paper proceeds as follows. In the first section I summarize the existing literature concerning the regional effects of the reforms. Section 3 provides a general description of the evolution of per-capita output in Mexico's regions from 1970 to 2001. In Section 4, I study the existence of absolute and of conditional convergence across Mexican states. Finally, Section 5 summarizes the main conclusions.

Section snippets

Previous research on the regional response to trade liberalization

The literature has identified several mechanisms through which the trade reforms had heterogeneous effects across Mexico's regions. Among these, the most important seem to be the following.

First, trade liberalization altered the optimal location choice of manufacturing firms, promoting a break-up of Mexico City's manufacturing belt and a movement of economic activity towards the border with the U.S. Indeed, after Mexico opened up to trade in 1985, a dramatic reduction in Mexico City's share of

A regional perspective of Mexico's recent growth performance3

Table 1 provides a summary of the recent evolution of per-capita output across Mexican regions. There are several features that can be observed from the data. First, there was no important mobility from 1970 to 2001 in the relative position that the states had in terms of their per-capita GDP ranking. In particular, most southern and central states, which tended to be the poorest in 1970, are still the poorest states as of 2001. Second, the periods from 1970 to 1985 and from 1985 to 2001 are

Convergence across Mexican states

The analysis made in the previous sections suggests that the convergent behavior across Mexican states' per-capita output that was observed before 1985 broke down afterwards. As a consequence, a process of increasing economic differentiation across regions seems to have characterized the country from 1985 on. In this section, I test formally this hypothesis and assess if this result holds after controlling for differences in initial conditions faced by each state.

Several authors have analyzed

Conclusions

This paper suggests that, after 1985, growth patterns across Mexican regions became more sensitive to new sources of growth. In particular, the kind of spillover effects implicit in many endogenous growth models seems to have gained relevance to determine growth patterns within Mexico after 1985. It is important to emphasize, however, that the results do not necessarily imply a rejection of the neoclassical growth model. The results could also be consistent with the transitional dynamics of a

Acknowledgements

I thank Kate Antonovics, Julian Betts, Richard Carson, Rodrigo García Verdú, Theodore Groves, Gordon Hanson, Mark Machina, James Rauch, Julio Santaella, Chris Woodruff, the participants at the Wider/Cornell/LSE Mexico Conference on Spatial Inequality in Latin America held at Universidad de las Américas, Puebla, during November 2–3, 2002 and two anonymous referees for very helpful comments. I also acknowledge the Consejo Nacional de Ciencia y Tecnología (CONACYT) for financial support. The

References (38)

  • R. Barro

    Determinants of economic growth

    (1997)
  • R. Barro et al.

    Convergence

    Journal of Political Economy

    (1992)
  • O. Blanchard

    Convergence across states and regions. Comments and discussion

    Brookings Papers on Economic Activity

    (1991)
  • D. Brown et al.

    North American integration

    The Economic Journal

    (1992)
  • Chiquiar, D., 2003. Essays on the Regional Implications of Globalization: The Case of Mexico. PhD Dissertation,...
  • Chiquiar, D., Hanson, G., 2002. International Migration, Self-Selection and the Distribution of Wages: Evidence from...
  • G. Esquivel

    Convergencia regional en México, 1940–1995

    El Trimestre Económico

    (1999)
  • G. Esquivel et al.

    Sources of regional (non) convergence in Mexico

  • R. García Verdú

    Income dynamics across states in Mexico: 1940–2000

    (2002)
  • Cited by (0)

    View full text