Regional income inequality in Mexico, 1895-2010

Abstract

The motivation of this dissertation is multi-fold. Firstly, regional income disparity is widely considered to be a central concern among economists and policy makers. This responds to many facts. Usually, when regional specialisation takes place, only a few regions are able to attract modern industrial activity and high value-added services, causing an increase in regional inequality over the long term, as these activities generally enjoy increasing returns (which, in turn, makes this pattern very difficult to be reversed). Moreover, regional inequality is, all things being equal, highly correlated to inequality among individuals, which is also a very relevant issue for economists and policy makers. Lastly, regional inequality has high political relevance because it may be a source of political instability, which can result in social and economic crisis. Therefore, there is a great deal of scientific literature concerned with the evolution and causes of regional inequality.

The interest in regional inequality is shared by the Economic History literature, especially by that based on quantitative methods, which has developed a number of innovative research strategies to analyse the main forces behind the long-term evolution of regional inequality. However, this line of research has mainly focused on high-income industrialised economies, such as the US and some Western European countries, and there is still a significant gap in our knowledge of the long-term trends of regional inequality in low and middle-income economies. This leads to the second motivation of this project. Even though there has been some recent work on long-term regional inequality in middle and low-income economies, this is still a rather understudied field, where new hypotheses and interpretations –different from those developed for the industrialised countries– need to be developed. For instance, in developing countries, industrial location and agglomeration economies may not have had such a central role as drivers of regional income disparities. On the contrary, the influence of institutions and the location of natural resources may be much stronger. Likewise, dual economic structures (i.e., the coexistence of modern and traditional economic sectors) are much more common in developing countries than in industrialized economies. All this may make it necessary to adopt different research strategies in the analysis of regional disparities in developing countries. The study of Mexican regional inequality is representative of middle-income economies, where economic growth has had different roots and dynamics than in industrialised countries. In addition, Mexico has some characteristics that make it a particularly interesting case study. While the northern regions in Mexico share a huge border with the biggest world market, the US, the southern ones limit with one of the poorest region in the world, Central America. Also, it is a case in which very different forces have affected the long-term evolution of regional income inequality, such as factor endowments, factor mobility, natural resources, structural change, market potential and regional and development policies, which have affected regional disparities with varying intensity across the different periods of the late modern history of Mexico. Finally, the last motivation of this research lies beyond the Economic History frontiers. Mexico is living a period of increasing regional divergence, according to different indicators. Although this problem has been object of harsh public debate between the mid 20th century and the present (actually, the current federal government has announced a huge program to encourage economic growth in the poorest regions) almost no progress has been reached. According to the National Institute of Statistics and Geography (INEGI), in 2010 the GDP per capita of the richest state was 5.2 times as high as in the poorest state. The National Council for Evaluation of Social Development Policy (CONEVAL) estimates that 43 per cent of the total population living in extreme poverty in 2010 were located in 4 southern states. These figures stand out globally, not only in comparison to high-income countries but also to most middle and low-incomes economies. In fact, the ECLAC (2014: 73) has recently pointed out that Mexico has the second highest income ratio between the richest and the poorest regions among Latin American countries, only surpassed by Ecuador. In this regard, although the Mexican economy has a deep-rooted and historically persistent high regional inequality, the literature on regional disparities has focused mainly on the period from the 1980s and the end of ISI policies. By contrast, very little research has been done for the State-led industrialisation period (1930-1980), and none for the previous years of the First Globalisation. This dissertation aims to provide new quantitative evidence on the long-term evolution of Mexican regional income inequality, covering the period from 1895 to 2010. With this research, we hope to contribute both to the literature on Mexico and to the international debate on the main forces that explain the historical evolution of regional inequality.

El objetivo de esta tesis es ofrecer nueva evidencia cuantitativa sobre la evolución de las desigualdades regionales en México en el largo plazo (1895-2010). Con esta investigación se espera contribuir al debate de la literatura internacional sobre las principales fuerzas que explican la evolución histórica de las desigualdades regionales. El estudio de las desigualdades regionales en México es representativo de las economías de niveles medios de ingreso, en las cuales el crecimiento económico en el largo plazo ha tenido una ruta distinta a las economías industrializadas. En el caso mexicano, distintos determinantes han afectado la evolución de las desigualdades regionales, como la dotación de recursos naturales, la movilidad de factores de producción, el cambio estructural, el potencial de mercado, entre otros. En este sentido, los principales resultados de esta tesis son los siguientes. La evolución de las desigualdades regionales en México ha seguido una forma de ‘N’ en el largo plazo, la cual está estrechamente vinculada a las distintas estrategias de crecimiento económico adoptadas en México desde finales del siglo XIX. Así, durante el periodo del modelo agro-exportador (1895-1930), y el periodo reciente de apertura económica (1980-2010), las desigualdades regionales han incrementado, mientras que, durante el periodo de la IDE (1940-1980), las desigualdades regionales experimentaron una fase de convergencia. La primera fase de divergencia (1895-1930) estuvo liderada por los estados ricos volviéndose más ricos, mientras que los estados pobres se volvieron aún más pobres respecto a la media nacional de ingresos. El periodo de convergencia subsecuente estuvo caracterizado por la caída relativa de los estados ricos hacia la media nacional. Por último, durante el último periodo apertura económica de 1980 a 2010, la divergencia regional ha estado guiada por el mayor dinamismo económico de algunas regiones en particular del centro y norte del país, y en particular, de la Ciudad de México. El análisis de clúster espacial muestra que el único clúster que aparece constante durante todo el periodo de estudio es el de los países con bajos niveles de ingreso del sur del país. Los principales determinantes detrás de los cambios de las tendencias de las desigualdades regionales en el largo plazo han cambiado en cada periodo histórico. Durante la primera globalización, un desigual proceso de cambio estructural a nivel espacial explica el incremento de las desigualdades regionales. Aquellas regiones que pudieron beneficiarse de la integración de los mercados internacional lograron una tasa mayor de cambio estructural, mayor ratios de capital-trabajo y, por consiguiente, mayores tasas de crecimiento de la productividad laboral. Por otra parte, la convergencia experimentada de 1940 a 1980 se explica por un fuerte proceso de movilidad de factores (mano de obra) entre las regiones de México. Por último, el periodo de divergencia experimentado a partir de 1980, ha estado determinado por una distribución espacial desigual de la IED (concentrada en la Ciudad de México y los estados del norte del país), y la concentración espacial de los servicios de alto valor agregado (en la Ciudad de México).