Modernization Theory valid for Latin America? A study from the 19th to the 21st centuries

Abstract

Introduction: We investigate the relationship between economic development, measured by real GDP per capita, and levels of democracy in Latin American countries from 1870 to 2010, empirically testing the Seymour M. Lipset hypothesis proposed in "Some Social Requisites of Democracy: Economic Development and Political", published in 1959 by The American Political Science Review. Materials and Methods: An econometric model known as an ordered probit was estimated. Democracy data were obtained from the measurement made by Polity IV. For the GDP per capita, the Angus Maddison Project database was used. The proxies for human capital were built by Morrisson and Murtin and are described in the article "The Century of Education" (2009). Results: The Modernization Theory is valid for countries with an intermediate level of democracy. For this group of nations, there was also a non-linear effect; that is, those poorer countries, when faced with an increase in GDP per capita, are more likely to have a lower level of democracy. However, this probability is inverted as GDP per capita expands. Discussion: The work simultaneously fills three gaps in the published works on the theme. (i) Contrary to the way most of them use the proxy for democracy, in our study it is not a binary variable. (ii) The time horizon is broader, since we have worked with data since the end of the 19th century. (iii) The regional scope, given that we take into account the particularities of Latin America. Our results partially prove the theory of modernization in Latin American countries, in addition to identifying which levels of democracy would be most likely in the face of a high level of income.