This question has attracted the attention of many writers in political science and political economy. In political science this issue is often referred to as the ‘modernization hypothesis’ and associated with a famous paper by S. Lipset (1959) who answered the question affirmatively. He also suggested that other variables also associated with a higher per capita income such as urbanization and education had an independent role in bringing about democracy.
In the 50 plus years since Lipset’s classic paper, subsequent writers brought to bear statistical analysis on this question. Earlier findings suggested a positive association between the level of per capita income and indexes of democracy across countries. This finding holds whether electoral democracy is measured directly in terms of political rights using Freedom House’s index of political rights or using indirect measures of political rights, related to balance of power considerations, such as in terms of the Polity indexes. Almost all writers focused on electoral democracy when looking at this issue. Thus, they concentrated on political rights. Nonetheless, the same finding of a positive association between per capita income and democracy held for liberal democracy when civil liberties are measured by Freedom House’s index of civil liberties in one of the few studies that did so, Barro (1999). Click here to read Barro’s study.
This finding was questioned from a number of perspectives. In political science some suggested the need for a more historically oriented perspective through the use of analytical narratives rather than statistical methodology. Advances in the latter area, however, led to the use of panel data and the adoption of more advanced statistical methods in order to address this topic. In a well-known book on democracy and development four political scientists, Przeworski, Alvarez, Cheibub and Lmongi (2000), concluded that economic development in terms of per capita income did not lead to bringing democracy into existence or democratization but it generated stability for both democratic and authoritarian regimes. In the political economy area a widely cited paper by Acemoglu, Johnson, Robinson and Yared (2008) [AJRY], relying on panel data and somewhat more advanced methods, showed that the simple correlation across countries vanishes when one controls for country fixed effects and the potential mutual causation between per capita income and democracy. This has led to numerous subsequent contributions trying to restore the conventional wisdom associated with the original ‘modernization hypothesis’. All of these recent contributions, however, have relied on the electoral democracy view of the world and measured democracy only in terms of political rights.
Recently BenYishay and Betancourt (2013), relying on more recent data and more recent statistical methodology, show that the same statistical result as in AJYR above holds for both the political rights dimension and the civil liberties dimension and, thus, for both electoral democracy and liberal democracy. Perhaps more importantly, however, they find that the prior levels of civil liberties have a positive effect on the current levels of political rights whereas the prior levels of political rights have no statistically discernible effect on current levels of civil liberties. Summing up, the current empirical evidence raises serious doubts about the idea that economic development generates democracy. Does empirical evidence say anything about the idea that democracy leads to development? We will discuss this issue in a subsequent posting.
Read Unbundling Democracy: Tilly Trumps Schumpeter Ariel BenYishay & Roger Betancourt* October 2013