Liberalization and Contingent Democratization

Matthew Wilson
Associate Professor and Director of Graduate Studies
Department of Political Science
University of South Carolina

In this talk, Wilson argues that one reason why the determinants of democratization remain unclear, despite decades of research, stems from scholars' failure to account for the fact that many non-democratic regimes became more politically open but that not all led to democracy. Using newly published data on failed and successful episodes of democratization, Wilson and his co-authors demonstrate that the start of democratization (liberalization) and its end (transition) may have different causes. They reevaluate major questions concerning the impacts of economic development and elections on regime change. They model the probabilities of democratizing and transitioning to democracy, separately, on income and growth and then use an instrument to estimate them in a selection model. They argue that economic development makes democratization less likely but that it makes a transition to democracy more likely, conditional on democratization. They show similar findings with elections, which helps to explain why a sample that compares all non-democracies against democracies would fail to find a consistent relationship between elections and democratization. These initial findings encourage us to go forward by developing research agendas that leverage distinctions in the data to refine conclusions about the drivers of democratization.