Failure or Success: The Resilience of Democratic Systems

• In Event: Panel II on Democratic Backsliding/Erosion

Abstract

The question who controls governments lies at the heart of a fundamental distinction between different types of democratic constitutions – parliamentarism, semi-presidentialism and presidentialism. The distinction between varieties of democratic constitutions has been a the core of how stable new democracies are and how we estimate their successful survival. Much of the extensive work on democratic survival suggests that this distinction affects the risk of democratic collapse although the literature focuses mainly on the perils of presidentialism, typically as compared to parliamentarism. The effects of constitutional variation within these regime types on democratic survival are generally not explored in this literature (Cheibub 2007). This focus of the is unfortunate because the latest wave of democratization has predominantly rendered semi-presidentialism – a much less studied regime type. As of 2018, 22% of the world’s democracies were semi-presidential (Cheibub 2018: 43). Moreover, the exclusive focus on differences between regime types is especially unhelpful with respect to semi-presidentialism because these constitutions are significantly more heterogeneous than presidential or parliamentary ones (Shugart 2006: 358).
This dearth of research is troubling because when semi-presidential democracies collapse the human consequences are often stark: The Weimar Republic’s demise was followed by the death and destruction caused by the Nazi dictatorship, the overthrow of democracy in Guinea-Bissau in 1998 precipitated a civil war that left thousands of civilians dead and displaced fully a third of the population, and the collapse of democracy in Haiti in 1999/2000, deprived the country of foreign aid, trapping Haitians in levels of poverty unparalleled anywhere else in the Western Hemisphere
In this project, I want to analyze if the constitutions of some semi-presidential democracies are less likely than others to contribute to such negative outcomes. Central to the survival of democracies is their constitutional capacity to process policy challenges and political conflicts effectively. Semi-presidential democracies combine a popularly elected president with an assembly to which the government and prime minister are accountable (Elgie 1999), so that control over the cabinet is a potentially contentious issue that can inhibit effective governance. Semi-presidential constitutions do not all regulate the difficult issue of government responsibility in the same manner, but divide into two sub-types: Premier-presidential constitutions make the government exclusively accountable to the assembly; while president-parliamentary constitutions provide for government accountability to both assembly and president. In what follows, I combine the emerging literature on bargaining about governments in these regimes with work on democratic survival and argue that premier-presidential regimes - in contrast to president-parliamentarism – give presidents incentives to bargain efficiently to secure the formation of cabinets that have mutual support from president and assembly. Moreover, when such cabinets cannot be formed, these constitutions limit the risk of crises of governability and institutional conflict by circumscribing the president’s power to challenge assembly control over the government. For both reasons, premier-presidential constitutions can be expected to process policy challenges and political conflicts better and to enhance the durability of democracy compared to president-parliamentary frameworks.
I want to test this argument using two structured, focused case studies and a large-n survival analysis of the full universe of semi-presidential spells of democracy from the first adoption of the regime type in 1919 to 2016. In identifying the universe of cases - semi-presidential democracies – I want to make use of the merits of both dichotomous and continuous conceptions and measures of democracy (Bollen and Jackman 1989). While dichotomous measures aim for conceptual clarity, continuous measures preserve flexibility, recognizing that constitutive features of democracy such as openness of participation or the fairness of electoral competition can vary by degree, and that the distinction between democracies and dictatorships can plausibly be drawn at a range of different levels. In investigating whether premier-presidential constitutions lower the risk of democratic failure, my dependent variable is the duration and outcome (failure or success) of each democratic episode measured in years. The explanatory variable is an indicator that records whether the constitution is premier-presidential. To the comparative literature on democratic survival my project should add a better and more nuanced understanding of semi-presidentialism and the findings should make clear that a fuller understanding of their impact on democratic survival requires attention to the constitutional variation within that regime type, in particular to the rules that structure government accountability.

Author

• Hartmut Lenz, SOKA University of Tokyo