Which Institutions Rule? Unbundling the Democracy-Growth Nexus

Over the past two decades studies of the causal impact of ‘institutions’ and ‘democracy’ on economic prosperity have occupied a prominent position in the cross-country growth litera-ture and within economics more broadly. While this body of work establishes a consensus that ‘institutions rule’ (over trade and geography) and that ‘democracy causes growth’, what has been missing in the debate is an attempt to systematically trace some tangible building blocks of these abstract ‘bundles’ driving the positive relationship with economic development. In this paper, we adopt an encompassing concept of ‘liberal democracy’, covering underlying po-litical and economic institutions, which we unbundle using the hierarchical data developed by the Varieties of Democracy project. We sketch how the incentives and opportunities as well as the distribution of political power created and shaped by these underlying institutions, in combination with the extent of the market, endogenously form an ‘economic blueprint for growth’, which is likely to differ across countries. Furthermore, political learning and insti-tutionalisation imply a non-linear growth effect of institutional change within countries over time. We overcome these challenges by adopting a heterogeneous treatment effects estimator which allows for non-parallel trends in the run-up to and endogenous selection into institu-tional change. Our results for each underlying institution are presented as a function of ‘time in treatment’ and conditioned on the evolution of ‘rival’ institutions, enabling us to interpret them as empirical horse-races. We find that freedom of expression, clean elections, and leg-islative constraints on the executive are the foremost institutional drivers of economic devel-opment in the long-run. Erosion of these institutions, as witnessed recently in many countries, may jeopardise the perpetual growth effect of becoming a liberal democracy we establish for the post-WWII period.We are grateful to Marc Chan, John Gerring, Simon Kwok, Daniel Treisman and conference session and seminar participants at Lund, Ghent, the 6th InsTED workshop in Nottingham, V-Dem Gothenburg, the 4th International Conference on the Political Economy of Democracy and Dictatorship in Münster, the Nottingham NICEP WIP seminar and the Case for Democracy conference in Brussels for useful comments and suggestions on earlier versions of the paper. The usual disclaimers apply