External Seminar by Markus Eberhardt

03-10-2019 from 11:30 to 12:45
Buffer room, Sint-Pietersplein 7, 2nd floor
Department of Economics
Add to my calendar

Title: Facets of Democratic Change and their Causal Effect on Economic Growth

By Markus Eberhardt (University of Nottingham)

Please register by Monday 30 September 2019 at https://doodle.com/poll/ehm6stswtm5w8xfc


What are the economic consequences of changes to the system of governance? Existing empirical studies of the growth effect of democracy have yielded mixed results, though more recently a consensus of moderate to large positive effects has emerged. While these studies cover a wide range of methods and specifications, we note the absence of three important aspects: First, the specification of a heterogeneous parameter model, which allows for a differential democracy-growth nexus across countries. Such heterogeneities can be motivated with arguments for democratic legacy or threshold levels in economic or human development as necessary conditions for positive democracy effects. Second, the adoption of methods which capture the effect of time-varying unobserved heterogeneity (strong cross-section dependence), which can arise from common shocks and/or from spillover effects. Third, flexibility in how the democracy-growth nexus is modelled: existing studies either assume a linear effect of democracy or pick ad hoc single thresholds to distinguish ‘democratic’ from ‘autocratic’ regimes. Our analysis of up to 159 countries (1960-2017) captures these aspects in dynamic heterogeneous panel regressions employing Varieties of Democracy (V-Dem) indices and regime dummies derived thereof. Our models assume (i) a linear democracy effect, (ii) a single threshold effect, or (iii) dual threshold effects. For the linear model we adopt the IVMG estimator by Norkute, Sarafidis, Yamagata, and Chu (2019) which uses de-factored regressors as instruments for the endogenous democracy index. For the single threshold model we adopt a difference-in-difference estimator by Chan and Kwok (2018) where endogenous selection into democracy and common shocks to ‘treated’ economies are accounted for by the inclusion of proxies for the latent factors constructed from ‘control’ economies (see Eberhardt, 2019). The dual threshold model is a simple extension of this estimator, allowing for an S-shaped relationship between democracy and long-run development, acknowledging the notions of a minimal level of and of diminishing returns to democracy.

Find more events: