Mastering Panel Metrics: Causal Impact of Democracy on Growth
- (pp. 77-82)
AbstractWe revisit the panel data analysis of Acemoglu et al. (forthcoming) on the relationship between democracy and economic growth using state-of-the-art econometric methods. We argue that panel data settings are high-dimensional, resulting in estimators to be biased to a degree that invalidates statistical inference. We remove these biases by using simple analytical and sample-splitting methods, and thereby restore valid statistical inference. We find that debiased fixed effects and Arellano-Bond estimators produce higher estimates of the long-run effect of democracy on growth, providing even stronger support for the key hypothesis of Acemoglu et al.
CitationChen, Shuowen, Victor Chernozhukov, and Iván Fernández-Val. 2019. "Mastering Panel Metrics: Causal Impact of Democracy on Growth." AEA Papers and Proceedings, 109: 77-82. DOI: 10.1257/pandp.20191071
- D72 Political Processes: Rent-seeking, Lobbying, Elections, Legislatures, and Voting Behavior
- C23 Single Equation Models; Single Variables: Panel Data Models; Spatio-temporal Models
- C33 Multiple or Simultaneous Equation Models: Panel Data Models; Spatio-temporal Models
- O43 Institutions and Growth