How Much Should We Trust the Dictator's GDP Estimates?

67 Pages Posted: 27 Dec 2017 Last revised: 6 May 2018

Luis R. Martinez

University of Chicago - Irving B. Harris Graduate School of Public Policy Studies

Date Written: May 1, 2018

Abstract

I study the manipulation of GDP statistics in weak and non-democracies. I show that the elasticity of official GDP figures to nighttime lights is systematically larger in more authoritarian regimes. This autocracy gradient in the night-lights elasticity of GDP cannot be explained by differences in a wide range of factors that may affect the mapping of night lights to GDP, such as economic structure, statistical capacity, rates of urbanization or electrification. The gradient is larger when there is a stronger incentive to exaggerate economic performance (years of low growth, before elections or after becoming ineligible for foreign aid) and is only present for GDP sub-components that rely on government information and have low third-party verification. The results indicate that yearly GDP growth rates are inflated by a factor of between 1.15 and 1.3 in the most authoritarian regimes. Correcting for manipulation substantially changes our understanding of comparative economic performance at the turn of the XXI century.

Keywords: GDP, nighttime lights, growth, autocracy, bias

JEL Classification: C82, D73, E01, H11, O47

Suggested Citation

Martinez, Luis R., How Much Should We Trust the Dictator's GDP Estimates? (May 1, 2018). Available at SSRN: https://ssrn.com/abstract=3093296 or http://dx.doi.org/10.2139/ssrn.3093296

Luis Martinez (Contact Author)

University of Chicago - Irving B. Harris Graduate School of Public Policy Studies ( email )

1155 East 60th Street
Chicago, IL 60637
United States

Register to save articles to
your library

Register

Paper statistics

Downloads
3,605
rank
2,379
Abstract Views
18,502
PlumX