POLITICIANS AT WORK: THE PRIVATE RETURNS AND SOCIAL COSTS OF POLITICAL CONNECTIONS

Authors


  • The editor in charge of this paper was Stefano DellaVigna.

  • Acknowledgments: We are grateful to the editor and three anonymous referees for very constructive feedback on an earlier version of this paper, and to Antonio Ciccone for constant help and advice. We also benefited from the comments of Tito Boeri, Juan Carlos Conesa, Pedro dal Bò, Francesco Decarolis, Christian Dustmann, Patricia Funk, Fabiano Schivardi, Giancarlo Spagnolo, Jaume Ventura, Melvyn Weeks, and seminar participants at Università Parthenope, EIEF, University of Milan, Collegio Carlo Alberto, EEA Meetings in Barcelona, University of Bologna, University College of London, Brucchi Luchino Workshop, Università Bocconi, and NBER Summer Meetings for very useful comments. All errors are our responsibility. The opinions expressed herein are our own and do not necessarily represent those of the Bank of Italy.

E-mail: federico.cingano@bancaditalia.it (Cingano); paolo.pinotti@unibocconi.it (Pinotti)

Abstract

We quantify the private returns and social costs of political connections exploiting a unique longitudinal dataset that combines matched employer–employee data for a representative sample of Italian firms with administrative archives on the universe of individuals appointed in local governments over the period 1985–1997. According to our results, the revenue premium granted by political connections amounts to 5.7% on average, it is obtained through changes in domestic sales but not in exports, and it is not related to improvements in firm productivity. The connection premium is positive for upstream producers for the public administration only, and larger (up to 22%) in areas characterized by high public expenditure and high levels of corruption. These findings suggest that the gains in market power derive from public demand shifts towards politically connected firms. We estimate that such shifts reduce the provision of public goods by approximately 20%.

Ancillary