Using a panel of 140 countries over the 1975–۲۰۰۷ period, we disaggregate democracies across five institutional dimensions (government forms, electoral rules, state forms, number of veto players, and age of democracies), to study the precise forms of democracy that may explain the lower economic growth volatility (EGV) in democracies compared to dictatorships, usually emphasized by the literature. We find that, while all government forms decrease EGV to the same extent, proportional electoral rules outperform majoritarian and mixed electoral rules, suggesting a role for a more inclusive political decision-making process. In addition, EGV is significantly lower in unitary states, suggesting a role for a limited separation of power between the central government and the local authorities, while the effect of the number of veto players and the age of democracies is significant only in developed countries. Consequently, the choice between various forms of democracy may not be neutral for EGV, and, possibly, for countries’ development path.

Introduction

A large literature investigates the effect of democracy on economic growth. Theoretical analyses lead to ambiguous predictions. On the one hand, democracy can discourage economic growth, because of the distortionary effect of redistributive policies, and the possibility of political gridlock and interest-group politics. On the other hand, democracy can promote economic growth, through fostering investment in education and public goods, and more constraints on political leaders that limit the ability of politically-powerful groups to absorb most of the lucrative economic opportunities (see Acemoglu et al., 2014, for a recent discussion). In the context of conflicting theoretical results, empirical studies did not manage to reach a consensus about the effect of democracy on economic growth, as pointed out by the meta-analysis of Doucouliagos and Ulubasoglu (2008). 1 On the contrary, there is a much larger consensus for a stabilizing effect of democracies, which were found to reduce economic growth volatility (EGV) compared to dictatorships (e.g. Acemoglu et al., 2003; Mobarak, 2005; Klomp and de Haan, 2009; Edward and Thames, 2010). Such a favorable effect may be explained by the ability of democracies to ensure a stronger control over political leaders’ decisions, limiting the implementation of distorsive public policies and as a result the occurrence of internal shocks (for example, high inflation episodes, see Acemoglu et al., 2003), and to better manage redistributive conflicts caused by external shocks (for example, trade shocks, see Rodrik, 1999, 2000). However, to the best of our knowledge, no study focused so far on assessing the precise political institutions that could explain the stabilizing effect of democracies. This issue is important since Acemoglu (2005) and Acemoglu and Johnson (2005) show that the concept of democratic regimes indiscriminately gathers a set of economic institutions (e.g. the limitation of government’s expropriation power, as related to property rights) and political institutions (e.g. the various constitutional rules in place). To have a better understanding of the political institutions channels linking democracies to EGV, a more in-depth approach of democratic regimes is necessary. The goal of this paper is to go beyond the simple opposition between democracies and dictatorships, in order to evaluate what are the specific democratic political institutions that matter the most to explain the favorable effect of democracies on EGV emphasized by the existing literature. According to constitutional economics and political science literature, different institutional setups of democracies lead to differences in the extent of constraints faced by political leaders when implementing public policies, and in the inclusiveness of the political decision-making process (see e.g. Persson and Tabellini, 2003). In turn, this can have significant macroeconomic implications, especially regarding the way democracies are able to deal with economic shocks. Therefore, we might expect that different institutional configurations of democracies could lead to differences in their capacity to reduce EGV compared to dictatorships.