This post was written by Pierre-Guillaume Méon & Khalid Sekkat, Centre Emile Bernheim – Université libre de Bruxelles (U.L.B.), on their ongoing research “A time to throw stones, a time to reap: How long does it take for democratic reforms to improve institutional outcomes?“
Democracy, institutions and growth
The Arab Spring by ousting authoritarian regimes raises hopes and expectations of better wealth and inclusiveness. Scientific analyses show, however, that democratization alone does not guarantee economic success. The better quality of institutions that is expected to follow democratization would improve economic performance, inclusiveness and effective accountability of rulers. While the outcome of the process started by the Arab Spring is still uncertain, studying other processes of democratization around the world may shed light on its potential impact on the quality of institutions in Arab countries.
A number of breaking path researches (e.g. Barro, 1991 and 1996 and La Porta et al., 1999) has shown that democracy does not guarantee economic success. At the same time, however, a flurry of studies established the importance of the quality of institutions for growth and development (Keefer, 1993 and Mauro, 1993). The relation is not simply a temporal or spatial correlation but reflects a causal linkage running from the quality of institutions to growth and development (Hall and Jones, 1999 and Acemoglu et al., 2001).
The apparent disconnection between the wealth of evidence of a positive impact of institutions on growth and the mixed evidence about the effect of democracy on growth has been clarified latter on (e.g. Glaeser et al., 2004) by distinguishing between the institutional change and its outcomes. Democratization is an institutional change that needs first to affect the functioning of institutions in order to affect growth and inclusiveness.
What international experience tells about democratic transitions and institutional outcomes
Our on-going research seeks to address the above issue. To do so, we apply a method of analysis of institutional quality change (Rodrik and Wacziarg 2005, Papaioannou and Siourounis, 2008, and Méon et al., 2009) that studies the evolution of institutional quality indexes around episodes of democratic transitions in a panel of 39 developing countries over the period 1984-2010.
The preliminary findings are summarized in Figure 1. Institutional quality is measured by eleven components of the International Country Risk Guide index. Democratic transitions are defined according to the PolityIV dataset, as the ending of an established polity and the beginning of a new and more democratic polity (Marshall et al., 2011).
The Figure displays the point estimate and the confidence interval (i.e. the margin of error) of the average effect of democratic transitions on institutional quality over time. It shows that democratic transitions are on average followed by an improvement in institutional outcomes. The effect is clearly positive and trends upward. Moreover, the effect becomes definitively significant six years after the transition. Democratic transitions therefore do improve institutional outcomes.
However, a finer analysis of the data reveals that all the eleven components/dimensions of the institutional framework do not react in the same way. Government stability and the role of the military in politics drive most of the effect. Conversely, bureaucratic quality, corruption, and ethnic conflicts seem to respond little.
Those results are to some extent good news. They suggest that democratic transitions on average improve institutional outcomes, even though it takes them a few years to do so. Furthermore, we could observe virtually no adverse effect of democratic transitions on institutional outcomes.
Those results nevertheless convey a word of caution. While democratic transitions on average improve some institutional outcomes, they are ineffective in improving other outcomes, such as bureaucratic quality, the control of corruption, or ethnic conflicts. Policy-makers wishing to improve those outcomes will therefore not be able to passively rely on the virtues of democratizations.