By Matteo Cervellati, Piergiuseppe Fortunato and Uwe Sunde
The mass movement that ousted Tunisia’s President Zine al-Abidine Ben Ali was the first case in modern history in which an Arab leader was toppled by a popular uprising. This unprecedented event has triggered similar protests in neighbouring countries and led to the overthrow of President Hosni Mubarak in Egypt, to turmoil in Bahrain, Algeria, and Yemen, and to an ongoing civil conflict in Libya.
The concept of a new “Arab awakening” as being the spread of democratic ideas has been evoked by Laurence Pope in The New York Times (2011)among others. But what are the chances of having truly democratic and inclusive societies in these countries? The historical evidence suggests that a key part of the answer may lie in the actual modes of regime change and, in particular, in the level of violence associated with the democratic transition.
Similar determinants but different paths to democracy
Tunisia’s “Jasmine revolution” and the upheaval in the other Arab countries have their roots in the growing discontent of the economically and politically disadvantaged segments of the population. In particular the fast population growth has led to widespread frustration among the young generations, with youth unemployment rates approaching 30% in many countries of the region. These tensions have been reinforced by the recent global economic crisis and the economic difficulties throughout the region. In a Vox column just over two years ago, Ciccone (2009) reported research suggesting that a negative income shock of 5% raises the likelihood of civil conflict by 15 percentage points. In the case of the Arab countries, the short-term economic downturn was also coupled with sharply rising food prices that have seriously threatened food security.
Despite sharing similar structural determinants of discontent, it is already apparent that the (attempted) regime transitions are taking very different paths in different countries. While in some cases the mass movements have succeeded in opening up the possibility for a rapid and peaceful regime change, in others the path towards democracy is faced with stronger resistance by parts of the ruling elites and appears longer, more uncertain, and stained with blood.1 Will these different transition modes make a difference for the features of the emerging democracies?
Assessing the impact of the transition scenario
Recent academic research suggests that, unfortunately, this may well be the case.
Samuel Huntingdon (1993) provides an analysis of regime transitions during, what he calls, the “third wave” of democratisation. He concludes that violent uprisings are less likely to lead to a regime change and, if it the shift does take place, they may lead to worse democracies. In a 2007 paper (Cervellati et al. 2007), we develop a theory of endogenous democratisation that analyses the determinants of different transition scenarios and their consequences for the features of emerging democracies.
This theory predicts that transitions to democracy under a broad consensus in the population foster the establishment of democracies with good institutional quality, because the individual behaviour can be coordinated along the transition process. In more recent work Cervellati et al. (2011) broaden this view by investigating the incentives of the different social groups to engage in violence, and by studying the role of openly violent conflict during the democratisation process for the features of the emerging democracies. The institutional quality of democracy being determined endogenously, the theory predicts that violent transitions to democracy are less likely to lead to “good quality democracies” with large protection of economic and political liberties. The results also suggest that a minimum consensus among the population is required for the emergence of a well-functioning democracy and that peaceful transitions are more likely if the economic interests of the population are fairly aligned. This is the case if, e.g., the lootable (natural) resources are not a major factor of the economy, and if economic power is not concentrated among very small oligarchic elites.
The prediction that the level of violence which characterises the regime transition may persistently affect the future prospects of democracy is consistent with the empirical evidence from the “third wave” of democratisation over the period 1970-2003. Using information about civil conflicts of different types and intensity from the Peace Research Institute in Oslo (Harbom and Wallensteen 2010) and information on the level of violence during democratisation provided by Freedom House (2005), we estimate the effect of violence during transition on the quality of the emerging democracy in terms of civil liberties.Extending the empirical strategy of Papaioannou and Siourounis (see their 2008 Vox column) considering the differential effects of violent and non-violent transitions and applying it to civil liberties, we are able to look at within-country variation in the timing of democratic transitions.
The comparison of civil liberties in countries that experience a peaceful democratisation or a violent transition to democracy to countries that do not experience a change in their political system reveals that the peaceful route leads to a substantially, and persistently, larger improvement in civil liberties.
Figure 1 illustrates this finding by plotting the evolution of average civil liberties before and after a permanent democratisation.2 Lower levels of the index depicted in the graph imply better civil liberties. The plot distinguishes between countries with a violent transition to democracy and countries in which the transition to democracy was peaceful. While the overall pattern in both groups is similar, there is a clear difference in the level of civil liberties after the transition. After a non-violent democratisation, civil liberties are about one index point better than the world average, corresponding to an improvement of about 1.5 index points.3 Countries with violent transition experience a much less pronounced improvement. The effect of violence during the transition does not vanish during the decade after the transition.
These findings are robust when conditioning on a rich set of control variables and emerge consistently throughout different specifications, and when conflict incidence in a particular year or lagged institutional quality are added as additional controls. In terms of the determinants of democratisation scenario, the evidence also suggests that higher inequality before the transition is associated with a significantly larger likelihood of violent civil conflicts during the regime change. These results complement previous research that predicts and documents significant interactions between inequality and political freedom for the quality of democracies (Cervellati et al, 2008 and Sunde et al, 2008).
Our research investigates the determinants and consequences of democratic transitions. The results suggest that peaceful transitions to democracy lead to democracies with better average protection of property rights and civil liberties as compared to democracies that emerge after violent conflicts. The evidence from the “third wave” of democratisation also suggests that countries that rely less on natural resources and have lower inequality are more likely to experience nonviolent democratic transitions. Concerning the prospects of the new Arab awakening, it is by now clear that the regime shifts will follow different transition paths in different countries. Taking the experience from the third wave of democratisation seriously, the prospects for the countries currently involved in the most violent conflicts do not look as bright.
Professor of Economics, University of Bologna; Research Fellow, IZA
Economist, United Nation Conference for Trade and Development
Professor of Economics, University of St. Gallen and CEPR Research Affiliate
The views expressed in this article are those of the authors and do not necessarily represent the views of the United Nations Conference on Trade and Development (UNCTAD) Secretariat or of UNCTAD Members.
Cervellati Matteo, Fortunato, Piergiuseppe and Uwe Sunde (2007), “Consensual and Conflictual Democratisation, Rule of Law, and Development”, CEPR Discussion Paper 6328
Cervellati Matteo, Fortunato, Piergiuseppe and Uwe Sunde (2008), “Hobbes to Rousseau: Inequality, Institutions and Development”, Economic Journal, 118(531): 1354-1384.
Cervellati Matteo, Fortunato, Piergiuseppe and Uwe Sunde (2011), “Democratisation, Violence, and the Quality of Democracies”, CEPR Discussion Paper 6618.
Ciccone, Antonio (2009), “Sudden impoverishment as a trigger of civil conflict” VoxEU.org, January 2.
Freedom House (2005): How Freedom is Won: From Civic Resistance to Durable Democracy, Karatnycky Adrian (ed.), New York.
Harbom, Lotta, and Peter Wallensteen (2010), “Armed Conflict, 1946-2009”, Journal of Peace Research, 47(4):501-509.
Huntington, Samuel Phillips (1993), “The Third Wave: Democratisation in the Late Twentieth Century”, University of Oklahoma Press, Oklahoma.
Papaioannou, Elias and Gregorios Siourounis (2008) “Democratisation and growth” VoxEU.org, 25 October.
Pope, Laurence (2011), “The Second Arab Awakening”, The New York Times, February 18.
Sunde, Uwe, Cervellati, Matteo, and Piergiuseppe Fortunato (2008), “Are All Democracies Equally Good? The Role of Interactions Between Political Environment and Inequality for Rule of Law”, Economics Letters, 99(3),552-556.
1 The UN estimate more than 300 deaths in Egypt and 200 in Tunisia. At the time of writing, the number of casualties in Libya appears substantially larger and already exceeds 1,000.
2 The index takes values from 1 to 7, with 1 representing the most free and 7 representing the least free. The data are available for the period 1972-2003. The civil liberties are time demeaned to capture global trends and then averaged over the respective pre- and post-transitional periods.
3 With a mean (standard deviation) of the index of civil liberties is 4.02 (1.94) in the sample, a transition to democracy improves the civil liberties by about three quarters of a standard deviation.