What Makes Democracy Endure? Kenya’s 2013 Elections and Lessons from Around the World
When Kenyans went to vote for their new president many around the world asked whether Kenya would descend again into the same violence which ensued 2007 election. This question in turn is inscribed in a wider dilemma: what allows a democracy to survive? If the survival of the democracy in Kenya was never seriously questioned continous post-electoral violence are one of the predictors of a democracy death. The striking point of any discussion about the fate of democracy is their powerlessness in predicting if a democracy will survive.
Law of Large Numbers. One of the problem with such analysis is that they zoom in on a specific country and they draw lessons that are said to be applicable to other context. It was the rise of Jihadists and the arrival of weapons from Lybia which degenerated the situation in Mali. It was the land issue which inflamed Kenyans communities in 2007, it was said. The problem with restricting the analysis to a small number of observations is that one cannot be confident enough that what one observe will play out again in other countries or in the future. To have a high degree of confidence in the average value of the sample on the other hand, one needs to have very big sample. This is what political scientist Adam Przeworski and colleagues did in their book Democracy and Development. They look at cross countries data and ask the following research questions: what are the conditions that determine whether democracy or dictatorship prevails? What causes political regimes to rise, endure, and fall? Can their transformations be explained generally, or are they caused by circumstances idiosyncratic to each country and period? This question in particular allows to understand whether one can predict the fate of democracies regardless of the context in which they are. When looks across countries and over time one finds out that what we know is very little.
Lessons from Around the World. Take for example income inequality: in democracies where income inequality is high, dominant social groups may seek recourse to authoritarianism the poor exercise their political rights to push for egalitarian reforms, as one of the hypothesis recites. The first finding is that there is no tenable finding: for many countries there is no information at all and for others only for scattered years; for other cases information about income inequality is collected with different methods and therefore is not comparable. When one wants to test if the level of income inequality causes regime transition one is left with only six observations of transition to dictatorship: keeping in mind the law of large number, one understands that any finding (e.g. democracy is more stable in more egalitarian societies) is very weak.
There are however some more robust findings: one is that democracy has the highest chances of surviving in affluent countries; in other words the level of income matters a great deal to predict the survival of democracy. What is not predictable on the other hand is what make democracy emerge: in fact it is not possible to identify any level of income threshold beyond which democracy emerges.
Another insightful observation is that when one put economic factors (i.e. level of income) in context by asking what is the impact of cultural, social and political conditions one finds out that economic factors continue to matter a great deal. As for social factors (e.g. ethno-linguistic fractionalization and religious heterogeneity) one finds out that their explanatory power is contigent upon the specification adopted (whether one includes one variable rather than the other). All in all however the authors conclude that the pattern is clear:
the level of economic development, economic performance, past regime instability (number of transitions to dictatorship in the past), and leadership turnover (changes of heads of government) tell almost all of the story.
Idiosyncratic Factors. Another important lesson is that such analysis works with chances and therefore necesseraly leaves out some other factors that matter. Some of these factors are idyosincratic that is specific to a particular context and therefore cannot be captured by a statistical analysis which looks for patterns across countries. In the case of Kenya political commentators have argued that the ability of political aristocracy to leverage identity has been a decisive factor in crystallizing the status quo and therefore in a certain sense to preserve the democratic order. When one asks whether democracy will survive in any specific countries one should keep these idionsyncratic factors.