A nice animation based on a paper by Adam Meirowitz (Princeton) and Joshua Tucker (New York University). The video explains why citizens who take action to overthrow bad governments might eventually lose hope  and decide to stay home.

The paper was published this year in the American Journal of Political Science. Here is the abstract:

In the aftermath of the Arab Spring, a crucial question is whether popular protest is now likely to be a permanent part of Middle Eastern politics or if the protests that have taken place over the past two years are more likely to be a “one-shot deal.” We consider this question from a theoretical perspective, focusing on the relationship between the consequences of protests in one period and the incentives to protest in the future. The model provides numerous predictions for why we might observe a phenomenon that we call the “one-shot deal”: when protest occurs at one time but not in the future despite an intervening period of bad governance. The analysis focuses on the learning process of citizens. We suggest that citizens may not only be discovering the type or quality of their new government—as most previous models of adverse selection assume—but rather citizens may also be learning about the universe of potential governments in their country. In this way, bad performance by one government induces some pessimism about possible replacements. This modeling approach expands the formal literature on adverse selection in elections in two ways: it takes seriously the fact that removing governments can be costly, and it explores the relevance of allowing the citizen/principal to face uncertainty about the underlying distribution from which possible government/agent types are drawn.

The authors’ highlighting of the need to “get it right” in the first transitional elections comes as a reminder of how frail electoral democracy may be in recently transitioned countries. A snapshot from the conclusion (ungated version):

In the aftermath of events like the Arab Spring and the Colored Revolutions, we are once again confronted with the question of what the long-range effects are likely to be of regime change ushered in by citizens taking to the streets. Such moments are often filled with optimism about the potential for people to seize control of their own destiny and finally demand accountability from their governments. The model we present in this article does not deny that this optimistic chain of events can come to fruition, but it does lead to important insights as to when such a scenario is more or less likely.

First, there is likely to be a fundamental difference between what happens following a protest that leads to regime change as opposed to one that leads merely to a change of government without changing the regime. So a protest in France that leads to a change in the government—for example, the 1968 protests and subsequent resignation of de Gaulle (Cerny 1970)—is not likely to change citizens’ minds about the overall distribution of the quality of governments available through French democracy. Citizens may feel that they have received a “bad” draw from this distribution, and thus it is worth trying to throw the current government out of power and replace it with a new draw from the same distribution. In these circumstances—provided the cost of protesting remains constant—there is no reason not to expect French citizens to take to the streets to protest against their government in similar numbers in thefuture.

In a new democracy, however, we suggest that citizens may be simultaneously updating about the quality of the current government and the potential quality of all governments available under a democratic regime. This learning may be rather dramatic during the first governments of the new regime. If these initial governments after a democratic transition should prove to be little better than previous nondemocratic governments, citizens might be unlikely to take costly actions in order to curtail new abuses or even deter a blatant disregard for the new rules calling for periodic elections. Thus, we are probably most likely to see a one-shot-deal scenario when the initial governments following a democratic transition are perceived as no better than the ones that preceded the transition. Conversely, repeated protest might be more likely if there are afew “good” governments—thus leading respondents to think that the average quality of government is indeed better under democracy—before a government that might threaten democracy appears again. The bottom line is that if the type of learning that underlines our model is indeed occurring in newly democratic regimes, then the stakes for these regimes during the initial post-transition governments are likely to be high indeed.

Taken together, these points suggest the potential for the following patterns to emerge. The size and strength of protests may increase over time in a nondemocratic regime, but once the initial goal of removing the old regime from office is accomplished, the ability of prodemocracy forces to bring their supporters to the street may diminish significantly. The intuitive explanation for such a pattern would normally be that “the goal has been met,” but our model suggests a different story: it may not be so much that the goal has been met as the goal no longer seems to be quite so valuable. Again, this points to the importance of “getting it right” in the first transitional elections.

You can read the full paper (ungated version) here [PDF].

HT The Monkey Cage.