Most attempts at democratic government end with a reversion to authoritarian rule. During the cold war, those reversals usually occurred at the hands of the military, via coups d’etat. Since the end of the cold war, however, the agents of those changes have usually been the winners of the last democratic elections.* In rare cases, incumbent officials have taken steps that effectively ended democratic government on the spot, usually by declaring a state of emergency and announcing plans to rule by decree. Alberto Fujimori did something like this with his 1992 autogolpe in Peru, and so did Boris Yeltsin with his aggressive response to Russia’s constitutional crisis in 1993. More often, though, elected officials have dismantled democracy in a piecemeal fashion, tweaking electoral procedures, haranguing opponents, and diminishing civil liberties to such a degree that ensuing elections can no longer be regarded as fair or even competitive.
What forces tempt democratically elected representatives to destroy or dismantle the system that put them in office in the first place? This question is crucial to understanding why and how attempts at democracy fail, but most studies of democratic breakdown don’t look very hard for answers to it. Instead, they either take it as axiomatic that office-holders maximize on political survival—in other words, that they want nothing more than to retain their posts as long as possible—or they wave in the direction of the material and emotional benefits that might accrue to the controllers of state power as a way to explain the predilection for political survival and move on from there.
Assumptions and vagaries do not provide a satisfying response to a question of enormous practical importance. If we can understand deeply why some elected officials choose to fix the game in their own favor while others don’t, we might be able to better support processes that help to reduce that temptation, or to recognize when it’s going to be especially difficult to do so. Politicians are people, too, so they presumably weigh the pros and cons of their decisions like the rest of us. State power can surely be lucrative, but the extent to which that’s true varies widely across countries and over time; the payoffs that make power directly lucrative to politicians are often illegal; and there may be better-paying careers outside of politics that well-connected individuals could also be pursuing. To deeply understand democratic breakdown and do better at preventing it, we need to try harder to understand how elected officials in new democracies weigh these trade-offs and wind up making the decisions they do.
Unfortunately, the illicit sources of the material benefits that sometimes accrue to office-holders make it not just hard but downright dangerous to research this question. Incumbents might pad their own incomes or convey valuable benefits to family and friends in many ways. Some, like pay raises for themselves and bureaucratic appointments for close associates, are legal and visible. The more lucrative ones, however, are often illegal—from taking bribes and skimming state funds to controlling real-estate markets or collaborating with organized criminal gangs. Because these transactions are illegal, they are usually hidden, and the parties involved may take extraordinary measures to keep them concealed. To discover these kinds of transactions and establish their effects on politicians’ behavior, we need the skills and courage of an investigative journalist, and those are risks most social scientists (including me) are rarely willing to take.**
Of course, the material benefits of holding office are not the only thing shaping elected officials’ behavior. My own theoretical model of democratic breakdown shows that incumbents’ perceptions of the threats posed by other organizations can also be very influential. Fear that political rivals or the military will snatch power through rebellions or coups can create strong incentives to strike first in order to preempt those rivals’ actions. Under certain circumstances, these fears can even loom large enough to tempt elected officials who would otherwise prefer to see democracy survive to rig the system in their own favor.
Again, though, empirical researchers confront a serious challenge. Now the object of interest is an elected representatives’ perceptions of political threats, and from the little I know about human psychology, I would guess that it’s virtually impossible to get reliable statements about those perceptions from interviews with principals. Politicians are often willing to talk about those perceptions—in fact, comments about the threat rivals pose to one’s own interests are an important part of normal political discourse—but it would be woefully naïve of us to take those statements at face value. That’s even true after the fact, when public figures may still have incentives to spin history in ways that favor their own reputation. So once again, our theories can identify something that ought to be an important contributor to the risk of democratic breakdown, but we can’t directly and reliably measure the thing in order to test the theories.***
I don’t know how to fix this, other than to call for more attention to the problem and hope that more students, scholars, and journalists interested in democratic breakdown take it upon themselves to do the kind of intensive case research that could identify these mechanisms in action. If we can deepen our understanding of how democratically elected officials benefit materially and socially from holding office and how they perceive threats from rivals, we might be able to give new democracies better chances for survival by shaping those incentives in smarter ways.
* See my recent guest post on The Monkey Cage for more information about the evidence behind these statements.
** For one academic study that digs much deeper in this direction than most, see Christoph Stefes’ comparative analysis at the effects of clientelism and corruption on the developmental trajectories of post-Soviet Armenia and Georgia, Understanding Post-Soviet Transitions: Corruption, Collusion, and Clientelism (2006). To try to get principals to talk honestly about official corruption, he used the “snowball” sampling method, which he describes on page 8 as follows: “I started with interviewing foreigners who dealt with corruption as part of their professional activities and who had lived long enough in either Armenia or Georgia to be acquainted with knowledgeable locals. These locals, who I interviewed in the second round, were usually businesspeople, human rights activists, journalists, and lawyers. These individuals further introduced me to colleagues who constituted the third round of interviewees. Finally, and most importantly, locals knew officials in the state apparatus who were willing to share their experiences—after a proper introduction and a promise that their names would not be revealed. ”
*** For a study that acknowledges these problems and still manages to use survey research and interviews with principals in a compelling way, see Gerard Alexander’s 2002 book on democratic consolidation in Europe, The Sources of Democratic Consolidation.