How do dictators control politics? How do they share power with the ruling coalition? The Politics of Authoritarian Rule takes up these questions in a fascinating account of the internal politics of dictatorships. Milan W. Svolik begins with two observations about autocratic politics and how it differs from democratic politics: Actors in autocracies cannot rely on an independent authority to enforce agreements between them, and violence is the ever-present and final arbiter of political conflict. These factors, he argues, shape how dictators consolidate personal power, why they create democratic-looking political institutions, and, most importantly, how they bargain with elite supporters, including the military. In short, the transaction costs associated with political exchange in dictatorships are incredibly high, and bargaining games occur in an environment where one or more actors can resort to violence to get what they want. The book thus draws on a series of incomplete information games and newly collected cross-national data on authoritarian ruling coalitions, and it offers innumerable historical accounts of real-world politics in dictatorships.
The second chapter introduces Svolik’s data. While much of this chapter describes data on formal political institutions that can be found in other data sets, the real nugget is his discussion of why it is crucial to move beyond “regime typologies.” He notes that concepts such as military power or the extent of personalism vary in degree across all regimes, and remains unmeasured when researchers use exclusive categorical indicators of “regime type.” However, despite this welcome critique, Svolik still resorts to a typology to circumscribe military rule in his own data set, categorizing regimes as civilian, indirect, personalist, or corporate.
Svolik also discusses the problem of using the tenure of individual autocratic leaders as a stand-in for authoritarian durability. As the Institutional Revolutionary Party (PRI) regime in Mexico illustrates, dictatorships that solve leader succession problems by institutionalizing peaceful rotation of the executive use this mechanism to prolong their rule. Therefore, short leader tenures in these cases should not be interpreted as losing power but rather as successful maneuvers for promoting regime durability. This critique leads Svolik to delineate the myriad ways in which autocratic leaders leave power, and he settles on the concept of a ruling coalition—“an uninterrupted succession in office of politically affiliated authoritarian leaders” (p. 42)—as the most appropriate for assessing the durability of autocracies. He is thus able to separate the loss of power for the ruling coalition from instances when the regime simply reshuffles its leaders—for example, by enforcing executive term limits or by removing the leader in a soft coup.
Political scientists know very little about how dictators amass personal power, but Svolik makes considerable progress. This endeavor is not trivial because personalist dictatorships are responsible for much of the bad behavior typically associated with nondemocratic rule, such as conflict initiation, investment in nuclear weapons, and volatile economic performance.
In the third chapter, Svolik examines a model in which the dictator and elite supporters both want to maximize their share of a fixed amount of power; the dictator can take power from the elites, and the elites can threaten to carry out a rebellion to oust the dictator. The rebellion is costly because it can fail, which becomes more likely as the dictator gets stronger relative to the elite. Information is the central feature of the game: While the dictator knows how much he is taking from the elites, they can only imperfectly observe his actions. Thus, the dictator can renege on a power-sharing deal and the ruling coalition will not know for sure if he is reneging or complying. The model shows how successful power grabs by the dictator can endogenously yield what Svolik calls an established dictatorship, whereby the elites cannot credibly threaten to oust the dictator and thus cannot deter his opportunistic behavior. Further, he demonstrates why an environment in which the elite coalition has better information is less favorable to the chances of the dictator consolidating personal power: Better information reduces the dictator's ability to opportunistically renege on a power-sharing arrangement.
Power is incredibly hard to measure, which leaves Svolik to examine two relatively modest implications: some established dictatorships exist; and the longer leaders remain in office, the less likely they will be ousted by insiders. The empirical tests in this chapter are carefully executed and generally corroborate the implications. However, Svolik uses this second piece of evidence to conclude that dictators in established autocracies “typically enjoy long tenures and die in their bed” (p. 56). While the first part of this conclusion stems directly from the model and is easily tested, the second does not necessarily follow from the first. That is, leaders in established dictatorships are, at least historically, those most likely to suffer a particularly nasty fate should their regimes fall, as shown by Abel Escriba-Folch (“Accountable for What? Regime Types, Performance, and the Fate of Outgoing Dictators, 1946–2004,” Democratization 20 [2013]: 160–85). While Muammar Gaddafi, Mobutu Sese Seko, and Rafael Trujillo—typical examples of dictators who amassed considerable personal power over their support coalition—certainly had long tenures, they did not die peacefully in their own beds. This empirical mismatch stems from equating coups—many of which are actually nonviolent—with violent leadership changes and dismissing the primary ways in which established dictators fall: assassinations and violent rebellions. A quick look at the author’s data on leader exit confirms that while the chances of being ousted in a coup declines as leader tenure increases, the chances of being ousted by assassination, revolt, or civil war also increase.
Svolik’s model points to one explanation for the greater likelihood of established dictatorships in particular contexts: Dictators with less initial power relative to their supporters are less likely to attempt power grabs because the elites' threat of ouster is more credible when the latter are stronger. This conjecture, however, can only be tested with detailed historical data on the initial bargaining positon of the dictator, which we currently lack.
The fourth chapter examines how political institutions contribute to autocratic stability. Svolik models an interaction between the dictator and elite supporters in which the dictator observes the total size of benefits from joint rule—things such as “government revenue, bureaucratic appointments, and favorable policy” (p. 102)—but the elites observe only an imperfect signal of the size of the pie. Importantly, the elites might observe fewer benefits because either some exogenous crisis reduced the size of the pie or the dictator is stealing from them—but they do not know which. As in the prior chapter, elites deter the dictator from grabbing power by threatening rebellion. However, their deterrent against the dictator is blunt because elites do not have the information necessary to differentiate bad times from the dictator's opportunism. Echoing Douglass North (Institutions, Institutional Change and Economic Performance, 1990), Svolik interprets institutions as information that reveals the true size of the pie and thus enable elites to better monitor power sharing.
The model shows that formal political institutions aid power sharing when the dictator's power vis-a-vis the ruling coalition is not excessively high or low because in such situations, political institutions are either infeasible or not necessary for reducing uncertainty in the interaction between the dictator and the elite. This empirical implication, however, is not tested. Instead, the analysis shows that autocratic legislatures are associated with more durable ruling coalitions and fewer coups and uprisings. A more complete test of the model implications would demonstrate that political institutions promote stability in regimes where the dictator has not consolidated power over the support coalition—that is, primarily in non-established or non-personalist dictatorships.
The next two chapters take up the problem of authoritarian control: How does the dictator repress and/or co-opt the excluded population? To answer this question, Svolik examines how two types of organizations—the military and support parties—implement strategies of repression and co-optation.
In the chapter on the military, Svolik highlights the danger posed by mobilizing repressive organizations to deter threats from regime outsiders: Repression as a strategy of authoritarian survival bears the risk that the men with guns will turn on the dictator. In this model, the military holds an informational advantage over the dictator because the men with guns know whether they are willing to accept the dictator's preferred policy, whereas the dictator does not know this information. Thus, the military can lie to the dictator by signaling their unwillingness to comply with the dictator’s preferred policy, even when they would accept it. In doing so, the military can induce the dictator to give them more of what they want without having to resort to a coup attempt.
The dictator provides resources to the military to deter rebellion from below. Crucially, these resources increase both the probability that a coup succeeds and the chances that the military defeats a rebellion. This assumption is the source of dictators’ dilemmas, forcing them to trade one threat for another. However, this premise ignores the possibility that dictators can invest resources in coup-proofing strategies—such as mobilizing a supportive political party, creating a private security force, or personalizing the security sector—that increase the coercive capacity of the regime vis-a-vis threats from outsiders without raising the risk of a coup.
In the final substantive chapter, Svolik argues that support parties can enhance authoritarian stability by hierarchichally assigning services and benefits. Similar to a gang (or academic departments), authoritarian parties structure benefits such that a large pool of junior cadres (or graduate students) perform costly service in hopes of obtaining a prized position in the senior ranks in the next period. By tying future benefits to current service, the party induces a large group of junior officials to join and perform costly service. Because junior cadres have sunk costs in party service, their future benefits are tied to the future of the party, giving them a stake in its survival. However, the churn from lower to higher ranking positions in the pyramid only keeps the system afloat if members of the senior rank retire, thus ensuring that the promise of future returns to costly junior service remains credible.
Svolik mercifully eschews a lengthy discussion of democratic transitions, but not because his theories lack implications for the way that we think about this process. By focusing on the internal dynamics of authoritarian power he helps readers understand how dictators share power and how they exert political control. Since democratization is only one outcome that befalls dictatorship when it ends and is often not the main threat faced by dictators, we must first understand how they rule before we can explain why they fall and what comes next when they do.
The Politics of Authoritarian Rule not only makes a valuable contribution to the literature on authoritarianism but is a model for combining rich theory with precise empirical work, all while providing the reader with fascinating anecdotes to illustrate Svolik’s main points with real-world politics. I have little doubt that it will remain on many graduate syllabi, including my own, for years to come.
