The book tackles an interesting puzzle: Why would pseudodemocrats invite international election observers and cheat in front of them if getting caught is costly? Susan D. Hyde examines the decisions of leaders to invite observers and shows that the international community offers real benefits to those who invite election monitors (Chapters 2 and 3). But observers make it difficult for leaders to cheat (Chapter 4), and moreover, getting caught cheating brings consequences. Pseudodemocrats may try to become more sophisticated in their cheating (Chapter 5), but this has not prevented the emergence of election monitoring as a norm. So a norm has developed not because of the efforts of democracy advocates or policy emulation, but because of leaders' attempts to signal to the international community their democratic credentials, even if they are not really committed to democracy. Hyde establishes a strategic tension for pseudodemocrats that is straightforward and compelling, and she provides an impressive array of empirical evidence to support her claims: cross-national time-series data, within-country experimental evidence, and qualitative accounts from hundreds of observer reports. The Pseudo-Democrat's Dilemma is representative of an increasing amount of work that bridges comparative politics and international relations. It is a must-read for those interested in elections and the development of norms.
To make her argument, Hyde uses a signaling model in which leaders are either “good” or “bad” types. “Good types” are those leaders who hold elections with the intention of abiding by the rules. Rulers are “bad types” because they hold elections with no intention of ceding power. Leaders want to signal their type to pro-democracy states within the international community because those states reward good types and sanction bad ones. Signaling one's type “requires inviting international observers and receiving their endorsement of the election” (p. 13). Signaling, however, induces bad types to also invite observers—in the hopes of passing themselves off as good types so that they may receive the benefits offered by the international community. Nevertheless, this type of pooling between the good and bad types leads to the spread of the practice of inviting election observers. Hyde's argument is potentially a bold one: It implies that leaders, in pursuing their self-interests, may unintentionally contribute to the construction of a norm that ends up significantly constraining their actions.
The book will also be important for the future work that it will likely stimulate—particularly in developing the idea of the signaling mechanism. Bad types have incentives to pool, but there are actors who should want to avoid this type of pooling equilibrium. States in the international community that provide the benefits for good behavior should have an incentive to advocate for more stringent monitoring and severe punishment for cheating (i.e., high costs of compliance) so that bad types will not pass themselves off as good types. Even good types should advocate for higher compliance costs: if benefits for good behavior are finite, an increasing number of states who pass the election observation test decreases the share of benefits for any one state. Good types should not be happy having to share their benefits with bad types posing as good ones. If correct, then Hyde may be overstating her claim: that the spread of election observation as a norm developed in the absence of strong advocacy by pro-democracy states within the international community. While in Chapter 3 her qualitative evidence illustrates the international community's seeming disinterest in shaping the practice and spread of monitoring, it is not clear why this should have been the case.
If the signaling mechanism works as it should and pooling occurs, can we correctly make sense of what we observe? Of the bad types who receive good evaluations, we may have two conceptually distinct types of leaders. There may be bad types who never intended to give up power and successfully fooled the observers into issuing a good evaluation. These cases should be difficult to distinguish observationally from good types (although Hyde discusses in Chapter 5 the forms of strategic manipulation that should make them distinct). There also may be leaders who initially had no intention of leaving office but as the election process unfolded, realized that holding onto power was futile. So they decide not to manipulate the election, and they receive a good evaluation. In this case, leaders were not good types all along, but rather “evolved” from bad to good type. Formal tools may make it difficult to consider these types of cases (i.e., types are given ex ante), but they offer the clearest instance of the norm's power—in that it actually affects behavior.
This still leaves the original puzzle: what to make of leaders who are bad types, invite observers, cheat, and get caught. Robert Mugabe in Zimbabwe, as discussed in Chapter 5, is perhaps the most prominent example. As Hyde argues, monitoring is imperfect, and so bad types may invite observers in the belief that they will get away with fraud. Another possibility is that some leaders with no intention of leaving power invite observers in the belief that they will be able to win without manipulation. During the campaign, they realize that the likelihood of losing or the cost of losing is much higher than expected. So they resort to fraud and get caught. In this case, there is room for a fuller domestic politics story that explains why leaders miscalculate the prospects of winning so badly or that compares the costs of losing the election (imposed by domestic actors) to the costs of cheating as imposed by the international community. Finally, it may just be that cheating is costless. Hyde argues against this idea, using examples showing that cheaters are sanctioned. Do the costs of cheating appear systematically? Do leaders believe that cheating will be punished? What is striking about examples such as Mugabe is just how often pseudodemocrats get away with cheating before they are actually punished. After Mugabe escaped punishment in the 2000, 2002, and 2005 elections, it may have been reasonable for him to believe in 2008 that he could cheat with impunity.
Also worth considering are the consequences and the quality of the signal. Is the threat of sanctions for cheaters credible, especially if sanctions requires a collective effort on the part of the international community? Since the Cold War, the pro-democracy contingent of international states has have been sufficiently powerful to make punishment credible. As countries that do not share democratic preferences become powerful in the world, will the norm dissipate? The issue of collective action appears again if we consider the observer groups. An election is usually monitored by various observer groups who do not always speak with one voice about the election's quality. Do observer groups coordinate in providing their evaluations, and if so, how? While observer groups may coordinate on the basis of normative arguments for universal standards in judging elections, they probably also have strategic considerations that influence whether their evaluations are similar. The degree to which observer groups issue the same message about an election will also affect the quality of the signal received by the international community. Mugabe in Zimbabwe may not have been sanctioned in earlier elections because those elections were not criticized by all of the monitors. An unclear signal may make the international community hesitant about enacting sanctions.
Finally, in Chapter 5, Hyde discusses how election monitoring increases the likelihood of electoral boycotts by opposition parties who fear that monitors may bestow unwarranted legitimacy on the elections. If elections are being monitored but the opposition boycotts, what type of signal does the international community receive? It would seem a noisy one at best.
On the empirical side, future work undoubtedly should expand the use of observational and experimental data in examinations of the causes and consequences of election monitoring. Hyde (with Nikolay Marinov) has put together a cross-national data set on various features of national-level elections (NELDA) that is already being used in the discipline. Expansion of these efforts by others would enable scholars to test the growing number of claims about elections in democracies and autocracies. The author also uses field experiments to properly assess the impact of monitoring on the extent of fraud by ensuring random assignment of observers to polling stations. Even with random assignment, we may need to exercise caution in comparing the effect of observers in elections under the auspices of bad and good types. In this research design, observer missions are “treatments,” but we cannot forget that observers are actors who have their own perceptions and motivations. If observers suspect that they are being “used” by bad types who want to fool the international community, do they become more motivated to find fraud? If so, then reductions in fraud may be correlated with observer effort, an unobservable and potentially confounding factor. Finally, experimental methods do not negate the importance of observational data: Information on a number of traits characterizing subnational electoral constituencies is needed in order to ensure balance between the control and treatment groups.
