Toppling Foreign Governments. Melissa Willard-Foster

as separate observations. Although these events are clearly related, each state’s decision to participate (or not) informs our understanding of when states pursue FIRC. To control for the nonindependence between multilateral cases, I use robust standard errors clustered on the target-state year.

       The Data Structure and Model

      I employ a logit model to test the probability of FIRC. The unit of analysis is the directed-dyad year, which means each observation contains a potential imposing state (Side 1) and target state (Side 2) for every year, starting in 1816.¹³ This produces an enormously large data set with many dyads for which the probability of conflict is almost zero. The large number of potentially irrelevant observations poses two problems. First, large data sets can be sensitive to very minor effects, which can overstate the importance of variables that have little explanatory power. Second, because my theory explains how states reconcile conflicts of interests, it assumes a population of dyads with conflicting interests. Testing the model on data that include many dyads with almost no potential for conflict introduces inefficiency and possible bias.

      I correct for these problems by limiting the scope of the data in three ways. First, I test the model using politically relevant dyads, commonly used to exclude dyads that have almost no probability of conflict. Politically relevant dyads are dyads in which Side 1 or 2 is a major power or the two states are contiguous.¹⁴ Twelve FIRC events fall short of these criteria. To include these cases, I perform a robustness check using an expanded definition of politically relevant dyads, which includes dyads located in the same region rather than only ones that are contiguous.

      Second, I test the model on interstate rivals. Given that rivals engage in repeated disputes, we know that they are likely to have conflicting interests. I use James Klein, Gary Goertz, and Paul Diehl’s rivalry data set, which defines rivals as states that have had at least three or more militarized interstate disputes (MIDs) concerning the same issue(s). Rivalries are coded as ending ten to fifteen years after the last dispute.¹⁵ The disadvantage of restricting the population of cases to rivals is that the data exclude dyads that reconcile their differences through negotiation without the use of force. These data also leave out FIRC events that involve nonrivals, such as the United States and Guatemala in 1954.¹⁶ Lastly, to account for these cases, I use a third approach in which I limit the population to dyads that have experienced at least one MID in their shared history. These dyads should have a higher probability of conflict than merely politically relevant ones, though they do not necessarily meet the criteria to be considered interstate rivals.¹⁷

      Measuring Domestic Political Opposition

      If my theory is correct, FIRC should become more likely as the targeted leader’s external or internal domestic opposition grows in power. Quantitative measures of domestic opposition can be either direct or indirect. Frequently used direct measures include counts of political-unrest events (e.g., riots, insurgencies, coups) or the number of disadvantaged sub-state groups. Direct measures are helpful because they are less likely to pick up effects other than domestic opposition. However, missing data can be a problem. In one of the most commonly used domestic unrest datasets, 75 percent of politically relevant dyads are missing.¹⁸ High levels of “missingness” can introduce bias.¹⁹ Likewise, data on disadvantaged sub-state groups or ethnic-religious heterogeneity is frequently only available for recent time periods. Such figures could produce misleading results if used to proxy for societal divisions further back in time.²⁰

      Direct measures of opposition also undercount internal and latent opposition. Leaders facing the threat of an internal coup will not necessarily encounter popular protests, insurgencies, or other forms of observable political unrest. Yet the fear of losing supporters to an internal rival could drive a leader to resist a foreign power, while the presence of internal opposition may tempt the foreign power to try regime change. Latent threats from groups that oppose the leader but are unwilling to fight under the status quo can also precipitate FIRC. Leaders may resist a foreign demand for fear of transforming this latent opposition into active opposition. At the same time, the foreign power may attempt to organize, equip, and train members of the latent opposition to carry out FIRC. By ignoring latent or internal opposition, we risk overlooking some of the conditions most likely to lead to FIRC.

      Indirect measures of opposition better account for internal and latent opposition. One such approach relies on the assumption that certain regime types are less adept at managing domestic opposition and, therefore, face greater domestic threats. H. E. Goemans, for example, argues that mixed regimes, which are neither fully democratic nor autocratic, are sensitive to domestic challenges because they can rely on neither repression nor democratic institutions to manage opposition.²¹ Abel Escribà-Folch and Joseph Wright take a similar approach, arguing, for example, that personalist authoritarian leaders, who exercise personal control over the political system, are more likely to be brought down by sanctions, which undercut their ability to buy off and deter domestic opponents.²² Indirect measures, however, can be prone to measurement error. For example, one problem with using regime type as a proxy for a leader’s domestic vulnerability is that this method overlooks potential differences within regime type. Older, well-established regimes may find ways to compensate for their institutional weaknesses that enable them to fend off foreign meddling. Newer regimes, in contrast, may be more prone to FIRC because they lack the experience, routines, and relationships that can help ensure their power.²³

      Another commonly used indirect measure captures this difference within regime type by using the amount of time that has passed since a major political transition to account for regime stability.²⁴ Political change, as Edward Mansfield and Jack Snyder argue, can set off struggles between new and old elites, making periods of change and their aftermath prone to political turmoil.²⁵ Regime transitions, liberal or otherwise, have also been shown to increase the risk of both civil and diversionary wars.²⁶ Yet, a problem with using regime transitions as a proxy for domestic opposition is that leaders may attempt to counter domestic threats, particularly latent or internal ones, by making minor institutional changes. Though the leader may succeed in the short term, these changes will not necessarily eradicate the domestic threat the leader faces. Indeed, these attempts to alter the political system could backfire and inspire greater opposition that later leads to a major transition. In short, leaders that institute minor political change may be doing so because they face significant domestic political pressure.