More Than Mixed Results: What We Have Learned From Quantitative Research on the Diversionary Hypothesis
Summary and Keywords
In the three decades since Jack Levy published his seminal review essay on the topic, there has been a great deal of quantitative research on the proposition that state leaders can use international conflict to enhance their political prospects at home. The findings of this work are frequently described as “mixed” or “inconsistent.” This characterization is superficially correct, but it is also misleading in some important respects. Focusing on two of Levy’s most important concerns about previous research, there has been substantial progress in our understanding of this phenomenon.
First, as Levy suggests in his essay, researchers have elaborated a range of different mechanisms linking domestic political trouble with international conflict rather than a single diversionary argument. Processes creating diversionary incentives bear a family resemblance to one another but can have different behavioral implications. Four of them are (1) in-group/out-group dynamics, (2) agenda setting, (3) leader efforts to demonstrate competence in foreign policy, and (4) efforts to blame foreign leaders or perhaps domestic minorities for problems. In addition, researchers have identified some countervailing mechanisms that may inhibit state leaders’ ability to pursue diversionary strategies, the most important of which is the possibility that potential targets may strategically avoid conflict with leaders likely to behave aggressively.
Second, research has identified scope conditions that limit the applicability of diversionary arguments, another of Levy’s concerns about the research he reviewed. Above all, diversionary uses of military force (though not other diversionary strategies) may be possible for only a narrow range of states. Though very powerful states may pursue such a strategy against a wide range of targets, the leaders of less powerful states may have this option only during fairly serious episodes of interstate hostility, such as rivalries and territorial disputes. A substantial amount of research has focused exclusively on the United States, a country that clearly has the capacity to pursue this strategy. While the findings of this work cannot be generalized to many other states, they have revealed some important nuances in the processes that create diversionary incentives. The extent to which these incentives hinge on highly specific political and institutional characteristics point to the difficulty of applying realistic diversionary arguments to a large sample of states. Research on smaller, more homogenous samples or individual states is more promising, even though it will not produce an answer to the broad question of how prevalent diversionary behavior is. As with many broad questions about political phenomena, the only correct answer may be “it depends.” Diversionary foreign policy happens, but not in the same way in every instance and not in every state in the international system.
The last three decades have witnessed a great deal of research on the long-standing argument that state leaders can use international conflict to enhance their political prospects at home. Papers dealing with this “diversionary hypothesis” often begin with the observation that quantitative research about it has thus far produced mixed or inconsistent results. Indeed, some of my own work on the topic has begun in just this way (Clark, Fordham, & Nordstrom, 2011). While this observation is correct as far as it goes, it is misleading in some important respects. For one thing, it implies that there is a single claim about which the evidence is mixed. In fact, there are actually many different diversionary hypotheses. General formulations of the argument like the one in the first sentence of this paragraph conflate several causal processes and omit a range of important qualifications about what state leaders may employ this tactic, what conditions may motivate them to do so, and what factors may affect their attentiveness to these conditions.
Another way in which the “mixed results” characterization is misleading is that it suggests that we have learned little from all the quantitative work on the topic. This impression is also mistaken. Jack Levy’s (1989) seminal review essay has informed much of the empirical research that has taken place since its publication. The problems he identified serve as a reasonable benchmark against which to measure progress. In this essay, I will focus on two of his most important concerns. First, he argued that existing quantitative research was not sufficiently clear and specific about the theoretical linkages between domestic and international conflict. Second, he held that the tests employed in this literature were too broad, setting aside likely scope limitations on the diversionary hypotheses they tested.
Quantitative research since Levy’s essay has made substantial progress on both of these problems. It has produced an array of more refined diversionary arguments, as well as several important countervailing processes, that apply within specific subpopulations. Continuing research in this vein remains promising, but it will not produce a clear conclusion about diversionary behavior in general. This is not a criticism of the research but rather of the goal. Because the use of international conflict for domestic political gain is not a single phenomenon, it is unrealistic to expect a single general conclusion about it.
What’s at Stake in Quantitative Research on Diversionary Conflict?
Arguments about diversionary conflict are especially well suited to quantitative behavioral research. Nearly all diversionary strategies require deception by the state leaders employing them. Revealing their diversionary intentions would both undermine the strategy and seriously damage the leader’s political prospects and historical reputation. The need to keep diversionary intentions secret poses a serious problem for research relying on evidence about the decision-making process drawn from archives or interviews. Leaders’ professed motives may be sincere, or they may be part of the deception a diversionary strategy entails. It is thus exceedingly difficult to know whether any particular use of military force was “diversionary.” By contrast, consistent diversionary behavior may well create patterns detectable though quantitative methods. The word consistent is important here. A single instance of diversionary conflict, even if we could recognize it as such, is unlikely to create a statistically significant pattern in a broader sample. The behavior has to be a reasonably consistent response to diversionary conditions in the sample.
Because quantitative work on the diversionary hypothesis is about consistent patterns, scholarly debate about this work has always concerned the extent of diversionary behavior, not whether it ever happens. There is little doubt that state leaders have sometimes used international conflict to shore up their domestic political position. The reasons they may do so are easy to understand. As Levy points out in his review essay, they have been the subject of historical and literary commentary for centuries. Scholars have developed a range of formal models that produce diversionary incentives from fairly uncontroversial assumptions (e.g., Downs & Rocke, 1993; Gent, 2009; Smith, 1996; Tarar, 2006). As with influence peddling, electoral fraud, and the murder of political opponents, we can understand why leaders may do these things, even as we hope that they will not. Diversion is more difficult to detect than these other abuses of public authority because it concerns the unobservable motives of leaders and because misleading the public (and perhaps oneself) about these motives is part of the strategy. Even so, it would be naïve to think that diversion, like these other phenomena, never happens.
At the same time, the notion that state leaders usually react to domestic political difficulties by turning to international conflict is equally implausible. Most leaders have other tools for managing domestic trouble, and diversion may be difficult or even impossible for many of them. Some state leaders have few or no military forces at their disposal, or they may face institutional restrictions on what they can do with the forces they have. Others are hemmed in by international circumstances that leave them with very few diversionary targets or only very risky ones. Still others face a domestic audience that does not respond positively to international conflict, or they face political trouble that constrains them in ways that make aggressive international behavior difficult or politically unwise. In sum, it makes little sense to suppose that most or all state leaders will turn frequently to diversionary conflict among the range of options that they have for shoring up domestic support. Discerning the special conditions under which diversionary behavior is likely is thus the principal task facing quantitative research on the topic.
With these considerations in mind, it is not surprising that empirical tests of broad linkages between domestic political difficulties and international conflict often fail to turn up much systematic evidence. These results are the source of concerns that diversionary behavior is relatively unimportant, if not entirely mythical. Levy (1989) has cited a range of such nonfindings, most notably Rummel’s work on the topic (Rummel, 1963). Empirical analyses published since his essay have reinforced these concerns. For example, in a sample of 18 developed democracies, Leeds and Davis (1997) found no linkage between economic growth or an approaching election and state leaders’ propensity to initiate militarized disputes. Using an even larger sample of all state leaders between 1919 and 1992, Chiozza and Goemans (2003) found that those facing a higher likelihood of losing office were actually less likely to initiate an international crisis than those who were more secure in their positions. Oneal and Tir (2006) found that weak economic growth made democracies somewhat more likely to initiate militarized disputes in a very large sample of states between 1920 and 1992, but they noted that the effect was quite small.
In these and other works, as in Rummel’s research, the proposition tested is whether a very broad sample of leaders was systematically more likely to engage in aggressive behavior when confronted with domestic political trouble of a fairly general sort. Like Rummel’s, such statistical tests frequently fail to reject the null hypothesis. However, even when they turn up support for this very general formulation of the diversionary argument, the questions Levy raised still apply. Not all forms of domestic trouble are likely to create incentives to engage in international conflict, and those that do may not work in the same way. More realistic formulations of the diversionary argument are likely to apply to a narrower subset of state leaders and thus imply more focused empirical tests.
Multiple Pathways to from Domestic Difficulty to International Conflict
One of Levy’s (1989, p. 269) principal critiques of quantitative research on the diversionary hypothesis is that “even if we were to restrict our attention to the relationship leading from internal conflict to external conflict, we would have to recognize several distinct causal mechanisms that could be involved.” He noted that some of these causal mechanisms may make international conflict more likely but that others may make it less likely. Research during the last three decades has elaborated and developed this intuition. It has uncovered several possible mechanisms linking domestic political difficulties to international conflict. Not all have the same behavioral implications.
The most commonly discussed causal mechanism linking leaders’ domestic political difficulties to external conflict works through in-group/out-group dynamics in public opinion (Coser, 1956; Levy, 1989, pp. 260–262; Simmel, 1955). Hostility toward out-groups is a powerful source of identity and an important force in public opinion about war (e.g., Berinsky, 2009). The essential argument is that the public will set aside its misgivings about the national leader when their country comes into conflict with a foreign power. This tendency is one basis for the “rally-round-the-flag” effect noted by Mueller (1973) and others. Because national leaders have a good deal of control over when they enter into military conflict, as well as how this conflict is presented to the public when it occurs, they may be able to manipulate the rally effect to their political advantage. Rally events need not be military successes—the September 11 terrorist attacks produced an enormous surge in public support for President George W. Bush—but research on the United States also suggests that not every dispute or crisis is visible enough to trigger this phenomenon either (Baker & Oneal, 2001; James & Rioux, 1994).
While most empirical research on diversion begins with some variant of the in-group/out-group argument, it is not the only process that could generate diversionary incentives. Research since the publication of Levy's review essay has explored at least three others. A second mechanism focuses on agenda setting. The logic here is that the government may benefit from increasing the salience of foreign policy, even in the absence of a rally effect. This is especially the case for leaders whose foreign policies are more popular than their domestic efforts. In these instances, the leader would benefit from the agenda-setting effect of international conflict, focusing greater public attention on an issue area in which they are relatively well-regarded (DeRouen & Peake, 2002).
Although this process generates similar diversionary incentives, it differs from the in-group/out-group dynamic in that the level of conflict involved in the diversionary activity need not be high enough to produce an actual rally effect. Indeed, no actual military conflict may be necessary at all. For instance, Kisangani and Pickering (2007, 2009, 2011) have shown peaceful deployments of military forces for purposes such as peacekeeping and disaster relief may respond to diversionary incentives under some circumstances. Even mere saber rattling may be enough to serve this purpose. Li, James, and Drury (2009) found that pro-independence rhetoric in Taiwan increases in response to declining presidential approval.
This process closely resembles what historians have called “social imperialism” in the 19th and early 20th centuries. As the franchise expanded, conservative parties in some states could not compete with the domestic social and economic policies offered by their opponents on the left. They instead stressed publicly appealing colonial enterprises, shifting the ground on which voters may evaluate them. Hans-Ulrich Wehler (1970, p. 149) made such a case about Wilhelmine Germany, arguing that “[i]n an age of critical social change and inhibited economic growth, Bismarck sought to pursue a conservative policy, even to the point of exploiting imperialism for the preservation of the internal structure of the Prusso-German state.”1 Other scholars have pointed to similar manipulation of the national agenda to build support for conservative political leaders elsewhere during the same era. Richard Hofstadter (1966 , pp. 148–149) contended that a constellation of social problems arising from industrialization produced a “psychic crisis” that drove the United States toward a more aggressive foreign policy in the 1890s. Political leaders who advocated it hoped that such a foreign policy would appeal to groups of people who were unresponsive to them on other public issues. In both cases, the diversionary effort did not rest on rally effects in response to specific instances of international conflict, but rather a sustained emphasis on national aggrandizement.
Formal models developed by Downs and Rocke (1993) and Smith (1996) rest on a third causal mechanism. Both these models begin with the leader’s need to demonstrate his or her competence to voters by producing foreign policy success. This is especially important when the leader is facing difficult economic times and cannot count on a favorable evaluation in this other policy area. These pressures create a bias toward military action when the economy is performing poorly. Indeed, Downs and Rocke suggest that a leader facing a hopeless domestic situation may even launch a military intervention that has a low probability of success in a “gamble for resurrection.” This process also diverges from the in-group/out-group argument in that it requires military success, something not necessary to produce a rally effect. An inconclusive conflict with a hated rival may produce a rally effect but not effectively demonstrate competence. Haynes (2015) explores the implications of these different processes, arguing that leaders who need to demonstrate competence will target more powerful states, while those who wish to benefit from in-group/out-group dynamics will prefer traditional enemies or rivals.
Yet another causal mechanism concerns scapegoating, the effort to blame a foreign enemy for the domestic political or economic difficulties besetting the country. This is essentially a rhetorical strategy that need not involve actual military conflict, though it is likely to exacerbate hostile political relations. It has received less attention in the literature than the other three mechanisms because it requires an enemy that may plausibly be responsible for the trouble that generated the diversionary incentive. Instances where a state faces sanctions from a powerful foreign enemy may produce this variant of diversionary policy. For instance, Davies (2012) found that the Iranian government has used American expressions of hostility to divert attention from domestic problems, exploiting the likelihood that the United States will not actually resort to force. Domestic political opponents or unpopular minority groups are equally plausible targets for scapegoating, and some research has found evidence of diversionary action against them (e.g., Klein & Tokdemir, 2016; Martinez Machain & Rosenberg, 2016; Tir & Jasinski, 2008).
These processes may appear only subtly different from one another, but each suggests somewhat different behavioral patterns. Nor are these the only possible mechanisms linking domestic political difficulties with international conflict. The range of possible linkages grows as one considers sets of states with special characteristics. For instance, some authoritarian states may use international tension simply to justify domestic repression that they would have undertaken anyway. Similarly, state leaders whose domestic opponents are known to be dovish may enjoy electoral advantages during periods of international tension, offering another motive for playing up these tensions.
Just as more than one process may create diversionary incentives, so there are also several ways in which domestic conditions—sometimes the same ones that create diversionary incentives—may also restrain state leaders’ conflict behavior. Research since Levy’s review essay has considered several of these countervailing processes as well. The most important of them concerns the strategic behavior of likely diversionary targets. First proposed by Smith (1996), the argument is that potential targets can observe the conditions that create diversionary incentives, especially when the diverter is a democracy. As Smith (1996, p. 150) explains, these likely targets have an incentive to avoid conflict at these moments because “nations undertake actions that lead to crises when they expect the least intervention from other nations.” In order to avoid providing a pretext for diversionary conflict, they may behave in a more conciliatory way toward a leader facing domestic political difficulties or an imminent election or at least avoid obvious provocations. A number of researchers have found evidence of strategic conflict avoidance (e.g., Chiozza & Goemans, 2004; Clark, 2003; Davies, 2007; Fordham, 2005; Leeds & Davis, 1997; Miller, 1999). If it is effective enough, it may prevent diversionary conflict from taking place.
Domestic conditions may also restrain leaders even as they create diversionary incentives if they affect the cost of using force or weaken the leader’s control over the necessary political or material resources. For example, legislative opposition, especially in states where the executive does not always control the legislature, may restrain leaders concerned about legislative interference with military operations (Clark, 2000; Howell & Pevehouse, 2005, 2007). Similarly, some economic conditions, such as inflation, may create diversionary incentives, on one hand, but may also increase the potential costs of using military force on the other (Fordham, 1998b).
Just as the strategic conflict avoidance argument suggests that states may adjust their behavior to avoid becoming diversionary targets, so the restraining effect of legislative opposition and other conditions raises the possibility that other states may exploit these conditions. Potential challengers can often observe the conditions that make it more costly than usual for a leader to engage in international conflict. They can time their demands to coincide with these conditions, taking actions that would otherwise provoke a forceful response or escalating their demands beyond a point that would otherwise lead to military conflict. This outcome arises from theoretical premises quite similar to those producing expectations about both diversionary behavior and strategic conflict avoidance, but it points to different behavioral patterns. Foster (2006a) and Clark, Fordham, and Nordstrom (2011) have explored this possibility in the case of the United States, finding some evidence for it.
The complexities uncovered in recent research on diversionary behavior underscore the importance of Levy’s concern about the mechanisms linking domestic and international conflict. The fact that these processes are not mutually exclusive but may instead take place simultaneously is one reason that tests of diversionary behavior in large samples of states may produce inconsistent results. Broad indicators of domestic trouble, like poor economic performance or a high probability of removal from office, may indeed create diversionary incentives. However, in some cases, they may also be correlated with countervailing conditions like strategic conflict avoidance, legislative opposition, or an unusually high cost of using military force. In developing convincing tests of diversionary arguments, scholars must not only specify the precise process they have in mind but also consider the other, related processes that may be going on at the same time. At a minimum, these should inform the selection of control variables.
Another of Levy’s (1989, pp. 276–277) critiques of earlier tests of the diversionary hypothesis concerns their “general failure to specify the conditions under which the hypothesis is likely to hold . . . There is good reason to believe that sweeping analyses that fail to control for these contingent conditions may be masking some significant empirical patterns.” In fact, identifying these scope conditions is as much a part of theory development as elaborating the various causal mechanisms discussed in the preceding section. The two activities are difficult to separate. The scope conditions arise from the theoretical argument. If the argument proposes a process that links a particular response to some antecedent condition, then actors who cannot adopt that particular response, cannot experience the antecedent condition, or are immune from the process connecting the two do not belong in a sample used to test the theoretical argument. The scope conditions differ somewhat for each of the theoretical mechanisms discussed earlier. For instance, many more state leaders can plausibly blame foreign scapegoats for domestic difficulties or play up hostile international relationships to increase the salience of foreign policy than can actually use military force to prompt a rally or demonstrate their foreign policy competence. Because it is the principal concern of most research on diversion, the focus here is on the more stringent scope conditions that apply to the actual use of force.
Research since the publication of Levy’s review essay has made substantial progress in identifying scope conditions associated with various forms of diversionary foreign policy. However, it has paid surprisingly little attention to what is arguably the most obvious scope condition: the capabilities required to use diversionary strategies that involve actual military action. Not every state leader can actually use military force. States like Iceland or Costa Rica have little or no military. Others, like Germany and Japan, have constitutional restrictions that rule out the use of force except under extreme circumstances. These facts have not prevented the inclusion of these states in many broad tests of diversionary arguments.
Simply having a military force and the authority to use it is not enough to make the use of force for diversionary purposes feasible. Few states have the capacity to use force in a way that is visible enough to prompt a rally effect or to demonstrate the leader’s effectiveness at foreign policy, yet unlikely to start a severe military conflict. Such a conflict is likely to wipe out any potential political gains from the use of force. Even among states that could do so in principle, this course of action may be possible only against a small number of targets and may be exceedingly dangerous even in these cases. The population of state leaders who could potentially engage in diversionary conflict with other states may be quite small.
Not surprisingly, research on variation in the effect of diversionary incentives in large samples of states has turned up evidence that not all states respond to these incentives in the same way. Capabilities, the nature of relations with other states, and domestic institutions all appear to influence whether state leaders respond to diversionary incentives. For instance, Mitchell and Prins (2004) found that poor economic performance made the initiation of interstate disputes more likely against rival states but less likely outside of these rivalries. Similarly, Foster (2006b) offered evidence that major powers in general, and the United States in particular, may target a wide range of states but that minor powers’ conflict behavior is confined mainly to rivals. Mitchell and Thyne (2010) found that states involved in disagreements over contentious issues were more likely to use force against one another when facing relatively high inflation. Tir (2010) found evidence of diversionary behavior among states involved in territorial disputes. Findings like these suggest that the leaders of states with limited military capabilities, few international rivals, or no serious ongoing disputes may have little or no way of using international conflict for diversionary purposes. This description fits most states in the contemporary international system.
One state that has clearly had the capacity for diversionary behavior is the United States since 1945. Even if the United States were the only state capable of diversionary behavior, its position in the international system would make this possibility important. Not surprisingly, the country has received the greatest empirical scrutiny, with many tests of diversionary arguments focusing exclusively on it (e.g., DeRouen, 1995; Fordham, 1998a, 1998b; Hess & Orphanides, 1995; James & Hriustoulas, 1994; James & Oneal, 1991; Meernik, 2000; Mitchell & Moore, 2002; Moore & Lanoue, 2003; Morgan & Bickers, 1992; Ostrom & Job, 1986; Prins, 2001; Stoll, 1984). The prevalence of these single-state empirical analyses is commonly seen as a shortcoming of research on the topic. For instance, Oneal and Tir (2006, p. 760) comment that “[m]ost tests of diversionary theory examine only the United States, but the United States is apt to be an exceptional case because of its ability to project its power globally.” Leeds and Davis (1997, p. 814) write that “the generalizability of tests performed on the case of the United States during the cold war should not be assumed.”
Is the stress on the United States in research on the diversionary hypothesis really a weakness? There is no denying that the United States is an unusual case, as the works just quoted indicate. At the same time, diversionary foreign policy is clearly not an option for all state leaders and perhaps is an option only for a small number of them. Postwar American presidents have had the capacity to use force for diversionary purposes, something we cannot say with any confidence about large samples of other states. Empirical tests on a limited but relevant sample are preferable to tests using a sample containing many observations that cannot experience the causal process being tested. Recent thinking on causal empiricism supports this position. For instance, Samii (2016, pp. 943–945) is quite critical of the “pseudo-generality” of studies in which the effective sample of cases actually able to experience the proposed causal effect is much smaller than the nominal sample of cases used to estimate this effect.
A closer look at the research on the United States suggests that focusing on a single state, or perhaps a small but clearly relevant sample of states, has at least two other underappreciated advantages. First, the limited setting makes it practical to collect more precise and appropriate data than one could obtain for a large cross-national sample. For example, scholars have subjected instances in which the United States has used force to a great deal of scrutiny. I have argued in my own work that militarized interstate disputes, the most commonly used source of data on international conflict, do not reliably identify the “uses of force” in the U.S. case that are most relevant to testing diversionary arguments (Fordham & Sarver, 2001). Other scholars have also proposed a variety of other sources of data that may provide more valid indicators of American leaders’ decisions to use military force (e.g., Blechman & Kaplan, 1976; Howell & Pevehouse, 2007; Meernik, 1994; Pickering & Kisangani, 2009). This level of attention to the data is prohibitively costly for large cross-national samples, so most scholars interested in broader patterns have simply had to set these problems aside.
Scholarly exchanges about data have produced substantively interesting conclusions about diversionary behavior by the United States. For example, Meernik (1994, 2000) and Meernik and Waterman (1996) have been quite critical of results suggesting diversionary behavior by American presidents. Meernik contended that researchers had failed to consider variation in opportunities to use force and gathered data on incidents that may constitute opportunities to support his argument. These data formed the basis for scholarly debate about what really constituted an “opportunity,” producing evidence that American domestic conditions could affect the frequency of these events (e.g., Fordham, 1998b; Meernik, 2001). Going even further, Howell and Pevehouse (2007) gathered similar data on opportunities to use force from different sources in order to show that there were many more of them than originally thought. Given the extremely high cost of collecting data even on actual conflict events for a cross-national sample, a scholarly exchange about opportunities to use force in such a large sample is probably impractical.
Beyond practical concerns about data availability, a second advantage of research focused on the United States concerns the possibility of testing highly specific and realistic theoretical mechanisms affecting diversionary behavior. For example, scholars have theorized and tested several effects of partisanship that are specific to the United States. Among the most important is the role of legislative opposition. Clark (2000) provides evidence that divided government in the United States has made American involvement in militarized disputes less likely. Howell and Pevehouse (2005, 2007) found similar effects using several different measures of legislative support. Continuing research is not unanimous that Congress restrains the president under all circumstances (e.g., Brulé, 2006, 2008), but the bulk of it points in that direction. Opposition parties may well exert important restraining effects in other states, but they are unlikely to work exactly as they do in the United States. Even setting aside the very different expectations about the use of force prevailing in other states, the institutional fact that the executive does not always control the legislature in the United States makes opposition support both more important and less certain.
Scholars focused on the United States have also explored the impact of party differences over economic and foreign policy on diversionary incentives. The foreign policy orientation of the two parties may influence presidents’ responsiveness to domestic conditions. Foster and Palmer (2006) hypothesized that Republican presidents, who have a relatively hawkish core constituency, are more likely to use force in the face of declining partisan approval, but they failed to find any evidence for either this pattern or the preference for approval within the president’s own party that Morgan and Bickers (1992) had suggested. Following up these results, Arena and Palmer (2009) found that party foreign policy orientation is more important than economic conditions for influencing the initiation of disputes, though both matter. The nature of the ruling party’s constituency may well be important in other national settings, but it is far from obvious that hawk-dove distinctions work in the same way outside the United States.
Research has also considered whether the two parties’ economic policy differences influence when Democratic or Republican presidents have diversionary incentives. Drawing on literature about the American political economy, Fordham (1998a) suggested that Republican presidents may be under pressure from their elite constituents to avoid expansionary fiscal policies to reduce unemployment. Democrats may face similar pressures to avoid deflationary policies that may increase unemployment when facing high inflation. Republicans may thus have greater diversionary incentives than Democrats when facing high unemployment and Democrats greater diversionary incentives than Republicans when facing high inflation. Subsequent researchers have explored the impact of other considerations on this dynamic, including strategic conflict avoidance (Clark, 2003), legislative support (Brulé & Hwang, 2010), and the basic foreign policy orientation of the two parties (Foster, 2008). Like the hawk-dove distinction, major parties’ economic policy orientations may well have similar effects in other national settings. However, they are unlikely to work in quite the same way in systems with multiple parties, let alone in authoritarian states or those with different degrees of control over fiscal and monetary policy.
None of these patterns readily generalize to a large sample of states. They point to the importance of specific national political and institutional characteristics in shaping diversionary incentives. The fact that these considerations make a difference in the U.S. case raises questions about whether it makes sense to expect broad cross-national patterns of diversionary behavior. Many states have institutional and political characteristics that could affect the mechanisms producing diversionary incentives in idiosyncratic ways that parallel those found in research on the United States. Generalizations about processes producing diversionary incentives in broad samples of states rest on the assumption that one can abstract away from these idiosyncrasies. Doing so is not impossible for every variant of diversionary foreign policy, but constructing cross-nation samples for testing these claims requires a good deal of care.
Efforts to assess the importance of regime type for diversionary foreign policy show how difficult it can be to theorize and test effects that apply to a broad range of states. In principle, democracy is a promising analytical category for cross-national generalizations about conflict behavior. The democratic peace proposition, while still the subject of some continuing debate and objection, has produced as much consensus as any finding in the field. If democracy, as a broad category of political regime, can affect conflict behavior directly, it is at least plausible that it could also modify diversionary incentives in ways that apply across a large sample of states. While many researchers have suggested such effects, there is no consensus about their direction. On one hand, some have argued that democratic leaders should be more responsive to diversionary incentives because they have to worry more about public opinion than autocrats do. Several researchers have presented evidence that this is indeed the case (e.g., Gelpi, 1997; Kisangani & Pickering, 2009; Oneal & Tir, 2006; Powell, 2014). On the other hand, autocratic leaders also need to manage public opinion and may be in a better position to control and manipulate public perceptions of foreign enemies than democratic leaders are. Moreover, the greater value of retaining public office may make autocrats more willing to resort to a diversionary strategy (Tarar, 2006). A number of researchers have found evidence for this line of argument as well (e.g., Heldt, 1999; Miller, 1995; 1999; Mitchell & Prins, 2004).
These apparently conflicting pieces of research rest on tests using different samples of states, different types of conflict events, and different diversionary conditions. The divergence in their findings may simply reflect the workings of the various theoretical mechanisms discussed earlier. However, other research suggests that cross-national findings about the effect of regime type are sensitive to narrower differences among the states in each sample. One possibility is that the control of information generally associated with autocracy also varies over democracies. Bell (2013) finds that democracies with less transparency are also more likely to respond to diversionary incentives, a result that could affect the apparent impact of autocracy in samples that include relatively more of these states. Another consideration that could affect estimates of the impact of democracy concerns the extent of a leader’s responsibility for domestic economic performance. Many things could affect whether the public imputes such responsibility. Johnson and Barnes (2011) suggest that leaders in relatively open economies where more actors can affect economic performance are less responsive to diversionary incentives. Kisangani and Pickering (2011) find that leaders in presidential systems are more likely to engage in diversionary behavior. Brulé and Williams (2009) present evidence that leaders in single-party parliamentary governments are more likely than coalition governments to use force when the economy is performing poorly. Set alongside the specific institutional features that appreciably affect diversionary incentives in the United States, these cross-national differences provide good reason to expect inconsistent results concerning broad institutional categories like “democracy” and “autocracy.”
Levy concluded his 1989 review essay by noting that “we need new methodological approaches that go beyond the previous generation of studies based on the Rummel paradigm and research designs that are more closely related to the theoretical questions being asked (Levy, 1989, p. 285).” Research since the publication of Levy's essay has followed this suggestion and has added a lot to our understanding of diversionary foreign policy. In doing so, it has moved toward more focused arguments and empirical tests intended to capture situations where we have genuine expectations about specific types of diversionary behavior as well as related behaviors, like strategic conflict avoidance and exploitation. In general, while it is often useful to speak of diversionary conflict in general terms, research on the topic may produce its most convincing results exploring theoretically grounded variants of the argument in relevant samples.
The contributions of such a large body of quantitative research produced by many authors over a period of several decades are not always easy to see amidst the many theoretical claims and research designs it contains, but they are nevertheless important. This essay has highlighted two of them. First, research has elaborated and tested several different causal mechanisms that bear on diversionary foreign policy. These include at least four processes that could produce diversionary incentives. These processes bear a family resemblance to one another, but each has somewhat different behavioral implications. In addition, researchers have proposed and tested several related processes that cut against leaders’ ability to carry out some types of diversionary foreign policy. This array of causal processes, some or all of which may bear on decisions about international conflict simultaneously, underscores the complexity of the relationship between domestic political trouble and international conflict.
Second, research since Levy's review essay points to the importance of the scope conditions on diversionary arguments. Not all of the causal processes elaborated in the research apply to all states. There is plenty of evidence that, for most states, the diversionary use of military force is possible only under limited circumstances, such as during periods of fairly serious ongoing hostility with other states. Diversionary foreign policy involving a wide range of targets is plausible only for major powers, especially the United States since 1945. For all potential diverters, a range of specific institutional and political features can affect diversionary incentives. These characteristics are not easily generalizable to a broad sample of states.
While we have learned a lot from research conducted since he wrote, Levy’s call for research designs more closely related to the theoretical questions being asked still bears repeating. The range of theoretical processes involved in diversionary foreign policy and the limits on the scope of most diversionary arguments, especially those involving the actual use of military force in international politics, make research designs focused on individual states, or small samples of states, especially appealing. The United States need not be the only subject of these studies. Studies of diversionary arguments in individual countries include work focusing on Britain (Morgan & Anderson, 1999), Israel (Kuperman, 2003), Iran (Davies, 2012), Latin America (Miller & Elgun, 2011), and Japan before World War II (Nicholls, Huth, & Appel, 2010). These works, like many of those focusing on the United States, take advantage of what we know about specific national circumstances to produce more interesting and compelling tests of specific diversionary arguments.
Scholars interested in diversionary foreign policy need not focus their attention entirely on single countries like the United States. Analyses of larger samples have produced interesting results and will continue to be appropriate in some cases. However, these samples should be appropriate for the argument one is testing. Most of the time, this sample will not be all states for which data on conflict and economic performance are available. This approach is bound to be unsatisfying to many scholars, for it will not provide a generalizable answer to the big question of whether state leaders frequently use international conflict to shore up their domestic political position. Unfortunately, no such answer may be available. As with many broad questions about social and political phenomena, the only reasonable general answer may be “it depends.” Diversionary foreign policy happens, but not in the same way in every instance and not in every state in the international system.
Arena, P. (2010). Why not guns and butter: Responses to economic turmoil. Foreign Policy Analysis, 6(4), 339–348.Find this resource:
Arena, P., & Bak, D. (2015). Diversionary incentives, rally effects, and crisis bargaining. Foreign Policy Analysis, 11(2), 233–250.Find this resource:
Arena, P., & Palmer, G. (2009). Politics or the economy? Domestic correlates of dispute involvement in developed democracies. International Studies Quarterly, 53(4), 955–975.Find this resource:
Baker, W. D., & Oneal, J. R. (2001). Patriotism or opinion leadership? The nature and origins of the “rally ’round the flag” effect. Journal of Conflict Resolution, 45(5), 661–687.Find this resource:
Bell, S. R. (2013). What you don’t know can hurt you: Information, external transparency, and interstate conflict, 1982–1999. Conflict Management and Peace Science, 30(5), 452–468.Find this resource:
Berinsky, A. J. (2009). In Time of War. Chicago: University of Chicago Press.Find this resource:
Blechman, B. M., & Kaplan, S. S. (1976). Force without War. Washington, DC: Brookings Institution Press.Find this resource:
Boehmer, C. R. (2007). The effects of economic crisis, domestic discord, and state efficacy on the decision to initiate interstate conflict. Politics & Policy, 35(4), 774–809.Find this resource:
Boehmer, C. R. (2010). Economic growth and violent international conflict, 1875–1999. Defence and Peace Economics, 21(3), 249–268.Find this resource:
Brulé, D. J. (2006). Congressional opposition, the economy, and U.S. dispute initiation, 1946–2000. Journal of Conflict Resolution, 50(4), 463–483.Find this resource:
Brulé, D. J. (2008). Congress, presidential approval, and U.S. dispute initiation. Foreign Policy Analysis, 4(4), 349–370.Find this resource:
Brulé, D. J., & Hwang, W. (2010). Diverting the legislature: executive–legislative relations, the economy, and US uses of force. International Studies Quarterly, 54(2), 361–379.Find this resource:
Brulé, D. J., Marshall, B. W., & Prins, B. C. (2010). Opportunities and presidential uses of force: A selection model of crisis decision-making. Conflict Management and Peace Science, 27(5), 486–510.Find this resource:
Brulé, D. J., & Williams, L. K. (2009). Democracy and diversion: Government arrangements, the economy, and dispute initiation. Journal of Peace Research, 46(6), 777–798.Find this resource:
Chiozza, G., & Goemans, H. E. (2003). Peace through insecurity: Tenure and international conflict. Journal of Conflict Resolution, 47(4), 443–467.Find this resource:
Chiozza, G., & Goemans, H. E. (2004). Avoiding diversionary targets. Journal of Peace Research, 41(4), 423–443.Find this resource:
Chiozza, G., & Goemans, H. E. (2011). Leaders in international conflict. New York: Cambridge University Press.Find this resource:
Clark, D. H. (2000). Agreeing to disagree: Domestic institutional congruence and U.S. dispute behavior. Political Research Quarterly, 53(2), 375–400.Find this resource:
Clark, D. H. (2003). Can strategic interaction divert diversionary behavior? A model of U.S. conflict propensity. Journal of Politics, 65(4), 1013–1039.Find this resource:
Clark, D. H., Fordham, B. O., & Nordstrom, T. (2011). Preying on the misfortune of others: When do states exploit their opponents’ domestic troubles? Journal of Politics, 73(1), 248–264.Find this resource:
Colaresi, M. (2007). The benefit of the doubt: Testing an informational theory of the rally effect. International Organization, 61(1), 99–143.Find this resource:
Coser, L. A. (1956). The function of social conflict. New York: Free Press.Find this resource:
Cramer, J. K. (2006). “Just cause” or just politics?: U.S. Panama invasion and standardizing qualitative tests for diversionary war. Armed Forces and Society, 32(2), 178–201.Find this resource:
Davies, G. A. M. (2007). US presidential popularity and opportunities to coerce North Korea: A quantitative test 1990–2000. International Relations of the Asia-Pacific, 7(2), 129–153.Find this resource:
Davies, G. A. M. (2012). Coercive diplomacy meets diversionary incentives: The impact of US and Iranian domestic politics during the Bush and Obama presidencies. Foreign Policy Analysis, 8(3), 313–331.Find this resource:
DeRouen K. R. (1995). The indirect link: Politics, the economy, and the use of force. Journal of Conflict Resolution, 39(4), 671–695.Find this resource:
DeRouen, K. (2000). Presidents and the diversionary use of force: A research note. International Studies Quarterly, 44(2), 317–328.Find this resource:
DeRouen, K., & Peake, J. (2002). The dynamics of diversion: The domestic implications of presidential use of force. International Interactions, 28(2), 191–211.Find this resource:
DeRouen, K., & Sprecher, C. (2006). Arab behaviour towards Israel: Strategic avoidance or exploiting opportunities? British Journal of Political Science, 36(3), 549–560.Find this resource:
Downs, G. W., & Rocke, D. M. (1993). Conflict, agency, and gambling for resurrection: The principal-agent problem goes to war. American Journal of Political Science, 38(2), 362–380.Find this resource:
Eley, G. (1976). Defining social imperialism: Use and abuse of an idea. Social History, 1(3), 265–290.Find this resource:
Fordham, B. O. (1998a). Partisanship, macroeconomic policy, and U.S. uses of force, 1949–94. Journal of Conflict Resolution, 42(4), 418–439.Find this resource:
Fordham, B. O. (1998b). The politics of threat perception and the use of force: A political economy model of U.S. uses of force, 1949–1994. International Studies Quarterly, 42(3), 567–590.Find this resource:
Fordham, B. O. (2002). Another look at parties, voters, and the use of force abroad. Journal of Conflict Resolution, 46(4), 572–596.Find this resource:
Fordham, B. O. (2005). Strategic conflict avoidance and the diversionary use of force. Journal of Politics, 67(1), 132–153.Find this resource:
Fordham, B. O., & Sarver, C. C. (2001). Militarized interstate disputes and United States uses of force. International Studies Quarterly, 45(2), 455–466.Find this resource:
Foster, D. M. (2006a). An “invitation to struggle”? The use of force against “legislatively vulnerable” American presidents. International Studies Quarterly, 50(2), 421–444.Find this resource:
Foster, D. M. (2006b). State power, linkage mechanisms, and diversion against nonrivals. Conflict Management and Peace Science, 23(1), 1–21.Find this resource:
Foster, D. M. (2008). “Comfort to our adversaries”? Partisan ideology, domestic vulnerability, and strategic targeting. Foreign Policy Analysis, 4(4), 419–436.Find this resource:
Foster, D. M., & Keller, J. W. (2014). Leaders’ cognitive complexity, distrust, and the diversionary use of force. Foreign Policy Analysis, 10(3), 205–223.Find this resource:
Foster, D. M., & Palmer, G. (2006). Presidents, public opinion, and diversionary behavior: The role of partisan support reconsidered. Foreign Policy Analysis, 2(3), 269–288.Find this resource:
Gelpi, C. (1997). Democratic diversions: Governmental structure and the externalization of domestic conflict. Journal of Conflict Resolution, 41(2), 255–282.Find this resource:
Gent, S. E. (2009). Scapegoating strategically: Reselection, strategic interaction, and the diversionary theory of war. International Interactions, 35(1), 1–29.Find this resource:
Gleditsch, K. S., Salehyan, I., & Schultz, K. (2008). Fighting at home, fighting abroad: How civil wars lead to international disputes. Journal of Conflict Resolution, 52(4), 479–506.Find this resource:
Haynes, K. (2015). Diversionary conflict: Demonizing enemies or demonstrating competence? Conflict Management and Peace Science, forthcoming. Published through OnlineFirst.Find this resource:
Haynes, K. (2016). Diversity and diversion: How ethnic composition affects diversionary conflict. International Studies Quarterly, 60(2), 258–271.Find this resource:
Heldt, B. (1999). Domestic politics, absolute deprivation, and the use of armed force in interstate territorial disputes, 1950–1990. Journal of Conflict Resolution, 43(4), 451–478.Find this resource:
Hess, G. D., & Orphanides, A. (1995). War politics: An economic, rational-voter framework. American Economic Review, 85(4), 828–846.Find this resource:
Hofstadter, R. (1966). Cuba, the Philippines, and Manifest Destiny. In R. Hofstadter (Ed.), The paranoid style in American politics and other essays (pp. 145–187). New York: Alfred A. Knopf.Find this resource:
Howell, W. G., & Pevehouse, J. C. (2005). Presidents, Congress, and the use of force. International Organization, 59(1), 209–232.Find this resource:
Howell, W. G., & Pevehouse, J. C. (2007). While dangers gather. Princeton, NJ: Princeton University Press.Find this resource:
James, P., & Hristoulas, A. (1994). Domestic politics and foreign policy: Evaluating a model of crisis activity for the United States. Journal of Politics, 56(2), 327–348.Find this resource:
James, P., & Oneal. J. R. (1991). The influence of domestic and international politics on the president’s use of force. Journal of Conflict Resolution, 35(2), 307–332.Find this resource:
James, P., & Rioux, J. S. (1994). International crises and linkage politics: The experiences of the United States, 1953–1994. Political Research Quarterly, 51(3), 781–812.Find this resource:
Johnson, J. C., & Barnes, T. D. (2011). Responsibility and the diversionary use of force. Conflict Management and Peace Science, 28(5), 478–496.Find this resource:
Jung, S. C. (2014a). Foreign targets and diversionary conflict. International Studies Quarterly, 58(3), 566–578.Find this resource:
Jung, S. C. (2014b). Searching for non-aggressive targets: Which states attract diversionary actions? Journal of Peace Research, 51(6), 755–766.Find this resource:
Keller, J. W., & Foster, D. M. (2012). Presidential leadership style and the political use of force. Political Psychology, 33(5), 581–598.Find this resource:
Kennedy, P. M. (1972). German colonial expansion: Has the “manipulated social imperialism” been ante-dated? Past & Present, 54, 134–141.Find this resource:
Kisangani, E. F., & Pickering, J. (2007). Diverting with benevolent military force: Reducing risks and rising above strategic behavior. International Studies Quarterly, 51(2), 277–299.Find this resource:
Kisangani, E. F., & Pickering, J. (2009). The dividends of diversion: Mature democracies’ proclivity to use diversionary force and the rewards they reap from it. British Journal of Political Science, 39(3), 483–515.Find this resource:
Kisangani, E. F., & Pickering, J. (2011). Democratic accountability and diversionary force regime types and the use of benevolent and hostile military force. Journal of Conflict Resolution, 55(6), 1021–1046.Find this resource:
Klein, G., & Tokdemir, E. (2016). Domestic diversion: Selective targeting of minority out-groups. Conflict Management and Peace Science, forthcoming, available through OnlineFirst.Find this resource:
Kuperman, R. (2003). The effect of domestic and foreign pressure on Israeli decisions to use limited military force. Journal of Peace Research, 40(6), 677–694.Find this resource:
Leeds, B. A., & Davis, D. R. (1997). Domestic political vulnerability and international disputes. Journal of Conflict Resolution, 41(6): 814–834.Find this resource:
Levy, J. S. (1988). Domestic politics and war. Journal of Interdisciplinary History, 18(4), 653–673.Find this resource:
Levy, J. S. (1989). The diversionary theory of war: A critique. In M. Midlarsky (Ed.), The handbook of war studies (pp. 259–288). Boston: Unwin Hyman.Find this resource:
Li, Y., James, P., and Drury, A. C. (2009). Diversionary Dragons, or 'Talking Tough in Taipei': Cross-Strait Relations in the New Millennium. Journal of East Asian Studies, 9(3), 369–398.Find this resource:
Martinez Machain, C., & Rosenberg, L. (2016). Domestic diversion and strategic behavior by minority groups. Conflict Management and Peace Science, forthcoming. Available through OnlineFirst.Find this resource:
Meernik, J. (1994). Presidential Decision Making and the Political Use of Military Force. International Studies Quarterly, 38(1), 121–138.Find this resource:
Meernik, J. (2000). Modeling international crises and the political use of military force by the USA. Journal of Peace Research, 37(5), 547–562.Find this resource:
Meernik, J. (2001). Domestic politics and the political use of military force by the United States. Political Research Quarterly, 54(4), 889–904.Find this resource:
Meernik, J. (2005). Foreign policy orientation, strategic interaction, and the initiation of international crises. Foreign Policy Analysis, 1(2), 165–180.Find this resource:
Meernik, J., and Waterman, P. 1996. The Myth of the Diversionary Use of Force by American Presidents. Political Research Quarterly 49(3), 573–590.Find this resource:
Miller, R. A. (1995). Domestic structures and the diversionary use of force. American Journal of Political Science, 39(3), 760–785.Find this resource:
Miller, R. A. (1999). Regime type, strategic interaction, and the diversionary use of force. Journal of Conflict Resolution, 43(3), 388–402.Find this resource:
Miller, R. A., & Elgun, O. 2011. Diversion and political survival in Latin America. Journal of Conflict Resolution, 55(2), 192–219.Find this resource:
Mitchell, S. M., & Prins, B. (2004). Rivalry and diversionary uses of force. Journal of Conflict Resolution, 48(6), 937–961.Find this resource:
Mitchell, S. M., & Thyne, C. L. (2010). Contentious issues as opportunities for diversionary behavior. Conflict Management and Peace Science, 27(5), 461–485.Find this resource:
Mitchell, S. M., & Moore, W. H. (2002). Presidential uses of force during the cold war: Aggregation, truncation, and temporal dynamics. American Journal of Political Science, 46(2), 438–452.Find this resource:
Moore, W. H., & Lanoue, D. J. (2003). Domestic politics and U.S. foreign policy: A study of cold war conflict behavior. Journal of Politics, 65(2), 376–396.Find this resource:
Morgan, T. C., & Anderson, C. J. (1999). Domestic support and diversionary external conflict in Great Britain, 1950–1992. Journal of Politics, 61(3), 799–814.Find this resource:
Morgan T. C., & Bickers, K. N. (1992). Domestic discontent and the external use of force. Journal of Conflict Resolution, 36(1), 25–52.Find this resource:
Mueller, J. E. (1973). War, Presidents, and Public Opinion. New York: John Wiley.Find this resource:
Nicholls, N. H., Huth, P. K., & Appel, B. J. (2010). When is domestic political unrest related to international conflict? Diversionary theory and Japanese foreign policy, 1890–1941. International Studies Quarterly, 54(4), 915–937.Find this resource:
Oakes, A. (2012). Diversionary war: Domestic unrest and international conflict. Palo Alto, CA: Stanford University Press.Find this resource:
Oneal, J. R., & Tir, J. (2006). Does the diversionary use of force threaten the democratic peace? Assessing the effect of economic growth on interstate conflict, 1921–2001. International Studies Quarterly, 50(4), 755–779.Find this resource:
Ostrom, C. W., & Job, B. (1986). The president and the political use of force. American Political Science Review, 80(2), 541–566.Find this resource:
Park, J. H. (2010). Structural change in U.S. presidents’ use of force. American Journal of Political Science, 54(3), 766–782.Find this resource:
Pickering, J., & Kisangani, E. F. (2009). The international military intervention data set: An updated resource for conflict scholars. Journal of Peace Research, 46(4), 589–600.Find this resource:
Powell, J. M. (2014). Regime vulnerability and the diversionary threat of force. Journal of Conflict Resolution, 58(1), 169–196.Find this resource:
Prins, B. C. (2001). Domestic politics and interstate disputes: Examining US MID involvement and reciprocation, 1870–1992. International Interactions, 26(4), 411–438.Find this resource:
Rummel, R. J. 1963. Dimensions of conflict behavior within and between nations. General Systems Yearbook, 8, 1–50.Find this resource:
Samii, C. 2016. Causal Empiricism in Quantitative Research. Journal of Politics, 78(3), 941–955.Find this resource:
Simmel, G. (1955). Conflict. (K. H. Worlff, Trans.). Glencoe, IL: Free Press.Find this resource:
Smith, A. (1996). Diversionary foreign policy in democratic systems. International Studies Quarterly, 40(1), 133–153.Find this resource:
Sobek, D. (2007). Rallying around the podesta: Testing diversionary theory across time. Journal of Peace Research, 44(1), 29–45.Find this resource:
Stoll, R. J. (1984). The guns of November: Presidential reelections and the use of force, 1947–1982. Journal of Conflict Resolution, 28(2), 231–246.Find this resource:
Tarar, A. (2006). Diversionary incentives and the bargaining approach to war. International Studies Quarterly, 50(1), 169–188.Find this resource:
Tir, J. (2010). Territorial diversion: Diversionary theory of war and territorial conflict. Journal of Politics, 72(2), 413–425.Find this resource:
Tir, J., & Jasinski, M. (2008). Domestic-level diversionary theory of war: Targeting ethnic minorities. Journal of Conflict Resolution, 52(5), 641–664.Find this resource:
Wehler, H.-U. (1970). Bismarck's imperialism, 1862–1890. Past & Present, 48, 119–155.Find this resource:
Williams, L. K. (2014). Hawks, doves, and opportunistic opposition parties. Journal of Peace Research, 51(1), 111–125.Find this resource:
Williams, L. K., Brulé, D. J., & Koch, M. (2010). War voting: Interstate disputes, the economy, and electoral outcomes. Conflict Management and Peace Science, 27(5), 442–460.Find this resource:
Wood, B. D. (2009). Presidential saber rattling and the economy. American Journal of Political Science, 53(3), 695–709.Find this resource: