Analytical Liberalism, Neoclassical Realism, and the Need for Empirical Analyses
Summary and Keywords
Two approaches currently enjoy widespread popularity among foreign policy analysts: Analytical Liberalism and Neoclassical Realism. On the surface, they seem remarkably similar. Both emphasize domestic factors, yet each claims to employ domestic variables in a distinct fashion. How do they differ? To answer that question, it would be helpful to reflect upon examples where scholars applying each approach have addressed the same case, allowing us to contrast their descriptions directly. Few such comparisons exist, however. Instead, as is apparent to even the casual observer, each approach fits neatly into its own niche. Neoclassical Realism appeals to scholars addressing security policy, whereas Analytical Liberalism dominates research in international political economy. Why would both approaches enjoy limited applicability? Here too, a direct comparison of their arguments might illuminate their comparative strengths and weaknesses. A review of how each approach works provides insight into their respective strengths and weaknesses. Under certain conditions, the key traits of the approaches can be revealed. These conditions identify a series of cases deserving closer empirical analysis, which would provide evidence concerning the relative utility of each approach.
Rival Theoretical Approaches
Two approaches currently dominate analyses of foreign policy: Analytical Liberalism and Neoclassical Realism. The two appear quite similar because both emphasize domestic causes—so much so that even specialists sometimes struggle to tell them apart. The best method for clarifying their differences would be to contrast them directly against one another. Applying them to the same cases would establish how they differ and would also enable us to evaluate their relative qualities. Is one consistently more accurate? Is one more parsimonious? In sum, is one more useful than the other? These questions can best be answered via empirical analyses, pitting theories from each approach against one another.
Although both approaches have been used empirically, surprisingly few works have employed both to analyze the same case.1 Even the casual observer of foreign policy analyses would note that scholars tend to apply each approach to a specific set of cases. Researchers working on security issues pioneered Neoclassical Realism; those interested in political economy developed Analytical Liberalism. Why does each approach appeal primarily to a select audience?2 Competing interpretations of specific cases would be useful for revealing each approach’s limitations as well as their strengths. Much can be learned by executing comparative empirical analyses of the two approaches. Prior research highlights each perspective’s relative advantages, and existing works play to each approach’s strengths. This may explain why each appeals so strongly to a limited set of scholars. But there are weaknesses with both approaches. These observations underscore our need for more empirical studies, if we are to appreciate the relative merits of each and continue to refine our theories.
Origins of Analytical Liberalism and Neoclassical Realism
The similarities between Analytical Liberalism and Neoclassical Realism did not arise by coincidence. Scholars developed the two approaches at roughly the same time, following a common intuition. Each perspective originated in the late 1980s and early 1990s, as researchers tackled important questions, especially as they struggled to explain the democratic peace and then wrestled with puzzles concerning the end of the Cold War. As paradigms (i.e., a group of theories built around similar core assumptions), the two rivals crystallized almost simultaneously. For Analytical Liberalism, that key moment came with Andrew Moravcsik’s 1997 piece, where he made the case for recognizing a specifically liberal approach to empirical analyses of international relations. Gideon Rose coined the term Neoclassical Realism in a review article published the following year. Rose noted how scholars working within the Realist framework had reintroduced domestic factors into their arguments, echoing the work of the earliest (i.e., Classical) Realists. In both instances, the author labeled research published earlier that decade. As the timing suggests, researchers had been seeking an explanation for why the Cold War ended with an unexpected fizzle. On the one hand, the failure of established approaches to anticipate this crucial event created the opening for new ideas to flourish; on the other, it simultaneously prompted scholars to look more closely inside countries in order to explain particular foreign policy choices, since the most powerful forces removing tension between the superpowers seemed to have arisen inside the Soviet Union.
Even though these two approaches emerged at roughly the same time and appeared to address the same questions by isolating similar causal forces, they differ in important ways. As their names suggest, they draw on different traditions. Their distinct traits can best be understood by examining their core assumptions. Moreover, these core assumptions shape the chief attributes of the theories within each grouping. Although both wish to draw on forces found in the domestic realm to explain foreign policy choices, each employs those factors to create a particular theoretical construct. These differences reflect competing views of the relationship between domestic and international domains; the role of the state; and the relative importance of domestic political institutions. To some extent, their rival assumptions can be largely deduced from their lineages. Less apparent, however, are some of the consequences that flow from these starting assumptions.
Both Analytical Liberalism and Neoclassical Realism aim first and foremost to explain foreign policy choices, and both do so by focusing on domestic factors. These similarities lead people to conflate the two, especially since individual scholars have employed both types of argument at different times. Readers would certainly err if they used the author as a rule of thumb to classify particular works.3 For the sake of clarity, the best place to begin is with definitions. Although the points below echo prior works, past descriptions of the two perspectives were mostly written to promote one approach over the other—they identified key characteristics but were not aimed at an impartial comparison. Rather than claim one set of assumptions is preferable, this issue remains unsettled. Judgment should be withheld until the two approaches have been tested against one another. The definitions list their core assumptions, which in turn establish the general traits found in each set of theories.
The term Analytical Liberalism denotes a version of liberal thought designed to execute empirical analyses rather than provide normative guidance. As a type of liberalism, this perspective builds off core assumptions typically found in other forms of liberal international relations theorizing: These theories presume that individuals are the most important actors and that individuals seek to enhance their own utility (i.e., each person is free to define his or her aims as each chooses). Outcomes arise from the interaction of individuals as they cooperate or compete, each chasing after his or her goals. To describe/explain/predict foreign policy, this approach focuses first on individuals in the domestic arena, as they seek to bend policies to meet their own ends. Domestic politics produces foreign policy.
The term Neoclassical Realism identifies a variant of realist thought, distinct from the form of realism that developed during the Cold War. Here the adjective signals that this work harkens back to early realist writings. The modifier especially differentiates this realism from the structural variant created by Kenneth Waltz. Waltz drew a distinction between theories created to explain systemic outcomes and theories generated to explain specific foreign policy decisions. The former could be done only through an emphasis on characteristics of the international system itself, he argued; the latter required such intense attention to the unit-level factors (causal forces that vary widely from one case to the next), that Waltz doubted we could ever successfully create generalizable theories based on them.4 Nonetheless, the key elements of Neoclassical Realism begin with core assumptions found in other strains of realist thought. Neoclassical Realists assume that states are the most important actors internationally and that states are primarily concerned with their security because of the nature of the international system. As Gideon Rose explained, “Its adherents argue that the scope and ambition of a country's foreign policy is driven first and foremost by its place in the international system and specifically by its relative material power capabilities. This is why they are realist” (1998, p. 146).
To explain a specific state’s foreign policy, then, Neoclassical Realists use that initial information (a state’s position in the system) especially to identify when and where states feel threatened. To describe/explain/predict a specific state’s response to systemic stresses and strains, however, Neoclassical Realists bring domestic factors into play. As Rose elaborated, “systemic pressures and incentives may shape the broad contours and general direction of foreign policy without being strong or precise enough to determine the specific details of state behavior” (Rose, 1998, p. 147). Rose also explained why these works should be seen as neoclassical: “the impact of such power capabilities on foreign policy is indirect and complex, because systemic pressures must be translated through intervening variables at the unit level” (Rose, 1998, p. 146).
These brief descriptions explain why some might find the two approaches confusing, difficult to distinguish, or even overlapping (Moravcsik & Legro, 1999). Both suggest that domestic factors shape an individual state’s foreign policy. Yet these starting assumptions also illustrate three significant features that separate the two. One point concerns the direction of the political process linking domestic and international affairs. A second centers on their assumptions regarding the state and the state’s relationship with other societal actors. A third point, widely overlooked, concerns how they depict domestic politics. There are further contrasting traits, which one cannot draw directly from these definitions but which emerge once we flesh out each approach more fully. Let us review these three obvious distinctions first, before delving into more subtle differences.
Clearly, each approach models the engagement between domestic and international politics in a particular way, observable in how each describes the direction of political processes. Analytical Liberals begin their arguments with the individual; all individuals reside in the domestic realm of a state. Everything begins with knowing what individuals want; as Moravcsik put it, this information is “analytically more fundamental” than any other (1997, p. 513). In order to shape foreign policy, individuals must organize onto groups, with competing aggregations then vying for control of the state. By dominating in the domestic arena, a group (or coalition) can then dictate to the state what its foreign policy should be. Moravcsik has described domestic politics as a pipeline or “transmission belt.” The state projects into the international system those interests first formulated and expressed from within its boundaries.
Neoclassical Realists start their theories at the other end of this array of relationships, with the state. The state focuses on the international environment because that is the source of the greatest possible threat; the state responds to changes in the international environment by shifting policies. The state’s choice reflects not only forces operating at the international level, but also constraints and opportunities arising in the domestic sphere. Variation arises from domestic factors because, as Rose stated, “leaders and elites do not always have complete freedom to extract and direct national resources as they might wish” (Rose, 1998, p. 147). Thus, the most obvious difference between the two is the direction of political processes. For Analytical Liberals, foreign policy begins in the domestic arena, flows “up” to the state, and from there goes “out” into the international system. In Moravcsik’s words, “Liberal theory rests on a ‘bottom-up’ view of politics in which the demands of individuals and societal groups are treated as analytically prior to politics” (1997, p. 517). Neoclassical Realists, in contrast, see foreign policy starting in the international system, exerting pressures that are transmitted “in” to the domestic context, especially when the state must reach “down” into its own society to amass the resources to support foreign policy.
An excellent way to distinguish the two approaches, then, is to consider how one observes political processes in the formulation and execution of foreign policy: Is it “bottom-up” or “top-down”? Neoclassical Realists expect key decisions to occur first within the state and then to flow from there; for Analytical Liberals, critical aspects of foreign policy originate with individuals, whose differences are resolved in the broader domestic political context, apparent in the lobbying or competition that goes on at that level. The two approaches generate competing expectations, which can be tested against each other empirically, especially via process-tracing. Who initiates policy? Who identifies and defines foreign policy goals?
This discussion feeds into the second important difference between the two approaches: their rival conception of the state. Neoclassical Realists believe it is useful to depict the state (or at least the core element of it responsible for foreign policy decisions, the executive) as something distinct from society (Ripsman, 2002). In this version, the state carries out unique responsibilities and exercises exclusive powers; the state does not share these obligations or capabilities with other domestic actors. Analytical Liberals conceive of the state as being embedded within society, an actor lacking significant autonomy. As Moravcsik expressed it, “In the liberal conception of domestic politics, the state is not an actor but a representative institution constantly subject to capture and recapture, construction and reconstruction by coalitions of social actors” (1997, p. 518). For Analytical Liberals, the state passively follows the instructions fed it from below, in clear contrast to the Neoclassical Realist view.
These rival assumptions concerning the relative autonomy of the state (i.e., its independence when developing and implementing foreign policy) naturally produce different notions about where, when, and how key decisions are made. Each approach relies on a view that resonates with the first point raised earlier. Neoclassical Realists assume that the system matters most and that at the core of the state there exists a relatively autonomous set of individuals developing policy. These independent actors (Ripsman’s Foreign Policy Executive, or FPE) monitor the international arena, establish foreign policy priorities, and then formulate possible options to achieve essential (and unchanging) goals (Ripsman, 2002). Their autonomy does not extend to having the means to implement their choice of policy. To gather the means required for policy, the FPE must engage domestic actors. This interaction occurs through particular processes, however. Neoclassical Realists do not view the state as simply one actor among many in the domestic arena; rather, they assume the state has special powers related to its claim to be sovereign. Neoclassical Realists expect the state to wield tools unlike those of its constituents (e.g., the state is the highest political authority; it alone enjoys the power to coerce legitimately, as when it collects taxes; perhaps most importantly, this executive may exert control over the domestic political agenda). In contrast, Analytical Liberals assume that politics originates in the domestic realm, as groups of individuals vie for dominance. Analytical Liberals see the victorious group (or coalition) controlling the state, telling it what to do, employing it as their instrument internationally.5
This brings us back to the need for empirical analyses, along the lines mentioned earlier. Does the foreign policy executive generate independent assessments of the systemic environment, as the Neoclassical Realists assume, or are its assessments provided through politicized domestic channels? This question again focuses on the process involved. Do dominant domestic actors clash with the foreign policy executive over policy goals? Can the state retain unpopular policies in the face of strong domestic opposition? These contrasting expectations should also produce some interesting patterns. Do states follow consistent goals over time, despite changes in government? Do elections signal significant shifts in foreign policy, as Analytical Liberals expect?
The third point of difference between the two, which we can deduce from their initial assumptions, concerns their contrasting conceptions of domestic politics itself.6 Analytical Liberalism treats domestic politics as largely a function of interests and institutions. While most students of international relations would readily attest that Analytical Liberalism prioritizes domestic politics, those same people do not always reflect on how that approach models competition between groups. Domestic politics comes across as surprisingly sterile because Analytical Liberalism emphasizes structural features in its depiction of domestic politics. Ironically, it does so for the very same reasons Waltz emphasized structural features in his theory of international politics. Structure provides a parsimonious, elegant, generalizable basis for theorizing. The trade-off, however, is the same as in Waltz’s famous theory: an emphasis on structure reduces the role of agency. Most Analytical Liberal applications present domestic politics as curiously straightforward and dull, absent any sense that internal politics involves strategic action. Interests, set prior to politics, explain behavior, as circumscribed by political institutions.
Neoclassical Realism tilts things in the opposite direction. Its focus on the state and top-down processes overemphasizes leadership. This approach assumes that policymakers act strategically: setting political agendas, cutting domestic deals, or even creating political coalitions. Neoclassical Realists’ key assumption (that a core part of the state exists with relative autonomy from society) demands that those running the state exercise leadership—these policymakers lead in the sense of moving first. This generates challenges for modeling, however, since it fails to establish useful boundaries concerning which issues are in play and which are not. If politicians may introduce other policies into domestic debates, with an eye toward reconfiguring domestic cleavages, then it becomes very difficult to isolate the relevant forces at work; endowing state leaders with strategic vision opens up too many doors to model with any ease. This feature of the approach explains why theorists have created a bewildering variety of arguments within Neoclassical Realism. Since Neoclassical Realists are not bound by any particular assumptions regarding how domestic politics works, they have deployed a plethora of causal variables at this level.7 As Waltz had warned, this may not be the best path to parsimonious or generalizable theorizing. It is just as problematic that this lack of boundaries also inhibits progressive theory building within the paradigm, since work in Neoclassical Realism tends to develop newly identified potential causal factors, rather than continually focusing on a handful of the more promising ones.
Empirical analyses would reveal the ability of the two approaches to describe the role of domestic politics in the formulation and execution of foreign policy. Do facets of foreign policy attain salience on their own, which are then forced on political leaders? Or does the importance of issues rise and fall owing to politicians’ tactical machinations? In terms of their descriptions of cases, we would be able to compare how well each captures the way domestic forces influence foreign policy. This in turn would improve our ability to predict policies and thus also inform our ability to offer more effective foreign policy prescriptions.
Analytical Liberalism’s Attributes
Based on the discussion thus far, some general conclusions can be drawn about the advantages and disadvantages associated with Analytical Liberal arguments. The Analytical Liberal approach enjoys a number of strengths because its simplified picture of domestic politics makes for parsimonious and generalizable arguments. It readily incorporates a great deal of work from comparative politics regarding the impact of institutions on domestic outcomes and from there on to foreign policy formulation. When looking at issues where interests are sharply divisive in the domestic realm and where the domestic cleavage that subsequently arises can be clearly described, this approach can draw on a wealth of other information. This information comes not only from painting portraits of domestic interests from work within political science, but also from drawing on findings from other disciplines (such as economics or sociology). These traits fuel vigorous debates within the confines of the paradigm. Such variety can be observed in work on several dimensions of foreign policy, especially within specific domains such as trade or monetary policy. In these instances, a wealth of models from economics have been developed to describe the distribution of interests on a particular policy prior to politics. This provides excellent material for informing the first stage in the “bottom-up” process of Analytical Liberalism.
Recent writings on foreign economic policies also illustrate how the Analytical Liberal framework leaves open the possibility for ad hoc modifications. Researchers have welded constructivist arguments, or models incorporating issue-framing onto more traditional Analytical Liberal arguments, in order to explain why observed outcomes vary from expectations (Hiscox, 2006; Sabet, 2014). These additions complicate the argument and thus may not signal a progressive development, but that depends on how much greater accuracy they generate, to balance out their impact on parsimony. A good deal of recent work in international political economy has followed this path; pushing the material interests as described by economic models to their logical conclusions within the framework, and now adding on to those arguments by looking at other forces that shape interest-formation. The main effort focuses on improving the accuracy of theories in terms of descriptions and predictions.
The more serious weaknesses in the Analytical Liberal approach remain largely unaddressed, however, because this framework begins with a premise that inadvertently obscures significant issues. One such potential problem springs directly from this approach’s simplified portrait of domestic politics. The starting point—the description of interests—often comes from an argument developed outside political science. Change in interests (as captured by another discipline) then carries most of the causal weight in the explanation. Put differently, the explanation for policy change rarely turns out to be political, since domestic institutions channeling interests change too infrequently to explain policy outcomes. How could a passive “transmission belt” determine change? This framework precludes the need to model strategic interaction in domestic politics. Analytical Liberalism therefore produces interpretations that downplay the relevance of domestic politics in favor of nonpolitical forces driving changes in domestic actors’ demands. Politicians (whether they be party leaders in government or the opposition) cannot alter individuals’ desires. While this can be advantageous (it simplifies the analysis by isolating a domain, such as trade or monetary policy), it also ignores the possible links that run across policy dimensions.
Lest one thinks this is not a significant theoretical problem, consider a rather common conundrum. If domestic groups can tell the state the policy for delivering their desire, why do these groups choose to alter a foreign rather than a domestic policy? Staying with the Analytical Liberal work on trade, for instance, consider their starting point regarding preferences. Analytical Liberals assume that individuals seek material gains; individuals’ interest in trade comes from a desire to earn additional wealth.8 Changes in trade policy occur because a group (or coalition of groups) emerges dominant, shifting state policy to serve its desires. But if individuals want material benefits, why adjust trade policy rather than taxes, or transportation costs, or any other feature that affects the wealth of domestic actors? As a group of top scholars on the subject put it, “Why do we observe trade policy at all given its well-known inefficiency in redistributing income?” (Alt et al., 1996, p. 691).
This discussion highlights a second problem in Analytical Liberalism—a logical inconsistency concerning political leaders. In the domestic sphere, politicians are assumed to be passive and unable to act strategically. Yet Analytical Liberals expect these same individuals, acting on behalf of the state, to think and act strategically when operating in international politics. Are leaders rational, calculating, and forward-thinking individuals, or are they not? Why should they act one way in one arena but differently in another?
Neoclassical Realism’s Attributes
The strength of the Neoclassical Realist approach lies in its ability to depict policymaking in the pursuit of a state’s well-defined, consensual interest—as with public goods. This statement is especially true if the desire for that public good persists over time. This explains why Neoclassical Realism appeals to those who study security, where one preference is often widely shared among the state’s constituents. What remains less clear with this approach, however, is how well it may address questions in policy domains beyond security. Other policies that resemble public goods include international aspects, such as environmental or monetary policies. Just as with the approach described here, researchers applying this perspective have largely worked to refine or strengthen the arguments from within, without considering what a comparison versus rival theories of foreign policy might reveal.9
Neoclassical Realist arguments examine decision making inside the state, and this scrutiny should yield more accurate descriptions of past events. Put differently, Analytical Liberals depict state decision makers as passive servants, downplaying their role. Since Neoclassical Realists insist that state executives behave strategically in both the domestic and international spheres, as the state operates in the two realms simultaneously, the role of leaders gains special significance. Neoclassical Realism stresses this point: individuals in the FPE formulate policies in the two areas at once. Decisions reflect the intimate intertwining of forces at both levels. Neoclassical Realists therefore must investigate the details of decision making inside the state. As a result, they should produce more insightful descriptions of past policy choices.
The knock-on effect should be to yield more accurate and detailed descriptions of domestic politics—though few scholars have attempted to do so. This approach does not ignore the factors Analytical Liberals emphasize (i.e., it still includes constituents’ interests and the institutions connecting the state with society), and it rearranges how these force interact. Since this approach grants independence and agency to the state, or at least to the people running the state, it can produce a very different picture of domestic politics. State leaders engage their constituents with an eye to shaping how interests are aggregated, in order to reach the leaders’ preferred policy outcomes. Leaders use their abilities (to package legislation together, to shape agendas) as they alter the make-up of domestic coalitions. Neoclassical Realists assume that international security takes priority, so leaders should attempt to build coalitions that provide the resources they need to achieve security.10
Neoclassical Realism also pictures domestic political institutions as more than a one-way, “bottom-up” transmission belt of interests. The notion that state–society relations exist in a “top-down” fashion brings state powers and the provision of services to the fore; the state delivers infrastructure, order, and a host of other benefits to constituents (along with defense from external threats). This emphasizes how the state may redistribute wealth. The state also reaches down into society, often coercively. States extract resources from society in a number of ways. Taxation or conscription represent important (and often unpopular) interactions between state and society, which Neoclassical Realists might be better at modeling than Analytical Liberals.
Neoclassical Realism, as an approach, also has weaknesses. As mentioned earlier, since it does not begin with an intense focus on interests and institutions at the domestic level, or draw the boundaries on those issues in the way Analytical Liberals do, it offers a different trade-off when constructing specific claims. Analytical Liberal theories tend to be more parsimonious, since they preclude the agency of the state in the domestic arena; Neoclassical Realists preserve state agency, but in so doing they sacrifice parsimony. Where would one draw boundaries, if we were to allow state leaders to act strategically in domestic politics? This shows up in the plethora of Neoclassical Realist depictions of domestic politics—some focus on political leaders’ beliefs, others on misperception or on strategic culture, society’s political narratives, still others on individual politician’s skills, and so forth.
As a result, Neoclassical Realism reflects a wider degree of experimentation. This diversity, however, imposes the first significant disadvantage: It undoubtedly hinders the development of progressive knowledge within the paradigm. In Analytical Liberalism, the steps of argumentation are clear; debate centers on how best to execute each stage. The ad hoc modifications introduced most recently concern ways to refine the information produced at particular steps (especially the initial one). In Neoclassical Realism, the tendency has been to introduce alternative methods for understanding domestic politics, rather than focus on an agreed method, and constantly tweak that style of argument. One author in this approach may have his preferred views on how to model domestic politics and apply his perspective consistently, while other Neoclassical Realists employ alternatives. With so many different options available for discussion, there may never be a “mainstream” or typical version of Neoclassical Realism.
By failing to circumscribe which domestic causal factors matter, leaving internal politics extremely fluid, and granting political leaders the ability to link policy across domains, Neoclassical Realism is clearly not easily employed to formulate predictions—one would have to know in advance when and where political leaders will move in domestic politics. Lacking that knowledge, we do not know what factors to include when making predictions. This in turn makes it difficult to reach consensus within the approach on how to formulate predictions or policy prescriptions. This problem may have a solution: insist on including not only the foreign policy domain in question, but also the budget. No policy can operate for long without financial support. Any policy must be consistent with management of the budget. As noted earlier, Neoclassical Realism assumes that the state’s goals are relatively clear and fixed over time; questions revolve around variation in the means used to pursue those goals. If the state’s financial position exercises a chief constraint across time and space, it ought to be a central consideration in Neoclassical Realist arguments.
A second major drawback arises from the assumption that leaders formulate policy inside the state. While earlier this was mentioned as a strength (since it forces close examination of decision-making processes), this information may not be available. Executives take decisions in secrecy, making crucial evidence frustratingly difficult to access. Without this material, politicians’ motives remain open to interpretation. Are foreign policy decisions really influenced primarily by system-level pressures? Can that essential question be answered without trustworthy, verifiable accounts of how executives formulated decisions? These challenges complicate the testing of Neoclassical Realist theories across cases and thus may constitute a serious obstacle to the further refinement of Neoclassical Realist theories. In contrast, Analytical Liberal arguments emphasize two tracks—one visible, the other less so—as groups vie for control over policy. The visible process involves overt contests between rival groups; in democracies, elections provide the opportunity to observe and analyze this competition. In the second track, groups express demands through lobbying. This second process is more difficult to observe, since much of it typically occurs out of view. Although one can definitely apply Analytical Liberal arguments to nondemocratic states, evidence of “bottom-up” demands influencing policy tends to be harder to gather than in democratic settings.
The ways Neoclassical Realists have responded to this problem may also be fragmenting their approach, rather than moving this school toward a more progressive research agenda. One response consists of allowing the perception of systemic constraints and opportunities to vary individually. This obviously pulls them away from their Realist roots, since individual misperceptions could account for any behavior. Moreover, this changes the focus from the state and places it on individuals. Another response has been to treat Structural Realism as providing the “correct” information for policymaking, in contrast to the errors undertaken by statesmen (Schweller, 2006). This, however, reduces Neoclassical Realism to a theory of mistakes. More often, Neoclassical Realists model policymaking from the viewpoint of decision makers, as rational actors under constraint. (This is not to deny that contrasting that starting point to one based on Structural Realism can be illuminating.) There appears to be little consensus on whether or not the assumption that policymakers are rational is useful. This disagreement again allows for variety and experimentation, but may hinder the accumulation of knowledge by impeding the development of an agreed-upon core approach.
In sum, then, both Analytical Liberalism and Neoclassical Realism exhibit strengths and weaknesses. Does Analytical Liberalism strike a more effective balance between parsimony, generalizability, and accuracy? Does the ill-defined nature of domestic politics in Neoclassical Realism add to its utility, or does it make it too open-ended? These questions are difficult to answer without considering what we have learned from past applications of the theories to actual cases. Without enough empirical applications of the right sort—where we can compare the two—these questions remain difficult to answer.
Less Apparent Traits—The Purpose of Theorizing
Other points distinguish the two approaches, though these distinctions would be more easily perceived if detailed comparative empirical work existed. Perhaps the most important—but far from obvious—difference lies in the chief objective of theorizing in each approach. While it is commonly recognized that each is a theory of comparative foreign policy first and foremost, the two conceive of the essence of foreign policy differently. Analytical Liberals concentrate on understanding state preferences. As Moravcsik eloquently put it, “For liberals, the configuration of state preferences matters most in world politics” (1997, p. 513). Later in this article he reiterated this point: “Variation in ends, not means, matters most” (1997, p. 522). Neoclassical Realists, however, focus on understanding the means states choose to pursue their goals (for an example, see the eschewing of a focus on preferences in Taliaferro, Lobell, & Ripsman, 2009, p. 31, n. 84).
This difference may seem slight at first, but it deserves greater attention. For Analytical Liberals, the key question about foreign policy centers on what states want. Variation in state’s goals must therefore be explained first. The answer to this crucial question lies, they argue, with domestic actors’ desires. This approach considers the selection of the means for achieving those ends less important. Since Analytical Liberals often begin their research with a single policy domain in mind (i.e., they pose questions about one dimension of policy, with explicit boundaries), this is surely convenient. Ultimately, it may be troubling, however, since this is exactly why we are left scratching our heads when we realize dominant domestic groups can manipulate either foreign or domestic policies. The approach is poorly suited to explaining why certain routes are employed rather than others, which also carries over into different dimensions of foreign policy. To continue with illustrations from the politics of trade, recognizing the dominant domestic actors’ desire for protection may not help explain whether the policy adopted employs an ad valorem tariff, a specific rate, a quota, or some form of non-tariff barrier.
In contrast, Neoclassical Realists rarely question what states prefer. Given the competitive, anarchic nature of the international system, Neoclassical Realists confidently assume that all states value security over other goals. The central question becomes not “what does the state seek?” but rather “how does the state attain its highest priority”? Neoclassical Realism therefore generates theories that reduce the key question about foreign policy to the choices concerning means. This works as long as one is focused on security. Neoclassical Realists are not troubled by possible variation in preferences regarding other aspects of foreign policy because domestic actors typically reach a consensus around the need for security, especially when international threats arise. Other facets of foreign policy are then subordinated to security needs. Citizens disagree about the best method for attaining the level of security desired because security has a price, and the means selected produces distributional consequences.
This difference in focus—means versus ends—helps explain why each approach tends to be employed in a specific subfield within foreign policy analysis. Analytical Liberalism dominates work on foreign economic policies. Neoclassical Realism fits better with security studies. Yet there are no obvious reasons one must be strictly limited to any particular issue-area. Some have tried to employ one approach to the other’s main policy domains. Analytical Liberals have argued that, within security policy, states may still struggle to set their goals (e.g., Narizny, 2003). Historically, some states have struggled to establish defense priorities. This is especially true when a state has clearly stretched its commitments beyond its available resources. Several of these episodes have been studied in detail, though usually from a single theoretical perspective. These cases can be revisited, with a more systematic application of rival theories.
On the flip side, Neoclassical Realists may address questions concerning the chief subjects Analytical Liberals have studied, such as foreign economic policies.11 There are fewer examples to cite, though it is easy to draw out what these would look like. For an issue such as trade or monetary policy, a Neoclassical Realist would assume that individuals ultimately desire wealth; what is contested, in practice, is each individual’s means for attaining wealth, rather than a specific foreign economic policy outcome. Since Neoclassical Realists employ a top-down process, they cede the initiative to political leaders. The starting question concerns the goals of the foreign policy executive. These leaders would then address domestic politics with an eye to promoting one policy over another, setting political agendas, and packaging policies together in order to create viable political coalitions.
By defining the core explanadum in a particular way, each approach poses a different question. Each perspective works toward comprehending a particular facet of a country’s foreign policy; thus, they do not produce equivalent analyses. This also explains why there is so little direct dialogue between the two. Yet much could be learned about each by taking them away from their “comfort zone.” To understand how far each could be taken into the domain of the other, direct comparisons are needed, where each approach is applied to the same empirical puzzles.
Choosing Cases for Comparison
Comparing the two theories by applying both to the same empirical cases is illuminating, in part because some particular cases may prove more valuable than other cases. Among those recommended here are those that showcase the strength of Analytical Liberalism and those that demonstrate the utility of Neoclassical Realism. But there are others, which remain puzzling precisely because all best efforts to explain the observed outcomes have been unsatisfactory, owing to shortcomings within the approach that has already been applied. Might the alternative approach offer new insight into these puzzles?
First, for cases demonstrating the utility of Analytical Liberalism, democracy is a crucial trait. Democratic practices offer much more transparent domestic politics. Not only would evidence be much more accessible, but elections provide obvious moments in time to focus on, when testing the two arguments against each other. More specifically, on the one hand, Analytical Liberalism posits that domestic political competition resolves disputes over foreign policy goals. Elections driven by foreign policy issues give clear predictions of how policy should have changed, since elections reveal the distribution of interests at a particular time. Foreign policy should change after a turnover in government. Neoclassical Realism, on the other hand, argues that systemic conditions drive changes. Neoclassical Realism applies to the decisions made in democracies, but it points to different causes behind policy change. Therefore, there are good reasons to concentrate on the foreign policies of democratic states.
Second, cases where the distribution of domestic interests clearly shifted, but systemic conditions did not, would provide specific examples for exhibiting the strengths of Analytical Liberalism. If a democratic state operates in a stable systemic environment, will policy shift owing to alterations in the domestic setting only? Along the same lines, there are instances where internal political institutions—the rules governing domestic competition—changed. For instance, a rapid shift in suffrage should redistribute the internal strength domestic actors enjoy. Alternatively, cases can be selected where changing economic circumstances altered the distribution of domestic actors’ interests; in trade policy, for instance, actors’ preferences shift when their country’s relative position in the international economy changes. These provide some of the best grounds for applying Analytical Liberalism.
Third, we should identify and analyze cases where domestic conditions remained stable, but the systemic environment fluctuated, in an effort to locate examples of when Neoclassical Realism is on favorable grounds. For these cases, internal groups, parties, and domestic political institutions exhibited stability, but systemic threats either rose or fell. These conditions provide the best grounds for exposing the strengths of Neoclassical Realism, while Analytical Liberalism should expect stable foreign policies.
Fourth, the analysis should cover multiple dimensions of foreign policy in the same time period. The discussion should not be restricted to security policy or to a single dimension of foreign economic affairs (such as trade or monetary policy). We know each approach is employed primarily by scholars interested in one dimension of foreign policy, so we often have half such work already. Moreover, most applications of Neoclassical Realism seem to have been focused on U.S. foreign policy. What we need are analyses of cases that challenge existing interpretations, especially when those interpretations are not particularly convincing.
A List of Suggested Cases
Quite a few interesting cases share some of these four traits. These are discussed in chronological order, rather than ranked in terms of their potential. In most cases, the only analyses that exist are quite one-sided. Typically, only one approach has been applied to explain the policy outcome.
British foreign policy shifted substantially in the second half of the 1840s, most famously when the country removed tariffs on grain. The repeal of the Corn Laws, however, represents the peak in a larger trend to reduce protection, involving a wide range of products and services. Past research has, predictably, used the Analytical Liberal approach. The well-known problem, of course, is that the party in power responsible for these changes had been elected with a mandate to defend the tariff. Neoclassical Realists could offer a reasonable alternative argument by focusing on the international systemic environment—particularly the danger posed by France and the impact of that threat on British statesmen. The protectionists, led by Robert Peel, won office in 1841 primarily to deal with the country’s chronic budget problems; Peel’s solutions included adjustments to the tariff schedules, which he continued later in the decade. Analytical Liberals tend to try to interpret the resulting legislative changes as a transfer of wealth from one group (landowners, or at least wheat growers) to others, whereas it might be much more accurate to describe Peel’s legislation as a redistribution of tax burdens.
A Neoclassical Realist interpretation might offer a keener explanation for the timing of the repeal of the Corn Laws than any Analytical Liberal explanation. Many historians have linked the outbreak of the potato famine in Ireland to changes in Peel’s attitude toward the Corn Laws. Some interpret Peel’s shifting stance on protection as a sign that he had to act consistently toward all parts of the realm; he could not justify protection for agricultural goods in England, Scotland, and Wales while using state money to provide food aid to Ireland. A Neoclassical Realist would reevaluate this material, using a different starting point: The Corn Laws were designed to ensure British self-sufficiency. Statesmen of this era knew the Corn Laws were needed for domestic sources of food, since the experience of Britain’s lengthy wars against Napoleon remained fresh. Self-sufficiency in this era, however, meant concentrating on a handful of crops; the potato famine demonstrated with staggering clarity the vulnerability of such crops to disease. Did Peel fear the British wheat crop might be wiped out? Might this offer the motive for a top-down explanation for Repeal of the Corn Laws?
The next case flows from this period: British foreign policy in the 1850s. Several questions persist about Britain’s involvement in the Crimean War. On the one hand, Analytical Liberalism may well capture the domestic interests said to be driving Lord Palmerston’s foreign policy decisions, especially in terms of Britons’ concerns centering on the fate of religious and ethnic minorities in eastern Europe and the Balkans, which Palmerston played upon to gain popularity. On the other hand, the war against Russia did not develop as intended. How did the Black Sea come to take priority over the Baltic? Britain dispatched forces to several regions. Did it do so because of the country’s war aims or because it was a reflection of the choice of means? Although this is an example of security policy, Analytical Liberals should be able to produce useful insights. Neoclassical Realists, however, would emphasize policy goals and the use of military force in a different manner, focusing on systemic competition and state leaders’ definition of security. The key features of the case they would explore concerned Britain’s apparent inability to raise adequate military forces quickly. Domestic constraints may have shaped practical war aims. This case offers promising opportunities for both approaches.
A third interesting case involves German foreign policy in the late 1880s and early 1890s. Germany dabbled with trade liberalization before t swinging back toward protectionism; its foreign policy shifted from trying to prevent the formation of tight alliances to binding itself to Austria-Hungary. While Imperial Germany was not a full-fledged democracy, it did have high levels of popular political participation and elections, providing researchers evidence regarding interests. Moreover, first the ruler and then the governing coalition shifted, allowing Analytical Liberalism to explain why both foreign economic and security policies might have changed. Neoclassical Realism, in contrast, would look first at the systemic changes occurring and then at the top-down political process to explain those same policy changes. Since Germany altered both its security and its foreign economic policy, both approaches had the opportunity to demonstrate their utility.
British foreign policy offers a fourth interesting case to consider, covering the period between 1900 and 1914. Several facets of the country’s foreign policy came into question, owing to systemic challenges on several fronts, setting up the opportunity to illustrate the usefulness of Neoclassical Realism. British policy in this era has been studied quite extensively because Britain was then struggling to reset its defense priorities. At the same time, economic challengers emerged, including Russia, Germany, Japan, and the United States. Britons therefore debated whether the country’s foreign economic policies needed to adapt, along with holding discussions over defense expenditures. Prior research within each perspective addresses parts of these debates (Brawley, 2010; Narizny, 2003; Taliaferro, 2008). These deserve closer comparison; moreover, the connections between security and foreign economic policy options need to be reexamined.
Although Britain emerged victorious from World War I, the country again faced rising economic competition and overstretched military commitments. The country needed to reestablish its economic foundation, previously based on trade and international investments, but it struggled to do so in the 1920s. In the 1930s, the rising German and Japanese threats caught the attention of British policymakers, noted when the Treasury relinquished its “Ten-Year Rule.”12 These events should provide excellent grounds for revealing the directions of domestic political processes, goal setting, and so on. While Neoclassical Realists have been drawn to this example, their interpretations deserve to be challenged by thoughtful Analytical Liberal rivals.
American foreign policy in the interwar period provides another excellent opportunity to examine the two approaches. The White House and Congress fought over security policy throughout Franklin Roosevelt’s first two terms. Domestic pressures flowing from the bottom up fueled isolationism. Congress responded to these demands with a series of Neutrality Acts—legislation constraining the country’s foreign policy. The White House—viewed through the Neoclassical Realist lens as the foreign policy executive—wanted a different set of policies. How did American policy evolve over the last years of the 1930s? Did political pressures flow from the bottom up or from the top down? Put another way, which was more important, public demands or presidential leadership? Answers to those questions will help determine which approach provides a better description of American policies in this period.
American foreign policy in the early phases of the Cold War provides another interesting opportunity to contrast the two approaches. How would Analytical Liberals explain the changing nature of American military expenditures and alliances in the late 1940s and early 1950s? Were domestic pressures guiding the Truman and Eisenhower Administrations, as they first demanded significant (and controversial) military contributions from American allies, then walked those demands back? Or were these really just decisions made by the foreign policy executive, absent shifting domestic pressures? Existing Neoclassical Realist interpretations offer telling explanations of these events (Brawley, 2010; Taliaferro, 2004), but since the largest shift in foreign policy came with Eisenhower’s adoption of the “New Look,” Analytical Liberal arguments might give a stronger interpretation of change. More in-depth comparisons of the arguments could be very useful here.
In the opposite way, the set of policy changes introduced by the Nixon administration in the early 1970s provides another excellent opportunity for comparing the two. In this instance, U.S. foreign economic and security policy changed abruptly, without the sort of flip in government just mentioned. Many of these changes were carefully orchestrated. For instance, consider the close timing of the new stance on the People’s Republic of China and the end of the Bretton Woods commitments. Previous work has exposed how weak the domestic pressures were for driving these economic policy changes, prompting the suggestion that new ideas or other less tangible forces were at work (Gowa, 1984; Odell, 1979). Perhaps, however, a Neoclassical Realist would find other processes at work, especially given how state leaders (primarily Henry Kissinger and Richard Nixon) prioritized the security challenges confronting the United States. Again, a serious comparison of the two approaches might provide some new perspectives on these policy decisions, but would also put us in a better position to evaluate how well each approach addresses policy decisions outside their favorite niche.
As noted in an earlier section, scholars have applied each of these approaches primarily within specific policy domains. Employing the approaches to the suggested cases may provide some interesting evidence that they work outside their preferred policy domains. This would directly address concerns about the generalizability of both approaches. More importantly, the comparisons would provide insight into the relative merits of the two in terms of accuracy as well as relative parsimony. In short, if we are to appreciate their strengths and weaknesses, we need these head-to-head empirical applications to evaluate the two approaches.
Creating a Productive Debate: A Research Agenda
These empirical applications can generate evidence for comparing the two approaches both clearly and consistently. They produce competing expectations for when policy should change, the locus where policies originate, as well as the direction of political processes. However, since they consider the essence of foreign policy to be different, their expectations are not entirely comparable—for example, Analytical Liberalism’s emphasis on preferences rather than Neoclassical Realism’s focus on means.
Moravcsik made a plea for this sort of empirical analysis long ago: “In the long run, comparative theory testing should be aimed at a clearer definition of the empirical domain within which each major theory performs best” (1997, p. 541). Perhaps Analytical Liberalism is the best approach for explaining some aspects of foreign policy, and Neoclassical Realism others. Have we arrived at that conclusion through rigorous testing or simply by applying each approach to cases where their strengths are clear and their weaknesses hidden? To improve in the future, their vulnerabilities must be grasped. Dialogue and debate will propel us in the right direction, but that dialogue requires further empirical work.
Hiscox, M. (2008). The domestic sources of foreign economic policy. In J. Ravenhill (Ed.), Global political economy (2d ed., pp. 95–133). Oxford: Oxford University Press.Find this resource:
Kitchen, N. (2010). Systemic pressures and domestic ideas: A neoclassical realist model of grand strategy formation. Review of International Studies, 36(1), 117–143.Find this resource:
Ripsman, N., Taliaferro, J., & Lobell, S. (2016). Neoclassical realist theory of international politics. Oxford: Oxford University Press.Find this resource:
Alt, J. E., Frieden, J., Gilligan, M. J., Rodrik, D., & Rogowski, R. (1996). The political economy of international trade. Comparative Political Studies, 29(6), 689–717.Find this resource:
Brawley, M. R. (2010). Political economy and grand strategy. New York: Routledge.Find this resource:
Gowa, J. (1984). State power, state policy: Explaining the decision to close the gold window. Politics and Society, 13(1), 91–117.Find this resource:
Hiscox, M. (2006). Through a glass darkly: Attitudes toward international trade and the curious effects of issue framing. International Organization, 60(3), 755–780.Find this resource:
Kupchan, C. (1996). The vulnerabilities of empire. Ithaca, NY: Cornell University Press.Find this resource:
Moravcsik, A. (1997). Taking preferences seriously: A liberal theory of international relations. International Organization, 51(4), 513–553.Find this resource:
Moravcsik, A., & Legro, J. (1999). Is anybody still a realist? International Security, 24(2), 5–55.Find this resource:
Narizny, K. (2003). Both guns and butter, or neither: Class interests in the political economy of rearmament. American Political Science Review, 97(2), 203–220.Find this resource:
Odell, J. (1979). The U.S. and the emergence of flexible exchange rates: An analysis of foreign policy change. International Organization, 33(1), 57–81.Find this resource:
Rathbun, B. (2008). A rose by any other name: Neoclassical realism as the logical and Necessary extension of structural realism. Security Studies, 17, 294–321.Find this resource:
Ripsman, N. (2002). Peacemaking by democracies. University Park: Pennsylvania State University Press.Find this resource:
Rose, G. (1998). Neoclassical realism and theories of foreign policy. World Politics, 51(1), 144–172.Find this resource:
Sabet, S. (2014). Feelings first: Non-material factors as moderators of self-interest effects on trade preferences. Unpublished manuscript at http://scholar.princeton.edu/sites/default/files/ssabet/files/sabet_feelings_first.pdf.
Schweller, R. (2006). Unanswered threats: Political constraints on the balance of power Princeton, NJ: Princeton University Press.Find this resource:
Snyder, J. (1991). Myths of empire. Ithaca, NY: Cornell University Press.Find this resource:
Taliaferro, J. (2004). Balancing risks: Great power intervention in the periphery. Ithaca, NY: Cornell University Press.Find this resource:
Taliaferro, J. (2008). The partisan origins of great power politics? International Studies Review, 10(2), 308–310.Find this resource:
Taliaferro, J., Lobell, S., & Ripsman, N. (2009). Introduction: Neoclassical realism, the state, and foreign policy. In S. Lobell, N. Ripsman, & J. Taliaferro (Eds.), Neoclassical realism, the state, and foreign policy (pp. 1–41). New York: Cambridge University Press.Find this resource:
(1.) Moravcsik and Legro (1999, p. 7) criticize Neoclassical Realists for failing to set up such tests. “This is obscured because most realists test their favored explanations only against other variants of realism—normally Waltzian neo-realism—rather than against alternative liberal, epistemic, and institutionalist theories, as they once did.”
(2.) Moravcsik noted a desire for such work (1997, p. 541): “In the long run, comparative theory testing should be aimed at a clearer definition of the empirical domain within which each major theory performs best.” Few examples exist.
(3.) That includes my own work; I have used, and continue to apply, both approaches.
(4.) Ripsman, Taliaferro, and Lobell claim Waltz’s distinction has been overdrawn in ways that are counterproductive (2016, pp. 1–2).
(7.) Ripsman, Taliaferro, and Lobell note that Neoclassical Realism has been criticized for including a “laundry list” of variables from the domestic level (2016, p. 60); their response is to categorize arguments into four broad clusters (2016, p. 9), though the real problem concerns whether there are any borders concerning what can and cannot be included in Neoclassical Realism.
(8.) Most approaches emphasize income as the focal point, though trade’s impact on individuals’ assets has also been modeled.
(9.) Kupchan (1996) took a step in this path: He examined whether domestic demands shaped state responses to overstretch, or whether policymakers set security priorities on their own. This work preceded the development of Neoclassical Realism, so Kupchan did not systematically reflect on how state leaders approached domestic politics (in terms of raising resources) when formulating policy.
(10.) Of course, policymakers can also find resources in the international system (i.e., allies).
(11.) Ripsman, Taliaferro, and Lobell rightly focus on this as one set of questions for which Neoclassical Realists can provide new interpretations rivaling existing arguments (2016, p. 72).
(12.) The “Ten-Year Rule” refers to the government’s assumption, for budgetary purposes, that Britain would not be involved in another military conflict in the coming decade.