Helen V. Milner. Interests, Institutions and Information: Domestic Politics and International Relations. Princeton, New Jersey: Princeton University Press, 1997. x + 309 pp. $32.95 (paper), ISBN 978-0-691-01176-9; $55.00 (cloth), ISBN 978-0-691-01177-6.
Reviewed by Erik Gartzke (Pennsylvania State University)
Published on H-Pol (June, 1999)
Interests, Institutions, and Information Reviewed
International politics is experiencing a major shift in focus coinciding with the collapse of the Soviet Union and the fall of the Berlin Wall. The 'unearthing' of domestic politics by students of international relations, and the observation that regime type is a factor important for understanding state behavior, has led to a movement to embrace and explore the impact of domestic political processes on happenings abroad. Helen V. Milner's book, Interests, Institutions, and Information: Domestic Politics and International Relations, is a strong, intellectually engaging voice in this movement.
Milner offers a theory of international politics with an emphasis on explaining cooperation between states. To do this, she argues, it is imperative to incorporate the competitive processes of politics inside, as well as between, borders. The book is organized into four sections; an introduction, three theory chapters, four case studies, and a conclusion. The text is excellently written and requires limited knowledge of the subject to enjoy (though some interest in international relations is necessary). The research involves some technical analytical tools (game theory), but presentation of this work is conveniently tucked away in an appendix. Let me begin by attempting to place the book in its context. I then offer a few reactions to Milner's theoretical arguments (I leave assessment of the case studies to more qualified readers).
Mainstream study of international relations in the post-World War II period emphasized conflict and systemic or state-level analysis. Realism sees domestic politics as hierarchical, necessarily involving leaders and followers. In contrast, international politics is anarchic. All states are nominal equals--lacking the protection of a higher central authority--and must fend for themselves. Realism largely ignores the domestic politics of states on theoretical grounds. Security--the preeminent concern of members of the anarchic international system--trumps competing objectives at home and reinforces the need to solidify the domestic political hierarchy. States must act as if they are unitary (even if they are not) because of international competition and because of the consequences of weakness and internal division.
In retrospect, it is not difficult to draw a link between cold war politics and the analytical biases of cold war political scientists. The advent of the 'new world order' and discoveries such as the democratic peace--the observation that liberal states seldom fight each other--undermine key elements of the realist paradigm even while they appear to demand a shift in attention toward neglected topics, such as international cooperation. With timing likely to be the envy of any stand-up comedian, Robert Putnam (1988) provided an analytical perspective that seemed to lend itself to allaying many of the weaknesses of realism. Putnam's two-level game posits a Janus-like leader, facing bargains at both international and domestic levels. To stay in office, the leader must satisfy his or her domestic constituency (In a democracy, it is a majority coalition of parties or voters. In an autocracy, it might be key members of the societal elite.). At home, the leader must compete with other hopefuls for his office. Abroad, the leader competes with other leaders facing their own domestic constituents. Playing domestic politics and ignoring international concerns may cost the leader his job (the realist point, e.g. Mohammad Mossadeq). However, playing the game of international politics to the exclusion of domestic bargaining may also propel the leader out of office (e.g. Woodrow Wilson).
REACTIONS TO THE BOOK:
Students of international politics have begun to evolve a new consensus that cold war realism, exemplified by the work of Kenneth Waltz, placed insufficient emphasis on the domestic level of Putnam's two-level game. Milner follows Putnam in arguing that domestic politics matters, that the interaction of domestic and international politics creates outcomes that are non-intuitive from the perspective of realism. "My central claim is that states are not unitary actors; that is, they are not strictly hierarchical but are polyarchic, composed of actors with varying preferences who share power over decision making" (p. 11). Three factors; interests, institutions, and information, define bargaining within the state and contribute to formulating foreign policy. Three actors are also modeled; an executive, a legislature and interest groups (there is also a foreign leader but not a true two-level game). These factors interact in a complex and contextual manner. Milner provides extensive deductive bases for her claims and then assesses the arguments with case studies. One of the most valuable contributions of the text are a series of insights about how informational asymmetries influence domestic and international politics.
Milner provides a stylized model of domestic and international processes. The executive bargains internationally with another executive. Agreements are referred to the legislature for ratification or rejection. Milner goes on to assess the effect of preference divergence between actors, information asymmetries, agenda control and other alterations of institutional structure. Surprisingly, she finds that the introduction of domestic politics often degrades the possibility of international cooperation. Milner also shows that information asymmetries can enhance the prospect for cooperation.
The richness and detail of the theoretical arguments mean that most readers are likely to quibble with the author on a few points. This is not a particularly constructive method for review, however. Elements such as what assumptions an author should make in modeling institutional procedures between branches of government are at least partially subjective (more below) and are adjudicated, ultimately, only through empirical verification. Neither is it likely to add much to the reader's discernment if I, say, differ about interpretations of Britain's motives for participation in NATO. Instead, I note a few concerns about larger questions facing the text in the hopes that doing so encourages others to examine this thought-provoking book for themselves.
i. Parsimony and the Critique of Realism:
Few would argue with the claim that states are made up of individuals and groups with diverse interests that are manifest in political competition. However, it is not clear why this insight necessarily challenges realism. Realists know that states are not really unitary actors. The unitary state is a simplifying assumption used for the sake of achieving other normative goals of theorizing. Every theory involves tradeoffs between assumptions and attempts to reflect greater empirical detail. A theory that explains everything explains nothing.
In normal science, the way to compare theories is to assess excess empirical content. Which theory accounts for more facts with the least theoretical bulk? Parsimonious theories are thus optimization problems. Realism's durability would seem to belie strength in addressing these tradeoffs. This means that assessment of the relative worth of theories is always contextual, depending on the question one seeks to answer. In physics, relativity was shown to account for things that the laws of motion did not; Einstein covers Newton. Innovations like super string theory are supposed to be thrilling. Still, I imagine that even the cognizenti probably stoop to the more mundane Newtonian logic on occasion (say, if one wants to determine the terminal velocity of a skydiver). Realism never claimed to grasp all the facts, just, from the perspective of realists, the important ones. Since all theories are stylized, we cannot accept or reject theories by comparing the assumptions.
I address not so much the merits of Milner's theory as the text's claims about the consequences for realism. The arguments posed in the text would seem to account for some things that realism does not. Yet, even if the text were to demonstrate excess empirical content (It does not. More on this below.), we might still concede that realism is simpler, and thus, more attractive for certain questions. Milner points out that "Domestic politics matters because the state is not a unitary actor" (p. 16). Neither is the state three actors. We accept assumptions because they are powerful, not because they are right.
The claims on behalf of the argument exceed what is demonstrated in the text. For example, the conclusion argues that, "Adopting the assumption that states are polyarchic retains parsimony and improves explanatory power" (p. 256). The claim is neither confirmed nor denied by the findings of the text. As noted above, we do not know whether the theory is parsimonious (The term is misused here. Parsimonious theories combine the attributes of simplicity and explanatory power.). We do know that including domestic politics makes the theoretical arguments more complex. Nor can we say anything about predictive power. As Milner notes, the case studies can offer plausibility, but they do not demonstrate the veracity of the theory. To do so, we must at least have information about how representative these case studies are of cooperation (or attempts at cooperation) in general. Milner herself points out that the cases selected are idiosyncratic (for example, they all involve democracies).
Milner takes a very constructive approach to these concerns, acknowledging the limited external validity provided by case study analysis and encouraging others to assess her hypotheses empirically. As such, my comments are not a critique as much as encouragement to others to consider testing Milner's hypotheses.
iii. The Curse of Domestic Politics:
One of the major implications of the work is that domestic politics is bad for international cooperation. Competition between actors in the domestic political realm is shown to often lead to bargaining failures abroad. The finding is powerful, but it is also troubling. Why should this be so? Is it true?
First, empirically, it does not appear to be correct that greater polyarchy leads to increased problems in cooperating. The number and diversity of international cooperative acts has increased geometrically in recent decades. At least, (I should say with the caveat below) the number of cooperative institutions has dramatically increased. Further, most of the cooperating is occurring among states that are the least hierarchical. Work on the democratic peace shows that democracies participate in more international organizations and that their participation is more extensive and binding.
Second, there are theoretical reasons not to expect cooperation to break down as supposed in the text. I suggest a few. When one gets non-intuitive results in modeling, this can either be very exciting or cause for concern. Since there are reasons to be suspicious on this account, it might be appropriate in future research to determine which aspects of the model's assumptions drive the non-intuitive results. The Nash bargaining game, for example, is not as appropriate as the text indicates. Nash bargaining assumes cooperation rather than allowing cooperation to derive (or not) endogenously from the model.
Instead of critiquing the assumptions, however, I will take them as given and offer a few other suggestions. Suppose that one lives in a polyarchic society. How could one address the problem of not being able to successfully bargain internationally? One approach would be to add to the institutional structure of the society in minimize the barriers to successful bargaining. The choice of decision rule between the executive and the legislature in the text is interesting since it parallels the fast track procedure in the U.S. Congress. Normally, proposal power is not limited to the executive and Congress does not face a closed-rule vote. However, in fast track, Congress voluntarily adopts such a procedure, presumably to enhance the ability of the executive to bargain abroad, but also to facilitate congressional goals. We do not know what the implications of other decision rules would be, but since it disperses some of the power of the legislature to the executive, presumably it increases polyarchy. It seems implausible to argue that the United States adopted a procedure repeatedly, over several decades, that was intended to make it a more effective bargainer internationally but that in fact made it less successful.
Governments that do not achieve desirable things for their constituents face trouble. Finding a way to make things work is an opportunity for credit taking. One of the things that polyarchy makes more likely is policy entrepreneurialism. If certain actors are unable to deliver preferred outcomes, others will step into the opportunity. As with markets, the competition implied by polyarchy brings with it additional incentives to achieve successful bargains abroad.
Finally, there is the problem of side payments. An actor, the legislature for example, that is disgruntled by a proposed agreement could be provided with a benefit elsewhere that is sufficient to "sweeten the pie." Students of Congress talk of "log rolling." Perhaps key members of Congress receive consideration for pet projects in return for approval of major foreign policy initiatives such as NAFTA. Milner discusses the possibility and impact of side-payments, but they are not reconciled with other aspects of the argument or with some of the case studies. Thus, even accepting that polyarchy makes international cooperation more difficult, methods exist to make cooperation more likely.
iv. Inside Out, or "Conflating Cooperation and Cooperative Institutions:"
The text does not distinguish between international cooperation and institutions designed to manage international cooperation. The two are arguably distinct. For example, the case studies all deal with cooperative agreements that involve institutions associated with administering the agreements (Bretton Woods and the ITO, the International Civil Aviation Agreement, the European Coal and Steel Community, NAFTA and the EMU). States could conceivably cooperate with or without institutions. Indeed, the presence of institutions may indicate challenges to cooperation. If institutions are costly, then they are only likely to be used when they provide sufficient value added. Thus, institutions are not an inevitable component of cooperation. A theory of cooperation is not necessarily a theory of international institutions and vise versa.
Domestically, institutions perform several functions. Perhaps the most salient of these for the discussion here is the role of institutions in altering future political calculations. As Milner notes, "Certain institutions privilege particular actors, and hence policy choices reflect their preferences more" (p. 19). The formation of institutions is strategic. One's interests are more likely to be maintained if one can institutionalize these interests. Thus, at least one motive for institutions domestically is inter-temporal constraint. An incumbent administration can use institutions to realize its preferences at the expense of future incumbents with different preferences. Yet, this must also be true internationally. International institutions constrain the range of future policy options, not just internationally, but domestically. Thus, the causal arrow may also go the other way. International cooperation may be hampered by domestic politics, but international institutions constrain domestic politics, at least in part. The implications of this argument are just touched on here, but I suspect that a general challenge for two-level games is to treat both domestic and international institutions in a consistent manner.
v. Unitary Actors and Cooperation:
The central tenant of the text is that one must typically have domestic politics to explain international cooperation. Claims to this effect pepper the book but are perplexing in light of widely recognized research documenting precisely the opposite. "The structure of domestic preferences is a central independent variable explaining cooperation" (p. 17). In The Evolution of Cooperation Axelrod offers a simple simulation of an anarchic international system and then endogenously induces cooperation from egoistic unitary actors. To cooperate, actors in the simulation need a payoff structure that does not discount future payoffs for cooperation 'too heavily.' Axelrod's intention was to show realists that anarchy is fully compatible with cooperative behavior, but the finding is also a puzzle for the text under review (it is strange that Axelrod's work is not cited). If cooperation can occur in a system without domestic politics, then the challenge posed by the text is really one of accounting for why states do not cooperate more. Realists are then in the odd position of being optimists about the potential for cooperation. This too is resolved through several ancillary arguments, including issue linkage, etc.
In conclusion, the text is a provocative and valuable addition to a growing literature on domestic contributions to international politics. The book provides a sophisticated though intuitive conceptual framework for understanding international interaction. Milner offers a provocative thesis and testable hypotheses that challenge the conventions of systems theories. It differs from a number of other attempts at two-level games in that it makes use of insights about the role of information in political bargaining. Where the text falls short, it offers avenues of pursuit for future research. In short, it is a text worthy of scholarly attention.
. Realism, like any durable paradigm, constitutes a big tent. Important divergences of view exist among realists. I simply follow Milner in pointing out some of the more prominent consistencies.
. Robert Putnam. 1988. "Diplomacy and Domestic Politics: The Logic of Two-Level Games." International Organization 42(3): 427-60.
. Kenneth N. Waltz. 1954. Man, the State, and War: A Theoretical Analysis. New York: Columbia University Press. Kenneth N. Waltz. 1979. Theory of International Politics. New York: McGraw-Hill Publishing Company. Waltz coins the term "neo-realist" to characterize a shift in emphasis from the state to the international system. Distinguishing realism from neo-realism in the context of this review complicates matters while adding little of substance.
. Use of the term "polyarchy" by Milner is likely to cause confusion since it is identified with Robert Dahl. Milner uses the term to represent political structures that stand the middle ground between anarchic and hierarchical.
. As Milner herself acknowledges, "[Realism and neoliberal institutionalism] are presented to show their contrasting hypotheses; they are not systematically tested in this book" (p. 23). I hasten to add that I am not a partisan of the realist cause. There are much better ways to skin the realist cat. In particular, realist explanations for international conflict fail to motivate the behavior of interest. See, James D. Fearon. 1995. "Rationalist Explanations for War." International Organization. 49(3): 379-14. Erik Gartzke. 1999. "War is in the Error Term." International Organization. Forthcoming.
. Bruce Russett, John Oneal, and David R. Davis. 1998. "The Third Leg of the Kantian Tripod for Peace: International Organization and Militarized Disputes, 1950-1985." International Organization 52(3): 441-68.
. The formal literature assessing fast track argues that it provides a Pareto improving move in the structure of decision making between the executive and the legislature because it is restrictive. See Elizabeth M. Martin. 1997. "An Information Theory of the Legislative Veto." The Journal of Law, Economics, and Organization 13(2): 319-43.
. Robert Axelrod. 1984. The Evolution of Cooperation. New York: Basic Books.
This review was commissioned for H-Pol by Lex Renda <firstname.lastname@example.org>
Copyright (c) 1999 by H-Net, all rights reserved. This work may be copied for non-profit educational use if proper credit is given to the author and the list. For other permission, please contact H-Net@h-net.msu.edu.
If there is additional discussion of this review, you may access it through the list discussion logs at: http://h-net.msu.edu/cgi-bin/logbrowse.pl.
Erik Gartzke. Review of Milner, Helen V., Interests, Institutions and Information: Domestic Politics and International Relations.
H-Pol, H-Net Reviews.
Copyright © 1999 by H-Net, all rights reserved. H-Net permits the redistribution and reprinting of this work for nonprofit, educational purposes, with full and accurate attribution to the author, web location, date of publication, originating list, and H-Net: Humanities & Social Sciences Online. For any other proposed use, contact the Reviews editorial staff at email@example.com.