Using the following attached reading titled "An Institutional Explanation of the Democratic Peace by Bruce Bueno de Mesquita, James D. Morrow, Randolph M. Siverson and Alastair Smith compose a paper length reading response on what it discusses about Democratic Peace Treaty. This paper must include a thesis, argument, and in text citations.

Please please help me with this question I will make sure to leave a good review!!!!

apsa An Institutional Explanation of the Democratic Peace Author(s): Bruce Bueno de Mesquita, James D. Morrow, Randolph M. Siverson and Alastair Smith Source: The American Political Science Review . Dec., 1999. Vol. 93, No. 4 (Dec., 1999). pp. 791-807 Published by: American Political Science Association Stable URL: https://www.jstor.org/stable/2586113 JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide facilitate new forms of scholarship. For use information technology and tools to increase productivity and arship. For more information about JSTOR, please contact supportejstor.org. Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at http://about.jstor.org/terms American Political Science Association is collaborating with JSTOR to digitize, preserve and extend access to The American Political Science Review ISTOR his content downloaded from 131.179 220.13 on Tee, 28 Dec 2021 02:43:32 UTC All use subject to https://about.jstor.org/termsAmerican Political Science Review Vol. 93, No. 4 December 1999 An Institutional Explanation of the Democratic Peace December 1999 An Institutional Explanation of the Democratic Peace international disputes of democratic states are in the hands comes of conflict between democratic states. The in- of individuals who have experienced the politics of com- ternational and domestic norms are induced from the BRUCE BUENO DE MESQUITA Hoover Institution peting values and interests and who have consistently responded within the normative guidelines of bounded observed patterns of behavior in international conflicts JAMES D. MORROW Hoover Institution competition. In situations where both parties to a dispute that these arguments seek to explain. That democracies RANDOLPH M. SIVERSON University of California, Davis are democracies, not only do both sides subs h sides subscribe to thes are willing to abandon their normative commitment to leaders of both are also fully cognizant that the peaceful resolution of disputes in the face of a ALASTAIR SMITH Yale University bounded competition is the norm, both for themselves and threat to their survival by another state that does not their opponents. adhere to those norms is entirely plausible. In order to W e examine formally the link between domestic political institutions and policy choices in the qualify as an explanation of the observation, however, context of eight empirical regularities that constitute the democratic peace. We demonstrate that A closely related argument is that citizens in democ- that assertion must be derived independently of the emocratic leaders, when faced with war, are more inclined to shift extra resources into the war racies abhor violence and so constrain their leaders observation, either from prior axioms or from unre- effort than are autocrats. This follows because the survival of political leaders with larger winning coalition from pursuing violent foreign policies. As succinctly lated empirical evidence, Otherwise, we cannot know hinges on successful policy. The extra effort made by democrats provides a military advantage over autocrats. explained by Morgan and Campbell (1991, 189), "the what the argument predicts about seemingly contradic This makes democrats up utional constraints cause them to mobilize key feature of democracy is government by the people tory patterns of evidence, For instance, analyses by resources for the war effort. In addition to trying harder, democrats are more selective in their choice of and . . . the people, who must bear the costs of war, are James and Mitchell (1995) and Forsythe (1992) of targets. Because defeat is more likely to lead to domestic replacement for democrats than for autocrats, usually unwilling to fight." Adherents of these perspec- covert operations suggest that, providing democratic democrats only initiate wars they expect to win. These two factors lead to the interaction between polities that tives also argue, however, that democracies are willing leaders can escape public scrutiny, they often under- is often referred to as the democratic peace. to set aside their abhorrence of violence or their take violent acts against other democracies. Does such respect for other points of view when they confront evidence contradict a norms-based arguments, or do authoritarian states, because the latter do not share the norms apply only to interstate conflict at the level of The study of international relations has produced few widely accepted generalizations, One of wars they initiate, democracies pay fewer costs in terms these common values. As stated by Maoz and Russett crises and war? of human life and fight shorter wars than nondemo- (1993, 625), "when a democratic state confronts a Second, a related difficulty is empirical. The histori- these, sometimes even asserted to be an empire cratic states (Bennett and Stam 1996; Siverson 1995). nondemocratic one, it may be forced to adapt to the cal record is replete with democratic states that fol- ical law (Levy 1988), is that democracies do not fight (6) Transitional democracies appear more likely to norms of international conflict of the latter lest it be lowed policies at variance with the norms argument. wars with one another. The empirical evidence for this fight than stable regimes (Mansfield and Snyder 1995; exploited or eliminated by the nondemocratic state Many democratic states pursued imperialistic policies claim is quite strong (Bremer 1992; Maoz and Abdolali Ward and Gleditsch 1998). (7) Larger democracies that takes advantage of the inherent moderation of and, in building empires, engaged in wars that were 1989; Oneal and Russett 1997; Ray 1995). Recent seem more constrained to avoid war than do smaller democracies." about subjugateon rather than self-protection. It may efforts to cast this empirical observation in doubt democracies (Morgan and Campbell 1991). We believe that any explanation of the democratic be correct to argue that democratic states resort to notwithstanding (Farber and Gowa 1995; Layne 1994; Although these observations about democracy and peace must satisfy two criteria. First, it must account realist strategies in the face of a powerful nondemo- Schwartz and Skinner 1997; Spiro 1994), extensive, war are part of an important pattern, they lack a for the known regularities that are often grouped cratic opponent who threatens their existence, but too rigorous statistical tests all show a significant propen- coherent explanation. Several possibilities have been together to define the de st significantly sity for democracies to be nearly immune from wars put forward, but none has gained broad acceptance. tion that accounts for all the regularities obviously is weaker states for this argument to be sustained as an with one another (Maoz 1998; Russett 1995). Associ- We propose a game-theoretic model that may help more comprehensive than those that account for only explanation for the democratic peace. It is difficult to ated with this so-called democratic peace are seven bring closure to the debate on the causal mechanism some. Furthermore, becau planations additional empirical regularities related to war prone- governing the eight regularities mentioned above. The are generally constructed in response to the observed ture. The explanation we propose, by contrast, explains ness and democracy, all based on empirical observa- model will be shown to account for the empirical regularities, the more patterns that are explained, the the willingness of democracies to undertake imperial- tions. (1) Democracies are not at all immune from record regarding the first six of the eight patterns. The more credible the explanation, provided that i stic or colonial conquest. This observation is the sort fighting wars with nondemocracies (Maoz and Abdolali regularities concerning transitional d I democracies not come at the expense of parsimony, Second, we of novel fact for which an explanation of the demo- 1989).' (2) Democracies tend to win a disproportionate (Mansfield and Snyder 1995; Ward and Gleditsch believe that a credible explanation also should suggest cratic peace should account, and we return later to our share of the wars they fight (Lake 1992; Reiter and 1998) and the constraints on great power democracies novel hypotheses that do not form part of the corpus of explanation of imperial wars by democracies. Stam 1998b). (3) When disputes do emerge, demo- (Morgan and Campbell 1991) are stated in monadic the democratic peace, Further credibility to the overall The institutional-constraints argument holds that cratic dyads choose more peaceful processes of dispute form. These regularities cannot be evaluated in terms explanation is added if these novel hypotheses are democracies are more deliberate in their decision settlement than do other pairings of states (Brecher of a model of strategic interaction without additional borne out by the evidence. making than autocracies because their procedures pre- and Wilkenfeld 1997; Dixon 1994; Mousseau 1998; information. Specifically, without data on the demo- The existing norms-based and institutional-con- clude unilateral action by leaders, This is thought to Raymond 1994). (4) Democracies are more likely to cratic state's adversary, we cannot evaluate the relevant straints arguments fail both tests. The model we pro- raise the costs of violence, Maoz and Russett (1993, initiate wars against autocracies than are autocracies institutional and resource relationships. We will pro- pose provides a direct explanation of six regularities 626) state: "Due to the complexity of the democratic against democracies (Bennett and Stam 1998). (5) In vide an interpretation of our model that is consistent and an account consistent with the two nondyadic process and the requirement of securing a broad base with both monadic results, but we cannot undertake a regularities mentioned earlier and suggests numerous of support for risky policies, democratic leaders are more rigorous assessment. novel hypotheses that are supported by evidence. reluctant to wage wars, except in cases wherein war Bruce Bueno de Mesquita is Senior Fellow and James D Senior Research Fellow, the Hoover Institution, Stanford University, Later, for instance, we show that our theory implies seems a necessity or when the war aims are seen as Stanford CA 94305-6010, Randolph Siverson is Professor of Politica that democracies d justifying the mobilization costs."? This suggests, how- Science, University of California, Davis, CA 95616, and Alastair THE DEBATE efforts than do nondemocracies. We cite independently ever, that democracies should be unlikely to wage war imith is Assistant Professor of Political Science, Yale University, derived evidence. In addition, we have shown else- New Haven, CT 06520-8301. generally, not just against other democracies. The debate over the war behavior of democratic states, The research reported here was supported by the National Science where that the institutional model we propose also The empirical record does not support such a con- Foundation under grant SBR-9409225. We are greatly indebted to and particularly the democratic peace, centers on accounts for variation in economic growth across re- clusion.' Rather, it shows that democracies do not fight Raymond E. Wolfinger for his facilitation of this research and to whether a normative or an institutional explanation gimes and explains why leaders with failed public one another but do indeed engage in wars with author- David Austen-Smith, Fiona McGillivray, Will Moore, Brace Russell. best accounts for the known facts (Thompson and policies tend to last in office longer than leaders with Kiron Skinner, Allan Stam, and several anonymous referees for their Tucker 1997). Normative accounts focus on several successful policies (Bueno de Mesquita et al. n.d.). helpful comments. Some studies suggest that democracies are, on the whole, more different presumptions about democracies. One such There are two difficulties with the norms-based Without some specification of war aims and mobilization costs, the pacific than autocracies (Benoit 1996, Ray 1995). Yet, such general supposition is that they share a common value system arguments in the literature. First, the arguments ap- last phrase of this statement allows anything with respect to demo- ratic war behavior. war avoidance cannot account for their relative propensity to fight including respect for individual liberties and competi- pear to be ad hoc. The presence and substance of 3 Benoit (1996) suggests that on average democracies are slightly less with different regime types. tion. As stated by Dixon (1994, 17): norms are established solely by reference to the out- war prome than other systems 791 792 This content Tug. 28 Dec 2021 02:43:32 UTC 131.179 220.13 on Tue, 28 Dec 20 131.179 220.13 - Tug, 28 Der: 2021 02:43:3 All use subject to https:/about jslor .orgterms All use subject to https:/about jslor.orgtermsAmerican Political Science Review Vol. 93, No. 4 An Institutional Explanation of the Democratic Peace December 1999 itarian regimes.* The argument based on the cheapness disenfranchised citizens with no say in the choice of tify whatever dependencies exist between regime type perforce cannot give a large amount of such benefits to of expressing opposition seems stronger than the other leadership. All members of S have the right to partic- and war participation and outcomes; and to evaluate each member of their winning coalition. putative institutional explanations, but it also has short- ipate in choosing the government. A subset of the the prospects that leaders are retained in office as a Democratic leaders are more likely to try hard to win comings, one of which is its failure to account for the selectorate forms a winning coalition, which we denote function of institutional arrangements. These issues their wars than are autocrats. If they do not expect to well-known rally-round-the-flag effect observed in de- as W. Members of the winning coalition are those influence whether leaders concentrate resources on the win, then they try to avoid fighting; This implies that mocracies at the outset of crises and wars (Mueller people whose support is required to keep the incum- pursuit of national goals or on benefits to key constit they pick and choose their conflicts carefully, which has 1973; Norpoth 1987). This effect suggests that there is bent in office. If the incumbent cannot find W members uents. In showing that this is true, we explain the several consequences. Democrats are more likely to not an inherent abhorrence of violence in de of the selectorate to support her, then she is removed empirical regularities that constitute the democratic win wars than autocrats for two reasons. First, if they Most important from a theoretical position, none of from office. Since we assume that leaders are keen to peace. need to, democrats try hard, spending resources on the the institutional-constraints arguments has a suffi- stay in power, they are eager to satisfy members of the Before turning to the formal representation, war to advance their public policy goals (Reiter and ciently well-developed theory of how and why demo- winning coalition. pause to describe the basic structure of our model and Stam 1998b). Second, fearing public policy failure, cratic institutions constrain leaders in the particular Typical categories of regimes can be related to the outline the intuition that leads to the democratic peace democrats try to avoid contests they do not think they way that produces the eight regularities observed, institutional variables on which we focus. In a democ result. We start by assuming that two nations, A and B, can win. Since two democrats in a dispute both try whereas other institutional arrangements do not. racy with universal suffrage, for instance, S is approxi- are engaged in a dispute. National leaders must decide hard, both can anticipate that, if they go to war, each Rather, these arguments generally just assert that mately equal to N, and W is large, typically being a whether to start a w g their will spend lots of resources in a risky situation in which democratic leaders are more constrained. majority of S. In monarchies and military juntas, W and neither is disproportionately advantaged by greater Bueno de Mesquita and Lalman's (1992) signaling S are typically small relative to N. In some authoritar- objectives or rely instead on a negotiated settlement. If effort. Therefore, democrats are generally inclined to explanation accounts for three of the eight observed ian states, S is quite large, and in others it is very small, war occurs, then leaders must decide how much effort to make to achieve military victory. By this we mean negotiate with one another rather than fight (Lake regularities, but it does not, for instance, explain why while W is always small in such systems. In a communist 1992; Stam 1996, 176-8). By contrast, autocrats typi- democracies win a disproportionate share of their wars state, for instance, $ has typically been a figure between what proportion of available resources a leader is prepared to allocate to the war effort rather than to cally reserve their resources for domestic uses, as their or why their costs are lower. Bueno. de Mesquita and just the members of the Communist Party (i.e., 57 Siverson's (1995) model accounts for these regularities million people in China) and all citizens, since univer- other purposes. Obviously, leaders who dedicate large political survival depends on satisfying key constituents through the distribution of private goods. Autocrats do but not for the democratic peace. Both those models sal suffrage (in largely meaningless elections) is com- quantities of resources to the war are more likely to have in common the assumption that democracies are monplace for such systems. Authoritarian regimes are win, but at the cost of not having those resources not have a great need to produce successful public available to reward their supporters. The citizens re- policies. Consequently, they try less hard than demo- more constrained than auto reasons o often characterized by rigged elections in w in which S is crats in war, and they sometimes fight wars in which theoretical parsimony, however, we prefer that this be large, but this has little effect on actual governance ceive payoffs based on the outcome of the crisis-be it a war or a negotiated settlement-and the rewards that their chances are poor because defeat does not so a deductive result of a general model rather than an other than to raise the risks for members of W if they accrue from resources not consumed in the war effort. greatly affect their prospects of political survival at assumption. That is, we wish to account for defect from the incumbent, as we explain below. home. Democrats, by their superior level of effort, empirical regularities without assuming that one type Incumbent leaders (as individuals or as a governing Given these payoffs, the winning coalition decides whether to retain or replace the current leader. often defeat autocratic foes and achieve successful of political system is more constrained than another. coalition) select and implement public policies, which policy outcomes. This helps enhance their reselection. Instead, we will demonstrate how institutional arrange- inevitably have a public goods component nd a private A polity's institutional arrangements shape the se- ments produce different levels of constraint in different goods component. Leaders have only limited resources lection criteria that supporters use to determine political systems and what effect those institutional to allocate to different policy goals and to help keep whether to retain the incumbent. Hence, political arrangements have on behavioral incentives and the them in office. They can put everything into public institutions determine which outcomes allow a leader THE GAME empirical generalizations of interest. policy that benefits everyone in the polity, everything to keep her job and which do not. As we shall see, these into private goods that are consumed only by m differences shape the policies that leaders choose. We assume that two nations, A and B, are engaged in Our explanation shows that the behavioral incen- tives-these could be called norms-are endogenous of the winning coalition, or any mix in between. If they Citizens in general enjoy the benefits of public a dispute, Our model examines the fundamental deci- to certain political institutions and the interests that spend resources on, for instance, providing defense for policies, whether or not they belong to the winning sions that national leaders make under this contin- gency. In the game, leaders choose to fight or to sustain them.' We make no claims about the citizens' the citizenry, then they cannot use those same re- coalition. The advantage that members of W have is negotiate a settlement. If the choice is to fight, then abhorrence of violence or the case with which they sources to provide special privileges to the members of that they also enjoy a share of whatever private goods might protest government policies. In fact, we assume the winning coalition. If they buy the necessities of are allocated by the leadership. On average, each leaders decide how many of their available resources tivated by the member of W receives a share of private goods equal to they are prepared to commit to the war effort. In that political leaders in all systems are m national defense only from cronies in the R/W, where R is the available pool of resources and W ourse, either side in a dispute can resort to same universal interest: They desire to remain in office. coalition, then the reduced competition to provide war, We, however, consider the restricted game in We make no normative assumptions about differences defense results in an inefficient provision of that public is the size of the winning coalition, in the values or goals of democratic leaders or their As the winning coalition increases, holding the bud- which the leader in nation A chooses between the use good while cronies skim money off the top for their of force and a negotiated settlement. If she decides to followers as compared to authoritarian leaders or their personal gain. Thus, scarcity necessarily requires lead- get constant, each member's share of private goods attack, then she also picks an effort level, by which we followers. ers to make choices over just how much to focus their decreases. This makes public policy benefits loom mean she allocates some proportion, g, of her avail- limited time and other resources on providing gener- larger in the overall utility assessment of members of if the war. Once POLITICAL INSTITUTIONS ally beneficial public policies and how much to focus on the winning coalition in more democratic polities a attacked, the leader in nation B also picks an effort just satisfying the wants of their core supporters. compared to autocracies. One consequence is tha level, ga: If nation A decides not to attack, then the All polities are characterized by institutional rules The main concerns here are to identify how re- democratic leaders, being just as eager to retain office dispute is settled peacefully through negotiations. pertaining to the role of citizens in influencing the sources are allocated during an international dispute, as their authoritarian counterparts, must especially The war's outcome is a function of the relative effort selection of government leaders. Let any polity consist given variations in institutional arrangements; to iden- concerned about policy failure. To reduce the risk of by each side. That is, who wins depends in part on how of N citizens. Let S be the subset of / that participates policy failure and subsequent deposition, they make a leaders choose to allocate their limited resources. in the selection of the political leadership, We call this * The theory we develop does not require that we assume minimal larger effort to succeed in disputes. This means that When the dispute is settled, either through negotiation subset the selectorate. Then N - S is the set of inning coalitions (Riker 1962). Indeed, the results generalize they are willing to spend more resources on war effort or war, the domestic audiences in A and B then decide straightforwardly to any defined magnitude for W. Yet, it makes and only engage in fights they anticip winning. In whether to retain or to depose the incumbent (Fearon " Although democracies do not fight wars with one another, they most sense to begin from a minimal winning context because that contrast, leaders with small winning coalitions reserve 1994; Smith 1998). To make this decision, they evaluate often become engaged in militarized disputes (Oneal and Russert appears to be incentive compatible with the objective of leaders to more resources for distribution to their supporters in their payoffs under each contingency and decide 197). See Senese (1997) for an alternative viewpoint. maximize their gains, A central question of interest to us here is how Earlier work demonstrates the endogenous relationship ch that it deviates from the form of private goods. As long as they can provide whether they are better off remaining in the incum- tions to foreign policy interests (Bueno de Mesquita and Siverson minimal winning. The comparative static analysis we undertake substantial private goods, they are at less risk of being bent's winning coalition or defecting to a prospective 1995; McGillwray 1997: Schultz 1996). focuses in part on that question. deposed than are their democratic counterparts, who new leader. 793 794 330 13 Tur. 28 Dee 2021 02:43:32 UTC 131.179 220.13 - Tue. 28 Der 2021 02:43:3 All use subject to https:/about jslor cupterms All use subject to https:/about jslor.orgtermsAmerican Political Science Review Vol. 93, No. 4 An Institutional Explanation of the Democratic Peace December 1999 Structure of the Game of modeling this process is to treat p, as the probability Reselection resources, she is uncertain of the qualities of a prospec- 1. The leader in nation A chooses between war and that a random variable e, with distribution b(e), is less tive domestic rival. We represent the distribution of than a function of the variables h(84, gm, M), Following the international dispute, the leaders in each negotiations. If she selects war, then she also common example of which is the probit model, where nation face reselection. The members of the selector- possible challengers with the cumulative density func- chooses how hard to fight, &, E [0, 1]. ate evaluate the payoff they received under the incum- tion FA(x), where F(x) = Pr(c S x). For technical 2. If war occurs, then the leader of B, having observed D(e) is the standard normal distribution, and h(84, 88. bent. They compare this with what they expect to convenience, we assume the distribution of challengers A's effort level, chooses how hard to fight, gg E [0, M) = M - + g - 8gr.7 receive if they depose the incumbent and choose a is exponential: F (x) = 1 - e-xu. What can a challenger credibly offer the selectorate? 1]- domestic challenger. Deposition is not simply a matter 3. Nature determines the outcome of the war. of concluding that the leader did a poor job during the He cannot make credible promises regarding how he 4. In each nation, the members of the winning coali- dispute. Rather, it is a question of whether the mem- will perform during a dispute or on other policy Negotiations tion, having observed the international outcome and bers of W believe they will be better off under alterna- questions. Knowing this, the selectorate's members the level of private goods allocated to them, decide When nations enter negotiations, we assume they have tive leadership. focus on the reservation value they expect if they whether to retain their leader or defect to a domes- expectations about the likely outcome of the bargain- Incumbents are deposed when they can no longer choose a new leader. We assume that the reservation tic political rival, thereby removing the current ing process. In particular we assume the expected convince W members of the selectorate to support value for picking the challenger is c + RAISA. leader from office. rewards for A and B from a negotiated settlement are them. If the package of benefits an incumbent offers to Incumbents can anticipate what they must give sup- x and 1 - x, respectively. We might suppose this deal, her supporters is better than the rewards any chal- porters to defeat challengers. They simply must pro- We assume that only A can start a war. This assump- tion is benign. The question of whether B wants to x, reflects the relative military balance, M, but it need lenger can credibly offer, then the incumbent can find vide more utility for their coalition members than that initiate is answered by simply flipping the labels A and not. Other factors can also be influential. As Fearon W members of the selectorate who will retain her in offered by the challenger, or B. Our central question is how political institutions (1995) points out in his bargaining model, the indivis- office. If, however, the incumbent fails to provide affect crisis behavior. ibility of the policy space, the sustainability of deals in benefits to the winning coalition in excess of what a (1 - BA) W + HA + Vi(2), the future, and the role of current negotiations in challenger can credibly promise to provide, then the altering the future military balance all influence which incumbent can no longer garner enough support to where u represents the performance of the leader on Settling Crises by War deals can be achieved through negotiations." form a winning coalition. At this point, supporters all policy dimensions other than the international We model war as a costly lottery in which each player's The payoff to A from a negotiated settlement is x. defect, and the incumbent is ousted. dispute. This utility term is quite intuitive. V(z) is the expected utility from the war depends on the probabil- which is enjoyed by all citizens in A. That is, this payoff Defection is risky. There is no shortage of challenge utility supporters derive from the outcome of the policy ity that its side will win or lose and the utilities is a public good. The citizens in B enjoy the benefit 1 - ers and prospective defectors. Consequently, a member of the leader in the international dispute." (1 - RA) is associated with each possible outcome, In this section x from negotiation. of the current winning coalition cannot be certain of the proportion of resources reserved for distribution as we develop our notions regarding the probability of being essential to the successor government. That is, b private goods to the winning coalition after spending victory (and defeat) and the attendant utilities. Ux(negotiations) = x; Up(negotiations) = 1 - x- defecting, members of the winning coalition place their go on the war effort. Of course, if there is no war, then The values of victory and defeat are normalized to private goods benefits at risk. We cha ize that risk The total pool of resources (R) is diminished one and zero, respectively. In addition, players pay a as increasing as the pool of available members for a Table 1 summarizes the policy payoffs associated by whatever portion goes to the war effort, if any. What per capita cost, k, associated with the war's destruction winning coalition increases and as decreasing as the remains is distributed evenly to the members of the and the risks of fighting. Therefore, the utility of victory with each possible outcome of the international dis- necessary size of the winning coalition in winning coalition so that each member receives (1 - equals 1 - k, and the utility for defeat is -k. pute. We let z represent the generic outcome. Every- that the risk increases as S increases and decreases as 8) R /W. Of note is that members of the winning Many factors shape the outcome of a a war. Observ- one in A receives the policy payoff (z) associated W increases, For simplicity's sake, we treat the proba- coalition are assumed to be certain to receive their able military capabilities certainly play an important with the outcome z, and all members of polity B receive bility of being a member of the successor wi share of private goods if they stay with the incumbent, part, as do short-term shifts in government priorities by the payoff Va(2). coalition in A as W/5 . Then, the expected utility but they only receive such goods probabilistically if putting more national resources behi a war effort from private goods in a new coalition if one defects to they defect. In addition, neither the incumbent nor the (Kim and Morrow 1992; Organski and Kugler 1980; TABLE 1. The Policy Payoffs Associated with challenger is simply (R/W ) (WAIS) - RAISA. challenger can promise to distribute any resources that Powell 1996a, 1996b). The probability of victory is International Outcomes Like the incumbent, the challenger proposes a mix of are destroyed or lost by the state during a war. This presumed to increase as the total military advantage Outcome VA(=) proves important later in understanding why autocrats dedicated to the war effort by one side grows relative to VB(2) public and private goods allocations, Of course, the Negotiations selectorate does not know what the challenger can do not make the same allocation decision as demo- the other side. Therefore, the victor is likely to be the A wins the war actually deliver, whereas they have observed the per- cratic leaders. nation with the most total military capabilities dedi- A loses the war formance of the incumbent. The selectorate must infer Leaders want to remain in office. At the time leaders cated to the war. the ability of the challenger (Riker 1996). We denote choose their actions, including whether to wage war We consider two types of military capabilities: the the competence, strength, or ability of the challenger in and how to allocate resources, they cannot be certain observable military balance, M, and the proportion of A as cy and use comparable notation for the challenger of the quality of p spective political rivals. Yet, given additional national resources committed to the war Other common examples used in international relations include in B." When the incumbent is making choices about the distribution of possible challengers, the incumbent effort, g- The subscript will refer to nation A or B, as Bueno de Mesquita, Morrow, and Zorkk (1997), who set whether to fight or negotiate and how to allocate can assess how outcomes influence her prospects for appropriate. The military balance, which takes values reselection. Leader A remains in office as long as she between 0 and 1, represents the ratio of observable 4()=1 1 ife20 " In autocracies and other nondemocratic regimes, campaigns may offers her winning coalition at least as much as a military assets of the two sides. Additional resources not be public but instead take such forms as secret discussions about (M. MAMS &Age) - MA + 2(M - 1) - my+ Bar - R potential challenger credibly promises. Hence, the dedicated to the war by either country are drawn from where in, and me are intangible military utilities, and the standard a coup, assassination, or other means to overthrow the incumbent. Success depends on attracting enough supporters to depose the probability the incumbent remains in office given out- the R, resources each leader has at her disposal. By ratio of forces model (see, c.g., Bueno de Mesquita 1981), which sets sitting government and replace it with one led by the challenger. come z and effort & is Pr((1 - 84) RAIW + VA(2) + choosing to devote the proportion g, of R, to the war 0 ifco Because of the risks as (captured by our notion that the probability of effort, she generates an additional gr in military assets, if c E [0, 1 being essential to a successor government is equal to WIS), challenge cracies general any operate in secret. It would be a mistake where r represents the exchange rate between re- sume here that all members of society benefit equally from to believe that challengers not always exist. Even in the most international outcomes. As car imes. As earlier work shows (Ba sources and military capability. and h(M. g.. go) = (M + rg.)/(1 + /g4 + 7g.)- Hirshleifer (1989) are always people cager to take over the role et al. 1997). however, foreign policy objectives are a function of The probability that A wins in a war, denoted p. = samines the implications of variation in these of leadership. One obvious indication is that following the sudden institutional arrangements. Countries with small winning coalitions PA(84: 8m, M), is increasing in military balance, M, and * Nations' private information about their capability, which may take death of an incumbent there is always someone willing to step in and typically fight for resources that can be targeted as private goods to the form of intangible military assets of morale and leadership assume leadership. Frequently, many people compete for the oppor- the regime's supporters. By contrast, countries with large winning A's effort, 54, and is decreasing in B's effort, ga- The (Reiter and Stam 1998a, 1998b; see also Morrow 1989), also influ- coalitions fight over policies that have larger public goods compo- probability B wins is pg = 1 - PA. A general method ences negotiations, although the closest supporters of the leader are the obvious nents. 795 796 This content downloaded from 131.179 220.13 - Tug. 28 Dee 3021 02:43:32 UTC 131.179 220.13 - Tue, 28 Der: 2021 02:43:3 All use subject to https:/about jslor ampterms All use subject to hups:/about jslor.orgtermsAmerican Political Science Review Vol. 93, No. 4 An Institutional Explanation of the Democratic Peace December 1999 HA Z CA + RUS ) = PR((1 - 84) RAWA - RUS, + The preceding discussion suggests the incentives of V(E) + HA = CA) - F(V (2) + (1 - & ) RAW - leaders. We now specify those formally by denoting the probability that nation B wins given an effort level of ate, or part of the disenfranchised portion of society. RASA + .). payoffs to leaders under each contingency they face. In "#- This probability of victory also depends upon A's Yet, increased effort comes at the expense of having All else equal, leaders who perform well on other particular, if the outcome of the international dispute is effort and the military balance, but since we are fewer resources with which to provide private goods for issues (large p) and those with large selectorates (large z and the leader's effort level is g, and assuming the currently focusing on B's decision, we suppress this supporters. S) find it easiest to remain in office. When the selec- leader also gets the payoffs received by any other notation. Similarly, we suppress all unnecessary sub- Resources not spent on the war effort go to the torate is large, challengers can draw their political member of the winning coalition, then the leader's scripting: winning coalition in the form of private goods. The backing from a large pool of potential supporters. This payoff is: value of these private goods depends upon how many means supporters of the current incumbent cannot be Y(ga) = Pu(ga) - k + (1 - g8) W people must be rewarded. Each supporter's share of certain of also being included in future coalitions. the private goods is R/W. As the leader allocates Hence, members of the current winning coalition only UA(z, 8.) = VA(2) + VF(VA(2) + (1 - 84) WA resources to the war effort, he reduces the private probabilisitically receive private goods under future RA + Pals.)F (vs + (1 - 80) w) goods rewards for his supporters. The rate at which leaders. By contrast, current members of the winning increased effort diminishes supporters' benefits de- coalition receive private goods with certainty from the 4 + (1 - 84) W pends upon W. When the winning coalition is small, incumbent. This forms the basis of the leader's incum- + (1 - p(85)F 4 + (1 - 86) w)) . each member's share of the resources is high. Given bency advantage. where is the leader's utility for remaining in office, these concentrated benefits, increased war effort dras- The size of the leader's incumbency advantage de- and the other terms refer to the payoff from member- tically reduces the utility of members of the winning pends upon the configuration of the polity's political ship in the winning coalition."? F(), recall, is the where vn = 1 - k - Ry/Sg + up and ly - -k coalition. In contrast, when W is large, each member institutions. The smaller the selectorate, the greater distribution of challenger types; hence, F (V (2) + receives only a small share of the private goods in the "are the private goods benefits members of the current (1 - 84) RA/W + u - RA/S,) is the probability that The term pa(ga) - k refers to the expected value of winning coalition can expect from any challenger and, the leader of A retains power given the international the war to the nation as a whole. (1 - go) R/W is the the reduction in supporters' utility from having re- therefore, the greater the private benefits the incum- outcome z and effort gA. leader's share of the private goods rewards. The final sources channeled into the war effort, instead of being bent must provide in order to remain in power. Simi- During a dispute, the incumbent leader in A decides term, "(p,(8 ) F(vs + (1 - 85) R/W) + (1 retained as private benefits, is small. It is incentive larly, as the size of the required winning coalition whether to fight and, if so, how hard to try. If attacked, p( ))F(l, + (1 - gm) R/W)), is the product of the compatible with a leader's goal of remaining in office to decreases, the number of people with whom private the leader in B chooses an allocation of resources to value of officeholding, Y, and the expected probability maximize his supporters' utility. In terms of the sup- benefits must be shared decreases, which makes the dedicate to the war effort. Following the end of the of surviving in office. This expected probability is made porters' rewards, the cost of improving the probability value of the benefits to each member that much dispute, members of the winning coalition decide up of two components. The first, Page)F(va + (1 - of victory increases as the winning coalition gets greater. Therefore, as the winning coalition becomes whether to remain loyal or to defect, thereby retaining g ) R/W), is the probability of victory (Pa(ss)). smaller, Thus, the marginal benefit of increased ef- larger, the incumbency advantage diminishes because or bringing down the incumbent leadership. This is the multiplied by the probability that B's leader survives in fort-the increase in the probability of victory-is the value of the private benefits to individual members sequence of play in the game. With the utilities all office given victory. F(vs + (1 - ga) R/W). The independent of political institutions but the cost of of the winning coalition gets smaller. When the win- specified, we can now turn to solving the game. second component, (1 = p(ga))Fla + (1 increased effort is dependent on those institutions, To ning coalition is small and the selectorate is large, R/W), is the probability of military defeat (1 - p(s,)). make the institutional comparison as stark as possible, supporters of the incumbent jeopardize t welfare if multiplied by the probability that B's leader survives in consider the following limiting case. SOLVING THE GAME they defect to a political rival of the incumbent, since office given defeat, Fig + (1 - ga) R/W). Suppose a leader chooses between making an all-out they face a high risk of being cut off from private We solve the game by finding subgame perfect equilib Leaders choose the effort level that maximizes their effort that guarantees victory (g = 1 and p, (1) = 1) benefits under the new leader. The risk of being ria, starting at the bottom of the game tree and working expected payoff. In general, the larger the winning and no additional effort at all, even though this makes excluded from the private payoffs of future coalitions sequentially backward through each decision. As such, coalition, the greater is this optimal effort level. This defeat inevitable (gg = 0 and pa(0) = 0). The grows as the size of the selectorate increases and as the the analysis of each stage is done with the knowledge of leads to our substantive conclusion that democrats try expected payoff of a leader who makes the all-out effort size of the winning coalition decreas actors' anticipated responses in subsequent decisions. harder in wartime than do autocrats. Before we explain is Ya(gs = 1) - 1 - k + VF(vg), where ve = 1 - s greatest in autocracies and smallest in democracies. Having already examined reselection above, we move the origins of portant - RalS, + up. Alternatively, by making no addi- The incumbent is advantaged in her ability to supply to the preceding decision: the level of resources dedi- qualifier. In some cases, the outcome of the war may be tional effort, the leader receives Y,(gs = 0) = private goods because current members of the coalition cated to prosecuting the war. so important relative to private goods that all leaders, R/W + VF(lg + R/W), where ly = -k - Ry/5, + are sure of receiving them. Given her advantage in If attacked, the leader of B chooses how many of his whatever their domestic political institutions, try all out Hy. Each of these equations has terms relating to the private goods, the incumbent survives so long as she available resources, Ra, to dedicate to the war effort: to win. At the other extreme, the salience of the war leader's direct rewards and terms relating to the pros- does not do such a poor job on public policy that she is gs E [0, 1). The advantage of spending more on the may be sufficiently low that no political arrangements pects of reselection. We examine these components judged grossly incompetent as compared to the chal- war is that he improves B's prospects of victory, but induce additional effort, Between these extremes, how- eparately. lenger. That is, as long CA is not especially large or trying hard also involves risks. By trying harder, B's ever, it is large winning coalitions that induce high In terms of direct rewards, an all-out effort gets the u strongly negative, the incumbent has the upper hand. leader reduces the amount of resources available to effort levels and small coalitions that induce the hoard- leader a payoff of 1 - & as opposed to no additional What constitutes sufficient policy incompetence by the reward his supporters through private goods. Political ing of resources for subsequent distribution as private effort, which generates a direct payoff of -* + R/W. leader to warrant deposition, however, depends on the institutions influence the effort decision because they rewards to the winning coalition. Therefore, if the winning coalition is larger than R, structure of the polity. If the leader has a huge determine the relative importance for political reselec- The game reveals that large winning coalitions en- then the leader makes an all-out effort (W > R). advantage over the challenger in her ability to supply tion of saving resources to use as private goods versus courage leaders to dedicate additional resources to the Although the direct benefits are illustrative of why private goods, then she can survive disastrous policy increasing the odds of military victory by spending war effort. As for all the technical results, the formal leaders with large W try hard, we think of our argument outcomes. Although leaders from systems with large those resources on the war effort. Let Y(g,) represent proof is in the Appendix. The key insight con as predominately reselection driven. The reselection winning coalitions, such as democracies, have some the expected payoff of B's leader from dedicating ge of coalition size influences the importance of private related rewards associated, respectively, with an all-out advantage in the supply of private goods, the magni- available resources to the war, where Pa(g,) is the goods as a means of rewarding supporters. As leaders effort and no additional effort are VF(vg) and tude of this advantage is small, so these leaders cannot increase their level of effort during a w "F(la + R/W). Since vs = 1 - - RalSo + 1 and tolerate policy failure as well as can autocrats or even 12 One might argue that the value of remaining in office, , is a the probability of victory. A military victory benefits In = -k - R,/Sg + u, and since F(.) is an increasing monarchs.11 function of regime type. As Goemans (1995) points out, being ousted everyone in nation B, including members of the win- function, the all-out effort term is greater than the is more often fatal for autocrats than democrats. Our assumption, ning coalition. The marginal benefit of increased effort no-effort term when 1 > R/W. Again, in this limiting however, is that the primary goal of all leaders is to keep their job. case, leaders make an all-out effort to win only when " Bueno de Mesquita and Siverson (1995) and King, Tomz, and Given this, the principal, component in every leader's ob is independent of political institutions, since everyone Wittenberg (1998) find empirical support for these predictions, function is reselection. benefits from the policy success, whether leader, mem- W > R. Otherwise, leaders hoard their resources for ber of the winning coalition, member of the selector- private goods provision and make no additional effort, 797 798 This content downloaded from 131.179 220.13 - Tug. 28 Dhee 2021 02:43:32 UTC 131.179 220.13 - Tug, 28 Der: 2021 02:43:3 All use subject to https:/about jslor.org/terms All use subject to hups:/about jslor.orgtermsAmerican Political Science Review Vol. 93, No. 4 An Institutional Explanation of the Democratic Peace December 1999 even though military defeat is inevitable. Such leaders in war differ as a function of their institutional settings. probability that A wins given effort levels g) and g;. If secondary consideration that is not driven by an auto- could improve their chances of victory by trying harder, All else being equal, institutional arrangements pro- A's leader chooses negotiation rather than conflict, crat's primary objective-to stay in office. As such, but this is not incentive compatible with their desire to vide democratic leaders with greater incentives to try stay in office. hard relative to autocrats. To show how this influences then her expected payoff is U(nego) = x conflict initiation depends upon an assessment of the WFA(HA + RAW ) + R /W, where ny = x + PA expected value of war relative to negotiations. As we Thus far we have shown how political institutions the empirical regularities associated with the demo- R 15. saw, autocrats find it easier to retain office than do affect the amount of resources dedicated to the war cratic peace, we now assess the incentives to negotiate A's leader only initiates conflict when the benefit of their democratic counterparts. Autocrats have an ad- effort for B. Similar logic applies for A. Although A's rather than fight. doing so exceeds what she expects from a negotiated vantage over challengers in their ability to provide leader's decision calculus is slightly more complicated, settlement: U(WARg). 8;) 2 U(nego)." The private goods. Since private goods figure predomi- because she must anticipate B's effort level, the same motivations persist. Again, the larger A's winning THE DECISION TO FIGHT OR TO more likely A is to win, the more likely it is that this nantly in the rewards given to supporters in autocratic condition is met. Given their political institutions, A systems, supporters risk much if they desert the incum- coalition, the less important private goods become NEGOTIATE and B exert effort levels g) and g, respectively. These bent. This incumbency advantage in the supply of relative to foreign policy success. Therefore, all else Because leaders in states with large winning coalitions effort levels influence the probability of A being victo- private goods means that as long as autocratic leaders being equal, the larger the winning coalition, the more cannot easily compensate for policy failure by doling rious, PA. In addition to structuring how hard A and B retain resources to provide private goods for their resources A dedicates to the out private good, they need to succeed in foreign and try, political institutions also influence whether leaders The game shows that democratic leaders, because domestic policy. Leaders in systems with small winning supporters, they survive. Hence, as shown above, even coalitions can more readily compensate for policy want to initiate conflict, given knowledge of the effort if she makes little additional effort, A's leader survives they require large winning coalitions, try harder in war than do autocrats, who only need support from a small levels that will follow in any subsequent war. We define n office. failure by providing private benefits to their few key P as the probability of victory that makes A indifferent Knowing that the leader in A makes no additional coalition to stay in office. One might think that auto- backers. Therefore, they do not try as hard in wars as crats have an interest in fighting hard to protect the between negotiations and war: do democrats. One consequence is that democracies effort and typically survives whatever the outcome of conflict, her expected value of fighting approximately pool of resources they need to distribute as private ss attractive targets than autocracies. By the same goods, but to stay in office they must only provide more token, democracies are reluctant to pursue wars they reduces to py - k + R /W, + , where pai than their challenger can credibly promise. The chal- do not expect to win. Their leaders are at great risk of probability of victory given no additional effort by A's lenger cannot promise to distribute any resources that political defeat at home due to failed policies. Auto- leader. Negotiated settlements leave A's resources have been lost as a consequence of defeat in the war. cratic leaders are not. untouched and available for distribution as private Therefore, the incumbent autocrat's comparative ad- It follows that democratic leaders generally attack (1+ 4(RAVA + (1 - 8)7) - F14 + (1- 207 goods. Again, having the incumbency advantage of vantage in distributing private goods and in reserving only if they anticipate victory. They are highly selective; guaranteed private goods provision, the typical auto- resources for that purpose remains unaltered following they prefer to negotiate when they do not anticipate A only initiates conflict if PA 2 P. Although this crat also retains office via negotiations. An autocrat's expected value for negotiations reduces to approxi- military defeat. military success. This does not mean they are unwilling expression is mathematically precise, it provides little The deduction from our model that democratic mately x + RA/W + V. Therefore, an autocrat's to fight. Democracies, because of their propensity to substantive interpretation of the incentives that leaders decision resembles that of a standard unitary actor leaders try harder in wars than do autocrats is, we try hard, can often overwhelm their foe. This carries an face. Yet, political institutions shape this believe, a novel theoretical result. It is interesting to important implication for the type of foe they model. An autocratic leader typically initiates fighting a systematic manner. We present examples for both note, therefore, empirical evidence that fits our deduc- and defeat. Because autocrats do not try as hard in war, when pa > x + k: The expected benefits of conflict autocrats and democrats to illustrate how political outweigh the expected value of negotiations, 15 tion. Bates et al. (1998, 7) report that Rosenthal they make attractive targets for democracies. In con- institutions structure the conditions under which lead- finds a "selection effect:" parliamentary governments, for trast, two democraci ers choose war rather than negotiations. These stan- example, fight fewer wars [than monarchies or autocra- Since both try hard, each minimizes to the extent dard cases show how autocrats and democrats differ in The Decision for Democrats cies]. They are only willing to fight wars that are profitable, possible, given its resource endowments, the chance their decisions to initiate conflict, Ge Unlike autocrats, leaders with large winning coalitions and they are more willing to adequately finance, and that the other will win. Since democrats need to be for an autocrat depends upon her ability to distribute have only a small incumbency advantage in the supply therefore more likely to win, the wars they choose to overwhelmingly confident of victory, it is difficult to sponsor. His conclusion is reinforced by the argument of private goods. Providing she does not expend resources of private goods. As already shown, this makes their satisfy the conditions necessary for democracies to fight Levi, which explores the impact of increased democratiza- on the war effort, she typically survives whether s survival in office harder and more contingent upon one another. tion and industrialization up wins, loses, or negotiates. Since her survival is not their public policy performance. Knowing that the Autocrats do not depend upon military victory to Faced by an increase in both variables, she argues, gov- strongly influenced by the war outcome, an autocrat's survival of de is upon their public policy ernments have to invest more in convincing their popula- keep their job. Of course, they prefer winning to losing, initiation decision resembles a standard realist calcula- but their political (and personal) survival is primarily a performance, we construct an example to show how tions of the importance of the war and in winning their tion of benefits and losses. For a democrat the situation function of satisfying their small band of supporters institutions structure the conflict decision of leaders in consent to fight. is different. Given the large number of supporters she large winning coalition systems, Having examined this rather than providing their citizens with succ must appease, she cannot buy political loyalty with That is, democratic leaders invest their effort and archetypal case, we analyze the limiting mathematical policies. They are more willing to gamble on war than private goods alone but must rely instead on public resources in mobilizing their societies to produce the case. Although we believe the stereotypical case is are democrats because the latter's political survival is public good of victory in war, as predicted. Lamborn policy success, For her, military defeat equates with on the line; the former's is not. Thus, it is straightfor- generally appropriate, there are plausible conditions (1991) presents additional direct evidence for the political defeat; so, where possible, she avoids fighting under which the results we generate from this example ward to see that democracies and autocracies can fight deduction that democracies try harder in war. He when defeat is a significant possibility. break down. These conditions are important because wars against each other and that autocracies can afford shows that, before World War I, Germany devoted a they predict the domestic political circumstance when o fight one another. larger percentage of its gross national product to the democrats become belligerent and when war between Stated formally, we characterized optimal effort de- The Decision for Autocrats military than did Britain or France. -Nevertheless, the democracies is most likely. Rather than interrupt the cisions for leaders in A and B once war occurs as g Autocrats initiate conflict when the. expected gains of flow of our general argument by continually referring latter mobilized greater resources once the war began and gi, respectively. All else being equal, the larger is conflict exceed what they expect to obtain through because they were better able than the Germans to W, the harder nation i tries. If A's leader initiates to these exceptional circumstances, which resemble negotiations. Yet, the decision to fight is largely a increase revenue extraction for the war effort. those of gambling for resurrection under the diversion- conflict with B, then A's payoff is its chance of winning We have demonstrated that the incentives of leaders ary war hypothesis (Downs and Rocke 1994; Goemans multiplied by its utility for winning, plus the chance it loses multiplied by its utility for losing, plus the value of If x k, then war is inevitable, since A would prefer losing to " In general, the interaction between the effort levels of A and B a negotiated settlement. At the other extreme, if A's expected de depends upon the precise function mapping effort into probability of the private goods retained after choosing a level of war through negotiations is greater than the best possible outcome from was case shows this clearly, As the winning coalition victory, P(ge gn). For the special case in which p(RA. ga) is the force 1)PACK+ (1 - 8A) effort: U (WARBA. 82) .) contracts, W -+ 0, the private goods that each supporter receives ratio model, however, increased effort by one side elicits increased of x as a division between 0 and 1, the first extreme is generally become extremely large, R/W -; hence, leaders make no effort, effort from the other. PA)FA( + (1 - 84) RAW,)), where PA is the impossible. To restrict ourselves to substantively interesting cases, we and the decision to fight resembles the unitary actor solution: lim,-. assume -k s x 1 - k. P-x+A. 799 800 330 13 Tue. 28 Dee 2021 02:43:32 UTC 131.179 220.13 - Tue, 28 Der: 2021 02:43:3 All use subject to https://about jslor apterms All use subject to hups:/about jslor.orgtermsPage 12 of 18 American Political Science Review Vol. 93, No. 4 An Institutional Explanation of the Democratic Peace December 1999 1995; Levy 1989; Richards et al. 1993; Smith 1996; they attack than is true for their autocratic counter- Under some circumstances autocrats may be more only if B is substantially weaker than A, taking both the Werner 1996), we consider them separately. parts. dovish, under other conditions they may be more prewar military balance and effort levels into account. As we have already demonstrated, in systems with The implications of the above theoretical result hawkish. Our institutional explanation has two oppo- Since B will try hard, A must have a great prewar large winning coalitions, leaders have only a small about the willingness of democratic leaders to use site components. First, leaders with large winning military advantage or a great advantage in overall incumbency advantage in the provision of private violence as the means to advance their objectives is coalitions typically try hard during conflict, dedicating resource endowments that can be put to use in the war goods. Instead, leaders survive on the basis of their noteworthy. Autocracies may engage in imperialist additional resources to the war effort, which allows effort. Aside from such asymmetric conflicts as charac public goods performance. With their sensitivity to expansion, for instance, out of a quest for additional them to overwhelm other states