1 This special issue on the economics of conflict was inspired by the following stylized fact: economics appears in almost every type of social conflict, from civil wars in Africa to the recent Arab spring, from the youth riots in the Parisian suburbs to the latest wave of civil unrest in Greece and looting in London – but until recently social conflict has been absent from mainstream economics. How can we explain the paradoxical relationship between our discipline and social conflict?

2 The origins of this paradox might be sought in the preface of Jean-Baptiste Say to his Treatise on Political Economy:

3 “For a long time the science of politics, in strictness limited to the investigation of the principles which lay the foundation of the social order, was confounded with political economy, which unfolds the manner in what wealth is produced, distributed, and consumed… Since the time of Adam Smith, it appears to me, these two very distinct inquiries have been uniformly separated the term political economy being now confined to the science which treats of wealth, and that of politics, to designate the relations existing between a government and its people, and the relations of different states to each other.” (Say [1821] 1964, p. XIV-XV).

4 This passage outlines two principles: 1) a field separation between “economics” and “politics” ever since Adam Smith, which precludes the treatment of conflicts and the State by economics; and 2) the confinement of economics to the study of productive activities. The subject matter of economics should be narrowed to the ways in which wealth is produced, distributed, and consumed.

5 Before scrutinizing these two principles, a preliminary remark regarding Smith’s position on the contours of political economy is in order. To our knowledge, Léon Walras ( [1874] 1954) was the first economist to oppose equating Smith’s definition of the objects of political economy with the definition advocated by Say in his first lesson of Elements of Pure Economics. For the author of Wealth of Nations, political economy acts as “a branch of the science of a statesman or legislator” with two distinct objectives: to provide plentiful revenue for the people and to supply the State with a sufficient income. To Walras, in direct opposition with Say’s definition of political economy as a science, the underpinning criteria of these two objects in Smith are “interest” and “equity” and not scientific “truth”. He wrote:

6 “It is evident at a glance that J. B. Say’s definition is not only different from Adam Smith’s, but, from a certain point of view, is its exact opposite. According to Adam Smith’s view, the whole of political economy is an art rather than a science; while, according to Say, it is entirely a natural science. From Say’s definition it would seem that the production, distribution, and consumption of wealth take place, if not spontaneously, at least in a manner somehow independent of the will of man” ( [1874] 1954, p. 54).

7 Contrary to Say’s view, Smith insists on the political dimension of political economy and asserts that the primary object of the political economy of every country “is to increase the riches and power of that country” ( [1776] 1961, Book II, Chapter V, paragraph 31).

8 Although the field separation between “economics” and “politics” as recommended by Say does not fall within the scope of Smith’s broad definition of political economy as an art, the separation has become the cornerstone of our field. In my opinion, this separation raises two questions:

9

1. If economics should not deal with the “relations existing between a government and its people”, what should be done with respect to the State and the collection of taxes?
2. If economics should be confined to the production of wealth, what should be done with “destructive power” of individuals, social groups, and nations?

10 These two questions are interrelated because the notion of “protection” (a product provided by the State for the citizens in return for their “taxes”) is ambivalent and includes both itself and its antinomy: aggression (Tilly [1985]). A State might instigate a war to “extortionate” its citizens; or it might use “guns” in a destructive manner to appropriate resources from neighboring peoples or states, and thus push out the production possibilities frontiers of the society. Haavelmo [1954] was a pioneer in modeling the redistributive impact of grabbing-protective activity on the general equilibrium of markets at an international level.

11 Similarly, Schelling [1963] highlighted the economic significance of a trade-off between the creative (productive) power and the appropriative power of agents within a State: the creative power of an ordinary healthy high-school graduate does not amount to more than $4,000 annually. This creative or economic power measures his/her capacity to produce or to exchange. However, as an extortionist, s/he can destroy a hundred times more. The extortionate threat can be used by a criminal, a brigand, or a revolutionary. Whatever the extortionist’s personality, s/he uses a destructive power: the conflictual power involving the destruction of use or exchange values or even human beings and nature (Vahabi [2004]).

12 The youth riots in the Parisian suburbs in 2005 provide a salient illustration. The crisis began in the suburbs on 3 November 2005 and culminated in two weeks of urban violence that rocked France. The damage contributed to some 70,000 incidents of urban violence (Wihtol de Wenden [2005]) with an estimated cost of approximately 200 million Euros (CNN, 2005). A comparison of the creative power of high-school students or graduates with their destructive power illustrates why our discipline must not ignore this enormous destructive potential. Hence, the question of how much an agent can destroy is as germane to political economy as the amount that he or she can create.

13 Destructive power has a twofold effect on wealth. On one hand, it affects the redistribution of wealth; on the other, it determines the sharing rule whenever the rule is contested by interested parties. “Social conflict has two different functions: appropriative and rule-producing… In its appropriative role, social conflict redistributes wealth without the mutual consent of all participants. In its rule-producing role, it is the source of institutional change” (Vahabi [2010a], p. 688). The impact of economic performance on institutional change is conditional on social conflict because it determines both the sharing rule and the redistribution of wealth.

14 The effect of conflict on economic activity notwithstanding, the question is whether the field separation between “economics” and “politics” allows us to analyze conflict. In other words, is it possible to abstract from the “political” dimension of conflict and reduce it to a mere economic relationship?

1. Conflict as an economic transaction

15 An analysis of conflictual process requires a theoretical distinction between three different moments: 1) conflictual interests; 2) conflictual behavior; and 3) conflictual action.

16 Economic science explicitly acknowledges conflictual interests due to the scarcity of resources. But conflictual or even antagonistic interests will not necessarily lead to conflictual actions. Suppose that there is a conflictual interest with regard to an economic change. According to the principle of potential Pareto superiority or the Kaldor-Hicks test (Coleman [1988], Chapter 4), a change is efficient whenever the winners win more than the losers lose, in the sense that, if the winners could in principle compensate the losers to their satisfaction, the winners would still be better off than they were before the change. In this case, the winners can avoid open conflict by redistributing part of their benefits to compensate the losers. Thus, the Walrasian general equilibrium integrates conflictual interests without assuming an open confrontation.

17 “With or without a fight, there is a settlement or compromise in which the rights are defined. Those who benefit from the activity gain the approval of those who object by giving them something to get them to agree. What I want particularly to stress is that the solution is essentially the transformation of the conflict from a political problem to an economic transaction. An economic transaction is a solved political problem” (Abba Lerner [1972], p. 259).

18 Accordingly, conflicts are reduced to “economic transactions” in the Walrasian general equilibrium or standard microeconomics. But what happens in the case of externalities or indivisibilities? The theory predicts a market failure and the need for State intervention (Pigou [1932]). However, this State is not a “political” one, because it behaves as a “benevolent” and “neutral” agent maximizing its collective utility function. This assumption of a “neutral” and “benevolent” state also appears in the Walrasian general equilibrium à la Lange (1936).

19 The so-called “Coase theorem” (Coase [1960]) undermines this Pigouvian result. State intervention is not necessarily warranted in the presence of externalities. If the transaction costs are nil, the institutional arrangements are irrelevant in the efficient allocation of resources. It suffices that the contending parties renegotiate the terms of contract so that the assets can be reallocated to the party that can maximize their value. Coase concedes the importance of “blackmail” and “bluffs” as integral parts of the bargaining procedure (Coase [1988a]). This kind of threat or conflictual behavior can be used to change the terms of contract. From this perspective, one can accept not only conflictual interests, but also conflictual behavior, but the latter should not necessarily degenerate into conflictual action. The Coase theorem provides a new foundation for microeconomics.

20 The development of a new microeconomics based on strategic interactions among agents and the use of game theory has contributed to an extensive body of literature about conflictual behavior or credible threat. The main objective of this kind of research is to clarify “rational conflict” and not real conflict. “Rational conflict” refers to threat power and can be defined as a bargaining procedure without any real clash or conflict between the parties, which are both partners and adversaries. The theoretical developments of Boulding [1962] and Schelling [1963], known as “strategic conflict theory” laid the foundation for “rational conflict”. One of its major implications is deterrence, and it can be used to analyze the balance of nuclear power. A particular version of strategic conflict theory was developed in the 1990s, and is known as “strategic bargaining models with destructive power” (Busch et al. [1998], Manzini [1999]). [3] Its particular fields of application are strikes and suspension of talks during labor contract negotiations (e.g., between the Canadian Auto Workers union and General Motors in 1998), commercial negotiations during the various rounds of the GATT or World Trade Organization (WTO), and negotiations between trade unions and firms. This research is concerned not only with efficient strikes, but also with the “threat” of strike, so one of its main areas of focus is the credibility of threat or the commitment to put a threat into effect. In all the above-mentioned cases, conflicts remain within the rules of the market economy and can be called “system neutral” or “pro-systemic” (Gupta [1990]). [4]

21 According to standard microeconomics, “protection/aggression” activity should be handled as a form of externality that warrants State intervention. However, if “externalities” are not sufficient to justify State intervention within the Coaseian paradigm, can the “exchange of threat” be considered a form of contractual relationship? In other words, can Coaseian bargaining be extended to coercive power and politics? This line of reasoning extends the meanings of “voluntary exchange” and “mutual gain from trade” far beyond their usual connotations. From this perspective, a robbery can be defined as an “implicit contract” between victim and robber, in which the robber agrees to save the victim’s life in return for a certain amount of money. The major assumption is that the market logic can be transposed to coercive power and involuntary transactions (for example, slavery) can be treated as if they are voluntary ones. Accordingly, the University of Washington branch of transaction costs economics (Cheung [1970]; Coase [1988b]; North [1990]) extends Coaseian bargaining to predatory activities and politics, thereby advocating a “Political Coase Theorem” (PCT).

22 According to this theorem, when political transaction costs are nil, political Coaseian bargaining between candidates (or state leaders) and constituents will lead to an efficient solution, regardless of the initial distribution of political entitlements. [5] The “neutrality” thesis of social conflicts is derived from this theorem. Neutrality of social conflict connotes a lack of need for real clashing or conflictual action, so a redistribution of wealth or reallocation of resources may occur despite conflictual interests among agents. In a sense, in standard economics, conflict is treated like money: it is neutral with regard to economic performance and disappears in equilibrium (Vahabi [2010a], p. 690). The application of the Coase theorem to appropriative and coercive activity necessarily excludes conflictual action and real destruction, and reduces conflict to an economic transaction. Paradoxically, this “pure economic” approach precludes “conflictual costs”.

23 In their excellent survey of the economics of conflict, Garfinkel and Skaperdas noted: “Conflict involves costs that are economically very important… It is thus surprising, if not shocking, that economists have not paid any attention to these costs until very recently” (2007, p. 704). At a theoretical level, it is not surprising that the costs of conflict have not been taken into account in economic modeling (Tornell and Velasco [1992]; Tornell and Lane [1999]; Stefanadis [2007]). Because conflict should disappear in equilibrium, why should the costs of conflict be included? Herein lies the discrepancy between the neutrality thesis in theory and the empirical evidence regarding different types of costs incurred in various kinds of conflicts, such as civil wars (Collier et al. [2003]; Collier and Hoeffler [2007]; Blattman and Miguel [2010]), warfare, and conflictual activity in general (Sandler and Hartley [1995, 2007]; Sturzenegger and Tommasi [2002]; Hess [2003]; Braddon and Hartley [2011]).

24 Similarly, it is not surprising that economists have not even started discussing the nature of conflictual costs: are they a sub-category of transaction costs (Anderson and Marcouiller [2002, 2005]; Skaperdas and Syropoulos [2002]; Garfinkel and Skaperdas [2007]), or do they constitute a idiosyncratic type of costs called “protection/aggression costs” (Lane [1979]; Bowles and Jayadev [2007]; Jayadev and Bowles [2006]; Vahabi [2011])? Finally, what are the various components of conflictual costs, particularly those involved in destructive dynamics of conflictual activity? What are the official methods of measuring these costs, and to what extent do they capture the real destruction costs?

25 The treatment of conflict as an economic transaction leads to a double omission: 1) the omission of conflictual costs; and 2) the omission of institutional identity of contending parties. I have already highlighted the shortcomings of this approach when analyzing conflictual costs. But there is a second major shortcoming: reducing the conflict to its mere appropriative function occludes its second role – rule-producing. Consequently, the institutional identity of the contending parties has no bearing on the dynamics of conflict. [6] Bandits, revolutionaries, and strikers are all lumped together under the title of “predators”, but ironically these “rational predators” do not act violently and do not destroy anything to collect their booty.

2. Conflict as a political problem

26 The preceding section showed that the economic literature acknowledges: 1) conflictual interests; and 2) conflictual behavior (credible threats). But under what conditions can these conflictual interests and behaviors lead to conflictual action?

27 A standard economic explanation attributes the outburst of open confrontation to traditional factors such as asymmetrical information, misperceptions, misunderstandings, or simply to the irrationality of agents and their basic instincts (Garfinkel and Skaperdas [2007], p. 677). But a new branch of literature explains the occurrence of conflicts as incomplete contracting under the menace of future conflicts (Fearon [1995]; Garfinkel and Skaperdas [2000]; Powell [2006]; McBride and Skaperdas [2007]). This contractual incompleteness stems from the impossibility of drafting a long-term contract regarding arming within an anarchic setting marked by the absence of a third party. To put it differently, this incompleteness is not caused by asymmetrical information or limited rationality of agents or the judge (what the incomplete contract literature dubs as “observable” but “unverifiable” terms of the contract by a third party). The incompleteness stems from the lack of credible commitment.

28 In contrast to voluntary transactions, contracting on the use of coercive power is by its very nature incomplete, because there is no authority beyond and above the authority of the one with the coercive means. As a result, the market logic cannot be simply transposed to coercive power. Accepting the idiosyncratic logic of force or coercive power as a starting point undermines the conception of conflict as an economic transaction, and resurrects the alternative approach of conceiving conflict as an unsolved political problem.

29 The main characteristic feature of coercive power is its destructive capacity. This destruction is not limited to the opportunity costs of foregone productive resources due to the production of conflict technology (Hirshleifer [1991]). It is not even the costs incurred by any other non-violent appropriative activity, such as “money laundering” and “lobbying” (Long and Sorger [2004]). Becker’s [1983] analysis of political (pressure groups) competition demonstrated that lobbying is a source of “deadweight losses” that tend to limit the extent of conflict.

30 Grossman and Kim [1995, 1996] expanded on Becker’s work to account for damage caused by fighting. They applied the concept of “collateral damage”, which measures the destructiveness [7] of appropriative activity. According to these authors, predation is destructive in the sense that in any appropriative interaction, the predator gains less than the prey loses. It is noteworthy that the destructiveness of predation deters predation.

31 Usher [1992] also invoked deadweight loss as one of the four costs of theft or “grabbing”. In his model of anarchy, Usher incorporated deadweight loss by distinguishing between types of goods (such as food) that must be defended against bandits and types of goods (such as clothing) that are intrinsically secure. In this model, deadweight loss is incurred because people produce and consume too much of the good that is safe from theft (clothing) and too little of the insecure good (food) (Usher [1992], p. 78-89).

32 In contrast to Grossman and Kim, Usher distinguished between deadweight loss and pure waste or destructiveness due to theft. Pure waste is another possible cost of theft, which may stem from the lack of a more amicable way of transferring property from victim to thief. For instance, when robbing someone, a thief may consider it necessary to assault the victim physically to reduce his/her ability to defend his/her property, or even go so far as to accidentally or intentionally kill the victim. Grossman and Kim [1995, 1996] excluded this type of cost because their model does not account for real destruction.

33 The costs of destruction impose an internal limit to the development of conflict. Following Lerner’s argument, if the sum of benefits does not exceed the total loss due to the increasing costs of destruction, the conflict can no longer be regarded as an economic transaction. Instead, it becomes a political problem. The solution to this problem is not just a redistribution of wealth, but rather the introduction of new sharing rules. Proponents of the destructive power approach insist on the destructivity of conflict to explain internal limits, and thereby argue for the need of institutional change: the need for new forms of governance and norms. This perspective is the exact opposite of a liberal approach, which maintains that trade or economic interdependence is the solution to conflict. It is not necessarily the “trade” but rather the “costs of raid” that impede the persistence of conflicts. However, use of these costs as an internal limit to conflict will be insufficient to prevent conflicts if new rules are not in place (Caruso [2010]).

34 A political approach to conflict avoids the two shortcomings of a pure economic approach: it captures the costs of conflict and focuses on the rule-producing function of conflicts. The dynamics of conflict shape the institutional identity of the contending parties and determine the type and size of their coalitions and group formation (Sandler and Hartley [2001]; Alesina and Spolaore [2006]; Bloch, Sánchez-Pagés, and Soubeyran [2006]).

35 Pioneering research by the public choice school bridged the analysis of “conflicts” and the formation of the State. Analyses of anarchy and Leviathan (Buchanan [1975]; Powell and Stringham [2009]), and of “stationary” versus “roving” bandits (Olson [1982]; McGuire and Olson [1996]) sparked a continuing research program on the emergence of the State (Findlay [1990]; Konrad and Skaperdas [2010]). Despite an early reference to the Coase theorem in developing a theory of the State (Buchanan [1973]), the public choice school has clearly adopted a critical standpoint with regard to the notion of an “efficient political market” as well as the transposition of market logic to coercive power (Buchanan [1987]; Olson [2000]).

36 Despite their opposite viewpoints, Hirschman [1970, 1974] and the founders of the public choice school advocated a political approach to conflicts. However, Hirschman’s starting point was the insufficiency of market coordination and especially the limits of the “exit” mechanism. He introduced “voice” as a complementary mechanism to overcome this insufficiency. While “exit” belongs to economics, “voice” is political action par excellence. He compared these two “correcting mechanisms” ( “exit” and “voice”) and argued that market failure is not limited to the presence of externalities. Where “exit” does not do a good job of stirring up management and restoring efficiency, “voice” can replace it. In other words, “voice” or conflictual activity within the existing rules is not necessarily counterproductive. It may contribute to improved economic efficiency, and should be thus classified as a positive externality. This theory can be applied to advocacy in civil movements and participative democracy.

37 Despite its richness, Hirschman’s concept of voice is ambiguous because he confuses two different things: 1) voice within the existing rules, e.g., a protest or a reformist movement within a political regime or an organization; and 2) voice against the existing rules, for example civil disobedience or revolution. While voice within the existing rules can be opposed to exit in terms of abandoning the existing rules or organization, voice against the existing rules cannot clearly be opposed to exit, because in this case the frontiers between voice and exit become blurred. In other words, voice within the existing rules comes within the scope of Hirschman’s dichotomy of voice versus exit, whereas voice against the existing rules escapes this dichotomy and cannot be opposed to exit.

38 Voice against the existing rules requires that actors also do a particular kind of exit, e.g., not remaining among loyalists and joining others or opponents. But this exit is not synonymous of emigration, because it implies active participation to overthrow the existing order. This explains why Hirschman has never distinguished between these two different senses of voice; such a clarification would have undermined his theoretical framework in terms of the dichotomy between voice and exit.

39 In my opinion, a third category should be added to Hirschman’s [1970, 1974] dichotomy of “voice” versus “exit” – “scream”, which is defined as voice against the existing rules. Revolution is a form of “scream”, whereas political emigration (voting by foot) is “exit” and political reform is “voice” (Vahabi [2004], p. 88-97). The use of “scream” could greatly improve economic efficiency. Much theoretical work is still needed to clarify its dynamics, particularly the roles of “fraternity” and “cooperation”. The “rationality” and “motivation” of revolutionaries remain controversial issues among economists, and cannot be explained by a reference to egoism without mentioning “mutual aid” and “cooperation” (Nessah and Tazdaït [2008]). The Berge equilibrium may be a promising avenue to clarify the logic of “cooperation” and to complete the Nash equilibrium (Colman et al. [2011]).

40 “Scream” can be transformed into “voice” if new rules are adopted. As Léon Gambetta reminded the members of the French National Assembly, if universal suffrage functions in the fullness of sovereignty, revolution is no longer possible because revolution can no longer be attempted – and a coup d’état need no longer be feared when France has spoken (Hobsbawm [1987]). A political approach to conflicts clarifies the role of conflicts in the genesis and evolution of rules and institutions. But can we consider the conflict as an institution in itself?

3. Conflict as an institution

41 Conflict is often equated with anarchy and an absence of rules. The proponents of orderly anarchy do not share this contention. Their research program pertains to a type of “order” wherein rules are not generally absent, even if State rules are absent or not enforced (for example, in the absence of a State or in the presence of a failed State). Private, associative, communitarian rules are generated through agents’ interactions, including their conflictual relationships, without the State’s interference as an arbiter or intermediary.

42 A salient illustration of orderly anarchy is the institution of “dueling for honor” within Europe from the sixteenth century to the nineteenth century. Dueling is one of the best indicators of political transition from anarchy to order. While aristocratic civil wars can be regarded as anarchy; and the state’s monopoly over violence as order or Leviathan, dueling for honor is an “orderly anarchy”, because it entails extra-legal or illegal strictly codified and regulated private conflict. Bush and Mayer [1974] demonstrated the theoretical possibility of orderly anarchy and Hirshleifer [1995] analyzed the conditions under which this type of order could break into. The public choice school extensively contributed to the definition and elaboration of this research program. The recent works of Cowen and Sutter [2007], Coyne [2005], and Leeson [2007, 2009a, b] elucidate the role played by private norms in the emergence and self-reproduction of this kind of order.

43 Borrowing from Polanyi’s triad [1944, 1968] of “reciprocity”, “redistribution”, and “exchange”, Vahabi developed a fourth form of social integration called “destructive coordination”. This type of coordination can be equated with orderly anarchy or “integration through coercion” (Vahabi [2009a]). It achieves coordination through intimidation, threat, and the use of non-institutionalized coercive means. Resources and human efforts are allocated in order to appropriate what other people produce. This concept has since been mobilized in analyses of social order in post-revolutionary Iran since 1979 under the Islamic Republic of Iran (Vahabi [2010b]) and that of Ivory Coast (Dago [2012]), as well as in descriptions of social organization among French rag-pickers (Lupton [2011]) and ethnic minorities in France (El Karouni [2012]).

44 A study of orderly anarchy (or destructive coordination) lends credence to the idea that under certain conditions, conflict is not only the source of institutions, but actually constitutes an institution in itself. In this case, conflict remains an unsolved political problem, and its rule-producing function predominates its appropriative function.

45 To sum up, the three theories of conflict can be classified in terms of how they address one of the two functions of conflictual activity:

46

1. Conflict as an economic transaction: appropriative function
2. Conflict as a political problem: rule-producing and appropriative function
3. Conflict as an institution: rule-producing function

47 As argued throughout this paper, the integration of conflictual costs and the institutional identity of contending parties calls into question the field separation of “economics” and “politics” with regard to conflict. Economics of conflict cannot be anything but a political economy of conflict.

4. Contributions of this special issue

48 The contributions in this special issue help clarify the economics of conflict. As shown in the bibliography and quotations cited in this introduction, the authors in this issue have all contributed to the development of this new field of economics.

49 The issue begins with Keith Hartley’s article, which provides an overview of how our discipline has contributed to the study of conflicts and defense. This paper, entitled “Conflict and Defense Output: An Economic Perspective,” examines the fundamental question of conflictual costs and highlights the discrepancy between the official measurement of these costs and the real costs involved in various forms of destruction that are not accounted for. By focusing on the increasing colossal conflictual costs, Hartley questions whether nations still have the means to pay for modern warfare, and whether the new economics means the end of war. To clarify the different components of conflictual costs, Hartley scrutinizes the product of conflicts, namely defense and peace. He finally underscores the limits of traditional methods of calculating defense products, and discusses the advantages of the new British method of measuring defense capabilities.

50 Michelle R. Garfinkel, Michael McBride, and Stergios Skaperdas study the problem of “Governance and Norms as Determinants of Arming.” Their simple static contest model tackles a new and important dimension of conflictual activity by focusing on the rule-producing function of conflict. Their findings are particularly original: societies with potentially conflictual relations can make either high or low levels of expenditures on security without any difference in the levels of security they actually enjoy. They also argue that the costs incurred in destruction impose an internal limit to the pursuit of conflict. Finally, they explore two factors that can limit arming and, more generally, the costs of enforcement within and across states: governance, or the formal organizations and institutions that help define and enforce property rights; and norms, or the informal arrangements used to settle potential disputes.

51 Peter T. Leeson, Daniel J. Smith, and Nicholas A. Snow analyze hooligans: rival football fans bent on brawling. This novel contribution provides a simple and robust model that illustrates a subtle case of orderly anarchy in which conflict constitutes an institution in itself, i.e., “fight clubs”. Legal penalties for conflicting with non-hooligans drive hooligans to form a kind of “fight club” where they fight only one another. This club makes it possible for hooligans to realize gains from trade, but it also attracts ultra-violent individuals that the authors call “sadists”. If the proportion of fight-club members who are sadists is sufficiently high, the fight club self-destructs. Like the two preceding articles, destructivity of conflict (in terms of self-destruction) imposes an internal limit to the conflictual activity. The rules governing “fight clubs” must not eliminate the conflict, but should keep the club from self-destructing even when populated exclusively by sadists. This requirement leads to strong pressures to develop private rules regulating conflict within the clubs.

52 Kai Konrad’s paper appears after the first group of articles devoted to the rule-producing function of conflicts, and just before other articles devoted to their appropriative function. His article, entitled “Dynamic Contests and Discouragement Effect,” describes the state of art of a particular field of dynamic game theory, which is frequently used by economists. This author of Strategy and Dynamics in Contests [2009] introduces his ideas in a clear and intuitive manner to ensure that the survey is accessible to a wide audience. The paper also includes a second line of inquiry: what is the driving factor behind player participation in different phases of a dynamic game? The article has thus a twofold focus: on one hand, it provides a detailed survey on dynamic contests (races, tug-of-war, elimination contests, and iterated incumbency fights). On the other hand, it highlights the discouragement effect, which explains why the sum of rent-seeking efforts often falls considerably short of the prize that is at stake. It may cause violent conflict in early rounds, but may also lead to long periods of peaceful interaction.

53 Raul Caruso’s contribution, “Differentials in Property Rights in a Two-Sector Economy,” provides an elegant model elucidating the appropriative function of conflict. In a contested sector, two agents struggle to appropriate the maximum possible fraction of a contestable output. In an uncontested sector, they hold secure property rights over the production of some goods. Agents split their resource endowment between “butter”, “guns”, and “ice cream” – the latter denotes productive activities secure from appropriation. As in preceding models, Caruso’s theoretical model integrates the effect of conflictual destructiveness and examines whether improvements in productivity can countervail the destructive impact of continuous conflict. Eventually Caruso extends his basic model to consider a government and a rival group. A redistributive government can boost production in the uncontested sector, but at a higher level of “guns”.

54 Like Raul Caruso, Antoine Soubeyran and Agnès Tomini tackle the appropriative function of conflict in their theoretical model entitled “Water Shortages and Conflict.” This paper investigates a particularly important topic: the possibility of future inter-state wars over water resources. The debate about “Natural resources: curse or blessing?” (van der Ploeg [2011]) may have underrated the potential of fresh water to create havoc in many areas under conditions of economic globalization and rapidly-increasing population growth. This paper brings this issue back to center stage by focusing on the risk of a conflict between countries sharing fresh water. The originality of the model lies in its “negotiation interval” whose size depends on water availability and heterogeneous productive capabilities among countries. This interval is equated with the probability-to-conflict, which decreases with size. The authors demonstrate that the risk of conflict increases with the level of water scarcity and the asymmetry of productive capabilities across countries. However, the paper does not account for the destructivity of conflicts.

55 The issue concludes with Derek Braddon’s paper that reconsiders the historical debate about the relationship between economic interdependence and conflict. In a sense, this contribution entitled “The Role of Economic Interdependence in the Origins and Resolution of Conflict” outlines an alternative approach to the thesis of the first paper of this special issue. While Hartley emphasizes conflictual costs as an impediment to conflict, Braddon maintains that the economic interdependence (trade effect) may be more influential in putting an end to conflict. In other words, Braddon’s argument comes within the scope of appropriative function of conflict, which can be channeled through trade. Of course, the author’s advocacy of economic interdependence as an avenue to end conflict is not one-sided and unconditional. The paper reflects the novelty and richness of the debate between the two opposing schools. One school argues that increased levels of economic interdependence encourage good political relationships and wards off possible conflict. The opposing school of thought argues that excessive interdependence may actually create resentment, intensify rivalry, and ultimately political discontent leading to conflict. The paper disentangles the main explanatory factors of the debate and then explores the role of economic interdependence in two different conflict scenarios: Africa and the Balkans.