Show Summary Details

Page of

Printed from Oxford Research Encyclopedias, Economics and Finance. Under the terms of the licence agreement, an individual user may print out a single article for personal use (for details see Privacy Policy and Legal Notice).

date: 19 June 2021

The Economics of Identity and Conflictfree

The Economics of Identity and Conflictfree

  • Subhasish M. ChowdhurySubhasish M. ChowdhuryDepartment of Economics, University of Bath


Conflicts are a ubiquitous part of our life. One of the main reasons behind the initiation and escalation of conflict is the identity, or the sense of self, of the engaged parties. It is hence not surprising that there is a consistent area of academic literature that focuses on identity, conflict, and their interaction. This area models conflicts as contests and focuses on the theoretical, experimental, and empirical literature from economics, political science, and psychology. The theoretical literature investigates the behavioral aspects—such as preference and beliefs—to explain the reasons for and the effects of identity on human behavior. The theoretical literature also analyzes issues such as identity-dependent externality, endogenous choice of joining a group, and so on. The applied literature consists of laboratory and field experiments as well as empirical studies from the field. The experimental studies find that the salience of an identity can increase conflict in a field setting. Laboratory experiments show that whereas real identity indeed increases conflict, a mere classification does not do so. It is also observed that priming a majority–minority identity affects the conflict behavior of the majority, but not of the minority. Further investigations explain these results in terms of parochial altruism. The empirical literature in this area focuses on the various measures of identity, identity distribution, and other economic variables on conflict behavior. Religious polarization can explain conflict behavior better than linguistic differences. Moreover, polarization is a more significant determinants of conflict when the winners of the conflict enjoy a public good reward; but fractionalization is a better determinant when the winners enjoy a private good reward. As a whole, this area of literature is still emerging, and the theoretical literature can be extended to various avenues such as sabotage, affirmative action, intra-group conflict, and endogenous group formation. For empirical and experimental research, exploring new conflict resolution mechanisms, endogeneity between identity and conflict, and evaluating biological mechanisms for identity-related conflict will be of interest.


  • Econometrics, Experimental and Quantitative Methods
  • Economic Theory and Mathematical Models
  • Micro, Behavioral, and Neuro-Economics
  • Public Economics and Policy

Violence is fomented by the imposition of singular and belligerent identities on gullible people, championed by proficient artisans of terror.

—Amartya Sen (2007, p. 2)


Conflict and competition are as common in day-to-day life as cooperation and acquaintance. People engage in conflict for various reasons. It can be due to the eagerness to win over others, to avoid losing, or to break away from a (bad) status quo. Conflict also emerges because of various behavioral factors of the involved parties, such as pride, spitefulness, and feeling of insult—to name a few. One of the most prominent motivations for engagement in conflict, however, is the identities of the involved parties. One of the oldest epics in the world, Iliad (ca. 1260–1180 bc), describes the conflict between people identifying themselves as either Greeks or Trojans. Even in the early 21st century, various identities—nationality, race, ethnicity, religion, language, immigration status, and economic class—epitomize some of the main reasons for conflict. This article surveys the economics literature and parts of the psychology and political science literatures on identity, conflict, and their interrelation.

As can be expected, the topic of conflict has been a very active area of research for a long time. Researchers from economics, political science, and psychology investigate various aspects of conflict with diversified tools and methods. Moreover, the way conflict is defined in different parts of the literatures varies substantially. Along with physical conflict and war, features such as conflict of interest, failure to cooperate, and bargaining are often considered as conflict (see, e.g., Bolton & Croson, 2012). For the purpose of this article, conflict is defined as situations in which engaged parties exert costly resources to resolve a dispute or to gain a reward. This definition aligns with the definition of a “contest” (Fu & Wu, 2019; Konrad, 2009), allowing the employment of the knowledge of contest theory to structure this article. Furthermore, although there exists an array of qualitative literature on conflict (see, e.g., Ashmore et al., 2001; Deutsch et al., 2011), this article defines the scope of the analysis on the relevant quantitative literature in economics, political science, and psychology.

On the same note, the literature on identity is vast and diversified. There is a long history of psychology research in identity, beginning in the 1950s (see, e.g., Foote & Cottrell, 1955; Sherif et al., 1961; Tajfel & Turner, 1979; Turner, 1978). In economics, Sen (1985) introduced the concept of identity, defining it as “how the person sees himself or herself.” Since then, a plethora of research has been carried out (including by three Nobel Laureate economists: George Akerlof, Amartya Sen, and Jean Tirole) on the topic of the economics of identity. The literature defines identity as a type of categorization: who a person thinks they are and what they think the other people are. Such categorization remains important because it affects one’s interaction with others, and the resulting welfare. More formally, Tajfel and Turner (1979) conjecture that people first categorize different identities (through various characteristics such as gender, race, religion, etc.), then identify themselves and others into such diverse categories, and finally compare themselves with others—while making decisions. Akerlof and Kranton (2010) summarize that this process not only leads to in-group and out-group biases, but also the utility itself gets affected by the categorization. The identity literature covers various research topics such as organization, education, social norm, behavior under risk, gender, and politics. This article limits the scope to the interaction between identity and conflict. Furthermore, because gender as an identity itself has a major significant part in the literature, gender as an identity remains out of the scope of this article.

In recent coverage of the literature on conflict (Garfinkel & Skaperdas, 2012; Wärneryd, 2014), various theoretical and applied aspects of conflict and conflict resolution have been discussed. Similarly, recent coverage on the economics of identity (Akerlof & Kranton, 2010) discussed how identity is intertwined in our everyday life. Charness and Chen (2020) summarized how identity can reinforce greater good, such as by improving cooperation, advocating equality, reducing discrimination, reducing crime, and so on. Shayo (2020), in another research, covered the policy-related issues of identity. The current article, on the other hand, focuses only on conflict, and points out possible areas for further research. In this sense, it complements the work by Charness and Chen (2020) and partially that of Shayo (2020).

A number of basic questions have motivated the research agenda in the field of identity and conflict. Both theoretical and applied research explore the way the salience of either a common or a different identity affects one’s decision making. Modeling identity and its effect on individual utility is another important area of theoretical research. This area of investigation is also extended to the endogenous choice of identity groups and to the tension between self-interest and the interest of the social group. The multidimensionality of identity and its implementation for possible conflict resolution also remains an active research area. Experimental research in identity and conflict employ both laboratory and field experiments to test these theories. Researchers also explore the effects of real and minimal identity in various tasks. Empirical research, on the other hand, is more focused on issues such as the effects of fractionalization, polarization, and political systems on conflict intensity.

Given the scope of this article, “Theoretical Background and Results” first provides a theoretical structure in the intersection of identity and conflict, and points out the main theoretical mechanisms and contributions in this area. In “Evidences of Identity and Conflict,” the focus then moves to the applications and evidences regarding the relationship between identity and conflict; this part is divided into “Experimental Evidences” and “Empirical Evidences.” The final section, “Discussion,” then summarizes the existing knowledge base, and points out some of the significant questions in this area that are yet to be investigated.

Theoretical Background and Results

Start with a contest theory model of conflict without introducing the conception of identity. For the sake of simplicity and intuition, employ only an individual conflict with symmetric players, which can easily be extended to asymmetric players or to group conflict settings. In this model, the engaged parties expend costly resources, such as money, effort, and time, in the conflict, and the winner is rewarded with a valuable prize. For the sake of brevity, call all such resources “effort.” The costly efforts are exerted in order to improve the probability of winning the prize. This prize can be viewed as the pride gained in winning a conflict; or it could be a piece of land (e.g., Kashmir in the context of India, Pakistan, and China) or the spoils of a war. As in the real-life situation, irrespective of the outcome of the contest, players forgo their efforts.

Following the same, consider a contest with N identical risk-neutral players. Player i (where i=1,2,3,N) spends their costly effort bi0 to win a prize of common value V>0. The level of effort, hence, is considered as the intensity of conflict—a higher effort level also reflects a higher conflict level. The probability that player i wins the prize (contest success function; [Skaperdas, 1996]) is pibibi, where bi is the vector of efforts of all players except i.

The contest success function has the desired properties: pibi0,pibi0, and Σipi=1. This means one’s own effort weakly increases one’s own probability of winning the contest, whereas opponent efforts weakly decrease it; and at least one of the players wins the contest. Various popular contest success functions such as the lottery (Tullock, 1980), where pibibi=bi/ΣibjifΣjbj0,and1/N otherwise; or the all-pay auction (Baye et al., 1996), where pibibi=1/kifbi is one of the k highest efforts, and 0 otherwise, can be employed in this setting. Without the concept of identity, player i tries to maximize their expected payoff:


It is possible to introduce the concept of identity in this structure in many different ways. A simple way to introduce identity in the utility function is to follow the same approach as in the “other regarding preferences” literature. The other regarding preferences literature is broadly divided into preference-based models, and belief-based models. Following the same path, consider one example from the preference-based identity model, in which identity is introduced through a group contingent social preference. This is a modified version of the social preference model introduced by Chen and Li (2009), which is employed in the identity literature also by Chen and Chen (2011).

In this model, a player’s utility is a combination of their own payoff, (relative) payoffs of the players with the same identity, and the (relative) payoffs of players with different identities. Formally, if there are two identity groups “In-group” (I) and “Out-group” (O) and player i belongs to the in-group, then their utility function can be written as


where fi and gi are functions measuring relative payoffs between the players. These functions, for example, can take the form of inequality aversion as originally introduced by Fehr and Schmidt (1999) and employed in contests by Riechmann (2007), Herrman and Orzen (2008), and Chowdhury et al. (2018). Presumably, a relatively lower payoff compared to others’ affects utility negatively, whereas a relatively higher payoff can affect utility either negatively (inequality aversion) or positively (spite). These effects are expected to be higher in comparison with those of the out-group players.

After imposing some specific functional forms for the contest success function and for the social preference functions, it will be possible to find closed-form solutions for the contest game with and without identity. Under certain reasonable regularity conditions, such as more spitefulness toward out-groups compared to in-groups, the results show higher equilibrium effort exerted when the identity is present versus when it is not. This, in turn, provides a micro-foundation regarding why the presence of an identity can increase the intensity of conflict.

Note that, such group contingent social preference can also occur when the identity of the winner is included in the utility. In the contest literature, Linster (1993) was the first to introduce such a structure with a lottery contest success function (Tullock, 1980), albeit without mentioning the term “identity.” In this model, Player i’s valuation of the reward depends on the identity of the winner of the reward. Hence, the valuation of the reward for player i is a vector vi=(v1i,v2i,,vNi), consisting of the valuation of player i when player 1 wins, when player 2 wins, and so on. It is assumed that everyone strictly prefers their own win over another’s win, but that they can prefer the win of some players over some other players—if self-win is not realized. This models an identity-dependent externality because the payoffs of the engaged parties not only depend on their own win but also on the identity of the winner.1 This can be easily translated into real-life phenomena such as multisided war or conflict among various ethnicities. This simple model gives some very intuitive results. It is found that as players value others’ winning more, they exert less effort, that is, identity-dependent externality may result in less conflict. However, if the players only value the winning of in-group rivals, and not the out-group ones, then the out-group players increase conflict effort.

This area of theoretical research was extended by Esteban and Ray (1999), Klose and Kovenock (2015a, 2015b), and Bagchi et al. (2019). Esteban and Ray (1999) also used a lottery-contest success function but provided a more general structure and focus on the distribution of the types of players in different characteristics. They considered many mutually exclusive groups consisting of individuals, an outcome of pure public goods, and a general cost function of effort. They proved the existence and uniqueness of equilibrium, and analyzed the effects of the distribution of players in different groups on the level and pattern of conflict.2 Klose and Kovenock (2015a, 2015b) investigated identity-dependent externality with an all-pay auction-contest success function. Klose and Kovenock (2015b) characterized the equilibria in such all-pay auctions. They found that the standard results of all-pay auctions hold only when identity-dependent externalities are “mild.” With sufficient externalities, however, equilibria may not be payoff equivalent, and even identical players may earn different payoffs. These results show that with the introduction of identity, the standard results in conflict and conflict resolution may not hold. Klose and Kovenock (2015a) used the same theoretical structure to investigate the interaction between extremists and moderates. They define extremism as a higher per capita effort by radicals compared to centrists. They found that under such identity-dependent externalities, extremism indeed can suppress moderates in the sense that even when the radicals are minorities, they exert higher effort than the moderates. Another theoretical application of identity in conflict was carried out by Bagchi et al. (2019), who explored the idea of proxy war with three sponsors of war and three proxies. They implicitly assumed that two of the sponsors were in-group, whereas the remaining sponsor was out-group. There exist multiple equilibria for this game. However, all the sponsors exert positive effort in equilibrium only when the winning reward is small. When the reward is high enough, the two in-group sponsors coordinate in a way such that only one of them engages in conflict—reflecting implicit alliance formation. The authors also worked out conditions for alliance to be formed explicitly. This study sheds light on the importance of identity in endogenous alliance formation in conflicts.

The group contingent social preference approach reported so far, however, is introduced here only to provide the conceptual background and to allow a micro-foundation of the results obtained from the laboratory or from the field. There are additional preference-based and belief-based approaches in the literature to model identity, but a detailed discussion of these is beyond the scope of this article. Hence, only some of those models are covered. For an excellent and comprehensive summary of those approaches, see Charness and Chen (2020, sec. 2) and Shayo (2020).

Among the preference-based identity models, Akerlof and Kranton’s (2000) introduced the idea of a norm of society intertwined with the identities. The players’ utility then depends on their own action, the action of the other players, and the degree of deviation from the norm. This structure is employed to understand various issues such as the labor market, schooling, and so on. The studies discussed up until now consider the existing identity groups as given. However, often people form such identity groups endogenously. Shayo (2009) considered the issue of one’s endogenous choice of groups. Players gain utility from their payoff, the status of their group, and their individual social distance from the group they identify with. Bernard et al. (2016) used a similar model, but endogenized the social distance as well. These models are also employed in various contexts, especially in public economics and trade. Another way an identity group can be evolved is through social interaction. Fang and Loury (2005), Freyer and Jackson (2008), and Akerlof (2017) took this approach and used various model specifications to endogenize group identity as well as the norms for such groups. All these models include identity into the players’ preference system. However, this leaves out a very important concept: identity is not only what a person thinks about themself, but also what they think about how others perceive them to be. Bénabou and Tirole (2011) introduced a belief-based identity model in which players have incomplete information regarding their own identity and make costly investments in building their identity.

Theoretical investigation on the interaction of identity and conflict are limited, and introducing all these diversified models to attain insight regarding behavior in conflict still remains an open question. One such attempt was made by Sambanis and Shayo (2013), who investigated whether a common identity among conflicting agents can mitigate the conflict. They also explored the relationship between the pattern of social identification and conflict. In this model, members of two groups are engaged in a group-specific public-good prize contest (Katz et al., 1990). The social identity has three building blocks: social groups, perceived distances among the social groups, and the relative value or status of the social groups. The status depends on the relative payoff and exogenous factors. The authors then characterized the social identity equilibria and showed that a vicious cycle might arise in which conflict persists and the unifying national identity is not salient. These results match well with the empirical observations from poor conflicting countries.

Evidences of Identity and Conflict

The evidences of the relationship between identity and conflict emerged even before the theories were worked out. Researchers in this area use two main methodologies: empirical methods employed in the field, and experimental methods employed in the laboratory as well as in the field. The empirical evidences from the field are obtained mostly by economics and political science researchers. These results provide us with a direct evidence relevant for the specific cases. Often, however, appropriate data for pursuing such analysis remain unavailable. Even when the data is available, an appropriate identification may not be possible. The experimental studies run by economists and psychologists, on the other hand, do not face such issues. The experimental researchers have a higher level of control and can employ straightforward identification strategies. Moreover, experimental researchers are able to investigate the effects of both real (e.g., race) and artificial (e.g., student ID number) identities. However, the results from experiments often face the criticism of lacking external validity and generalization. In summary, both methods have their own benefits and are viewed as complementary tools in better understanding the research topic. In what follows, first “Experimental Evidences” and then “Empirical Evidences” are reported.

Experimental Evidences

As defined by the scope of this article, only experiments that use the features of contests are covered here. However, other families of games, such as the prisoner’s dilemma, are often employed to understand conflict (see, e.g., Bornstein, 1992). For a comprehensive survey of contest experiments in general, see Dechenaux et al. (2015), and for conflict experiments beyond the structure of contests, see Abbink (2012). Note also that instead of classifying the experimental results in terms of the methodology (laboratory or field) used, in this article they are organized in terms of the research questions asked.

Experimental researchers employ both real (often termed as “natural”) and artificial (termed as “minimal”) identities to investigate research questions. Clear methodological protocol has evolved in the literature implementing either of the identities. While using natural identity, researchers make such identity salient by either mentioning the identity (see, e.g., Chowdhury et al., 2016) or by priming on such identity (see, e.g., Shih et al., 1999). For minimal identity, however, the protocol is different. Here, the subjects are assigned with some individual-level task that is irrelevant to the conflict game, and depending on the outcome of this task they are anonymously assigned to mutually exclusive groups. An example of such a task is to choose between a pair of paintings (Chen & Li, 2009). It has been observed in both economics and psychology literature that even the minimal identity often has significant effects on subject behavior. For a broader coverage of minimal identity and non-conflict literature, and further detail about the methodology, see Charness and Chen (2020).

The experimental conflict literature started with the famous Robbers Cave experiment by Sherif et al. (1961). This experiment was run in 1954 in the Robbers Cave State Park, Oklahoma (hence the name) in three stages. In the first stage, 22 pre-teenage boys were randomly assigned to two groups, and the group identity was made salient by performing group work. In the second stage, the groups were engaged in various camp games in which only one group could win. In the third stage, the groups were integrated, and joint teamwork-based tasks were run to enforce mitigation. This is a real-effort group contest experiment with minimal group identity. The researchers observed that at stage 2 there were hostility and aggressive behavior against the out-group members by both the groups. This result hints at the effect of identity in the escalation of conflict.

The Robbers Cave experiment was run among otherwise homogenous subjects with no prior history of conflict. Diab (1970) replicated the Robbers Cave experiment in Lebanon with 10 Christian and 8 Muslim pre-teen subjects divided equally into two groups (5 Christian and 4 Muslims in each group). There was an existing historical conflict between Christians and Muslims in Lebanon. Hence, the within-group composition was not homogenous. The two groups were named “blue ghosts” and “red genies.” In the second stage (the conflict stage) of the experiment, a physical and violent fight broke out between the subjects, and the third stage was cancelled. Surprisingly, the conflict broke out between the two artificially created groups, and not between the two religious groups. This demonstrates that salience of even minimal groups can escalate conflict.

The observation from Diab (1970), only partially in line with the views portrayed in Sen (2007). Sen (1985, 2007), coined the idea of multidimensional identity. A person can be categorized into various groups in terms of their gender, age, language, nationality, race, religion, and so on. But when only one dimension of such identities becomes salient and dominant, then it initiates and escalates conflict. Sen notes:

What is done to turn that sense of self-understanding into a murderous instrument is (1) to ignore the relevance of all other affiliations, and (2) to redefine the demands of the “sole” identity in a particularly belligerent form. This is where the nastiness as well as the conceptual confusions are made to creep up.

(Sen, 2007, p. 176)

If one interprets the color-coded group in Diab (1970) to be the salient and dominant identity over and above the religious identities of the subjects, then it can easily explain the outcome of the experiment.

However, Sen (2007) also noted that not all the identities have the same effect. Some temporary identity that he terms “classification” may be salient but will not have the same effect as the real identities. Some examples of classification are shoe size, shirt color, and so on. This idea is supported in Basu, who documented:

We hear of religious wars, ethnic tensions and the coming clash of civilizations but we do not hear of friction between short and tall people, between the bald and the hirsute, or between those who can do mathematics and those who cannot.

Basu (2005, p. 222)

If one considers the color-coded identity in Diab (1970) as classification, and religion as real identity, then the experimental results do not match with Sen’s hypothesis.

Chowdhury et al. (2016) ran a laboratory experiment to test the hypotheses placed by Sen (2007). They conducted a lottery contest between two groups and employed three between-subject treatments. In each treatment, a group of three white subjects were matched with a group of three East Asian subjects in a partner-matching repeated group contest. In the baseline treatment, no information regarding the group composition was revealed. In the classification treatment, following the structure of Diab (1970), the groups were called either a blue group or a green group. In the identity treatment, the race composition of the groups was common knowledge. The authors found that the minimal color-coded identity does not significantly increase the conflict intensity over the baseline. However, when the real identity is revealed, then the conflict intensity increases and within-group free-riding (defined as exerting zero effort) is reduced. These results support the hypotheses of Sen (2007) that a salience of a real identity increases conflict, but that of a classification does not.

Following a similar structure to that in Chowdhury et al. (2016) with lottery-contest success function, Chakravarty et al. (2016) ran a lab-in-the-field experiment of individual contest in India, with Hindu and Muslim subjects. Their treatment manipulation involves information regarding the religions identity of the subjects and the degree of religious fragmentation in the subject village. They hypothesized that the effort level in homogenous religion pairs would be lower than in heterogeneous religion pairs in fragmented villages. Although they found support for this hypothesis for the majority (Hindus), they could not find support for the same for the minority (Muslims).

The effects of identity in conflict intensity, hence, are overall positive. Whereas laboratory experiments suggest support for the effect in the direction of the theory, field experiments provide mixed results—and they depend on the subject demographics. One question that remains is what drives some subjects to exert a higher level of effort against out-group subjects in a conflict. This can occur because of their parochial altruism, that is, in-group love or out-group hate, or a combination of the two (Bernhard et al., 2006; Choi & Bowles, 2007). Abbink et al. (2012) investigated this with a sequence of a prisoner’s dilemma game and a group contest. In one treatment, they also allowed group members to punish each other after the effort provisions were met. They found that effort provision increases with peer punishment. They also found that subjects who cooperate in the prisoner’s dilemma are likely to exert more effort in the group contest as well. Moreover, subjects who exert low effort are punished by peers. Because standard rational models, or in-group love or out-group hate in separation, cannot explain these results, the authors conclude that overall parochial altruism drives the level of conflict. In the context of identity as well, it can be inferred that overall parochial altruism increases the level of conflict in the presence of identity. This proposition, however, is still to be supported with data.

Although not strictly within the contest paradigm, two very relevant experiments are worth mentioning. Weisel and Zultan (2016), using a modified prisoner’s dilemma game, were able to distinguish the motivation of individuals to contribute to the group effort, or to free-ride. Klor and Shayo (2010) divided their subjects into two different groups in terms of their area of study (a real identity). They were then allocated some individual income and had to vote on the redistribution of the income. It was found that when the group identity is revealed, a significant proportion of the subjects vote to support their in-group, even at a cost to their own income. This, however, is not the case for the baseline (when the identity is not revealed). A similar type of analyses to that in these two studies can be run in a contest structure as well.

Can the inclusion of identity compensate the reduction of effort in a group setting compared to an individual setting in conflict? Huang et al. (2018) ran an online all-pay auction experiment in the M-Turk platform to answer this question. Their treatment manipulation came from the subject identity and the game they play. In one set of treatments, the subjects were simply allocated in two groups, whereas in the other treatments the groups consisted of Democrats and Republicans. The all-pay auction was either run individually or between three payer groups. It is well known that subjects exert less effort while in a group compared to while contesting alone. The results of this experiment show that even when the political identity is made salient, the effort provision in a group setting remains lower than in an individual setting.

Empirical Evidences

The main questions investigated in the empirical literature in identity and conflict are the identity-related determinants of conflict and whether those can be separated from economic determinants. Compared to the theoretical and experimental investigations, the empirical literature on identity and conflict with field data is relatively new and still developing. This is mainly due to the lack of availability of appropriate data and lack of control over the variables. An important exception is Collier and Hoeffler (1998), who investigated whether civil wars are caused by economic reasons or because of ethnolinguistic identity. They build a simple cost benefit model of civil war and use civil war data from 1816–1992 for various countries to test the hypotheses from the theoretical model. Specifically, using a probit and a tobit regression, they determined the occurrence and duration of civil wars. The results show that both identity and economic reasons are responsible for the occurrences of civil wars. Economic determinants such as initial income, initial population size, and available resources, along with the identity-related determinant ethnolinguistic fractionalization, turn out to be significant determinants of conflict (civil wars). Interestingly, the authors found that the degree of identity fractionalization has a non-monotonic effect on the conflict; that is, a moderate level of fractionalization affects the conflict the most.

In the same line, Reynal-Querol (2002) investigated the effects of religion and language-related identities in the intensity of civil war, and the role of the existing political system in conflict resolution. She gathered and combined data from various sources for 138 countries for the time frame of 1960–1995 on civil war, level of democracy, characteristics of political system, rebellion, education, export, religious fragmentation, linguistic fragmentation, religious polarization, and so on. Running a logistic regression on the incidence of ethnic civil war, Reynal-Querol found that both animist diversity—a measure of identity especially in Africa—and religious polarization are significant predictors of conflict. The results also show that religious polarizations are more important than linguistic differences as a predictor of civil war. Finally, a consociational political system significantly reduces the occurrence of civil wars.

Esteban et al. (2012a, 2012b) investigated the relationship between ethnicity and conflict. Unlike Collier and Hoeffler (1998) or Reynal-Querol (2002), these authors did not restrict conflict only to civil war, but rather defined it on the basis of death toll. Esteban et al. (2012a) used the three measures of ethnic distribution—polarization, fractionalization, and the Gini-Greenberg index introduced in Esteban and Ray (2011)—to investigate their effects on conflict (while controlling for various other factors such as population, GDP per capita, democracy, etc.). This is important because these measures essentially show the social distances among identity groups. Using data from 138 countries over the time period of 1960–2008, Esteban et al. (2012a) found that both the ethnic polarization and the fractionalization measures have significant positive effects on ethnic conflict, whereas the effects of the Greenberg-Gini index are significant and negative. These results suggest that conflict is affected by both public and private components. This investigation was extended in Esteban et al. (2012b), in which similar data, research aim, and empirical strategy were followed. These authors reaffirm that polarization and fractionalization jointly influence conflict. Moreover, polarization is a more significant determinant of conflict when the winners of the conflict enjoy a public good reward; but fractionalization is a better determinant when the winners enjoy a private good reward. Hence, this study points out the relationship between various measures of social distance of the identity groups, type of conflict, and conflict intensity.

Mitra and Ray (2014) revisited the issue of economic versus identity-related causes of conflict. In a theoretical model of inter-group conflict, they showed that as the income of a group increases, the conflict intensity against that group increases while the conflict intensity of that group decreases. The authors then tested this theory empirically in the context of Hindus and Muslims in India for the time period of 1979–2000. Ethnic conflict is defined in three ways: the number of casualties, the number of deaths, and the number of riot outbreaks. They ran a Poisson specification (and robustness checks with negative binomial and ordinary least squares). The results show that indeed an increase in the per capita expenditure (a proxy for income) of the Muslims significantly explains the occurrences of conflict, whereas an increase in the expenditures of the Hindus has the opposite effect. Specifically, the authors found that an 1% increase in the per capita expenditure by the Hindus decreased the likelihood of conflict up to 8.1%, whereas an 1% increase in the per capita expenditure by the Muslims increased the likelihood of conflict up to 9.9%. The results hint at the possibility that identity-related conflict may occur not purely as a result of identity differences, but also because economic reasons perceived by the involved parties.

In a related study, Girard (2017) found inconclusive effects of inequality in religious conflict in India. Finally, Dasgupta and Pal (2021) extended this idea into caste-related conflict, especially the untouchability issue (higher caste people avoiding physical contact with the lower castes) in India. They first used a contest theory model and then the data from the India Human Development Survey 2011–2012 to arrive at a related but somewhat orthogonal conclusion to Mitra and Ray’s (2014). Dasgupta and Pal (2021) found that although untouchability and related conflict depends on the inter-group distribution of resources across both caste and religious divides, it becomes less prominent with a relative increase in the collective resource endowment of the lower castes.


This article is a survey of the literature on identity, conflict, and their interrelation. Conflicts are modeled as contests, and the theoretical, experimental, and empirical literature, mostly from the areas of economics, political science, and psychology, are included. A large part of the theoretical literature focuses on the behavioral aspects—such as social preference and beliefs—to explain the effects of identity on behavior. The article also investigates issues such as identity-dependent externality, endogenous choice of joining a group, and so on. The experimental literature begins with the result that a salience in difference in identity increases conflict. Further investigations in the laboratory and in the field explore the idea of real versus minimal identity, multidimensional identity, and parochial altruism, as well as their effects on conflict behavior. The empirical literature focuses on the various measures of identity, identity distribution, and other economic variables on conflict behavior. This whole field of literature is still emerging and can be extended to various important avenues. There still remains a list of topics to which the theoretical and applied literature can be broadened.

First of all, the current literature confines itself to the traditional aspects of conflict—in which engaged parties exert unidimensional effort to win the conflict. However, important aspects of conflict covered in the contest theory literature can provide further insights to the identity and conflict field. In many conflict situations, the players not only exert effort to improve their own likelihood of winning, but they also exert unproductive effort to damage the likelihood of their rivals winning. Example of such acts are the “scorched-earth” policy, economic sanctions, destroying the supply line of the rival, to name a few. In the contest theory literature, these acts are termed “sabotage.” Although there is a long literature on sabotage (see Chowdhury & Gürtler, 2015, for an overview), no existing study analyzes the effects of identity on sabotage. Both theoretical and applied investigations in this area will be interesting.

A vacuum also exists in the identity and conflict literature in the area of attack and defense. The existing literature views the parties engaged in conflict as “similar,” in the sense that they exert effort with the same objective—to win the conflict. In real life, however, often there are “attackers” (e.g., terrorists) who exert efforts to win over “defenders” who exert efforts to defend (e.g., the government).3 Chowdhury and Topolyan (2016a, 2016b) employ lottery and all-pay auction-contest success functions in group attack and defense games and show that the results become very different compared to the standard contests. The idea of attack and defense is also highly relevant in identity conflict (e.g., Mitra & Ray, 2014), but research is yet to be conducted in this area.

From a theoretical point of view, many other extensions are possible. There is an existing body of literature on endogenous group or alliance formation in identity and in contest theory. But currently there are not many studies that intersect both areas. Moreover, all the identity-dependent externalities research was conducted with the assumption of complete information. The effects of ambiguous information will make the topic closer to real life. It may also be possible in cases of identity-driven group conflict that there are fragmentations within a group because of multidimensional identity, which leads to intra-group conflict (see, e.g., Choi et al., 2016; Hausken, 2005). Although there are empirical investigations in this area, a solid theoretical investigation would be highly appreciated. Finally, the controversial topic of affirmative action (Chowdhury et al., 2020) in contests is aptly interrelated with the idea of identity. One can observe from the field that identity-driven affirmative action may lead to or mitigate identity-driven conflict. Girard (2020) provided a first empirical result from India that such affirmative action increases the number of murders of members of lower castes. However, a theoretical investigation in this area is yet to be conducted.

In a laboratory experiment, Cadsby et al. (2013) found that priming about professional (compared to personal) features encourages females to join competition more. It is well known that females are often averse to being involved in conflict. Hence, it will be interesting to investigate whether priming can affect female conflict behavior as well. Experiments can help in understanding the components of conflict resolution. Kimbrough and Sheremeta (2013) showed that side payments help to resolve conflict even without commitment. Kimbrough et al. (2014) showed that mediation also helps to resolve conflict in the laboratory. However, Fisher (2001) argued that mediation could not mitigate the identity-driven ethnic conflict in Cyprus. There is no experiment involving identity-driven conflict that tests the effectiveness of conflict resolution tools in the laboratory. Note also that there is a recent neuroscience literature on the economic games of conflict (see, e.g., De Dreu & Gross, 2019; Rojek-Giffin et al., 2020). It will be interesting to understand the neuro-economic results on identity and conflict as well (Huettel & Kranton, 2012).

A great deal of interesting applied empirical research can also be conducted in this area. How a macroeconomic or global event such as the COVID-19 pandemic affects the inter-relation between identity and conflict (see, e.g., Chowdhury, 2020) is an important and compelling area of research. This may also help to solve a core question that remains unanswered: whether an identity instigates conflict, or a conflict makes identity salient. Further possible extensions of critical empirical research are both sabotage and affirmative action under the shadow of identity. The issue of multidimensional identity and its effects on conflict resolution will also be appealing. As it is not possible to theoretically or experimentally evaluate the effectiveness of public policies to mitigate identity-related conflict, empirical research in this area will be very welcome.


The author thanks Jordi Brandts and Joo Young Jeon for useful comments. He retains the responsibility for any remaining errors.


  • Abbink, K. (2012). Laboratory experiments on conflict. In M. R. Garfinkel & S. Skaperdas, (Eds.), Oxford handbook of the economics of peace and conflict (pp. 532–555). Oxford University Press.
  • Abbink, K., Brandts, J., Herrmann, B., & Orzen, H. (2012). Parochial altruism in inter-group conflicts. Economics Letters, 117(1), 45–48.
  • Akerlof, R. (2017). Value formation: The role of esteem. Games and Economic Behavior, 102, 1–19.
  • Akerlof, G. A., & Kranton, R. E. (2000). Economics and identity. Quarterly Journal of Economics, 115(3), 715–753.
  • Akerlof, G. A., & Kranton, R. E. (2010). Identity economics: How our identities shape our work, wages, and well-being. Princeton University Press.
  • Ashmore, R. D., Jussim, L., & Wilder, D. (2001). Social identity, intergroup conflict, and conflict reduction (Vol. 3). Oxford University Press on Demand.
  • Bagchi, A., Faria, J. R., & Mathews, T. (2019). A model of a multilateral proxy war with spillovers. Public Choice, 179(3–4), 229–248.
  • Basu, K. (2005). Racial conflict and the malignancy of identity. Journal of Economic Inequality, 3(3), 221–241.
  • Baye, M. R., Kovenock, D., & De Vries, C. G. (1996). The all-pay auction with complete information. Economic Theory, 8(2), 291–305.
  • Bénabou, R., & Tirole, J. (2011). Identity, morals, and taboos: Beliefs as assets. Quarterly Journal of Economics, 126(2), 805–855.
  • Bernard, M., Hett, F., & Mechtel, M. (2016). Social identity and social free-riding. European Economic Review, 90(C), 4–17.
  • Bernhard, H., Fischbacher, U., & Fehr, E. (2006). Parochial altruism in humans. Nature, 442(7105), 912–915.
  • Bolton, G. E., & Croson, R. T. (Eds.). (2012). The Oxford handbook of economic conflict resolution. Oxford University Press.
  • Bornstein, G. (1992). The free-rider problem in intergroup conflicts over step-level and continuous public goods. Journal of Personality and Social Psychology, 62(4), 597–606.
  • Cadsby, C. B., Servátka, M., & Song, F. (2013). How competitive are female professionals? A tale of identity conflict. Journal of Economic Behavior & Organization, 92(C), 284–303.
  • Chakravarty, S., Fonseca, M. A., Ghosh, S., & Marjit, S. (2016). Religious fragmentation, social identity and conflict: Evidence from an artefactual field experiment in India. PloS One, 11(10), e0164708.
  • Charness, G., & Chen, Y. (2020). Social identity, group behavior, and teams. Annual Review of Economics, 12, 691–713.
  • Chen, R. W., & Chen, Y. (2011). The potential of social identity for equilibrium selection. American Economic Review, 101(6), 2562–2589.
  • Chen, Y., & Li, S. X. (2009). Group identity and social preferences. American Economic Review, 99(1), 431–457.
  • Choi, J. K., & Bowles, S. (2007). The coevolution of parochial altruism and war. Science, 318(5850), 636–640.
  • Choi, J. P., Chowdhury, S. M., & Kim, J. (2016). Group contests with internal conflict and power asymmetry. Scandinavian Journal of Economics, 118(4), 816–840.
  • Chowdhury, S. M. (2019). The attack and defense mechanisms—perspectives from behavioral economics and game theory. Behavioral and Brain Sciences, 42, e121.
  • Chowdhury, S. M. (2020). Conflict in the time of (post-) corona: Some assessments from behavioral economics. Peace Economics, Peace Science, and Public Policy, 26(3), 20200052.
  • Chowdhury, S. M., Esteve-González, P., & Mukherjee, A. (2020). Heterogeneity, leveling the playing field, and affirmative action in contests. SSRN Working paper 3655727.
  • Chowdhury, S. M., Jeon, J. Y., & Ramalingam, A. (2016). Identity and group conflict. European Economic Review, 90(C), 107–121.
  • Chowdhury, S. M., Jeon, J. Y., & Ramalingam, A. (2018). Property rights and loss aversion in contests. Economic Inquiry, 56(3), 1492–1511.
  • Chowdhury, S. M., & Gürtler, O. (2015). Sabotage in contests: A survey. Public Choice, 164(1), 135–155.
  • Chowdhury, S. M., & Sheremeta, R. M. (2011). A generalized Tullock contest. Public Choice, 147(3), 413–420.
  • Chowdhury, S. M., & Topolyan, I. (2016a). The attack-and-defense group contests: Best-shot versus weakest-link. Economic Inquiry, 54(1), 548–557.
  • Chowdhury, S. M., & Topolyan, I. (2016b). Best-shot versus weakest-link in political lobbying: An application of group all-pay auction. Social Choice and Welfare, 47(4), 959–971.
  • Collier, P., & Hoeffler, A. (1998). On economic causes of civil war. Oxford Economic Papers, 50(4), 563–573.
  • Dasgupta, I., & Pal, S. (2021). Touch thee not: Group conflict, caste power, and untouchability in rural India. Journal of Comparative Economics.
  • Dechenaux, E., Kovenock, D., & Sheremeta, R. M. (2015). A survey of experimental research on contests, all-pay auctions and tournaments. Experimental Economics, 18(4), 609–669.
  • De Dreu, C. K., & Gross, J. (2019). Revisiting the form and function of conflict: Neurobiological, psychological, and cultural mechanisms for attack and defense within and between groups. Behavioral and Brain Sciences, 42(e116), 1–66.
  • Deutsch, M., Coleman, P. T., & Marcus, E. C. (Eds.). (2011). The handbook of conflict resolution: Theory and practice. John Wiley.
  • Diab, L. N. (1970). A study of intragroup and intergroup relations among experimentally produced small groups. Genetic Psychology Monographs, 82(1), 49–82.
  • Esteban, J., Mayoral, L., & Ray, D. (2012a). Ethnicity and conflict: An empirical study. American Economic Review, 102(4), 1310–1342.
  • Esteban, J., Mayoral, L., & Ray, D. (2012b). Ethnicity and conflict: Theory and facts. Science, 336(6083), 858–865.
  • Esteban, J., & Ray, D. (1994). On the measurement of polarization. Econometrica, 62(4), 819–851.
  • Esteban, J., & Ray, D. (1999). Conflict and distribution. Journal of Economic Theory, 87(2), 379–415.
  • Esteban, J., & Ray, D. (2008). On the salience of ethnic conflict. American Economic Review, 98(5), 2185–2202.
  • Esteban, J., & Ray, D. (2011). Linking conflict to inequality and polarization. American Economic Review, 101(4), 1345–1374.
  • Fang, H., & Loury, G. C. (2005). “Dysfunctional identities” can be rational. American Economic Review, 95(2), 104–111.
  • Fehr, E., & Schmidt, K. M. (1999). A theory of fairness, competition, and cooperation. Quarterly Journal of Economics, 114(3), 817–868.
  • Fisher, R. J. (2001). Cyprus: The failure of mediation and the escalation of an identity-based conflict to an adversarial impasse. Journal of Peace Research, 38(3), 307–326.
  • Foote, N. N., & Cottrell, L. S., Jr. (1955). Identity and interpersonal competence; a new direction in family research. University of Chicago Press.
  • Fryer, R., & Jackson, M. O. (2008). A categorical model of cognition and biased decision making. BE Journal of Theoretical Economics, 8(1), 1–44.
  • Fu, Q., & Wu, Z. (2019). Contests: Theory and topics. In J. H. Hamilton (Ed.), Oxford research encyclopedia of economics and finance. Oxford University Press.
  • Garfinkel, M. R., & Skaperdas, S. (Eds.). (2012). The Oxford handbook of the economics of peace and conflict. Oxford University Press.
  • Girard, V. (2017). Does inequality matter in conflicts? A case study of religious conflicts in India. Revue d’Economie Politique, 127(1), 93–117.
  • Girard, V. (2020). Stabbed in the back? Mandated political representation and murders. Social Choice and Welfare.
  • Hausken, K. (2005). Production and conflict models versus rent-seeking models. Public Choice, 123(1), 59–93.
  • Herrmann, B., & Orzen, H. (2008). The appearance of homo rivalis: Social preferences and the nature of rent seeking (No. 2008-10). CeDEx discussion paper series.
  • Huang, L., DeScioli, P., & Murad, Z. (2018). Pulling for the team: Competition between political partisans. Mimeo.
  • Huettel, S. A., & Kranton, R. E. (2012). Identity economics and the brain: Uncovering the mechanisms of social conflict. Philosophical Transactions of the Royal Society B: Biological Sciences, 367(1589), 680–691.
  • Katz, E., Nitzan, S., & Rosenberg, J. (1990). Rent-seeking for pure public goods. Public Choice, 65(1), 49–60.
  • Kimbrough, E. O., & Sheremeta, R. M. (2013). Side-payments and the costs of conflict. International Journal of Industrial Organization, 31(3), 278–286.
  • Kimbrough, E. O., Sheremeta, R. M., & Shields, T. W. (2014). Resolving conflicts by a random device. Journal of Economic Behavior and Organization, 99, 96–108.
  • Klor, E. F., & Shayo, M. (2010). Social identity and preferences over redistribution. Journal of Public Economics, 94(3), 269–278.
  • Klose, B., & Kovenock, D. (2015a). The all-pay auction with complete information and identity-dependent externalities. Economic Theory, 59(1), 1–19.
  • Klose, B., & Kovenock, D. (2015b). Extremism drives out moderation. Social Choice and Welfare, 44(4), 861–887.
  • Konrad, K. A. (2009). Strategy and dynamics in contests. Oxford University Press.
  • Linster, B. G. (1993). A generalized model of rent-seeking behavior. Public Choice, 77(2), 421–435.
  • Mitra, A., & Ray, D. (2014). Implications of an economic theory of conflict: Hindu-Muslim violence in India. Journal of Political Economy, 122(4), 719–765.
  • Reynal-Querol, M. (2002). Ethnicity, political systems, and civil wars. Journal of Conflict Resolution, 46(1), 29–54.
  • Riechmann, T. (2007). An analysis of rent-seeking games with relative-payoff maximizers. Public Choice, 133(1–2), 147–155.
  • Rojek-Giffin, M., Lebreton, M., Scholte, H. S., van Winden, F., Ridderinkhof, K. R., & De Dreu, C. K. (2020). Neurocognitive underpinnings of aggressive predation in economic contests. Journal of Cognitive Neuroscience, 32(7), 1276–1288.
  • Sambanis, N., & Shayo, M. (2013). Social identification and ethnic conflict. American Political Science Review, 107(2), 294–325.
  • Sen, A. (1985). Goals, commitment, and identity. Journal of Law Economics & Organization, 1(2), 341.
  • Sen, A. (2007). Identity and violence: The illusion of destiny. Penguin Books India.
  • Shayo, M. (2009). A model of social identity with an application to political economy: Nation, class, and redistribution. American Political Science Review, 103(2), 147–174.
  • Shayo, M. (2020). Social identity and economic policy. Annual Review of Economics, 12, 355–389.
  • Sherif, M., Harvey, O. J., White, B. J., Hood, W. R., & Sherif, C. W. (1961). Intergroup conflict and cooperation: The Robbers Cave experiment (Vol. 10). University Book Exchange.
  • Shih, M., Pittinsky, T. L., & Ambady, N. (1999). Stereotype susceptibility: Identity salience and shifts in quantitative performance. Psychological Science, 10(1), 81–84.
  • Skaperdas, S. (1996). Contest success functions. Economic Theory, 7(2), 283–290.
  • Tajfel, H., & Turner, J. C. (1979). An integrative theory of intergroup conflict. In M. J. Hatch & M. Schultz (Ed.), Organizational identity: A reader (pp. 56–65). Oxford University Press.
  • Tullock, G. (1980). Efficient rent seeking. In J. M. Buchanan, R. D. Tollison, & G. Tullock (Eds.), Toward a theory of the rent-seeking society (pp. 97–112). Texas A&M University Press.
  • Turner, J. (1978). Social categorization and social discrimination in the minimal group paradigm. In Henri Tajfel (Ed.), Differentiation between social groups (p. 140). Academic Press.
  • Wärneryd, K. (2014). The economics of conflict: Theory and empirical evidence. MIT Press.
  • Weisel, O., & Zultan, R. (2016). Social motives in intergroup conflict: Group identity and perceived target of threat. European Economic Review, 90(C), 122–133.


  • 1. Note that such identity-dependent externality is different from outcome-contingent spillover (e.g., Chowdhury & Sheremeta, 2011), where depending on the conflict-outcome opponent effort enters into one’s utility.

  • 2. See also Esteban and Ray (1994) for a measure of polarization that is relevant for identity-wise fragmented society; Esteban and Ray (2008) for the interaction of class and ethnicity as identity in the context of conflict; and Esteban and Ray (2011) for an interpretation of social distance and inequality as identity.

  • 3. See Chowdhury (2019) for a summary of the literature.