Abstract
The lack of a shared national identity is often blamed for ethnic conflict and low economic development. This raises the question: does a society’s modernization (in particular, industrialization) lead to a shared identity, thereby bringing good outcomes in conflict and development? This paper theoretically examines the question using a contest model of conflict augmented with multiple production sectors and social identification.
The analysis shows that as modernization proceeds, a society shifts to an equilibrium with a universal national identity, a low level of conflict, and high output if national pride is high, ethnic differences are not salient in people’s minds, resources are not abundant, or institutions are of good quality. Otherwise, the society shifts to an equilibrium with a universal ethnic identity and worse economic and political outcomes. The analysis suggests that nation-building policies play a critical role in overcoming the negative effects of modernization under the latter situation.
Introduction
Evidence suggests that ethnic divisions in a society lead to negative outcomes in civil conflict (Esteban et al., 2012) and economic development (Montalvo and Reynal-Quero, 2005). The lack of a shared national identity, that is, the dominance of ethnic over national identity, is often blamed for the negative outcomes (Collier, 2009, Michalopoulos and Papaioannou, 2015). If shared national identity is important for positive outcomes, then how can it be realized? This study focuses on the effect of the modernization of a society, in particular, industrialization and resultant greater ethnic integration in the workplace. Does modernization contribute toward shared identity, as the classic thesis in political science argues (Deutsch, 1953, Gellner, 1964, Gellner, 1983, Weber, 1979)? Or does it lead to heightened ethnic identity and conflict, as a competing thesis asserts (Melson and Wolpe, 1970, Bates, 1983)?
This paper explores the effect of modernization on social identity, ethnic conflict, and economic development theoretically. To do so, it develops a contest model of conflict augmented with multiple production sectors and social identification, drawing on the model by Sambanis and Shayo (2013).
In a standard contest model of conflict, groups compete for exogenous resources used to produce group-specific club goods (e.g., public services and infrastructure benefiting specific groups), the members of each group choose contributions to conflict, taking into account the cost and benefit of their action, and the total contributions of group members determine the amount of resources the group obtains.
Sambanis and Shayo (2013) augment the standard model with socio-psychological factors. Individual utility depends not only on the material cost and benefit of contributing to ethnic conflict but also negatively on the perceived distance between oneself and the group one identifies with (their ethnic group or the nation) and positively on the group’s status. In other words, one bears a large cognitive cost when they are very different from other members of the group in relevant aspects, but takes pride in being a member when the group’s status is high. These socio-psychological components are major determinants of social identification and intergroup behaviors, according to influential theories in social psychology (Tajfel and Turner, 1986, Turner et al., 1987) and empirical evidence (Manning and Roy, 2010, Sambanis et al., 2015).23 Importantly, because these components differ depending on whether one identifies with their ethnic group or the nation, the social identities of people influence individual contributions to conflict and thus the level of conflict. Further, social identity is endogenously determined: one chooses the identity that brings them higher utility. Hence, social identity and individual and aggregate outcomes interact with each other.
To examine the effect of modernization on social identity, conflict, and development, this study modifies Sambanis and Shayo’s model in two ways. First, private good production is introduced into the model. There are multiple sectors—ethnically segregated traditional sectors (e.g., traditional agriculture and the urban informal sector in the actual economy) and the ethnically integrated modern sector—and the individuals work in the sector that maximizes their utility. In the model, the modernization of a society is the sectoral shift of labor and production from the traditional sectors to the modern sector, and it is driven by the increased (total factor) productivity of the modern sector. Second, sectoral affiliation, along with ethnicity, is a component of the perceived distance. In other words, one feels strong discomfort when their ethnicity and sectoral affiliation, which represents the type of job (e.g., white collar, informal sector, and agricultural jobs) they have, differ significantly from those of a typical person of their identity group.
Although the present model is a simple extension of Sambanis and Shayo’s model, it can analyze not only the interaction between social identity and conflict, the focus of Sambanis and Shayo (2013), but also the interactions of social identity with economic variables—sectoral compositions of labor and output and aggregate output. By examining how productivity-driven modernization changes these interactions, this paper shows how modernization affects social identity, conflict, and development and derives policy implications.
The model has multiple equilibria that differ in the proportion of individuals identifying with the nation. First, to see how social identification itself affects conflict and economic outcomes, different equilibria are compared for given parameters and exogenous variables.4 The analysis shows that an equilibrium with more prevalent national identity has a lower level of conflict, higher modern sector shares in employment and production, and, under plausible conditions, higher levels of private good production and aggregate material payoff (the value of private and club good consumption net of the cost of conflict). In other words, national identity is associated not only with a lower level of conflict, shown in Sambanis and Shayo (2013), but also with higher modern sector shares and higher output.
Then, a driving force of modernization—the increased productivity of the modern sector—is introduced into the model. The productivity growth raises the modern sector wage, induces a higher proportion of workers to work in the ethnically integrated sector, and increases the sector’s output share. How does the productivity-driven modernization affect identity, conflict, and development? Analysis shows that the effect depends on the status difference, the difference between the status of the nation and that of ethnic groups. The national status and the status difference would be high when the people of a nation believe that they share a glorious history, rich culture, or a “right” sense of values because they feel proud of belonging to such a nation. When ethnic groups are distinctive in these aspects, the ethnic status is high and the status difference is low.
If the status difference is very high (low), everyone identifies with the nation (one’s ethnic group) and the level of conflict is low (high) all the time. Otherwise, when the status difference is relatively high (low), the society typically shifts from an equilibrium in which the national identity is more prevalent among modern sector workers to an equilibrium with a universal national (ethnic) identity and a low (high) level of conflict. Given the productivity level, the society with a high (low) status difference has relatively large (small) modern sector shares and, under plausible conditions, relatively high (low) levels of private good production and aggregate material payoff. Hence, having sufficiently high national status relative to ethnic status is crucial for achieving good outcomes in development as well as in identity and conflict.5 An exogenous change that makes ethnic differences less salient in people’s minds also exerts effects similar to those of a rise in the national status. Similar results also hold for contested resources. Specifically, given the status difference, when the amount of contested resources is large (small), the society ends up in the equilibrium of a universal ethnic (national) identity. This indicates that both the abundance of resources and the absence of strong political and economic institutions (e.g., weak rule of law), which leads to abundant contested resources, are obstacles to good outcomes.
The above results are consistent with the classic thesis on the effects of modernization on identity in political science if the national status relative to the ethnic status is high, resources are not abundant, institutions are of good quality, or ethnic differences are not salient in people’s minds. Otherwise, the results are consistent with the competing thesis. The results also have important policy implications. Under the former conditions, policies promoting modernization, such as policies stimulating the technological progress of the modern sector and the reform of institutions supporting the sector’s activities, might be sufficient for achieving good outcomes. Conversely, under the latter conditions, they exert negative effects on national identity and conflict, thereby becoming less effective for development. This highlights the vital role of policies that raise the national status, improve institutional quality, or make ethnic differences less salient. Miguel, 2004, Collier, 2009, and Blouin and Mukand (2019), based on a case study or statistical analysis, argue that national identity is effectively strengthened through nation-building policies, such as school education and government propaganda that emphasize common history, culture, and values and the promotion of a national language.6 These policies may be considered as ones that lift the national status or deemphasize ethnic differences and thus are critical under the adverse conditions.
This paper adds to the literature on contest models of conflict (Sambanis and Shayo, 2013, Esteban and Ray, 2008, Esteban and Ray, 2011, Besley and Persson, 2010, Sambanis et al., 2015, Mariani et al., 2018). Besley and Persson (2010) examine the interaction between conflict and development, as in this study, but focuses on capacities of the state to raise revenue and provide market-supporting services. Sambanis et al. (2015), partly drawing on Sambanis and Shayo (2013), present a model in which leaders may initiate interstate war anticipating that victory raises the national status and induces national identification beneficial to the people.
The paper also adds to the theoretical literature examining interactions between identity and economic or political behaviors (Fearon and Laitin, 2000, Akerlof and Kranton, 2000, Akerlof and Kranton, 2010, Shayo, 2009, Benabou and Tirole, 2011, Bisin et al., 2011, Gennaioli and Tabellini, 2019, Grossman and Helpman, 2021).7 By generalizing the pioneering work of Akerlof and Kranton, 2000, Shayo, 2009 constructs the basic framework, which motivates Sambanis and Shayo (2013) and this study, and applies it to analyze the political economy of income redistribution. Shayo’s (2009) framework has been applied to various issues. For example, motivated by a recent reversal of trade policies in some western countries seemingly influenced by rises of populism and ethnic tensions, Grossman and Helpman (2021) construct a political economy model of trade policy with social identification and examine how policies are affected by changes in the identification patterns triggered by events such as increased ethnic tensions.
Further, the paper is related to theoretical work on nation-building policies, such as Alesina and Reich, 2020, Almagro and Andrés-Cerezo, 2020, and Alesina et al. (2021). Theses papers explicitly model the determination of nation-building polices and examine how the implemented policies depend on factors such as the threat of democratization and the initial distribution of identities.
Finally, the paper contributes to the literature theoretically examining the modernization of an economy, such as Lewis, 1954, Banerjee and Newman, 1998, Proto, 2007, Vollrath, 2009, and Yuki, 2007, Yuki, 2008, Yuki, 2016. For analytical tractability, this study models the inefficient sectoral allocation of workers in a simplest manner and considers the modernization induced by exogenous productivity growth. Conversely, these studies model factors leading to the inefficient allocation more explicitly and examine economic mechanisms of modernization in detail.
The rest of the paper is organized as follows. Section 2 presents the model, Section 3 examines equilibria, and Section 4 analyzes the effects of modernization on identity, conflict, and development. Section 5 examines how the results in Section 4 are affected by the abundance of contested resources and discusses how the results depend on several assumptions. Section 6 concludes. Appendix A presents the existence conditions for equilibria, and Appendix B contains proofs.