Assumptions

We consider an extensive game where the sovereign either maintains more restrictive institutional rules (R) or adopts less restrictive institutional rules (L). These rules determine the barriers to entry into different sectors of the economy and, therefore, the number of these sectors and the allocation of individuals across them. Here, ‘sectors’ should be understood as institutionally protected economic niches rather than conventional branches of economic activity. For example, the less restrictive rules can represent the weakening of the power of the guilds, reducing barriers to migration of labour, or liberalization of access to public-sector positions.

In practice, the extent to which such restrictions were enforced varied across regions and over time. We abstract from this heterogeneity by interpreting institutional regimes as capturing the overall degree of effective barriers to entry faced by individuals. In this sense, the model reflects the de facto restrictiveness of the institutional environment rather than its formal legal structure. Historical research on guilds supports this interpretation, showing that entry into various kinds of industries was far from free prior to the emergence of national states (Kadens, Reference Kadens2015; Ogilvie, Reference Ogilvie2005; Reference Ogilvie2014). Accordingly, external intervention by the sovereign can be understood as reducing existing entry barriers, thereby reshaping the structure of economic opportunities and patterns of interaction.

Under more restrictive rules, we assume that there are K _R sectors, while under less restrictive rules, there are K _L sectors, with K _L < K _R . More restrictive rules mean higher barriers to entry, which results in more fragmented markets and a greater number of sectors, an inefficient allocation of individuals to sectors, and a lower level of output. Liberalization reduces such artificial segmentation. Denote n _i (r) the number of individuals in sector i under the rules r and N is the total number of individuals. Therefore, we have ${\sum }_{i=1}^{K_{r}}n_{i}(r)=N$ , for r = R, L.

Assume that each individual is described by a set of characteristics. To simplify the analysis, we consider one such characteristic (‘ethnicity’), e ∈ {D, L}, where D denotes the dominant group and L the local group. Throughout the model, the dominant group refers to the ethnic group that initially enjoys greater access to economic and political opportunities within the existing institutional framework, whereas the local group denotes the subordinate ethnic group seeking to improve its position. These labels are purely relative and serve only to distinguish the two groups analytically; the model does not depend on any intrinsic characteristics of either group.

For analytical tractability, we treat ethnicity as given. This should be understood as a simplifying assumption rather than a claim that ethnic identities themselves are fixed or primordial. In line with the constructivist literature (e.g., Laitin, Reference Laitin2007), we view ethnicity as socially constructed and historically contingent. However, we do not model the processes through which identities are formed and become politically relevant.

The formation of ethnic-based coalitions depends on whether ethnicity is a salient characteristic. Formally, Nature determines the state s ∈ {0, 1}, where s = 1 denotes that ethnicity is salient and s = 0 that it is not. We treat the salience of ethnicity as an exogenous state variable, allowing us to focus on the mechanism of coalition formation conditional on identity salience.

Individuals can choose whether to invoke a common ethnic identity. Let A denote not invoking ethnicity and B invoking ethnicity to form a coalition.Footnote ¹ If s = 0, ethnicity is not salient and individuals choose A. If s = 1, individuals may choose between A and B.

Consider a sector i with the rules r. The utility of an individual j in this sector who does not invoke ethnicity is given by:

${u}_{ij}^{e}\left(r,A\right)={\left(1-t\right)Y_{i}\left(r\right) \over n_{i}\left(r\right)}{\rm \ for\ } e=D,L$

where Y _i (r) is the output produced by the sector i under the rules r and t is the tax rate.Footnote ²

If ethnicity is salient and individuals of a given group e ∈ {D, L} coordinate on invoking it, they redistribute the output in their favour and exclude others. We assume that if it is beneficial for an individual of a given ethnicity to choose B, then all individuals of this ethnicity choose B. In such a case, the utility of an individual of ethnicity e, is given by:

${u}_{ij}^{e}\left(r,B\right)=\cases{ {(1-t)(Y_{i}(r)-c_{i}) \over {n}_{i}^{e}\left(r\right)}{\rm \ if\ }{n}_{i}^{e}\left(r\right)\gt {n}_{i}^{e\prime}\left(r\right){\rm \ or\ }{n}_{i}^{e}\left(r\right)= {n}_{i}^{e\prime}\left(r\right){\rm \ and\ } e =D\ \cr 0{\rm \,\ \,\,\ \,\,\ \,\,\ \,\,\ \,\,\,\ \,\,\,\,\,\,\,\,\,\, if\ }{n}_{i}^{e}\left(r\right)\lt {n}_{i}^{e\prime}\left(r\right){\rm \ or\ }{n}_{i}^{e}\left(r\right)= {n}_{i}^{e\prime}\left(r\right){\rm \ and\ }e=L }$

e′ ≠ e denotes the other ethnic group. c _i represents the cost of conflict, or more specifically, of forming the coalition and excluding the other ethnic group from sharing the output. This specification abstracts from classic collective action problems (Olson, Reference Olson1965). Although individuals are the payoff-relevant actors, we model the decision to invoke common ethnicity as a collective action at the sectoral level. We assume that when invoking common ethnicity is individually beneficial for members of an ethnic group, coordination within the group is possible and participation is enforceable.

In particular, we abstract from intra-group free riding by assuming that ethnic groups possess informal enforcement mechanisms, such as social sanctions, exclusion from future cooperation, or the use of coercion, that make non-participation unprofitable. The cost c _i should therefore be interpreted broadly as incorporating both the costs of inter-group conflict and the internal costs of organizing and enforcing collective action. Since sectors differ in size, productivity, and ethnic composition, c _i may implicitly reflect variation in the number of excluded individuals. Allowing c _i to increase explicitly with the size of the excluded group would strengthen the trade-off between exclusion and efficiency, but would not alter the qualitative results of the model.

The assumption of feasible coalition formation applies only when ethnicity is salient. When ethnicity is not salient, individuals may still coordinate along other social or economic dimensions, but ethnic identity does not serve as the basis for coalition formation. Such non-ethnic forms of collective action are outside the scope of the present analysis.

n _i ^e (r) is the number of the members of the ethnic group e in the sector i. We have ${\sum }_{i=1}^{K_{r}}{n}_{i}^{e}(r)=n^{e}(r)$ and n ^D (r) + n ^L (r) = N, where n ^e (r) is the total number of the members of the ethnic group e. We assume that the ethnic group that is in the majority shares the output among its members. The minority obtains nothing. If the number of group members is equal for both coalitions, we assume that the group D shares the output. While we assume that the dominant ethnic group is more numerous than the local group, i.e., n ^D (r)> n ^L (r), we may have n _i ^L (r)> n _i ^D (r) for some sectors, especially if the sectors are formed on the basis of the location and the local ethnic group is geographically concentrated.

If n _i ^e (r)> n _i ^e′(r) or n _i ^e (r)= n _i ^e′(r) and e = D, the members of the group e will invoke common ethnicity if u _ij ^e (r, B) > u _ij ^e (r, A), i.e., if:

${Y_{i}(r)-c_{i} \over Y_{i}\left(r\right)}\gt {{n}_{i}^{e}\left(r\right) \over n_{i}\left(r\right)}$

That is, if the cost of conflict is sufficiently low and if the number of members of the group e is not too large. Figure 2 illustrates the combinations of conflict costs and group size under which ethnic-based coalitions emerge.

Figure 2.

Formation of ethnic coalitions.

If n _i ^e (r)< n _i ^e′(r), or equivalently n _i ^e (r) < n _i (r)/2 group e is the minority in sector i. In this case, its members will not invoke a common ethnicity because they cannot secure control over the sector’s output. These situations are represented by the white area above the line c _i = Y _i (r)/2, which is obtained from the coalition condition when n _i ^e (r) = n _i (r)/2. By contrast, if n _i ^e (r) > n _i (r)/2, or n _i ^e (r) = n _i (r)/2, and e = D, the members of the group e may or may not form a coalition, depending on Y _i (r) and c _i . The light-grey area, F, represents situations where, for given n _i ^e (r), the additional output gained by forming a coalition does not compensate for the cost of conflict. The dark grey area, G, shows the cases where coalitions are formed.

It is assumed that under more restrictive rules, individuals never form coalitions, i.e., u _ij ^e (R,A) > u _ij ^e (R,B). The rationale is that if it was profitable to form coalitions under the existing rules, they would be formed already. Under less restrictive rules, coalitions may or may not be formed, depending on whether ethnicity is a salient characteristic and on the parameters Y _i (r), c _i , and n _i ^e (r).

A nation emerges if the number of ethnic-based coalitions of locals is sufficiently large. That is, if the number of these coalitions exceeds some threshold $\overline{K}$ . This threshold represents the minimum number of ethnically based economic coalitions required for national identity to become socially and politically salient at the aggregate level. It captures coordination and visibility constraints: below $\overline{K}$ , ethnic mobilization remains localized and fragmented, whereas above $\overline{K}$ it becomes sufficiently widespread to sustain shared institutions and political claims. For simplicity, $\overline{K}$ is treated as exogenous, reflecting slow-moving factors such as communication technologies, transport infrastructure, and the broader political environment. Denote the number of sectors where the local group prevails with K _LL . Thus, a nation is formed if $K_{LL}\geq \overline{K}$ .

Our aim is to present a simple and tractable framework that identifies the conditions under which ethnic-based coalition formation emerges. To do so, the model abstracts from richer dynamic processes of bottom-up interaction. Alternative approaches, such as social interaction models (e.g., Durlauf, Reference Durlauf1999; Young, Reference Young1998), evolutionary dynamics, or spontaneous-order frameworks (Hayek, Reference Hayek1973), may provide more general accounts of decentralized institutional emergence. Our contribution is not to develop a general theory of spontaneous social order, but to identify a particular mechanism capable of explaining the emergence of nations under specific historical conditions.

The framework assumes a one-directional causal mechanism in which economic conflicts generate ethnic-based coalitions, which may subsequently contribute to the emergence of a nation. In historical settings, however, this relationship can become bidirectional once nationalist movements gain momentum, as emerging national identities may reinforce patterns of coalition formation. We abstract from such feedback effects in order to isolate the initial process through which economic conflict generates nationalist mobilization. Accordingly, the model should be interpreted as explaining how coalition formation can give rise to nationalism, rather than treating nationalism and coalition formation as interchangeable phenomena.

Let us now consider the payoff of the sovereign:

$V\left(R\right)={\sum }_{i=1}^{K_{R}}tY_{i}\left(R\right)=tY\left(R\right){\rm \ and\ }V\left(L\right)= {\sum }_{i=1}^{K_{L}}t\left(Y_{i}\left(L\right)-C_{i}\right)=t\left(Y\left(L\right)-C\right)$

where C _i = c _i if an ethnic group is formed in sector i, and C _i = 0 if an ethnic group is not formed in sector i. This formulation abstracts from the possibility that conflicts may arise even when ethnicity is not salient and coalitions form along other dimensions; such conflicts are not explicitly modelled, and their costs are normalized to zero, as the focus of the model is on ethnic-based coalition formation. The sovereign’s payoff consists of total tax revenues collected from economic output. It is assumed that Y(L) > Y(R), i.e., the aggregate output is larger under less restrictive rules. However, if ethnic coalitions form in a sector under less restrictive rules, output is reduced by the conflict cost c _i , which reflects both direct losses from conflict and enforcement as well as foregone efficiency due to exclusion. The sovereign maximizes his payoff V(r) by choosing and implementing the set of institutional rules, anticipating coalition formation and the associated conflict costs, but not the long-run political or ideological consequences of nation formation.

The model does not explicitly introduce a technological or geopolitical shock as shown in Figure 1 as a separate variable. Instead, such shocks are captured implicitly through their effects on the economic environment, in particular on the relative productivity of institutional regimes, as reflected in Y(L) and Y(R). A shock that increases the gains from market integration or institutional reform can be interpreted as raising Y(L) − Y(R), thereby making liberalization more attractive to the sovereign. In this sense, the shock provides the trigger for reform by shifting the relative benefits of institutional regimes. The model abstracts from the source of such shocks and focuses instead on their consequences for institutional choice and subsequent coalition formation.

Timing and equilibrium

Figure 3 summarizes the equilibrium outcomes of the model and the conditions corresponding to the four scenarios. The timing of the game is as follows. First, Nature determines the salience of ethnicity by selecting the state s ∈ {0, 1}. This state is observed by the sovereign, who either keeps more restrictive rules or chooses less restrictive rules. Based on the chosen rules and on whether ethnicity is salient, individuals in each sector either invoke common ethnicity or not. The resulting equilibrium outcomes depend on the realized state and the relative magnitude of the gains from reform and the cost of conflict. These outcomes are described in the following scenarios.

Figure 3.

Equilibrium outcomes.

Scenario 1. Ethnicity is not salient. In this situation, conflicts do not arise. Therefore, there is an equilibrium in which the sovereign chooses the less restrictive rules as Y(L) > Y(R). In this scenario, nations do not emerge.

Scenario 2. Ethnicity is salient, and Y(L) − Y(R) ≤ C. In this situation, there is an equilibrium in which the sovereign keeps the more restrictive rules. Therefore, common ethnicity is not invoked, and nations do not emerge.

Scenario 3. Ethnicity is salient, Y(L) − Y(R) > C and $K_{LL}\lt \overline{K}$ . In this situation, there is an equilibrium in which the sovereign adopts less restrictive rules, as the increase in output by relaxing the rules outweighs the cost of conflict. Since ethnicity is salient, ethnic-based coalitions are created. However, since $K_{LL}\lt \overline{K}$ , a nation is not formed.

Scenario 4. Ethnicity is salient, Y(L) − Y(R) > C and $K_{LL}\geq \overline{K}$ . This scenario is the same as Scenario 3, except that a nation is formed, as $K_{LL}\geq \overline{K}$ .

Note that nation formation is a by-product of growth-oriented institutional reform rather than its objective. In the model, the emergence of a nation refers to the formation of a sufficiently large number of ethnically based economic coalitions, reflecting the consolidation of a shared national identity and collective capacity. Importantly, nation formation does not imply political independence or the loss of sovereign authority. Ethnic coalitions may coexist with continued imperial rule, taxation, and administrative control.

In the following section, the model is applied to account for the emergence of the Czech nation in the late Habsburg Monarchy. This case is an example of Scenario 4.

Overview

The emergence of the Czech nation can be understood within the framework developed in Section Model as a sequence of interconnected processes triggered by exogenous shocks and subsequent institutional change. The relevant period spans from the late 18^th century to the early 20^th century, during which the Czech Lands were part of the multiethnic Habsburg Monarchy and later the Austro-Hungarian Empire, accounting for 26% of the area of the Austrian part of the monarchy and 35% of its population (Jindra and Jakubec, Reference Jindra and Jakubec2015).

The process was initiated by a series of geopolitical and economic shocks, most notably the loss of a substantial part of Silesia in the mid-eighteenth century, followed by the loss of the Austrian Netherlands and the disruptions associated with the Continental Blockade. These events weakened the fiscal and economic position of the Empire, exposed structural weaknesses in its economic and administrative system, and, in combination with the onset of industrialization, significantly altered the incentives of the sovereign.

In terms of the model, such shocks can be interpreted as increasing the gains from institutional reform, that is, raising Y(L) − Y(R), thereby making liberalization more attractive to the sovereign. At the same time, the loss of Silesia shifted the economic centre of gravity toward the Czech Lands, which became the new economic powerhouse of the Empire (Jakubec et al., Reference Jakubec, Efmertová, Szobi and Štemberk2008; Jindra and Jakubec, Reference Jindra and Jakubec2015).

In response, the Habsburg monarchy implemented a series of institutional adjustments, including the Theresian and Josephine Reforms in the second half of the 18^th century, followed by measures such as the abolition of serfdom, constitutional reform, and the Trades License Regulation later in the 19^th century. These reforms, among other effects, reduced barriers to internal migration and market entry while lowering transaction costs, corresponding in the model to a shift from more restrictive to less restrictive institutional rules.

These changes fundamentally altered the structure of the economy by increasing interaction across groups and setting the stage for intensified competition over economic and political resources, as discussed in the following section.

Reforms

In the mid-eighteenth century, the Czech Lands had a predominantly agricultural structure, with peasants organized in communes and bound by manorial obligations (Ogilvie, Reference Ogilvie2005). This system was decentralized and granted substantial authority to landlords and commune leaders. Beginning in the late eighteenth century under Maria Theresa and Joseph II and continuing into the nineteenth century, a series of reforms fundamentally altered this structure.

These reforms were aimed at centralization, administrative rationalization, and market integration (Evans, Reference Evans2006; Good, Reference Good, Teichova and Matis2003). They included legal harmonization, the standardization of currency and measurements, the gradual dismantling of internal trade barriers, and the reform and eventual abolition of guild privileges (Ogilvie, Reference Ogilvie2014). At the same time, reforms of the manorial system – culminating in the abolition of serfdom in 1848 – substantially increased the mobility of labour and reduced barriers to entry into economic activities.

In terms of the model, these changes correspond to a shift from more restrictive to less restrictive institutional rules, undertaken in response to increased gains from reform reflected in Y(L) − Y(R). At the same time, fiscal and administrative reforms strengthened state capacity and increased the sovereign’s ability to extract revenue.

These reforms led to a profound restructuring of the economy. The reduction of barriers to movement and market entry increased interaction across previously segmented groups and facilitated the expansion of urban labour markets and commercial activity. In the framework of the model, this corresponds to a reshuffling of individuals across sectors and a reduction in the number of sectors as markets became more integrated.

As a result, competition for economic and political resources intensified. In particular, migration from rural areas brought Czech-speaking workers into cities where German-speaking populations had previously been dominant, increasing competition for urban employment. Similar tensions arose in access to credit and positions within an increasingly professionalized state administration. These developments illustrate how institutional reforms generated new forms of interaction and conflict across groups, setting the stage for the emergence of ethnic-based coalition formation.

The emergence of the Czech identity

Although the historical roots of the Czech nation date back to the mediaeval period, the political and cultural position of the Czech population weakened significantly under Habsburg rule. Following the Battle of White Mountain in 1620, the Czech ruling class was largely replaced by foreign nobility, and German gradually became the dominant language in administration, education, and urban life (Bláhová et al., Reference Bláhová, Frolík, Profantová, Čornej, Bobková, Klimek, Borovička and Bělina1999). By the eighteenth century, Czech was used primarily in rural areas and lacked institutional support, with few elites engaged in its cultivation (Bělič, Reference Bělič1953).

The second half of the eighteenth century saw a revival of scholarly interest in Czech language and culture, corresponding to Phase A of national movements in Hroch’s (Reference Hroch1985) framework. Scholars actively reconstructed the language using historical sources and vernacular usage, expanding its vocabulary to meet the needs of a modernizing society. In our framework, this process represents the existence of a latent cultural characteristic that can later serve as a basis for coordination.

Industrialization and associated institutional reforms altered the conditions under which this latent identity became relevant. Increased labour mobility brought Czech-speaking workers from rural areas into urban centres, where German had previously been dominant. This migration intensified interaction between linguistic groups and corresponds in the model to a reshuffling of individuals across sectors. As economic competition increased, language became a more salient marker of group identity.

At the same time, segments of the Czech population – particularly emerging economic elites – had incentives to resist German dominance in economic and administrative spheres. In cities such as Prague, where interaction between groups was most intense, linguistic and cultural boundaries became increasingly important in structuring economic and social relations (Auty, Reference Auty1956).

Over time, these developments translated into institutional change. Czech speakers gradually gained representation in administrative bodies, and language regulations adopted in 1880 and 1897 granted Czech formal equality with German in official communication. In terms of the model, these changes reduced the cost of coordination and conflict for Czech-speaking individuals, as linguistic barriers that had previously limited access to economic and administrative positions were weakened.

Nationalism and economic interests

As economic development and industrialization progressed, industrial entrepreneurs began to play an increasingly important role in the Czech national revival. Their participation reflected not only cultural or political motivations but also economic incentives, as national mobilization provided a means of advancing group-specific economic interests (Hroch, Reference Hroch, Jančík and Kubů2011).

A key constraint for emerging Czech entrepreneurs was access to finance. Most prospective founders came from relatively modest backgrounds and lacked the capital required to enter industrial sectors. Established German and Jewish banks primarily served large, incumbent firms and offered limited access to smaller entrants. In response, Czech actors established their own savings banks, which channelled funds to Czech entrepreneurs (Kubů and Šouša, Reference Kubů, Šouša, Jančík and Kubů2011).

In terms of the model, this represents a case of resource conflict over access to capital, where existing financial institutions functioned as an entry barrier. The creation of Czech savings banks can be interpreted as an instance of ethnic-based coalition formation, in which individuals coordinated on shared identity (i.e., invoking ethnicity) to overcome exclusion and reallocate resources within the group. By pooling resources and restricting access, these coalitions both facilitated economic activity and reinforced national identity.

Similar dynamics can be observed in the political sphere. As institutional reforms expanded opportunities for participation, entrepreneurs became increasingly involved in local governance, particularly in industrializing regions (Kladiwa et al., Reference Kladiwa, Pokludová and Kafková2008). Control over municipal institutions allowed these actors to influence the allocation of public resources and administrative decisions in ways that favoured their own group.

This development reflects the institutionalization of ethnic-based coalitions. Once established, such coalitions did not operate only through informal coordination but increasingly shaped formal rules and decision-making processes. In this sense, nationalism and economic interests became mutually reinforcing, as economic competition generated incentives for coalition formation, while successful coalitions strengthened the salience and persistence of national identity.

Tensions between Czechs and Germans

The model developed in Section Model predicts that when a shared cultural characteristic becomes salient, exogenous shocks and institutional reforms generate ethnic coalition formation, leading to segmentation along national lines. It also predicts that nationalist mobilization is more likely to emerge in regions characterized by both cultural diversity and relatively high levels of economic development, where the gains from coordination outweigh the costs of conflict.

These predictions are consistent with the historical development of the Habsburg Monarchy. Nationalist mobilization contributed to the emergence of parallel Czech and German economic and social structures. Professional societies and associations, while sometimes formally open to all, increasingly operated along ethnic lines, with smaller, culturally homogeneous organizations proving more effective in coordinating economic activity. In line with the model, such groups faced lower internal coordination and enforcement costs, making ethnic-based coalitions more efficient in securing access to local resources than broader, multiethnic organizations.

There is also evidence that nationalist activity was strongest in economically developed and culturally mixed regions. Hroch (Reference Hroch1985) shows that areas with above-average national mobilization were concentrated near the Czech-German language boundary and in regions with significant economic activity, such as grain and linen production. In these areas, the middle classes played a particularly important role in nationalist movements, and the number of Czech patriots was higher than elsewhere.

Economic conflicts increasingly translated into nationalist practices, including consumer boycotts and exclusionary market behaviour, particularly from the 1880s onward (Albrecht, Reference Albrecht2001; Jančík and Kubů, Reference Jančík and Kubů2011). In the framework of the model, these practices can be interpreted as deliberate exclusionary strategies used by ethnic coalitions to reallocate resources in their favour by restricting access to markets for competing groups. These dynamics reinforced ethnic boundaries and further encouraged coalition formation along national lines.

As these processes intensified, the costs associated with maintaining a multiethnic political structure increased. In terms of the model, this can be interpreted as a rise in aggregate conflict costs associated with widespread ethnic coalition formation. Once these costs became sufficiently large relative to the benefits of maintaining the existing institutional arrangement, the stability of the system was undermined. The eventual collapse of the Austro-Hungarian Empire and the emergence of an independent Czechoslovak stateFootnote ³ in 1918 can thus be understood as the outcome of these cumulative processes.