Identity and State Capacity

Regular readers of Broadstreet will note that we write frequently about two topics: state capacity and identity. Previous posts delved into the intellectual history of the study of the state,  and how scholarship is evolving to incorporate new thinking around conceptual issues, measurement, political agency and agents, and the iterative nature of state-building. On identity, posts from both our editors and guest contributors have covered identity persistence, transmission, propaganda, victimhood, cleavage formation, the challenges of collecting ethnographic data  and the role of ethnicity in bureaucratic development. And much much more!

In this post, I want to discuss recent work in historical political economy, some of which has appeared on this blog, that sits at the nexus of identity and state capacity.

(A quick aside: when I speak of identity, I mean “ethnic identity” and I follow broad convention in defining ethnic identity as membership into a descent-based or more subjectively as “self-identification around a characteristic that is difficult or impossible to change, such as language, race, or location” (Birnir 2006, 66). The range of ethnic identities I focus on include tribe, religion, caste, language, and race across a range of countries and cases.)

Historically oriented projects have some unique advantages in establishing how identity might matter to institutional development. First, in many instances, key institutional features—like per capita taxation, bureaucratic agencies or state census undertakings—were introduced for the first time following franchise expansion, conquest, or war, allowing for causal claims-making on how identity mattered to state-building .  Second, identity-based instruments of state-building, like a poll-tax on minority ethnic/racial groups or religious tithes, were explicit and observable, unlike implicit bias in contemporary outcomes such as the study of public goods or tax burdens. Third, recent HPE research deploys a wide range of quantitative and qualitative data to consider the specific mechanisms through which identity matters to the development of strong/weak bureaucratic institutions.

Rethinking war and state building through the lens of identity

But it is not entirely straightforward why identity politics should matter to state capacity. In fact research on state capacity has largely been silent on the role of identity in state-building. Take, for example, studies of fiscal capacity, which are core to the literature on state capacity.  Much of this literature that focuses on the Western European cases, tends to be macro-historical in orientation, focused on factors like wars, geography, and colonialism. Institutions are a happy accident of the churning of the wheels of history.

One reason for this omission is that the extent to which identity concerns can shape capacity is somewhat limited by the inability of rulers to ramp up institutions substantially in the short run (see Hilel Soifer’s work). Another related reason is that the sheer will and marshaling of resources required to ramp up capacity is so extensive,  that only existential threats like wars are seen to matter.  To the extent that identity has been studied in historical development it is focused on the material  or territorial concerns of groups and elites rather than on a non-economic identity.

So let’s begin with how by focusing on identity, recent work has problematized what we think about war-focused accounts of state-building. In her new book, Anna Gryzmala-Busse shows how the roots of state-building can be traced to the medieval era, a period prior to the early modern bellicist accounts of state capacity. The pivotal actor was the Roman Catholic church that used key resources like its literate clergy and land ownership to deliberately fragment state authority and consolidate its religious and social power. By focusing on the church, Gryzmala-Busse asks us to consider motivations beyond territorial expansion or the material enrichment of rulers.

Lisa Blaydes and Christopher Paik (2016) show the role of religious crusades in the development of the extractive state. As economic elites left cities across Europe to fight the religious crusades, rulers were more easily able to build institutions of extraction. They also show that places with greater numbers of religious fighters experienced more political stability, and these places also experienced greater extraction through crusade tithes – one of the first forms of “per-head” taxation in Europe.

This is a story that repeats across contexts. In Turkey, Yusuf Magiya (2022) shows using local-level tax data from the late 19th century in the Ottoman Empire that during inter-state wars, administrative units with more ethnically homogenous populations were able to collect more taxes and had more investments in fiscal capacity. This suggests that the conventional trope that “war makes the state” was mediated through the ethnic composition of the ruled.

In China, Peng Peng (2022) finds that Chinese Emperors during the Qing dynasty’s reign between 1644 to 1722 were more likely to deploy co-ethnic bureaucrats to regions of the kingdom experiencing conflict. In contrast, they relied on meritocratically selected officials in areas where security concerns were less heightened. In this way, ethnic considerations shaped key bureaucratic decisions during times of external and civil wars.

Rulers might seek to emphasize identity concerns even if they lose revenues in the process. Mohamed Saleh and Jean Tirole (2021) show how that following the Arab conquest of Christian Egypt, the Arab caliphate levied both a non-discriminatory tax and a poll-tax on non-muslims that would be eliminated upon conversion. When faced with declining revenues, as large numbers of Christians converted to Islam, pious rulers called for even more conversions suggesting religious motives dominated economic ones. In the United States, Nancy Qian and Marco Tabellini (2022) argue that identity-based discrimination shaped the military state because the lack of inclusiveness in institutions discouraged voluntary conscription, weakened the war effort, and intensified identity-based cleavages. They find that Black Americans enlisted in the military at lower rates after the attack on Pearl Harbor in counties with greater level of anti-Black discrimination.

Why does identity matter to state-building?

Beyond these challenges to “war built the state,” new research has sought to understand the mechanisms through which identity shaped institutions.  One strand of research focuses on the distinctive challenges ethnic or racial diversity posed for elites.  Identity can shape the quality of information the state can collect about its subjects (see this excellent post by Broadstreet editor Volha Charnysh on this topic). Research has shown that the ability to collect information is often contingent on sharing a language, eliciting the trust of subjects, or having reliable bureaucrats. Rulers can more accurately collect information about their populations when ethnically homogenous populations are more legible to bureaucrats. Magiya finds that in the late Ottoman Empire administrative units were more likely to complete a census when they were homogenous. Steven White and I show that census collectors collected worse data from white populations in enumerator districts with greater proportions of African Americans in 1880.

Identity can also explain who becomes a bureaucrat. Peng Peng’s research on 17th century China shows that rulers made trade-offs between ethnic loyalty and competence, and chose co-ethnics when loyalty was more prized as a way to secure information in perilous times. In Nigeria, Johnson-Kanu (2022) shows that ethnic groups that were more English educated had a first-mover advantage within the colonial bureaucracy. These ethnic groups continued to be over-represented in the bureaucracy a century later.

Research on  identity typically focuses on either individual-level behavior or preference formation; or group-based characteristics and how they shape political outcomes like violence, voting or public goods. The studies discussed here show that we need to take seriously the role of identity at a meso-level: in shaping institutions.

These findings also suggest that there is much work yet to be done to understand how other types of identity are at work too — notably gender — in a literature that has for too long focused on macro-history.