Undead Past: What Drives Support for the Secessionist Goal of the Indigenous People of Biafra (IPOB) in Nigeria?

Abstract

This study examined the effect of perceived ethnic marginalization, perception towards Nigerian democracy, and socioeconomic condition on support for secession among members of the Igbo ethnic group. Perceived ethnic marginalization and negative perceptions toward Nigerian democracy were found to positively correlate with support for secession. Socioeconomic condition was measured at the household and communal levels. The household measure had no effect on support for secession, but the communal measure did: socioeconomic condition at the communal level positively correlated with support for secession. Igbo ethnicity increased the likelihood of supporting secession, while belonging to the Hausa/Fulani and Yoruba ethnic groups reduced the likelihood of supporting secession.

1. Introduction

Since Nigeria’s transition to civilian rule in 1999, it has contended with various groups agitating for the independence of the predominantly Igbo Eastern Region to form the Republic of Biafra. This sentiment is tied to the defunct Republic of Biafra, which was in existence from May 30, 1967, to January 15, 1970. The secession of the Eastern Region from Nigeria marked the beginning of the Nigerian Civil War, which is also known as the Biafra War. The collapse of Biafra and its subsequent reincorporation into Nigeria marked the end of the war. Pro-Biafra agitations have largely centered on the topic of marginalization faced by members of the Igbo ethnic group in postwar Nigeria. Although secession had been contemplated by the predominantly Hausa/Fulani Northern Region and the predominantly Yoruba Western Region prior to the Biafra War, these considerations never morphed into action (Harnischfeger 2019; Imuetinyan 2017, p. 216; Orobator 1987). The Igbos were the first to challenge the entity called “Nigeria” by both words and action. Despite losing the war and being reintegrated into Nigeria, Achebe (1983, p. 50) warned that unless all Nigerians (especially the Igbo) were treated fairly, the polity risked retrogression and instability. Two prominent pro-Biafra groups are the Movement for the Actualization of the Sovereign State of Biafra (MASSOB), which was established in 1999 (Okonta 2018, p. 361), and the Indigenous People of Biafra (IPOB) movement, a splinter group of MASSOB, which was established in 2014 (BBC 2017). Although both groups are still in existence, MASSOB has diminished in prominence compared to IPOB. IPOB was created to rejuvenate the drive for secession when the head of MASSOB, Ralph Uwazurike, was thought to have compromised in his commitment to the goals of the movement (Adangor 2018, p. 144).

The Nigerian Government has been resolute in its commitment to keep the country united and has often employed brute force through its security agencies to quell pro-Biafra agitations (Amnesty International 2016, 2021). Nigeria’s former President, Muhammadu Buhari, who was in office from May, 2015, to May, 2023, and who had fought on the side of the Nigeran military during the Biafra War, proscribed IPOB as a terrorist organization in 2017 (Ogbonna 2017; Ezea and Olaniyi 2017), a move that Adangor (2018) strongly criticized for its politicization of the fight against terrorism. The leader of IPOB, Nnamdi Kanu, is currently in the custody of Nigeran authorities and has been charged with treasonable felony (Jalloh 2021, The Cable 2021). This is not the first time he has been arrested on such charges. He was first arrested in 2015 and released on bail after two years of incarceration. The conditions of his release prohibited him from granting interviews, participating in rallies, and being in gatherings of over ten people (Richards 2017), conditions he flouted when he fled the country shortly after his release and continued his campaign from abroad (Ojoye 2017; Nasiru 2018). His lawyers and IPOB, on the other hand, contend that he did not defy his bail conditions, but was rather compelled to flee the country because his life was under threat (Nwachukwu 2021; Yusuf 2021; The Cable 2017).

IPOB’s first objective, as stated on its official website, centers on discrimination: “To promote human rights advocacy and protect the rights of Indigenous Peoples in all parts of the world who are facing persecution and discrimination.” ¹ The relationship between the various ethnic groups in postcolonial Nigeria, especially the major three (Hausa/Fulani, Yoruba, and Igbo), has been very competitive, and to some degree adversarial. The overlap between ethnicity and religion adds another layer of complexity to the mix. The Igbos are predominantly Christian, the Hausa/Fulani are predominantly Muslim, and both religions are almost equally represented among the Yoruba (Laitin 1986, p. 8). After six decades of independence, the goal of national integration continues to elude Nigeria. Commenting on the prominence of ethnic cleavages after Nigeria gained independence from British colonial rule, Falola and Heaton (2008, p. 159) observed: “when Nigeria became an independent sovereign state in 1960, in many ways it was a state without a nation.” In his little book entitled, The Trouble With Nigeria, Achebe (1983, p. 5) pointed out that “Nothing in Nigeria’s political history captures her problem of national integration more geographically than the chequered fortune of the word tribe in her vocabulary.” Whenever the topic of marginalization is mentioned in Nigerian public discourse, it often pertains to the Igbos and their relationship with the other two major ethnic groups (Adewole 2021; Ede 2021; Njoku 2019). Ikpeze (2000, p. 90) observed: “As a people the Igbo have been systematically disempowered politically, economically, militarily and socially by the Hausa/Fulani and Yoruba groups.”

Although the state of Biafra ceased to exist in 1970, its memory persists in the minds of many, including those who were born after the war (Smith 2007, pp. 191–220; Maier 2000, p. 271). Despite the government’s heavy-handed approach to suppressing IPOB’s activities, it has remained active, and its ideology has kept spreading. The government’s fixation on the group has enhanced its popularity and turned its leader, Nnamdi Kanu, into a “cult hero” (Maiangwa 2021). Emphasizing the futility of the forceful approach adopted by the Nigerian government, Idachaba and Nneli (2018, p. 56) observed that it “only strengthens ethno-nationalist movements, radicalize some of her members and attract public sympathy to such groups.”

Many Igbos migrate from their Eastern homeland to other regions within Nigeria. Such movement is driven by their competitive, individualistic, and entrepreneurial nature, which enables them to perceive and take advantage of opportunities (Ede et al. 2021; Nnadozie 2002; Coleman 1958, p. 333). Also, the infertility of the land in Eastern Nigeria, which makes it unsuitable for agricultural purposes, coupled with its scarcity and a high population density, further drives emigration from the region (Achebe 2012, pp. 74–75; Ikpeze 2000, pp. 105–106; Coleman 1958, p. 332). Given such interregional dependence, not every member of the Igbo ethnic group would support the secession of the Eastern Region from Nigeria. Relying on the horizontal inequalities theory, this study seeks to investigate how perceived ethnic marginalization, attitudes toward Nigerian democracy, and socioeconomic condition influence support for secession among members of the Igbo ethnic group. Moreover, it examines how belonging to Nigeria’s three major ethnic groups—i.e., Igbo, Hausa/Fulani, and Yoruba—influences support for secession.

Although much research has been conducted on the demand for secession by neo-Biafra groups, most of them are qualitative. With the exception of the study by Lewis (2022), the few studies that use quantitative data often employ it descriptively and focus on the leadership of the movements and the response of the Nigerian government, while paying scant attention to the perceptions of the larger Nigerian population (e.g., Idachaba and Nneli 2018; Obi-Ani, Nzubechi, and Obi-Ani 2019; Chiluwa 2018). Relying on survey data, Lewis (2022) found that exposure to neo-Biafran conflicts positively correlated with support for secession among the subsample of respondents who belong to ethnic groups that are autochthonous to Biafran territory. Moreover, his regression results showed that support for secession was strongest among ethnic groups that would dominate the state of Biafra; the smaller ethnic groups that would be dominated were not supportive of secession.

This study differs from the one conducted by Lewis (2022) in the following ways: First, it relies upon the framework of the horizontal inequalities theory and pays particular attention to how political and socioeconomic horizontal inequalities influence support for secession. Second, recognizing the tripoidal nature of Nigeria’s ethnic landscape, this study zooms in on Nigeria’s three major ethnic groups—i.e., the Hausa/Fulani, Yoruba, and Igbo—and investigates how belonging to each of these groups influences support for secession. Third, this study, which relies heavily on historical analysis, provides a more detailed analysis of the Nigerian case study than Lewis (2022). Lastly, unlike Lewis’ study, which focuses on the ethnic groups that are autochthonous to Biafran territory, this study focuses specifically on members of the Igbo ethnic group. It is important to focus on Igbos because the IPOB movement is essentially an Igbo affair: IPOB members are primarily Igbos, most IPOB-related conflicts occur in Igbo territory, support for IPOB is highest among Igbos, and IPOB’s claims mostly pertain to the political and socioeconomic exclusion of Igbos in postwar in Nigeria.

This study finds that among Igbos, perceived ethnic marginalization at the group level and negative perceptions toward Nigerian democracy increase the likelihood of supporting secession. This is likely because of the exclusion of Igbos from political power at the center in Nigeria. Socioeconomic condition was measured at the household and communal levels. The household measure had no effect on support for secession, but the communal measure, which I proxied using the literacy rate of the population in the local government area (LGA) (i.e., municipality) where the respondent resided, positively correlated with support for secession. A plausible explanation for this finding could be that Igbos feel that their association with Nigeria holds them back and they could do better alone. Belonging to Hausa/Fulani and Yoruba ethnic groups reduced the likelihood of supporting secession while belonging to the Igbo ethnic group positively correlated with support for secession. This study contributes to the broader literature on the determinants of secession, especially those with an empirical focus (e.g., Lewis 2022; Cunningham and Sawyer 2017; Jenne, Saideman, and Lowe 2007; Cunningham 2013).

This study proceeds as follows: Section 2, which relies on historical analysis, provides an overview of the relationship between Nigeria’s three major ethnic groups from the 1950s up to 1970 when the Biafra War ended. Section 3 discusses the literature on inter-group conflict with particular emphasis on the nexus between horizontal inequalities and secessionist conflict. Section 4 discusses the trend of Biafra-related conflicts in Nigeria. Section 5 operationalizes the variables that will be used to estimate the regression model and specifies the general form of the model to be estimated. Section 6 presents the regression results and discusses them, while Section 7 summarizes the paper and concludes.

2. Ethnic competition and the Biafra War

Nigeria, like most African countries, plummeted into civil war after gaining independence from colonial rule. Nigeria was created by the British when they merged the Northern and Southern protectorates, previously distinct entities administered by the British, on January 1, 1914. The amalgamation report written by Sir. Frederick Lugard, Nigeria’s first Governor-General, shows that the merger was driven by financial expediency and administrative convenience, with little consideration for the cultural differences between the peoples who were brought together (Lugard 1919, pp. 7–8). Despite the amalgamation, the British continued administering the Northern and Southern Regions differently. In the Northern Region, which had been an Islamic caliphate for a century until its capture in 1903, the British did not change much when they took over. They appropriated the existing institutions and even employed the local Hausa language for administrative purposes. This contrasts with the Southern Region where English was adopted as the administrative language, and the strategies of Westernization and Christianization were pursued ardently (Coleman 1958, pp. 46–47; Nafziger and Richter 1976, pp. 92–93). The division of the Southern Region into the Eastern and Western Regions in 1939 entrenched ethnic consciousness in the minds of Nigerians. This is because the three administrative divisions were closely associated with each of Nigeria’s three major ethnic groups: The Northern Region was dominated by the Hausa/Fulani, while the Eastern and Western Regions were dominated by the Igbo and Yoruba, respectively (Ake 1993, p. 3; Imuetinyan 2017, pp. 208–209). This put the minority ethnic groups in the uncomfortable position of having to fit into the mold of the dominant ethnic groups in the regions where they resided (Achebe 2012, p. 47).

The heterogenous peoples who constitute Nigeria were able to transcend their differences to confront a common foe—European rule, but then turned against each other after the goal of independence had been achieved. This explains why Geertz (1973, p. 237) asserted that “removing European rule has liberated the nationalisms within nationalisms.” Having established their dominance over the colony through brute force and demonstrated their willingness to employ violence in quelling dissent, the British created a superficial semblance of stability within the colony (Falola 2009, pp. 1–25). In the middle of the twentieth century, when Nigeria began taking bold strides toward independence, the ethnic cleavages between the various ethnic groups that had been lurking beneath the surface became more prominent.

Besides the adversarial relationship between Nigeria’s three major ethnic groups, the smaller ethnic groups also feared domination from the bigger ones (Nigeria comprises of 250 ethnic groups). As the period of independence drew nearer, the minority ethnic groups became apprehensive about their status under the majority ethnic groups in post-independence Nigeria. They appealed to the British government to create more states to mitigate their concerns about ethnic domination. The British Government set up the Willink Commission in 1957 to examine these concerns (Akinyele 1996).

During the hearings that followed, the minority ethnic groups accused the majority ethnic groups of occupying most of the top positions in the civil service, using the institutions of the state for their benefit, and bias in the allocation of infrastructure and social amenities (Akinyele 1996, pp. 77–78). Although the commission acknowledged that the minority ethnic groups had genuine concerns, it did not support the creation of more states because this could create new ethnic minorities, thus failing to address the fundamental problem of ethnic domination. The commission recommended constitutional safeguards as a viable tool for protecting ethnic minorities (Imuetinyan 2017, p. 217). Nevertheless, the Nigerian government created more administrative regions after independence. The Mid-west Region was created in 1963, which increased the number of administrative units to four. In the wake of the Biafra War in 1967, the military government divided the four regions into twelve states. The number of states increased to nineteen in 1976, twenty-one in 1987, thirty in 1991, and thirty-six in 1996 (Alapiki 2005).

Nigeria gained independence from British colonial rule on October 1, 1960. Politics in post-independence Nigeria was characterized by stiff competition between the three regions, and by extension the three major ethnic groups. This seriously undermined national integration. Each region had a major political party whose support base largely consisted of the dominant ethnic group residing there. Northern People’s Congress, as can be inferred from the name, was the main party in the Northern Region. It was akin to the party of the Hausa/Fulani. Action Group in the Western Region was closely associated with the Yoruba, and the National Council of Nigeria and the Cameroons was closely tied to the Igbos (Akinyele 1996, p. 75; Laitin 1986, p. 6). The fear of domination was not peculiar to the minority ethnic groups. The Hausa/Fulani feared that they would be dominated by the Igbos and Yoruba because they lagged in the area of education and did not have a large pool of educated people to fill up positions in the civil service. The Igbos and Yoruba, on the other hand, feared domination by the Hausa/Fulani because of their influence in the political sphere (Falola and Heaton 2008, pp. 165–166; Siollun 2009, pp. 76–77; Diamond 1988, pp. 49–50; Laitin 1986, p. 6).

Nigeria’s first six years after independence were tumultuous. Like Falola and Heaton (2008, p. 159) concisely put it, “Official corruption, rigged elections, ethnic baiting, bullying and thuggery dominated the conduct of politics in the First Republic, which existed from 1960 to 1966.” Frustrated with the inability of the civilian government to unite the peoples of the various regions and maintain stability over the polity, a group of army officers, mostly of Igbo ethnicity, launched a coup on January 15, 1966, which toppled the civilian government. The coup resulted in the deaths of Nigeria’s Prime Minister, who was from the Northern Region, the Premiers of the Northern and Western Regions, and some senior military officers, most of whom were from the Northern Region. However, the Premiers of the Eastern and Mid-West Regions, who were Igbos, were not killed (Falola and Heaton 2008, p. 172; Achebe 2012, p. 64; Siollun 2009, p. 79). The coup succeeded in toppling the civilian government, but the plotters were arrested.

The most senior army officer at the time, Major General Johnson Aguiyi-Ironsi, who was of Igbo ethnicity, took over power and appointed military governors to head the four administrative regions. Lt. Colonel Chukwuemeka Ojukwu was appointed as the Governor of the Eastern Region. Since the perpetrators of the coup were mostly Igbos, it was interpreted as an attempt by the Igbos to dominate the other ethnic groups (Achebe 2012, p. 66; Siollun 2009, p. 79). Arguiyi-Ironsi’s regime lasted for only six months because he was killed in a counter coup orchestrated by soldiers from the Northern Region. This led to the ascension of Lt. Colonel Yakubu Gowon, a Northerner, as Head-of-State. The Northern soldiers’ thirst for revenge on the Igbos was not allayed by the killing of Arguiyi-Ironsi and the transfer of power to a Northerner. They systematically targeted and killed their Igbo colleagues. Subsequently, they moved into the civilian sphere, alongside blood-thirsty hoodlums, and unleashed their barbarity on Igbo civilians. Tens of thousands of Igbos were killed, and their properties looted and destroyed. This led to the mass exodus of the Igbos to their homeland in the Eastern Region (Siollun 2009, pp. 117–138). Between 80,000 and 100,000 lives were lost during the pogrom (Ekwe-Ekwe 1990, p. 12). Conservative estimates put the death toll at 30,000 (Achebe 2012, p. 82).

The Igbos no longer felt safe in a united Nigeria. “It was not until 1966–7 when it [anti-Igbo sentiment] swept through Northern Nigeria like ‘a flood of deadly hate’ that the Igbo first questioned the concept of Nigeria which they had embraced with much greater fervor than the Yoruba or the Hausa/Fulani.” (Achebe 1983, p. 45). This led to the Governor of the Eastern Region, Lt. Colonel Odumegwu Ojukwu, proclaiming the Republic of Biafra on May 30, 1967. Although Ojukwu was at the forefront of the fight for secession, the decision to secede was a collective one that was supported by the Igbo people (Achebe 2012, p. 91). Diplomatic attempts to forestall the secession like the Aburi Summit in Ghana and a visit to the Eastern Region by members of the National Reconciliation Commission were unsuccessful. The secession of Biafra was soon followed by the Biafra War, which was primarily about the reincorporation of Biafra into Nigeria. After 30 months of fighting and the deaths of over a million people, mostly from starvation as a result of the economic blockade imposed on Biafra by the Nigerian military, the war ended on January 15, 1970, with the surrender of Biafra (Falola and Heaton 2008, p. 180; Achebe 2012, pp. 222–228).

The Biafrans and the Federal Military Government (FMG) viewed the conflict from different perspectives: The Biafrans saw the war as a fight for the continued survival of the Igbo people and an effort to resist the genocidal tendencies of the FMG. In his speech commemorating the one-year anniversary of Biafra, Ojukwu observed: “We have convinced the aggressors that force alone cannot subdue a people, that Biafrans have chosen independence, the only guarantee of their survival, and are prepared to die defending it.” (Biafran Government 1968, p. 3). The FMG saw the war as a battle to preserve a united Nigeria. The dynamics of the war were influenced by external actors, especially the British and French Governments. Nigeria may have gained independence from British rule in 1960, but it was not economically independent because Britain still played a central role in its economic life. The British had significant investments in the financial sector, oil sector, and extractive industries. Also, Nigeria was an important market for manufactured goods from Britain. The close ties between the two countries made it difficult for Britain to remain neutral in the war, especially because its investments were at stake and the unity and stability of Nigeria were in its economic interest. The six-day Arab-Israeli war, which resulted in the closure of the Suez Canal, further threatened British oil supplies. These factors prompted the British Government to renege on its initial decision to not sell ammunition to the FMG. British support skewed the war in favor of the FMG, leading to the eventual defeat of Biafran forces in 1970 (Ekwe-Ekwe 1990, pp. 27–36). The duration of the war was also influenced by the support that the French offered Biafran forces, which enabled them to better resist attacks launched by the Nigerian military (Ekwe-Ekwe 1990, p. 46).

After Biafra’s surrender, Nigeria’s Head-of-State, Yakubu Gowon, declared an amnesty to ensure that vengeance was not taken out against the Igbos (Okafor 2006, p. 160; Achebe 1983, p. 45). In his speech after the war, he declared, “no victor, no vanquished,” and referred to the conflict as a “war between brothers.” (Hill 2012, p. 113). Gowon subsequently launched the Reconciliation, Reconstruction, and Rehabilitation program, which was aimed at reintegrating the Igbos back into Nigerian society. However, many Igbos today still feel marginalized politically and socioeconomically (Campbell and Page 2018, p. 34; Smith 2007, pp. 191–220; Ikpeze 2000).

3. Horizontal inequalities and conflict

Not every multiethnic society is embroiled in conflict. The horizontal inequalities theory explains why cultural diversity leads to conflict in some instances but not in others. The crux of this theory is that cultural differences among groups, say along ethnic or religious lines, do not lead to conflict. What causes conflict is the horizontal inequalities that exist among these culturally defined groups (Stewart 2000). By portraying inequalities among groups as the source of conflict rather than cultural differences, this theory challenges the “Clash of civilizations” perspective, which contends that conflict occurs when different cultural groups come into contact (Huntington 1996). Horizontal inequalities could be economic, social, political, or cultural. Social horizontal inequality constitutes unequal access to services like education, healthcare, housing, etc. Cultural horizontal inequality could emanate from discrepancies in the recognition ascribed to different languages, norms, customs, and practices (Stewart 2000, p. 249; 2010, pp. 1–2). These inequalities, which have the capacity to cause grievances among marginalized groups, are crucial in the mobilization process that precedes the onset of conflict. For a group to be cohesive, its members need to share certain characteristics like having a common language, tradition, ethnicity, religion, and the same source of hardship. The leaders of these groups often employ the strategy of “reworking historical memories” to accentuate the identity of the group and strengthen cohesion within it (Stewart 2000, p. 247).

Toft (2012) contends that ethnicity is central to secessionist conflicts: “Whereas an ethnic group is a latent nation, a nation is a politically active ethnic group, which tends to demand greater cultural autonomy or self-determination.” (p. 584). She also points out that exclusion precipitates the emergence of nations because it “forces groups of individuals to identify themselves in relation to that non-membership.” (p. 584). In the mobilization process that precedes conflict, perceived inequality is just as important as objective measures of inequality. Stewart (2000, p. 252) notes that “a poor ‘objective’ situation in terms of group inequality may not translate itself into conflict…if ideological elements are such that the inequalities are not widely perceived.” Similarly, Brown and Langer (2010, p. 30) observe: “If groups do not perceive the prevailing socioeconomic and political inequalities as unfair or unjust, severe objective horizontal inequalities might not provoke conflict.” The capacity of the marginalized group to challenge the state also depends on its size: “Where groups are small numerically, their potential to cause conflict on a substantial scale is limited, even when they suffer persistent discrimination.” (Stewart 2000, p. 254).

Brown and Langer (2010) acknowledge that similarities exist between horizontal inequalities theory and Ted Gurr’s “relative deprivation” theory, ² but they also contend that both theories differ because horizontal inequalities theory allows for the possibility that the group instigating conflict might not necessarily be disadvantaged. As they concisely put it: “[I]f an economically privileged group is geographically concentrated, it may seek more autonomy or even independence in order to maintain or improve its relatively advantageous position.” (p. 33). Similarly, Horowitz (1985, pp. 249–250) argues that when the population in the affluent region of a country feels it is contributing disproportionately to the state’s coffers—thus subsidizing the poorer regions—it might seek secession to gain greater control over its resources.

Although horizontal inequalities theory is predicated upon the constructivist view that identity is malleable, socially constructed, and changes over time, it acknowledges that some aspects of identity are quite stable and difficult to change. It is often on the basis of these difficult-to-change identities that group leaders stir up grievances in the mobilization process. These relatively stable aspects of identity also tend to make ethnic boundaries more salient (Stewart 2008, pp. 10–11). Sen (2006, p. 2) supports this argument: “A strong—and exclusive—sense of belonging to one group can in many cases carry with it the perception of distance from other groups.” He also highlights the tendency for identity to be exploited for violent ends: “Violence is fomented by the imposition of singular and belligerent identities on gullible people, championed by proficient artisans of terror.” (Sen 2006, p. 2). A crucial element in Sen’s view of identity is the attribution of responsibility to the individual. He points out that while an individual can belong to several groups and have multiple identities, he or she has the capacity to decide on the amount of importance to attach to each of these identities (Sen 2006, pp. 5–6). Nevertheless, he also acknowledges that people may encounter difficulty in disentangling themselves from certain categories that are ascribed to them by the larger society, for instance in the case of race (Sen 2006, pp. 6–8).

Griffiths (2021) has developed a theory where he asserts that secessionist movements are essentially the same since their ultimate goal is to obtain recognition from both the home state and the international community. Secessionist movements employ two main strategies to achieve this goal: compellence and normative appeal. He defines compellance as “the use of assets to coerce the home state and/or international community; it is direct action designed to increase the costs of not complying with secessionist demands.” (p. 30), and normative appeal as a “set of tactics that are designed to showcase the grievances and demands of the aspiring nation and either change preferences on the issue or bring into the game previously uninvolved parties.” (pp. 30–31). Both strategies, which tend to complement one another, are often used simultaneously by secessionist groups. Englebert and Hummel (2005) contend that the decision on whether or not to pursue secession depends on a cost-benefit analysis: Regional leaders compare the benefits of having partial control over the sovereign state’s institutions with the potential benefits that could accrue from secession without international recognition. Only when the dividends from the latter exceed the former do regional leaders pursue secession. They tout this as one of the reasons for the deficit of secessionist conflicts in Sub-Saharan Africa. Horowitz (1992, p. 122) makes a similar assertion when he observes that “whether a group is integrationist or secessionist depends, in large measure, on its assessment of its prospects in the undivided state.”

Cunningham (2013) argues that self-determination campaigns are likely to turn violent when the group seeking secession is large, marginalized in the political sphere, economically disadvantaged, seeking independence, and operating in a country characterized by a low level of economic development. Cunningham and Sawyer (2017) have identified three conditions for a group to make a self-determination claim: First, the group needs to see its ethno-nationalist identity as relevant, especially in relation to other ethnic groups. Second, there needs to be a common grievance among members of the group, this grievance could be economic, political, or even embedded in the lack of recognition of traditional structures that are peculiar to the ethnic group (e.g., language). Lastly, there needs to be an expectation among members of the group that the goal of self-determination is achievable. Cunningham and Sawyer (2017) have also noted that governments address secessionist claims either by making concessions to accommodate the demands of the group or by repressing the group. Both approaches are likely to foment the demand for secession because making concessions could embolden the group to make further demands; repressing the group could heighten the feeling of grievances among its members, lending credence to their claim of maltreatment by the government.

Members of the Igbo ethnic group and IPOB meet most of the criteria stipulated in horizontal inequalities theory. The Igbos constitute the third largest ethnic group in Nigeria (i.e., after the Hausa/Fulani and Yoruba), which makes them politically relevant and large enough to contend with the Nigerian state. ³ Since Nigeria’s transition to democracy in 1999, the Igbos have been excluded from political power at the center. Presidents and vice presidents have emerged from the Hausa/Fulani and Yoruba ethnic groups but never has an Igbo person held any of these offices. For most of Nigeria’s history, political power at the center has been controlled by the Hausa/Fulani group (Harnischfeger 2019, pp. 329–330; Ukiwo 2013, pp. 182–184; Mustapha 2009). ⁴ Harnischfeger (2019) contends that even though the rhetoric of the early neo-Biafran movements (especially MASSOB) was radical, its ultimate goal was not to secede from Nigeria, but rather to use the threat of secession to gain leverage in the political landscape and eventually produce a president of Igbo ethnicity.

IPOB, through its campaigns, has kept memories of the defunct Republic of Biafra alive and constantly reminded the Igbos of their marginal position in postwar Nigeria. Choi and Piazza (2016) have shown that political marginalization positively correlates with the incidence of domestic terrorist attacks. In a microlevel study conducted in Turkey, Sarigil and Karakoc (2016) found that Kurds who feel that they are treated unfairly by the Turkish Government are more supportive of secession and autonomy. Choi and Noll (2021) have pointed out that “ethnic inclusiveness is part of everyday perceptions of democracy,” and when politically relevant ethnic groups are excluded from the political process, the risk of intrastate and interstate conflicts rises. In a more recent study, Choi (2022) has shown that when political leaders uphold their legitimacy by appealing to the ethnicity of their supporters, disfavored ethnic groups, whose members fear being dominated, might resort to ethnic terrorism as a means of survival.

In a country like Nigeria, where the population attaches more importance to its ethnoreligious identity than its nationality (Tuki 2023; Agbiboa 2013; Agbiboa and Maiangwa 2013), it is not surprising that neo-Biafran conflicts have religious undertones. ⁵ Igbos often frame their resentment over perceived marginalization by the Nigerian state in religious terms by likening themselves to the Jews and the Nigerian state to the Egyptian Pharaoh who held the Jews captive in biblical times (Smith 2007, p. 206). IPOB, on its website, associates Igbo ethnicity with Christianity: “We, the Indigenous People of Biafra, are seeking the restoration of the sovereign, independent, and Judeo-Christian Nation-state of Biafra.” The group also alludes to the threat of Igbos (i.e., Christians) being dominated by a predominantly Muslim national government: “We are also fighting against radical, state-sponsored, Islamic terrorism that has slaughtered millions of our people and is hell-bent on the occupation of our land and Islamization of our people.” ⁶

Given the political exclusion of the Igbos, I expect that they would have a higher level of perceived ethnic marginalization at the group level and invariably more negative attitudes toward Nigerian democracy than members of the other major ethnic groups—i.e., the Hausa/Fulani and Yoruba. This, in turn, should increase their likelihood of supporting secession. The discussion so far leads to the first set of hypotheses that this study seeks to test:

1. H1: Belonging to the Igbo ethnic group positively correlates with support for secession.

2. H2: Perceived ethnic marginalization at the group level positively correlates with support for secession among Igbos.

3. H3: Negative perceptions toward Nigerian democracy positively correlate with support for secession among Igbos.

While the existence of grievances—imagined or real—is crucial in linking horizontal inequalities to conflict, some scholars have challenged the notion that grievances cause conflict. Collier and Hoeffler (2004, 2000), in their Greed and Grievance series of publications, argue that most rebellions are driven by greed; the narrative of grievances is merely a façade used by rebel leaders to conceal their ulterior motive of self-enrichment. One of the main predictors of conflict onset in their model is socioeconomic condition, which they proxy with income. Low income reduces the opportunity cost of joining a rebel group, which in turn increases the risk of conflict. In a later publication, in which they zoom in on the determinants of secessionist conflicts, they conclude that “secessionist movements should not in general be seen as cries for social justice. Those few secessionist movements that are able to scale-up to being organizations with a serious political or military capability are likely to occur in rich regions and contain an element of a resource grab.” (Collier and Hoeffler 2006, pp. 52–53).

Choi and Luo (2013) have shown that poverty emanating from economic sanctions creates an avenue for opportunistic leaders to exploit the grievances of the poor by attributing their suffering to the actions of external governments. This, in turn, increases the risk of international terrorism. In a study conducted in Ivory Coast, Langer (2005) finds that political horizontal inequality among elites and socioeconomic horizontal inequality among the masses were crucial in the mobilization process preceding the country’s descent into violent conflict in the 1990s. This is because the exclusion of some elites from the political process prompted them to mobilize their ethnic kin/supporters toward conflict; the mobilization process was successful because of the socioeconomic horizontal inequality that existed among the masses. Using disaggregated data for Sub-Saharan Africa, Østby, Nordås, and Rød (2009) have shown that regions characterized by low levels of education, relative deprivation in terms of asset ownership, and intraregional inequalities are susceptible to conflict. Conversely, Deiwiks, Cederman, and Gleiditsch (2012) contend that advantaged groups might seek secession because they contribute disproportionately to the state’s coffers and think they could do better on their own. A case in point would be the pursuit of secession by Indonesia’s natural resource-rich regions of Aceh, Riau, Papua, and East Kalimantan (Tadjoeddin 2011).

Toft (2012) has questioned the potential for economic grievances to cause secessionist conflicts: “In short, it generally takes something more than economic grievances to motivate groups to challenge the existing order and demand greater autonomy or secession.” (p. 587). These arguments could be tied to the Nigerian case: IPOB’s leaders might court the support of poor Igbos by attributing their poverty to the Nigerian government’s ineffectiveness and portraying secession as the key to improving their socioeconomic condition. It is also possible that the Igbos have a better socioeconomic condition than the national average; this disparity makes them feel that their association with Nigeria holds them back, and they can do better if the Eastern Region secedes. The latter mechanism is especially plausible given that Nigeria relies heavily on crude oil exports for its revenues and a significant proportion of Nigeria’s oil reserves are within the proposed Biafran territory. ⁷ Moreover, relying on gridded data on literacy rate in Nigeria (Bosco et al. 2017), I find that the average literacy rate for members of the Igbo ethnic group is higher than those for the Yoruba, Hausa/Fulani, and the Nigerian population. To capture these different possibilities, I will test the following three hypotheses:

1. H4a: Socioeconomic condition negatively correlates with support for secession among Igbos.

2. H4b: Socioeconomic condition positively correlates with support for secession among Igbos.

3. H4c: Socioeconomic condition does not correlate with support for secession among Igbos.

4. Trend in pro-Biafra agitation

Except for the period of rule by Alhaji Shehu Shagari, which lasted from 1979 to 1983, and the 83-day rule of Chief Ernest Shonekan in 1993, Nigeria was ruled by military dictators from 1970, when the Biafra War ended, until May 1999, when the military officially handed over power to a civilian government. ⁸ Nigeria has remained under civilian rule since then. The transition to civilian rule marked the beginning of pro-Biafra agitations. The absence of agitation prior to 1999 may be explained by the repressive nature of the military governments and their willingness to employ brute force to keep the polity under control. Moreover, the way the Biafra War ended almost certainly shook the resolve of the Igbos. Biafran forces were defeated, the Biafran leader fled to Ivory Coast, and the war had caused both material and psychological carnage. The transition to civilian rule probably created a semblance of expanded freedoms, which allowed pro-Biafra sentiments that had been simmering beneath the surface to erupt.

Figure 1, which is based on data obtained from the Armed Conflict Location and Events Database (ACLED) (Raleigh et al. 2010), shows the trend in Biafra-related conflicts and the accompanying fatalities from 1997 to 2022. ⁹ A limitation of the ACLED dataset is its heavy reliance on media reports. It is possible that pro-Biafra agitations were present when Nigeria was under military rule, but they were not reported in the media because press freedom was stifled. Nevertheless, the ACLED dataset still remains invaluable in understanding the trends in and nature of Biafra-related conflicts because of its disaggregated nature and the fact that it is updated in real-time. The blue dashed curve shows the annual trend in conflicts where at least one of the actors was a neo-Biafran group; the black dotted curve shows only the incidents involving IPOB. The latter curve is a subset of the former. The red dashed curve shows the annual trend in fatalities associated with Biafra-related conflicts.

Figure 1. Biafra-related conflicts and fatalities, 1997–2022 (ACLED).

A total of 409 Biafra-related conflicts occurred between January 1, 1997, and December 31, 2022. These incidents caused 761 fatalities. IPOB accounted for 73 percent of the total Biafra-related conflicts. The first incident, which involved MASSOB, was recorded in February 2000. Prior to 2015, MASSOB was the main pro-Biafra movement, but this changed with the advent of IPOB, whose first incident was recorded in 2015. The advent of IPOB has been associated with an increase in the number of fatalities. In 2016, a year after IPOB entered the scene, 176 fatalities were recorded. This corresponds to a growth of 1,157 percent when compared to the 14 fatalities recorded in 2015. The year 2021 was the most violent year between 1997 and 2022, both in terms of the incidence and intensity of the conflict. There were 114 incidents and 220 fatalities. The proximity of the black dotted curve to the blue dashed curve from 2019 to 2022 indicates that IPOB alone accounted for almost all Biafra-related conflicts during this period. The total Biafra-related incidents (n = 409) were categorized as Battles (40%), Protests (23%), Violence against civilians (20%), Riots (7%), Explosions/Remote violence (1.7%), and Strategic developments (8%). Total of 32 incidents were recorded in 2015, of which 75 percent were protests. A total of 220 incidents were recorded between 2021 and 2022, of which 79 percent were either Battles or Violence against civilians. There were only nine protests during this period. This indicates that Biafra-related conflicts have become violent.

Since the ACLED dataset is georeferenced, I rendered the geolocations of the Biafra-related conflicts on a map showing Nigeria’s 36 states, the federal capital territory (i.e., Abuja), and Nigeria’s three major regions. ¹⁰ Relying on the Georeferencing of Ethnic Groups dataset (Weidmann, Rød, and Cederman 2010), I also show the spatial area occupied by members of the Igbo ethnic group. ¹¹ As shown in Figure 2, Biafra-related incidents are clustered in Nigeria’s Eastern Region—especially the Igbo spatial area. There were seven incidents in Abuja, which is not surprising since it is the seat of the Nigerian Federal Government (See point B on map). Four incidents were recorded in Lagos State (see point A). This may be explained by the concentration of Igbos in the state (See panel 1 in Figure 3).

Figure 2. Spatial distribution of Biafra-related conflicts (1997–2022).

Figure 3. Ethnicity and settlement patterns.

Ethnic settlement patterns have persisted over time in Nigeria. Moreover, ethnicity was considered in the state-creation process in postcolonial Nigeria. Toft (2003) has highlighted the tendency for ethnic identity to be closely associated with the spatial area occupied by members of an ethnic group. She has also pointed out that structural factors like settlement patterns—i.e., the concentration of members of an ethnic group in a region—could lead to secessionist conflicts because it “a) makes political organization easier over a compact territory; b) facilitates military operations; and c) defines the territory over which claims can be made.” (Toft 2012, p. 590). “In Nigeria, …territories have traditionally been associated with ethnic groups and the expression, ‘owners of the land,’ remains viable even to the extent of identifying cities as ‘belonging’ to particular ethnic group” (Plotnicov 1972, p. 001).

Because this study relies on the Afrobarometer survey dataset (BenYishay et al. 2017), which is georeferenced, I rendered the geolocations of the respondents who belong to Nigeria’s three major ethnic groups (i.e., Hausa/Fulani, Igbo, and Yoruba) on a map showing Nigeria’s three major regions and the spatial areas they occupy. Figure 3 shows that settlement patterns among Nigeria’s major ethnolinguistic groups have persisted over time. Panel 1 shows the geolocations of the 252 survey respondents who belong to the Igbo ethnic group, 83 percent of whom reside within the Eastern Region and the spatial area occupied by their ethnolinguistic group. As Panel 2 shows, 95 percent of the Yoruba respondents reside within the Western Region and the Yoruba spatial area. Panel 3 shows that 99.5 percent of the Hausa/Fulani respondents reside in the Northern Region and the Hausa/Fulani spatial area. The relatively higher number of Igbos residing outside the Eastern Region and the Igbo spatial area mirrors their tendency to emigrate to other regions within Nigeria.

5. Data and methodology

This study relies primarily on the Round 7 Afrobarometer survey data for Nigeria which was collected in 2017 (BenYishay et al. 2017). ¹² It contains 1,600 observations and is representative of Nigeria’s population. Observations were drawn from each of Nigeria’s 36 states, plus the federal capital territory—i.e., Abuja. The data spanned 147 of Nigeria’s 774 local government areas (LGAs) (i.e., municipalities). ¹³ The dominance of Nigeria’s three major ethnic groups was reflected in the data, with the Hausa/Fulani (25.62%), Yoruba (23.19%), and Igbo (17.22%) together accounting for 66 percent of the total respondents. Respondents were at least 18 years old. The data contain information about the ethnicity of the respondents, which enables me to breakdown the data based on ethnic affiliation. Table A1 in the appendix reports the summary statistics of the variables used to estimate the regression models, while Tables A2 and A3 report the cross-correlations between all the variables for the Igbo subsample of respondents and the full sample, respectively. The variables used to estimate the regression models are discussed below.

5.1. Dependent variables

Support secession . This measures the degree to which respondents support IPOB’s secessionist goal. It was derived from the question, “The Indigenous People of Biafra or IPOB, should be given the right to secede from the federation?”, with response options on a scale with five ordinal categories ranging from “1 = strongly disagree” to “5 = strongly agree.” In the original Afrobarometer dataset, higher ordinal values denoted more disagreement with IPOB’s secessionist goal and vice versa. For easy interpretation of the regression results, I inverted the ordinal values assigned to the response categories by subtracting each of them from six, which allows higher values to denote greater support for secession and vice versa. I treated the “don’t know” and “refused to answer” responses as missing observations. I applied this rule to all variables derived from the Afrobarometer survey dataset.

Figure 4 plots the responses to the question regarding support for IPOB’s secessionist goal on a stacked bar chart. The y-axis shows the number of respondents who belong to the various major ethnic categories and the total number of respondents from the full sample who answered the relevant question. The x-axis shows the percentage of respondents who chose a particular response category. The figure shows that Igbos are more supportive of secession than the Yoruba, Hausa/Fulani, the minority ethnic groups combined, and the national average.

Figure 4. Ethnic distribution of support for secession.

5.2. Explanatory variables

Ethnic marginalization . This measures the extent to which respondents think members of their ethnic group are marginalized. It was derived from the question, “How often, if ever, are [Insert respondent’s Ethnic Group] treated unfairly by the government?”, with the responses measured on a scale with four ordinal categories ranging from “0 = never” to “3 = always.” A strength of this variable is its focus on marginalization at the group level rather than the individual level, which fits snugly with the horizontal inequalities theory. Its limitation is that it refers to marginalization in broad terms without focusing on any specific dimension of horizontal inequalities. As shown in Figure 5, Igbos have a higher level of perceived marginalization than the other ethnic categories and the national average.

Figure 5. Ethnic distribution of perceived marginalization.

Democracy. This measures the degree to which respondents think the political system in Nigeria is democratic. It was derived from the following question, “Now let us speak about the political system in this country. In your opinion how much of a democracy is Nigeria today?” The responses were measured on a scale with four ordinal categories ranging from “1 = not a democracy” to “4 = a full democracy.”

As shown in Figure 6, Igbos have the most negative perceptions toward Nigerian democracy compared to the other ethnic categories and the national average. While 25 percent of Igbos contend that Nigeria is not a democracy, the estimates for the other ethnic categories are all below 10 percent. The Hausa/Fulani have the most positive attitudes toward democracy in Nigeria: only 3.2 percent of them contend that Nigeria is not a democracy.

Figure 6. Ethnic distribution of perceptions toward democracy.

Socioeconomic condition . I measured socioeconomic condition at the household and communal levels. The household measure for socioeconomic condition is based on a deprivation index derived by summing the responses to the following four survey questions: “Over the past year, how often, if ever, have you or anyone in your family: (a) Gone without food to eat? (b) Gone without enough clean water for home use? (c) Gone without medicines or medical treatment? (d) Gone without fuel to cook your food?” The responses were measured on a scale with five ordinal categories ranging from “0 = never” to “4 = always.” I summed the ordinal values across the four items to create an indicator that ranges from 0 to 16. Higher values denote a higher level of deprivation and vice versa. The four items had a Cronbach Alpha statistic of 0.78, which shows internal reliability. I proxied socioeconomic condition at the communal level using the mean literacy rate (Bosco et al. 2017) in the LGA where the respondents reside. ¹⁴ Because this dataset is gridded, I computed the relevant statistic for the respective LGAs using QGIS software. The literacy rate is expressed in percentage and measures the number of men and women aged between 15 and 49 years in the LGA who were literate in 2013. Since the raw data is gendered, I computed the estimates for males and females separately and then took the average. Higher values indicate a better socioeconomic condition and vice versa.

Igbo. This is a dummy variable that takes the value of one if the respondent belongs to the Igbo ethnic group and zero otherwise. I developed a similar variable for the Yoruba, Hausa/Fulani, and the minority ethnic groups combined.

5.3. Control variables

Since most of the variables are measured at the individual level, I included control variables for the demographic attributes of the respondents—i.e., age and gender—in the regression models. Gender took the value of one if the respondent was male and zero if female. I also controlled for political instability, which I measured using the total number of violent conflict incidents that occurred in the LGA where the respondents reside between 1997 and 2016. Based on the ACLED dataset (Raleigh et al., 2010), I define violent conflicts as incidents that fall under any of the following three categories: Battles, Violence against civilians, and Explosions/Remote violence. Although the ACLED data is available starting from 1997 and is updated in real-time, I excluded incidents that occurred after 2016 while developing the variable. This lags the measure for political instability since the dependent variable is measured in 2017. The persistence of violent conflict in Nigeria signals the inability of the Nigerian Government to protect its citizenry, which in turn has precipitated the establishment of ethnic paramilitary organizations. For instance, IPOB established the Eastern Security Network (ESN) in 2020 to address rising insecurity in Eastern Nigeria—especially because of the violent clashes between nomadic Fulani herders and the Igbo resident communities (Opejobi 2022; Njoku 2021; Campbell 2021). ¹⁵ The Nigerian Government has voiced its opposition to the group; there have even been violent clashes between ESN and the Nigerian army (Gabriel 2023; Ugwu 2023; Chukindi 2023). ¹⁶ Given that ESN is a subset of IPOB, the government’s heavy-handed approach toward the group might attract sympathy for IPOB’s secessionist goal. Moreover, the Afrobarometer survey dataset shows that 30 percent of Nigerians agree that if the violence perpetrated by extremist groups cannot be resolved, Nigeria should be split into two countries.

5.4. Analytical technique

The general form of the model to be estimated could be expressed thus:

(1) $${Y_t} = {\beta _0} + \;{\beta _1}D{'_t} + {\beta _2}X{'_t} + \;{e_t}$$

Where ${Y_t}$ is the dependent variable, which measures support for secession, ${\beta _0}$ denotes the intercept, and $D{'_t}$ is a vector of explanatory variables measuring perceptions toward democracy, perceived ethnic marginalization, socioeconomic condition, and Igbo ethnicity. $X{'_t}$ is a vector of control variables measuring political instability and the demographic attributes of the respondents. ${\beta _1}$ and ${\beta _2}$ denote the coefficients of the explanatory and control variables, respectively, ${e_t}$ is the error term, and $t$ denotes the year in which the variables are measured. Since the dependent variable has five ordinal categories, I estimated the model using ordered logit regression, which is based on maximum likelihood estimation. To allow for the possibility of correlation between observations within the same LGA (i.e., district), I clustered the standard errors at the district level. ¹⁷ Because Hypotheses 2, 3, and 4 pertain specifically to members of the Igbo ethnic group, I will test these hypotheses using the subsample of respondents who belong to the Igbo ethnic group. I will test Hypothesis 1 using the representative sample for Nigeria’s population.

6. Results and discussion

6.1. Regression models using the Igbo subsample

Table 1 presents the regression results of models that were estimated using the subsample of Igbo respondents. I considered only ethnic marginalization in model 1. It carried a positive sign and was significant at the one percent level. This supports Hypothesis 2 which states that Igbos who feel that members of their ethnic group are treated unfairly by the Nigerian Government would be supportive of secession. In model 2, where I considered only democracy, it carried a negative sign and was significant at the one percent level. This suggests that Igbos who have positive attitudes toward Nigeria’s political system are less supportive of secession. Put differently, Igbos who think Nigeria’s political system is undemocratic are more supportive of secession. This is consistent with Hypothesis 3. As shown in model 3, the household measure for socioeconomic condition—i.e., deprivation index—had no statistically significant effect on support for secession. This supports Hypothesis 4c, which states that socioeconomic condition does not correlate with support for secession. However, as shown in model 4, the communal measure for socioeconomic condition—i.e., literacy rate—was significant at the one percent level and carried a positive sign, which supports Hypothesis 4b that socioeconomic condition positively correlates with support for secession. A plausible explanation for this finding is that Igbos feel that their association with Nigeria holds them back socioeconomically and they could do better if the Eastern Region secedes. A closer inspection of the literacy rate variable shows that Igbos outperform the other ethnic categories and the national average: Igbos had a mean literacy rate of 86 percent; the estimates for the Hausa/Fulani, Yoruba, and the national average were 35, 80, and 64 percent, respectively. The discrepancy in the findings between the household and communal measures for socioeconomic condition indicates that the effect of socioeconomic condition on support for secession depends upon the level of aggregation. Moreover, these two variables do not measure the same thing. In fact, the correlation between the deprivation index and literacy rate was 0.13. In model 5, where I considered all the explanatory variables simultaneously, the results were consistent with those in the baseline models. As shown in model 6, these results are also robust to the inclusion of control variables for political instability and the demographic attributes of the respondents. ¹⁸

Table 1. Correlates of support for secession among Igbos

Note: Clustered robust standard errors are in parentheses, † is the dependent variable which has five ordinal categories, ***p < 0.01, **p < 0.05, *p < 0.10. All models are estimated using ordered logit (Ologit) regression.

To illustrate the magnitude of the effects reported in Table 1, I plotted the predicted probabilities for the baseline models—i.e., models 1, 2, 3, and 4—in Figure 7. A cursory look at the four panels shows that the effect size is biggest for the “strongly agree” response category of the dependent variable. Panel A shows that a one-unit increase in the perceived ethnic marginalization increases the probability of an Igbo respondent “strongly agreeing” with secession by 13 percent and reduces the probability of him/her “strongly disagreeing” with secession by 5 percent. Panel B shows that a one-unit increase in perceived democracy reduces the probability of an Igbo respondent choosing the “strongly agree” response category by 13 percent when asked whether Eastern Nigeria should be allowed to secede. Consistent with the results in model 5, the average marginal effects of the deprivation index on the respective categories of the dependent variable were all statistically insignificant (i.e., Panel C). Panel D shows that a one percentage point increase in literacy rate in the municipality where an Igbo respondent resides increases the probability of him/her “strongly agreeing” with secession by one percent.

Figure 7. Average marginal effects of the explanatory variables on support for secession.

Note: Panels A, B, C, and D show the average marginal effects of perceived ethnic marginalization, democracy, deprivation index, and literacy rate, respectively, on the five ordinal categories of the dependent variable which measures support for secession. These results are based on the baseline regression models (i.e., models 1, 2, 3, and 4) reported in Table 1. Confidence intervals are at the 95 percent level.

6.2. Regression models using the full sample

To test Hypothesis 1 regarding whether Igbo ethnicity increases the likelihood of supporting secession, I estimated a bivariate regression model. To better understand the attitudes of members of the other major ethnic categories (i.e., the Hausa/Fulani, Yoruba, and Ethnic minorities) toward secession, I estimated additional bivariate regression models where I considered each of them. Table 2 reports the regression results. In model 1, Igbo ethnicity carried the expected positive sign and was significant at the one percent level. This suggests that compared to non-Igbos, belonging to the Igbo ethnic group increases the likelihood of supporting secession. This is likely due to the high level of grievances among Igbos compared to members of the other major ethnic categories. As shown in models 2 and 3, respectively, Yoruba and Hausa/Fulani ethnicities both reduced the likelihood of supporting secession. The opposition toward secession among the Hausa/Fulani and Yoruba might be because the secession of Eastern Nigeria, which houses most of Nigeria’s oil resources, would be a loss for them. This is especially plausible when one considers the fact that the Hausa/Fulani and Yoruba have been well represented in political power at the center since Nigeria’s transition to civilian rule in 1999. Moreover, given the centralized nature of government in Nigeria, controlling power at the center is tantamount to controlling the country’s vast oil wealth (Smith 2007, p. 192). It could also be that the Hausa/Fulani and Yoruba think secession is not the appropriate way for the Igbos to address their grievances.

Table 2. Correlates of support for secession among the major ethnic categories

Note: Clustered robust standard errors are in parentheses, † is the dependent variable, ***p < 0.01, **p < 0.05, *p < 0.10. All models are estimated using ordered logit (Ologit) regression.

Model 4 shows that belonging to a minority ethnic group had no statistically significant effect on support for secession. However, pooling the heterogenous minority ethnic groups into a single category might be problematic because they differ considerably in terms of their relations with the Igbos and whether or not they are autochthonous to the spatial area within Biafran territory. As Lewis (2022) points out in a recent study, large ethnic groups in Eastern Nigeria like the Igbos and Ijaws are supportive of secession, while smaller ethnic groups like the Efik and Ibibio that will be dominated should the Eastern Region secede, oppose secession. In model 5, where I included the three major ethnic groups in the same model and used the ethnic minorities as the reference category, the results were consistent with those reported in the baseline models.

Figure 8 shows the predicted probabilities for the baseline models reported in Table 2—i.e., models 1, 2, 3, and 4. Unlike the predicted probabilities reported in Figure 7, where the effect size of the four explanatory variables was largest on the “strongly agree” response category of the dependent variable, the effect of ethnicity on support for secession was most salient in the “strongly disagree” response category of the dependent variable. Panel A shows that compared to non-Igbos, belonging to the Igbo ethnic group reduces the probability of a respondent “strongly disagreeing” with secession by 43 percent and increases the probability of him/her “strongly agreeing” with secession by 23 percent. Panels B and C show that belonging to the Yoruba and Hausa/Fulani ethnic groups increases the probability of “strongly disagreeing” with secession by 15 and 17 percent, respectively.

Figure 8. Average marginal effects of ethnicity on support for secession. Note: Panels A, B, C, and D show the effect of Igbo, Yoruba, Hausa/Fulani, and minority ethnicities, respectively, on the five ordinal categories of the dependent variable which measures support for secession. These results are based on the baseline regression models (i.e., models 1, 2, 3, and 4) reported in Table 2. Confidence intervals are at the 95 percent level.

7. Conclusion

Relying on the horizontal inequalities theory, this study examined the effect of perceived ethnic marginalization, perceptions toward Nigerian democracy, and socioeconomic condition on support from secession among members of the Igbo ethnic group. It also investigated how belonging to the Igbo, Yoruba, and Hausa/Fulani ethnic groups, respectively, influenced support for secession. Perceived ethnic marginalization at the group level and negative attitudes toward Nigerian democracy were found to positively correlate with support for secession. These results may be explained by the political exclusion of the Igbos in postwar Nigeria, as some commentators have argued (Uroko, Obinna, and Inibong 2022; Ojoko 2022; Njoku et al. 2022; Akubo 2021). Socioeconomic condition was measured at the household and communal levels. The household measure had no effect on support for secession but the communal measure did: socioeconomic condition at the communal level was found to positively correlate with support for secession. A plausible explanation for this finding is that Igbos feel that their association with Nigeria holds them back and they could do better if they secede. Igbo ethnicity was also found to positively correlate with support for secession while Yoruba and Hausa/Fulani ethnicities reduced the likelihood of supporting secession.

It is unlikely that the Nigerian Government would allow Eastern Nigeria to secede. The Nigerian Government relies heavily on oil exports for its revenue, and the oil-rich Niger-Delta Region is located within the proposed Biafran territory. The Nigerian Government may also be hesitant to allow Igbo secession because it might set a precedent, prompting other ethnic groups to make similar demands. Even if IPOB achieves its goal of seceding from Nigeria, it would have to contend with the problem of gaining the trust of the minority ethnic groups in Eastern Nigeria, many of whom have distanced themselves from the movement (Godwin 2021; Wahab 2021). The persistence of IPOB despite the Nigerian Government’s heavy-handed approach toward the group highlights the necessity for the government to adopt a nonviolent approach in dealing with the group. Proscribing IPOB as a terrorist organization, which puts it at par with radical Islamist groups like Boko Haram and the Islamic State West Africa Province (ISWAP), is counterproductive. This cuts off the channel for dialog, which could be a precursor to the peaceful resolution of the conflict.