Why do people in oil-rich autocracies, like Azerbaijan, rarely protest? They have many grievances, yet civil disobedience – the peaceful and purposeful breaking of laws – is uncommon. I argue that a strategic silence happens when three factors come together: repressive capacity, opportunity costs, and coordination thresholds. These factors make civil disobedience, whether by one person or many, likely to have a negative outcome.
Each potential protester considers: (1) the moral and political value of successful protest; (2) the chance of success based on others joining; (3) the cost of repression per person, which goes down as protest size increases but rises with the regime’s strength; and (4) the opportunity cost linked to relying on the state. I show that in Azerbaijan these parameters align to produce a single, stable equilibrium at which no one finds protest rational. This model sees solitary acts of civil disobedience as the same as mass mobilization. They differ only in scale and face the same strategic challenges.
Using this model to analyze Azerbaijan’s protest timeline from 2000 to 2024 shows its effectiveness in explaining the situation. Declining turnout, preemptive arrests of organizers, and constant surveillance, along with economic sanctions against public-sector workers, create a “low-𝑥 (low participation rate), high-𝔞(high repressive capacity)” balance. In this state, the expected utility of civil disobedience is lower than zero (UR𝒾 < 0) for almost all potential protesters. This equilibrium not only accounts for the absence of mass rebellion but also the near-total absence of lone acts of conscience.
Azerbaijan as a Rentier State
Rentier state theory explains how states that get much of their money from natural resources, mainly oil and gas, form unique political and economic systems. Such states often see less political liberalization. This happens because they can fund government spending without needing to tax their citizens. This gap between state revenue and what citizens contribute hurts democratic accountability. It also limits political participation. Resource wealth lets regimes invest in military and security forces. This helps them suppress dissent and keep power. The result is a paradox where economic prosperity, instead of fostering democracy, strengthens authoritarian rule (Ross, 2012).
Azerbaijan is a classic example of a rentier state. The country’s oil and gas profits drive the economy and influence political systems. Ibadoghlu (2019) states that hydrocarbon resources make up over 80% of the country’s exports. They also provide nearly 60% of government revenues. This means the national budget relies heavily on changes in global oil markets. The government’s reliance on resource rents has helped it maintain public spending. This also reduces dependence on direct taxes, which weakens democratic accountability.
Managing oil income depends much on the State Oil Fund of Azerbaijan (SOFAZ). SOFAZ was established to ensure fairness and economic stability for coming generations. It gathers a large portion of Azerbaijan’s oil wealth. In practice, the government often uses SOFAZ transfers to fill budget gaps. The government also funds big infrastructure projects this way. This focus on short-term needs means less investment in long-term economic diversification (Ibadoghlu, 2019). Budgeting mostly focuses on energy-sector earnings. Non-oil revenues have a small impact on overall fiscal planning. This way of sharing revenue corroborates that the rentier state model fits Azerbaijan. In this model, the ruling elite controls national wealth and limits political competition. Ibadoghlu (2019) argues that centralizing oil-revenue management leads to rent-seeking and corruption. This, in turn, strengthens authoritarian rule. The state’s reliance on oil wealth has slowed economic reforms. High government spending on subsidies and public sector jobs is used for political gain, not for efficiency.
The consequences of this rentier economy are evident in Azerbaijan’s economic volatility. When oil prices fall, like in 2015 and 2020, the country’s fiscal stability suffers (Ibadoghlu 2019). This can cause currency devaluations, inflation, and cuts to social spending. Even with some reform attempts, the oil sector still blocks growth in non-hydrocarbon industries. So, Azerbaijan’s economy relies heavily on oil revenue. This dependence strengthens authoritarian rule and restricts democratic accountability.
The Rationalist Model of Protest Participation
I use a rationalist approach based on public choice theory to explain protest behavior in authoritarian settings like Azerbaijan. This approach sees individuals—like voters, bureaucrats, or protesters—as rational agents. They act in their self-interest while facing institutional limits. In this view, political participation—like civil disobedience and mass protest—is seen as a smart choice. It’s not about identity or feelings. Instead, people weigh the costs and benefits before acting.
Protest in authoritarian regimes shows a clear example of a collective action problem. Successful rebellions offer benefits like regime change, more freedoms, and better governance. These rewards are public goods. Once achieved, the benefits include everyone. However, individuals alone bear the costs of participation, such as arrest, violence, social ostracism, or economic loss. Rational agents tend to free-ride. They hope others will take the risk while they enjoy the benefits if the protest succeeds (Olson, 1965; Wright, 2022). People who want change might still avoid protest. This is because they fear personal loss and doubt their impact on the outcome. This leads to political inaction. It is not that people support the regime; rather, they see that rebelling is not worth the cost. This happens especially in authoritarian situations. People make political choices by weighing expected costs and benefits.
Formally, the utility an individual 𝒾 expects to receive from participating in a protest, UR𝒾, is defined as:
This equation captures three critical components of the protest decision:
- Probability-weighted benefits: 𝓅(𝑥)𝒱𝒾represents the expected benefit from participating in protest.𝒱𝒾 is the individual’s valuation of a successful protest outcome. It can include freedom, democracy, accountability, or improved economic conditions. 𝒫(𝑥)is the probability that the protest will succeed, which is assumed to increase in the share 𝓍 of the population expected to join. The more people participate, the higher the likelihood that the protest will bring about meaningful change (Kuran, 1991; Chwe, 2001).
- Repression cost:
represents the expected cost of repression. The term 
reflects the fact that the individual risk of being punished decreases as more people participate. When protests are large, the state cannot effectively punish every participant. This means the risk for each person drops. The constant 𝑐 captures the severity of punishment (e.g., prison time, torture, fines). 𝑎 is a multiplier reflecting the regime’s overall capacity and willingness to repress. In highly authoritarian systems like Azerbaijan, 𝑎 is large (Wintrobe, 1998; Edmond, 2013). - Opportunity cost: 𝑂𝒾 refers to the individual’s opportunity cost of participating in protest. It can include lost wages, family risks, loss of status, or added stress. Including this lets the model consider two things. First, it looks at how much the economy depends on the regime (Karl, 1997; Ross, 2012). Second, it accounts for differences in what drives people to protest (de Mesquita and Shadmehr, 2023).
The model’s decision rule is simple: a person will protest if UR𝒾 > 0 but they will stay passive otherwise. In equilibrium, the level of protest participation, 𝑥, relies on how many people feel their personal utility is positive. This depends on their expectations about what others will do and how strong they think the regime is. This model highlights a key idea from public choice theory for authoritarian regimes. It shows that mass political inaction isn’t always about regime legitimacy or ideological agreement. Instead, it’s often a smart reaction to the system of incentives. Authoritarian regimes, like Azerbaijan, keep participation low and stable. They do this by raising 𝑎, and distributing selective benefits that increase 𝑂𝒾, and preventing the emergence of protest coordination (thus keeping 𝑥 low).
The formal model above shows the main reasons people join protests. But it misses a key issue: the coordination problem that affects mass disobedience. Protest spreads through common knowledge. People need to want change. They also must know others want change and know that others know it too. In tightly controlled places like Azerbaijan, civil society is weak. Independent media faces limits, and public squares are watched. As a result, this common knowledge is kept hidden. The coordination problem also explains why regimes strategically target early or prominent dissenters. The state quickly punishes potential leaders. This stops critical mass from forming and delays or prevents cascade effects. Even if people are very unhappy and deal with some personal costs, protesting seems pointless if they think others won’t join in. That is why there is a big gap between real grievances and visible mobilization in authoritarian settings.
The Low-Participation Equilibrium
Repression, opportunity cost, and coordination failure create a low-participation equilibrium. In this state, most people rationally decide not to protest. As a result, the system stays stable, even with deep discontent beneath the surface. Most people find that the expected utility of protest (UR𝒾) is lower than zero.
At the center of this equilibrium is the self-reinforcing nature of expectations. When the anticipated level of participation 𝑥 is low, the expected success, probability 𝓅(𝑥) is close to zero. Thus the individual risk of repression 
becomes extremely high. In this situation, even those with strong dissatisfaction and high values (𝒱𝒾)will likely stay passive. They will only go out if they believe many others will protest too. Authoritarian regimes aim to maintain this condition. They suppress coordination, which keeps 𝑥 low. This leads to UR𝒾 < 0 for almost everyone, ensuring that actual protests stay minimal.
The state actively helps keep this balance by adjusting each part of the protest calculus. It keeps a strong and clear repression capacity (𝑎), so the cost of protest stays high even as more people join in. Regimes like Azerbaijan target key activists and arrest dissenters quickly. They also use selective punishment. This shows that repression is real and costly. As a result, it boosts the deterrent effect on the population.
Second, authoritarian regimes often inflate individual opportunity costs (𝑂𝒾). They do this by linking jobs and basic security to loyalty to the regime. In rentier states like Azerbaijan, many people depend on government jobs and support. If someone voices dissent, they may face serious consequences. This can mean losing access to important services, getting fired, or being expelled from school. The more dependent individuals are on the state, the more they stand to lose by participating in protest. This creates a climate of strategic silence. This is especially true for the middle class and state-employed workers.
Third, the government invests considerable effort in keeping 𝑥—the perceived likelihood of others protesting—as low as possible. This happens through censorship and media control. It also includes the targeted suppression of protests. This can involve student demonstrations, environmental activism, or independent journalism. The regime punishes these acts quickly and openly. This deters others from joining future actions and stops a critical mass from forming.
From a public choice perspective, this is a stable and rational outcome. Each person acts based on their own cost-benefit analysis. The regime sets up the environment so that the costs of protest are higher than the expected benefits. The result is not collective apathy, but a rational equilibrium of mutual inaction. The cost of joining a rebellion is often too high. Plus, the potential benefits are unclear. So, many individuals hesitate to take the risk (Wright 2022).
In this context, protest only becomes viable when some exogenous factor alters one or more of the variables in the utility function. A sudden economic shock might reduce the opportunity cost (𝑂𝒾). A powerful political event might raise the perceived probability of success (𝓅(𝑥)). Also, international pressure might temporarily reduce the repression coefficient (𝑎). Unless things change, the authoritarian regime keeps a low-participation state. It looks stable, but dissatisfaction is brewing underneath.[1]
A clear sign of Azerbaijan’s recent political path is the big drop in protest participation over time. In the early 2000s, especially in 2003, many people protested in Baku. They took to the streets after the presidential elections were contested. Tens of thousands joined the mass mobilizations. By the mid-2010s, national protests mostly turned into local gatherings. These events focused on specific issues and often had only a few dozen people involved. Protests are still happening, however, participation is more fragmented now. Protests are spread out over larger areas and tend to be smaller in size. The average protest size in 2021 was just 29 people, with a median of 10 participants per event (Kamilsoy, 2021).
The key moment seems to be the aftermath of the Arab Spring in 2011. During this time, protests inspired by regional movements did not attract many people in Azerbaijan. The government acted quickly against organizers. They also set strict limits on gatherings. This created a chilling effect. The typical Azerbaijani protest in 2021 involved 10 to 50 participants and was quickly dispersed. A low participation rate makes repression more focused and effective (Kamilsoy, 2021).
In the rationalist model, the drop in people willing to protest leads to two main outcomes. First, when only a few people join, individuals think their protest will not matter. Second, isolated protesters are cheaper and easier to punish than large, coordinated crowds. In Azerbaijan, two things are clear. Individual protesters face state violence. Also, the belief in the power of protests has faded over time. These factors create a low-participation situation.
Azerbaijan’s authoritarian government has built a strong system of control. This system discourages large protests by punishing some people selectively. Azerbaijan’s repression is now more targeted, preemptive, and strategic.
This strategy was clear after the failed mass mobilization attempt in 2011. On March 11 of that year, opposition activists tried to organize a Facebook-led “March 11th-Great People’s Day” protest in Baku. Seventeen organizers were arrested just days before the protest. The demonstration was violently broken up, and many more were detained at the scene (Geybullayeva, 2011). This preemptive action signaled the regime’s ability to detect and suppress dissent at the planning stage. Such early suppression functions as both a deterrent and a public demonstration of the futility of protest.
The trend intensified in subsequent years. In June 2023, villagers in Soyudlu protested against toxic waste. They faced harsh responses. Tear gas and rubber bullets were used to disperse them. At least eight people were jailed. Journalists at the event faced assaults or were removed (Human Rights Watch, 2024). The crackdown didn’t just stop at the event. Social media commentators who spoke out against the repression were also detained days later. Even small protests that focus on specific issues face harsh punishment. This is done to discourage others from trying the same.
In the rationalist framework, 𝑎 operates as a credible and visible signal. The regime targets early actors and sets a high cost per person, even if 𝑥 is low. This strategy discourages others from joining the protest at all. Organizers can learn from past events. Arrests aren’t random; they follow a pattern of strategic incapacitation. Selective repression keeps the cost of joining a protest high. The record from Azerbaijan supports this mechanism. Even without big protests, high 𝑎 keeps things steady. It makes people think that even small dissent will face harsh punishment.
Preemptive censorship and the arrest of visible dissenters create a tense atmosphere. In this environment, few people expect others to join them. This leads to a situation where 𝑥 is less than the critical mass threshold. As a result, the probability 𝒫 (𝑥) is close to zero. For successful coordination, it’s important to have both private support and shared knowledge of that support (Chwe 2001). In Azerbaijan, repeated early suppression prevents common knowledge from forming. Citizens notice that protests do not grow, so they choose to stay inactive.
Opportunity Costs and Dependency on the State
In Azerbaijan, relying on the state increases the opportunity cost (𝑂𝒾) of protesting for individuals. This is especially true for those involved in public-sector networks. Azerbaijan is a rentier state. It employs many workers. State ministries, SOFAZ-funded projects, and state-linked companies lead this effort. Public-sector workers, such as teachers, healthcare staff, and city employees, may lose their jobs or pay. They may also lose housing and utility subsidies if they join unsanctioned protests.
The early 2016 protests in Azerbaijan were sparked by currency devaluation, rising food prices, and job cuts. These events show how the government’s economic control increases the cost of speaking out. The protests happened in at least ten areas, including Guba, Lankaran and Siyazan. They were mostly spontaneous and aimed at urgent socioeconomic issues (Institute for Reporters’ Freedom and Safety, 2016). These protests are important for showing the explanatory power of the rationalist model because of who joined. Many participants were small business owners, laid-off state workers, or citizens relying on public welfare. In Siyazan and Guba, residents asked for subsidized food. They also wanted unpaid wages repaid and an end to selective arrests of protest organizers. Authorities responded not just with police force but with punitive economic pressure. In many cases, protesters lost food rations. Social payments were delayed. Also, some faced informal retaliation, like being denied jobs or housing benefits in their communities.
In contrast, people outside the state system—such as unemployed youth, informal vendors, or rural smallholders—face a lower 𝑂𝒾, but remain highly exposed to repression (𝑎). Even for those with minimal state ties, the threat of arrest, fines, or blacklisting discourages protest. The regime takes advantage of opportunity-cost differences. It uses harsh economic sanctions on those who have the most to lose. At the same time, it allows low-stakes, depoliticized protests in less important areas. This careful use of incentives explains why privileged groups rarely rebel, even when they have political and economic complaints.
Conclusion
The patterns we see in Azerbaijan match the predictions of the rationalist utility model. First, protest participation (𝑥) has been low since 2003. Second, the state’s repressive capacity (𝑎) remains high. Third, the chance of success (𝓅(𝑥)) seems almost zero. Fourth, opportunity costs (𝑂𝒾) rise for state-dependent workers. Citizens think protest is too risky, so they choose to stay silent. This leads to a stable situation where few speak out.
Individuals consider four main factors before taking action: (1) the personal value of successful resistance; (2) the probability of success, influenced by others joining; (3) the expected cost of repression, which only slightly decreases as the group gets larger; and (4) the opportunity cost from relying on the state for economic support.
The model shows why lone dissidents and would-be organizers in oil-rich autocracies, like Azerbaijan, often choose to do nothing. It does this by combining key elements into one utility function. High repressive capacity comes from resource rents. This leads to preemptive arrests, digital surveillance, and selective punishment. This makes the cost of dissent seem too high to bear. At the same time, low and declining protest turnout depresses the expected probability of success to near zero. Relying on public-sector jobs increases the opportunity cost for people. This makes them less likely to lead or join civil disobedience. These factors create a steady state of “low participation, high cost.” In this situation, the expected benefit of civil disobedience is negative.
Civil disobedience in rentier authoritarian settings is not just about moral beliefs or organizational skills. Instead, it is a strategic reaction to the regime’s disincentive structure. The rarity of both individual and group acts of conscientious lawbreaking shows a balance of silence. It does not mean there is widespread apathy or acceptance of the status quo. To break this deadlock, we need big changes in the model’s parameters. This could mean a credible drop in repressive capacity, a clear rise in coordination signals, or a major cut in opportunity costs for dissenters.
Notes and References
Chwe, M. S.-Y. (2001). Rational ritual: Culture, coordination, and common knowledge. Princeton University Press.
De Mesquita, E. B., & Shadmehr, M. (2023). Rebel motivations and repression. American Political Science Review, 117(2), 734–750.
Edmond, C. (2013). Information manipulation, coordination, and regime change. Review of Economic Studies, 80(4), 1422–1458.
Geybullayeva, A. (2011). From Facebook to the streets of Baku. OBC Transeuropa. https://www.balcanicaucaso.org/eng/Areas/Azerbaijan/From-Facebook-to-the-streets-of-Baku-93251
Human Rights Watch. (2024, January 11). World report 2024: Rights trends in Azerbaijan. https://www.hrw.org/world-report/2024/country-chapters/azerbaijan
Institute for Reporters’ Freedom and Safety. (2016, January 15). Azerbaijan braces for protests. https://www.irfs.org/news-feed/azerbaijan-braces-for-protests/
Ibadoghlu, G. (2019). Azerbaijan: A rentier state. SSRN.
Kamilsoy, N. (2021). Protest dynamics in Azerbaijan and their relation to authoritarian regime stability. Baku Research Institute. https://bakuresearchinstitute.org/en/protest-dynamics-in-azerbaijan-and-their-relation-to-authoritarian-regime-stability/
Karl, T. L. (1997). The paradox of plenty: Oil booms and petro-states (Vol. 26). University of California Press.
Kuran, T. (1991). Now out of never: The element of surprise in the East European revolution of 1989. World Politics, 44(1), 7–48.
Morris, S., & Shadmehr, M. (2023). Inspiring regime change. Journal of the European Economic Association, 21(6), 2635–2681.
Olson, M. (1965). The logic of collective action: Public goods and the theory of groups. Harvard University Press.
Ross, M. L. (2012). The oil curse: How petroleum wealth shapes the development of nations. Princeton University Press.
Wintrobe, R. (2000). The political economy of dictatorship. Cambridge University Press.
Wright, R. E. (2022, March 29). Why not rebel? The Daily Economy. American Institute for Economic Research. https://thedailyeconomy.org/article/why-not-rebel/
[1] The rationalist utility model provides a general explanation for the absence of mass protest. However, it is important to acknowledge that not all individuals face the same protest calculus. Individuals with high 𝒱𝒾—like unemployed youth, marginalized ethnic groups, or political dissidents—often see more personal gain from changing the regime. In contrast, people with secure jobs, social connections, or ties to the government often see less value in protest success (Morris & Shadmehr, 2023).
