Rent Distribution Modes in Azerbaijan and Belarus: Implications for the Opposition

Abstract The study uses the theoretical framework of rentier states to compare the political economies of Belarus and Azerbaijan and the differing conditions created for the political opposition. It conducts a comparative analysis of the structural and agency factors affecting these conditions, focusing on the sources, control and the mode of distribution of rents in the two states. While both states combine post-Soviet legacies with rentier economies, they have different sources of rents. The study shows a ‘converging divergence’ of these two rentier states, based on the analysis of their type (pure rentier or semi-rentier state) and demonstrates the role of power transition in determining the distribution of rents. The combination of these variables, in turn, shapes different types of regimes—exclusive patronage in Azerbaijan and inclusive distributive rent-seeking in Belarus—and creates different tasks for the political opposition in terms of ideology and forming Western alliances.

The paths of the two states started to diverge after the collapse of the USSR. Azerbaijan fought Armenia over the enclave of Nagorno-Karabakh (1988-1994 and experienced the parallel partial evolution of economic institutions with the curtailment of political pluralism after an initial liberalisation. After a year of liberal government and reforms of the Popular Front (Azerbaijan Khalq Jebhesi), it started to move towards autocracy as the communist-era leader, Heydar Aliyev, assumed the presidency in 1993. Azerbaijan's ability to capitalise on exports of crude oil and the extraction of rents while avoiding large-scale privatisation allowed President Aliyev to strengthen his grip on power by distributing the wealth to his cronies. This 'crony contract' survived the transfer of power to his son Ilham in 2003.
Unlike Azerbaijan, where autocracy went hand-in-hand with some economic liberalisation, Belarus experienced a simultaneous backsliding, with the reversal of both political liberalisation and structural economic reforms in 1995, one year after Alyaksandr Lukashenka's rise to power. Winning the office with populist slogans and an explicit antinomenklatura agenda, and enjoying high social support, Lukashenka replaced economic restructuring with the rents from Russia. The rents acted as an economic anaesthesia, allowing the economy to stay unreformed through investments into the old social welfare. Ways were found to extract significant rents from Russia: gas and oil subsidies, access to the Russian market and the ability to abuse the terms of the integration agreements made with Russia in 1996, leading to the launch of the Union of Belarus and Russia the following year.
Overall, both countries represent what Freedom House calls 'not-free' regimes and could be labelled as fully-fledged political autocracies (see Table 1). 1 Both countries, with similar-sized populations, are middle-income economies: Belarus's GDP is US$59,015 billion and that of Azerbaijan is US$42,603 billion. 2 Furthermore, their growth levels have been similar, as both economies largely depend on international oil prices. Whereas Azerbaijan exports crude oil (86% of total exports), with petrochemicals playing a secondary role, 3 Belarus's oil and petrochemicals industry exists thanks to Russia. The Soviet-era Sumgait petrochemical complex, along with the refineries in Baku, lost their significance after the collapse of the Soviet Union when the market demand became for crude rather than refined oil. Two oil refineries built in Belarus during the late 1970s, Novopolotsk and Mozyr, produce four times the annual domestic demand. Given that Russia has been providing crude oil to Belarus at domestic prices whilst the world market price for oil has been significantly higher, that made Belarus's oil exports highly marginal. The oil sector thus used to be the main producer of hard currency and the biggest locomotive of the national economy.   While the economies of Azerbaijan and Belarus depend largely on oil revenues, they differ in many other indicators. First and foremost, Belarus is still a socialist-mixed economy in which more than 70% of the GDP is produced in the state sector. Lukashenka maintained state ownership of most Soviet-era enterprises and avoided laying off labour, insider privatisation and other negative effects associated with the transfer of property and structural reforms in the immediate post-communist period. The unprofitable segments of the Belarusian economy were, thus, supported through a complex system of crosssubsidies and soft budget constraints, in which the state sector was the recipient of aid and the oil sector often served as a donor. The situation is different in Azerbaijan, where the oil industry is divided between foreign oil companies, which generate most of the country's GDP, and the state sector is dominated by the president and his cronies.
Throughout the post-Soviet years, Belarus's spending on education and health greatly exceeded that of Azerbaijan. 4 Yet, this statement does not fully capture the difference in social welfare systems. The average salary of US$510 is the key to understanding the Belarusian social contract. 5 Since a large number of Belarusian citizens are employed in the state sector, and many state-owned enterprises are loss-making, the average salary should be regarded as a form of social welfare. In other words, the level of employment was artificially maintained, despite market pressures, in the name of social stability. The main function of the Belarusian state is extracting rent from Russia, using it to ensure low employment and a relatively high income.
In spite of its abundance of oil, the average wage in Azerbaijan-US$293 as of February 2017-is one of the lowest in the former Soviet Union (Israfilbayova 2017). However, officially low salaries are partly compensated by the significant share of informal income. In the state sector, these are 'envelope' payments, which occur when employees receive informal income from their employer according to the logic of political patronage, while in other sectors it is a special type of social contract, where government 'while not providing the population with the welfare, leaves open semi-formal avenues to achieving its goals at one own's risks … ' (Sayfutdinova 2015).
Finally, both countries have a high ratio of rents to GDP, a phenomenon we will approach using the rentier-state theory. Since he came to power in 1994, President Alyaksandr Lukashenka's political strategy has centred on creating revenue from external economic rents and using them to underwrite a social contract with the population, namely, the provision of sustained social welfare in exchange for public loyalty. The external rents, mostly from Russia, come in the form of oil and gas subsidies, but also in the form of special conditions in regional customs agreements and support in deterring international pressure for implementing reforms. Russian military installations and a joint air defence, as reflected in the Agreement of 3 February 2009, which created a Unified Regional Air Defence System between the Russian Federation and the Republic of Belarus (Rogozinska & Olech 2020). Overall, Belarus has been able to exploit the asymmetries in its relationship with Russia: promises of integration, loyal foreign policy-making, the taming of the anti-Russian democratic opposition in Belarus led to significant economic yield, paid for by the Russian government.
In this article we aim to contribute to the theoretical debate on the nature of different types of rentier states, while bringing forward some practical conclusions regarding the implications of these different types for the political opposition. By comparing the two types of rentier states we will show how the type of political economy determines the conditions under which the opposition must function and, thereby, the specific obstacles to democratic development. We will start with the analysis of rentier states and their types, then discuss our two case studies in depth, focusing on their particularities, and conclude with the implications of our findings for the functioning of the opposition. We argue that, due to different sources of rents, combined with the mode of power transition, the political opposition in each country is deprived of the assets that usually constitute the operational strength of parties in such regimes: Western support and an attractive welfare agenda.

Theoretical framework of rentier states
The theory of rents was developed in several stages. First, there was the influential piece by Mahdavi (1970), which set out a new concept of the rentier state: a state that receives substantial rents from foreign individuals, companies or governments, thus freeing it from dependence on the surplus production of the population (Karl 2005).
The second explanatory step was the popular petrostate variation of the theory, which introduced into the political realm the 'Dutch disease' economic effect present in oil-exporting countries, leading to Michael Ross's 'oil hinders democracy' assumption (Ross 2001). The debate on the political economy mechanisms of rents juxtaposes structural factors with the agency of policy action, as well as of the range of its effects over state institutions, economy or democracy. Beblawi and Luciani (1987) suggested that the commodity structure of the economy has a determining impact on both political and economic developments. Autocratic regimes and distorted economies with an uneven distribution of property rights are the usual outcomes in resource-abundant economies that generate rents from raw commodity exports, usually oil. Rents are believed to hinder democracy through the following mechanisms: first, an uneven concentration of power, namely, those who control rent extraction and distribution can afford costly coercive and co-optation tools (Fish 2005); second, economic statism (Luong & Weinthal 2001), weak institutionalisation and corruption; and third, lack of accountability, that is, the state can spend without taxation and thus, any need to justify spending to the public (Beblawi & Luciani 1987).
The third theoretical development was the assumption that the effects of the oil revenues may not be unique, but similar to other externally generated revenues, such as foreign aid (Moore 1998;Bräutigam 2000). Finally, Acemoglu and Robinson (2006) held that non-taxed state revenues (including foreign aid, borrowing from abroad, profits of state-owned companies) can act as stabilisers for the regime; since the state can spend without taxing its citizens, it can pursue its own interests without constraint and secure regime survival by redistributing these rents. Further developing the theory, specifically as regards the connection between natural resources and democracy, Michael Ross (2001) identifies three effects of the oil-rich state: the rentier effect, the repression effect and the modernisation effect. The rentier effect is expressed in lower, or total absence of, taxation to neutralise the public's possible demand for democracy: accountability and transparency. Consequently, there is less political participation, leading in turn to the unlimited abuse of power by the ruling elite, in terms of misappropriation of state funds and revenues. Ross further describes the causal mechanisms of the rentier effect, among others-the 'spending effect'. Under the 'spending effect', the government spends a lot of funds on popular social welfare programmes, subsidising certain sectors (energy and agriculture) and opening employment opportunities in the energy sector to legitimise its regime. Rentier states also have the capacity to invest in their security institutions to suppress possible protests and opposition, hence the 'repressive effect'.
The theory also suggests that there are two subcategories of rentier states-first-grade or pure rentier states, and second-grade rentier states or semi-rentier states, which do not possess large resources (Beblawi & Luciani 1987). According to Smith (2004), pure rentier states are much more likely to become autocratic than semi-rentier states. This suggests that, structurally, Azerbaijan faces greater obstacles to establishing and preserving democracy and democratic institutions than Belarus, as it suffers additionally from the 'paradox of plenty', which surfaces when resource abundance leads to slow economic growth and authoritarianism.
The rentier-state literature was recently extended by the introduction of the concept of 'post-Soviet rentier states' (Franke et al. 2009). The authors refer to the 'double curse': Soviet legacies and a rentier economy. While patrimonialism is considered a common post-Soviet legacy, in some countries, such as Azerbaijan and Kazakhstan, the authors also regard the role of regionalised groupings of elites as a key pre-Soviet legacy.
While all these characteristics are relevant to Azerbaijan and Belarus, we highlight three main structural aspects. The first is the source of rents (whether internal or external) that shapes the overall incumbent's survival strategies (coalition-building tactics, internal and external policies). The second is the ownership of rents: whether the state (the incumbent) or the private groups are the key owners of rents. The third is the mode of distribution of rents: the proportion of rents utilised for self-enrichment relative to that spent to buy loyalty through social transfers.
Our framework borrows Mahdavi's original definition of a rentier state and treats rents as non-taxed revenues as far as the budget is concerned, that is, not as the surplus of the productive sector but rather unearned income obtained via various mechanisms. Moreover, we are particularly interested in both the oil rents and the rents that the former resource-poor post-Soviet republics (both Belarus and Ukraine fall under this definition) obtained from external sources, mainly Russia.
Since the economic disintegration of the Comecon, Russia has continued to act as a sponsor for various post-Soviet republics, issuing currency for their non-reformed economies and thus bearing the inflationary burden for them, supplying them with negative interest credit (usually paid to the borrowers, rather than to the lenders), fuelling them with subsidies and, most importantly, providing energy for significantly below market price. Certain post-Soviet countries were able to establish beneficial links with Russia at Russia's expense. Consequently, interstate rent-seeking enabled various state and non-state actors in the post-Soviet republics to convert external public losses into internal private benefits. Herein, Russian leverage and linkage could change donor-recipient relations. The origins of those rents were multiple: the windfall profits contained in the interstate trade and customs agreements; administratively stimulated demand for exports; foreign aid; price imbalances; smuggling; and barter agreements.

Sources of rents: external versus internal
Overall, the source of rents represents the first variable to explain the divergent paths of the two countries. The presence of the internal rents in Azerbaijan makes it a classic first-grade rentier state confirming all its three effects as described by Michael Ross (2001): rentier, repression and modernisation effects. From the signing, in 1994, of the oil contract to develop the off-shore oil field Azeri-Chirag-Guneshli in the Azerbaijani sector of the Caspian Sea, 6 and especially after the rise of world prices in 2000 and the construction, in 2006, of the pipeline bypassing Russia Baku-Tbilisi-Ceyhan, 7 the share of oil rents over total GDP grew steadily, reaching a 50% average growth in the period 2000-2013 (Gurbanov et al. 2017). This rise surfaced notwithstanding the creation of the oil fund SOFAZ, which was primarily designed to preserve the wealth derived from oil production. In 2011 the share of oil and oil products constituted 92% of the country's exports. This high dependence on oil rents led to the phenomenon of 'no taxation, no representation'. In short, tax revenues constituted no more than 14% of GDP in a few consecutive years of the oil boom-2010-2014, 8 compared to the share of oil revenues in the GDP reaching 61.73% in 2011, 9 which meant that the leadership did not depend on the public for its survival but had an independent source of income. With the exclusive role of the state in the distribution of rents, as well as due to the vertical patron-client logic of rentierism, the neopatrimonial authoritarian type of regime provides a convenient framework for such a system (Gyene 2016).
The structural effects of the oil curse have been exaggerated by the Soviet legacies, where the institutions of checks and balances are weak or non-existent, being characteristic of many post-colonial states (Ishiyama 2008). In 2006, the rise of 'big oil', stimulated by production increases in the Azeri-Chirag-Guneshli oil field, coincided with the parliament's complete transformation into a rubber stamp body, while the judicial branch became fully subservient to the executive as the result of policies of repression and co-optation. By 2018 there was no opposition party represented in the national parliament. The 'spending effect' as part of the rentier effect expressed itself in the large amount of resources directed to subsidies-less to health and welfare programmes, as the analysis below reveals, with most going to big infrastructure projects, the industrial (mining) sectors, the repressive apparatus and the patronage system.
As an indicator of patronage, the number of employees in the public sector in Azerbaijan increased significantly, especially in comparison to neighbouring Georgia and Armenia (Guliyev 2013). The salaries of the repressive agencies were the only ones to be raised regularly throughout the oil boom, which confirms the validity of 'repressive effect' (Franke et al. 2009;Almaz 2015). With regard to the 'modernisation effect', the other indicator of the rentier economy, the statistics of 2008 show that employment in unskilled occupations has been much higher in Azerbaijan compared to neighbouring republics, suggesting a lack of qualified labour in specialised occupations (Shaw 2013). Unlike Azerbaijan, Belarus had to seek rents by adjusting its foreign and internal policies to maximise the inflows of rents from Russia. The exchange of rents for concessions was not new: the Soviet Union had distributed much of its petrochemicals towards its Eastern European satellites to secure their loyalty and maintain economic dependency. The same approach was used by the Belarusian elites. In this regard, using a brotherly pro-Russian rhetoric, engaging in the integration process with Russia and cracking down on anti-Russian nationalist elites often served Lukashenka as an efficient instrument for securing rents from Russia. To be more specific, there were three main sources of rents for Belarus. The first were oil-based rents that originated from the price difference between the price of Russian crude and the world market prices for refined oil.
Belarus was equipped with modern oil refineries. Thus, the Mazyr refinery was transformed into a shareholding company in 1993 and the Russian oil companies had important interests there. In 1994 the Slavneft Joint Company was established as a Russian-Belarusian joint venture. The major stakes were held by the Slavneft Oil Company (based in Russia), with 42% of ownership; the Belarusian government, 42%; workers and managers, 8%; and private investors, 8%. 10 The Naftan refinery in Navapolatsk became one of Belarus's most attractive assets: it was closer to Western markets than Russian refineries; it was the meeting point of two pipelines, the Surgut-Polatsk and the Samara-Ventspils; and the Samara-Polatsk oil pipeline also passed through it. Naftan's profits boomed during the Chechen War in 1997, its output helping to compensate for the loss of the Grozny oil refinery. Of all the sources of income related to Russian energy, none proved more profitable to Belarus than the export of refined oil products to the Western markets, especially between 2002 and 2006 (Balmaceda 2004). Belarus imported crude oil from Russia at low prices and free or almost free of any Russian export duties, refined it, and subsequently sold it to Europe. As a result, Belarus profited from the difference between the special low price paid to import the crude oil and the much higher market price it charged for the refined products, and it received related export duties as well. Thus, in the mid-2000s, the export duties on oil products provided about 10% of the total revenue for the Belarusian budget (Manenok 2009). According to Pikulik's (2010) calculations, oil-based rents comprised 11% of Belarus's GDP in 2006 and 8% in 2008.
There were three ways in which Belarus was able to extract rents from Russia in the area of gas business. First, it acquired generous discounts on gas from Russia, diminishing the prices of the finished goods produced. Second, the government sold gas to internal producers and households with 20-25% markup added to the price. 12 Third, Belarus made extra rent on the fees paid by Russia to pay for transiting the gas (25% of the total supply to Europe) and oil (36% of the total supply to Europe) through Belarusian territory. These transit fees alone earned Belarus around US$1 billion a year. 13 According to Pikulik's (2010) calculations, gas-based rents totalled 12% of the country's GDP in 2006 and 9% in GDP in 2008.

Who controls rents?
The second and third variables namely the ownership and the distribution of rents are deeply intertwined. This presumes that external rents have divergent impacts on political and economic developments, depending on who is the main owner of the rents: public/state actors (incumbent, government), various private actors or alliances of private-public actors. For those countries in which the governments are the chief owners of rents, the political cost of rent-seeking (hereinafter, costs of rents) and the perceived stability of rent flows (hereinafter, stability of rents) are highly relevant.
Conversely, in those countries where the private actors own the external rent flows, the design of political and economic institutions will depend on, first, the survival strategy of the incumbent institutional channels through which the incumbent obtains/retains the office and the modes of interaction with the rent-seeking cronies and, second, the (in)ability of cronies to reach a compromise and put self-limitations on the looting.
Having spelled out the structural factors it is necessary to show how structural variables are interrelated with agency variables. We argue that, besides the structural factor of rents, the ways in which the incumbent wins the office and maintains power influence the consequent system of rent distribution. The presence of the internal rents simplifies the relations between the business and power: the incumbent establishes the parameters within which the oil rent is distributed between cronies and society and balances elite interests. Another dilemma is to what extent foreigners should be made stakeholders in the rent-seeking industry by being allowed in the capital of the energy companies. Alternatively, when external rent-seeking is the main strategy of the political regime, certain political costs are necessarily attached. Trading too much sovereignty in exchange for rents can jeopardise the political survival of the incumbent internally, as this may cause significant dissent. On the other hand, the presence of some vocal (but weak) opposition with strong anti-sponsor identity can increase the legitimacy of the political incumbent in the perception of the rent-provider. The right balance is one of the keys for political survival when external rents are the main source of economic anaesthesia meant to substitute for market reforms.
In the views of Gubad Ibadoglu, there is practically no real sector of economy in Azerbaijan 14 and more than half of the non-oil sector is the result of capital investment in infrastructure and transportation industries by unknown companies registered in off-shore zones. Ibadoglu refers to data highlighting that, in the period 2005-2014, budget expenditures in the industrial sector grew 43.2 times, while for health it dropped by −5.4, and for education −3.9. Even the leading industries in the non-oil sector, such as construction and transport, are indirectly funded by the oil economy. 15 Much of the private sector depends on government contracts from public investment. Azerbaijan failed to diversify the economy after independence and major investments went into the industrial sector, supporting oil and gas production. Corruption increased, confirming the idea that the extraction of resources boosts illegal rent-seeking behaviour (Leite & Weidmann 1999). Applying mirror statistics, local experts calculated the difference between two sets of data of Azerbaijani imports in year 2015 from three countries only-Russia, Georgia and Turkey-as US$1.6 billion. 16 Independent experts reported in 2015 that, in the oil industry, the difference between official customs data and the statistics of the State Oil Company was 2.16 million tonnes  economy, leading to a significant drop in agriculture's share of GDP, from 16% in 2000 to 5.3% in 2013. 19 Around 58% of the budget for 2014 was made up of SOFAZ transfersdirect transfers to the state budget from the State Oil Fund of Azerbaijan, which testified to the increasing direct dependence of the budget on oil rents (Aslanli 2015).
Control over strategic hydrocarbon resources and, respectively, rents in Azerbaijan is organised in the following way. Political power is highly centralised with legislative and judicial branches subordinated to the executive, or to the president, who enjoys the support of a small group of bureaucracy-ex-colleagues, relatives or people from the same region (Nakhichevan and Armenia)-connected in patron-client relations. Control over oil is based on legal documents, such as the presidential decree on the SOCAR of September 1992, which renders it almost a monopoly on the management of the oil industry (Bayulgen 2003). In practice, SOCAR functions 'as a vertically integrated state company and as the government ministry … where the SOCAR president reports to the president as a minister' (Bayulgen 2003, p. 214). Consequently, the president enjoys exclusive rights in the decisionmaking of every Production Sharing Agreement (Bindemann 1999). Azerbaijani President Heydar Aliyev even introduced the unique rule according to which every significant oil contract in the form of a Production Sharing Agreement is given the power of law through adoption by the parliament, rather than being regulated by common investment law (Bayulgen 2003). Thus, all the external actors who signed the Production Sharing Agreement had vested interest in the stability and continuity of this particular political force, as proved by the 'dynastic' power transfer from the father Aliyev to his son in 2003. By the end of the 2000s, the SOCAR became the 'best tool' of the Azerbaijani presidency (Lussac 2010, p. 36).
In Belarus, according to the results of the national referendum of 1996, the president has the right to rule by decree and state bodies can thus be financed in a non-transparent way. Thus, there is no obligation to finance state bodies solely from the official budget, opening the way for the creation of the Presidential Fund, rumoured to be bigger than the country's entire budget (Pikulik 2010). Money from the oil sector, gas, barter and illegal smuggling accumulated in this fund and was distributed to finance certain public projects. Differently from the patterns established in Azerbaijan, in the Belarusian context, resources were allocated not to enrich the incumbent but to maintain the social contract.
Oil resources are usually associated with a high level of corruption. Pleines (2012) described how regional leaders form large patronage networks through which they reward supporters and co-opt rivals, bypassing the institutions and the norms of formal political decision-making. In Belarus and Azerbaijan, the access of ruling elites to state resources enables their survival through the creation of patronage networks, resulting in endemic, systemic corruption.

Distribution of rents
The ownership and the distribution of oil rents is a key issue for our research. Structurally, the rentier economy leads to the formation of rent-seeking elites. The mode of rent distribution thus tends to be exclusive and according to the political logic of rewards in exchange for loyalty. In this regard, patrimonialism appears to be the basis for antiwelfare rentierism and rent-seeking behaviour, with a high level of corruption as part of the rent-seeking behaviour, thus forming the basis of political patronage (Franke et al. 2009).
In Azerbaijan, during the years of oil exploration, billions of dollars have been directed into the presidential circle. Ministries form a line of patronage, depending on the closeness to the president (Auty 2006). This, however, is the point where the two states diverge. While Azerbaijan's expenditure on social welfare constitutes an insignificant share of rent distribution, with the oil revenues concentrated in the hands of a small group of the president and his cronies (ministers and oligarchs), this is not the case in Belarus. In order to explain the difference in the distribution of the oil rents between the two states we need to look at the debate on the political economy of rentier states and the sources of legitimacy of a given mode of rent. The rentier state is characterised by a strong elite structure and a functionally weak party system, which fails to provide adequate checks and balances.
Patronage networks as a structural effect of the rentier economy are well covered in the literature. There are studies proving that the rentier economy does hinder the development of institutions and conditions of democracy, among others, an independent civil society (Ross 2001). The rent distribution system of the semi-rentier state (with external sources of rents) is not as exclusive as that of a rentier state, as the former needs much more democratic legitimation and tends to be politically less stable than the latter (Gyene 2016). Other authors argue that the provision of mass welfare in oil-rich states is strongly conditioned by whether ruling elites have faced subversive threats at the critical junctures of regime formation (Hertog & Eibl 2017). In Azerbaijan, due to the absence of such threats, the pattern of rent distribution followed the logic of the rentier state, strengthened by the path dependency as it continued the practice of informal rewards in exchange for loyalty typical for the Soviet system. Below we will demonstrate it on the rent distribution via budget expenditures.
The state budget of the Azerbaijan Republic consists of transfers from the Ministry of Taxes, the State Customs Committee and the State Oil Fund (SOFAZ). The Ministry of Taxes provides a lower share of the budget than the State Oil Fund-although the transfers from this agency from 2011 to 2014 increased from 34.1% to 38.7%-compared to 49% of the SOFAZ transfer share in the budget in 2014 (Hajizada et al. 2015). SOFAZ was created as a sovereign wealth fund in 1999 and was worth US$1.7 billion by 2007; at the time of writing, however, the country's per capita wealth did not reflect its resource wealth.
In fact, in spite of the creation of SOFAZ, the government of Azerbaijan began to finance the budget through transfers from this fund. Between 2003 and 2013, the transfers increased from 8.2% of the state budget to 58.2%, increasing from 42% of fund expenditure to 92% of its expenditure (Aslanli 2015). This risky overspending undermined the sovereign wealth fund's major function: saving resources for future generations. This trend was caused by the political economy of the rentier state reinforced by the weakness of institutions of public control, the centralisation of political and economic power, and the minimal or non-existent transparency in government decisions.
The literature review shows that there are sufficient structural explanations of the difference between the modes of rent distribution between the two states. First of all, Belarus is a semi-rentier state, if a broader understanding of rents is taken that includes, for instance, remittances as rents. Compared to energy-rich states, semi-rentier states are spared the effect of the 'paradox of plenty' and experience less pressure of associated structural justification of neopatrimonial framework, as the most suitable for the patronclient relations as a characteristic of rent-seeking behaviour. At the same time, rentseeking elites are present in many ex-Soviet republics, and not only the energy-rich states. The other direction of rent distribution in rentier states is the support for politically relevant groups: the security apparatus and the bureaucracy. This is especially true in the case of Azerbaijan (Franke et al. 2009, p. 128). Statistical data confirm that its judiciary and law enforcement sectors experienced a steady increase in real terms from 2011 to 2016 at a real annual average rate of 30%, which represents one of the highest growth rates of any expenditure (Hajizada et al. 2015). All post-Soviet rentier states show the following traits: the purchase of support through the specific and strategic allocation of rents through the powerful presidents and mechanisms of neopatrimonialism, due to weak parliaments and to the Soviet culture of social benefits (Franke et al. 2009), which is characteristic of rent-seeking behaviour. Lukashenka's regime in Belarus is a special case: the president prefers to maintain his popularity by using external rents to fund the social contract rather than purely for self-enrichment, behaving as a benevolent autocrat.

Change of power in rentier states: the agency?
The need for additional explanatory factors makes us consider the political conditions before the arrival of rents in the country, or the timing of the arrival of rents. Regime change and the nature of the power base are useful concepts in this analysis. While both leaders used the Soviet legacies-institutions, societal attitudes and patrimonial structure-as their power base and sources of legitimacy, Lukashenka came to power as a result of relatively free and fair elections, winning 47.7% of votes in the first round and winning in the second round against his opponent Vyacheslav Kebich. In the case of Azerbaijan, elections took place following the 1993 coup and in the presence of the tanks of the rebel colonel Suret Huseynov. In the absence of a real alternative candidate, as major opposition candidates were persecuted, in the 1993 elections Heydar Aliyev formally won 98.7% of votes, while his opponent Kerar Abilov won just 1.1% of votes cast. While Lukashenka had a coherent ideology and was opposed to liberal reforms, maintaining state-owned industries and drawing on Soviet-era symbols and sentiments, especially concerning World War II, Heydar Aliyev could not appeal to the same values, as they had little social resonance and would not be welcomed by Western investors. Thus, the transfer of power took place through a combination of repression, population apathy and the support of the former party affiliates of Heydar Aliyev, who served as the basis for the future ruling party, the New Azerbaijan Party (Yeni Azerbaijan Partisi-YAP). While his appeal was devoid of ideology, the vague slogan 'Stability and Development', which he declared soon after coming to power, appealed to both foreign investors and the local elite and population, who were exhausted by the Nagorno-Karabakh conflict. Signed in 1994, and followed an agreement stipulating a ceasefire in Karabakh, the so called 'contract of the century', by regulating the exploration of the Azeri-Chirag-Guneshli off-shore oil field, soon reversed the initial mistrust of Heydar Aliyev by European and US politicians and quickly gained him the reputation of a pragmatic and visionary leader. In turn, the opposition, such as the Musavat leader Isa Gambar, who visited the United States in the years of Azerbaijan elections-presidential in 2003 andparliamentary in 2005-were surprised to encounter harsh criticism by the representatives of some leading conservative 'think tanks' during their presentations in Washington,DC. 20 The other direction of rent distribution in rentier states is the support for politically relevant groups: the security apparatus and the bureaucracy. This is especially true in the case of Azerbaijan (Franke et al. 2009, p. 128). Statistical data confirm that its judiciary and law enforcement sectors experienced a steady increase in real terms from 2011 to 2016 at a real annual average rate of 30%, which represents one of the highest growth rates of any expenditure (Hajizada et al. 2015).
The stability of the oil flow, guaranteed by Heydar Aliyev, made the transfer of power after his death strategically important for both the local elite, who benefitted from the regime, and for the external actors and investors, who were interested in preserving the political status quo. Thus, the rigged elections of 2003, which brought his son Ilham Aliyev to power 21 and were recognised by the international community, although with criticism. Moreover, then Deputy Secretary of State Richard Armitage congratulated President Ilham Aliyev even before the official results were announced. 22 Pre-election polls put opposition leader Isa Gambar at 45% compared to Ilham Aliyev's 24%. A group of 188 IDEE (Institute for Democracy in Eastern Europe) observers, who participated in the OSCE-ODIHR mission, regarded as overly soft the criticism expressed by Western states and institutions as regards these fraudulent elections and the post-election violence. 23 In Azerbaijan, with the arrival of oil money, a social contract was developed between the president and a small group of ministers/oligarchs with resources being channelled to the repressive apparatus, the only sector where salaries were increased (Almaz 2015). At the same time, the inadequate official salaries and pensions extended the informal economy to the wider layers of the public sector, such as health and education spheres, where the collection of 'informal' fees at the lower levels-schools and hospitals-were distributed at upper levels in respective ministries.
However, the decline in world oil prices-which went from an average price of US$108 per barrel in 2013 to US$57 in 2015-hit the Azerbaijani economy hard. Budget revenues for the first half of 2015 declined sharply. 24  spent near US$3 billion to support the currency. The currency's dependency on oil rents and the underdevelopment of the non-oil sector were the major reasons for the crash of the national currency, the manat, 25 which by December 2015 had dropped 33.5% against the US dollar. 26 Economic hardship caused a wave of social protests in the regions and raised concerns regarding political stability. 27 In response, the government increased the distribution of rents to the wider population, as reflected in the budget for 2016, which compared to previous years was more socially oriented (Hajizada et al. 2015).
Two more traits of the rentier economy have developed in the country to influence conditions for opposition: underdevelopment of the non-oil sector of the economy and shrinking share of SMEs in GDP. The high dependency on the oil and gas sector has prevented the development of the non-oil sector. Similarly, the development of a small and medium business sector has been lagging behind. According to World Bank assessments, the SME comprised only 10% of GDP in 2016. 28 The same source reminds us that the official number in 2014 was more than 23%, while according to independent experts it did not exceed 3% of GDP.
The other trend observed was that the sector of state employees was expanding-by 2013, there were 29,710 people working in the state service, an increase by 3,500 people from the 2005 baseline. 29 According to the state statistical committee, in 2015 the people employed in the public sector exceeded those working in the private sector-877,100 compared to 629,200 in the private sector. 30 Along these trends, the rent distribution that followed the logic of the power change in the country ensured, on the one hand, the strengthening of the power base of the country's leader and, on the other, further undermined the potential financial sources of the opposition.

Implications for the opposition: Azerbaijan
A series of interconnected factors-the decreasing shares of small and medium enterprises in the economy; the increase in the number of state employees; rising repression against representatives of small and medium businesses who donated to the opposition, 31 as well as widespread poverty-contributed to the narrowing down of the financial and economic basis of the opposition parties in Azerbaijan.
Besides the opposition's principled stance of not accepting foreign funding, Article 17 of the national Law on Political Parties of 1992 (amended a few times since its adoption) prohibits foreign funding to political parties (Alieva 2014). The so called 'crackdown on civil society' of 2014-the unprecedented attack by the government on civil society with the persecution of dozens of leading NGOs and their leaders-paralysed civil society and also affected, indirectly yet not insignificantly, local political parties, as some of the activists were also benefitting from the NGO's projects. By 2016, Musavat, one of two biggest opposition parties in Azerbaijan, had only one salaried staff member. 32 The political opposition has been weakening not only due to the decline of the economic basis of the parties, caused by greater centralisation and monopolisation of resources and greater dependence of both population and business on the state, but also due to the increasing legal and political restrictions of their activities, which began after Xalg Cabhasi (the Popular Front) lost to the former communists after a year in power in 1992-1993. The democratically elected president, Abulfaz Elchibey, started liberal democratic reforms and achieved the full withdrawal of Soviet bases from Azerbaijan, but this cost him the leadership. After losing the war in Nagorno-Karabakh he was forced to resign under pressure from the rebel colonel, Suret Huseynov. The experienced KGB general Heydar Aliyev, who took over the presidency, used natural resources as a foreign policy asset and satisfied the major Western interests in the region by signing, in 1994, the contract of the century-an oil contract with major shares for mainly Western oil corporations and only 10% left for the Russian oil company Lukoil. He played a key role in building alternative pipeline projects that bypassed Russia and had strategic significance for the West. The pro-Western orientation of the democratic opposition was successfully 'hijacked' by an authoritarian leader and thus secured external sources of legitimacy for as long as he was loyal to the energy security (and security) of the United States and the European Union. This characteristic of rentier states, when the recycling of rents domestically is paralleled by the external recycling of rents to build a power base abroad, contributing to regime continuity, was well described by Adeel Malik (2017). After the entry into line of the Baku-Tbilisi-Ceyhan pipeline, accompanied by high world oil prices, Azerbaijan's SOCAR became a major investor in the economies of its neighbours; 33 through the Heydar Aliyev Foundation, it also entered many cultural restoration projects elsewhere in Europe. 34 Domestically, the agenda of opposition political parties began to target, throughout the late 2000s and beyond, a series of topics all connected with the increasing authoritarianism of the regime. 35 However, external support was not only significantly 32 Leila Alieva's interview with Isa Gambar, leader of Musavat Party, Baku, 25 September 2016. 33 'SOCAR Invests Over $1b. in Georgian Economy', Trend.az, 25 September 2019   limited by the legal prohibition of foreign funding but also by lack of contacts, cooperation and integration of the political parties in the European networks. The Eastern Partnership (EaP) integration index in the section of EU linkages reports only three Azerbaijan parties that participate as associates or as observers to the European political parties as compared to eight from Moldova, or six from Ukraine. 36 We observed last minute cancellations of a few planned visiting parties throughout the 2000s. This was usually perceived by the opposition and in the society at large as an unwillingness to upset ongoing EU energy cooperation with the incumbent regime. In this situation, the pro-Western orientation of the major opposition parties Musavat and Popular Front did not empower them either politically or financially. This is in spite of the fact that the opposition parties were the main drivers of the campaign, launched by the NGO National Committee for European Integration, to include a clause on European aspirations in the national draft ENP Action Plan, which was leaked to the public during EU-Azerbaijan negotiations in 2006. A decade after however, societal attitudes and expectations in regards to the West have changed-the major international corruption schemes which were disclosed by the media, revealing the effectiveness of 'caviar diplomacy', compromised the integrity of the Council of Europe, especially during the vote in January 2013 when the critical report on political prisoners in Azerbaijan by Christoph Strasser was not adopted. Together with the perceived weak reaction of the European governments and of the international organisations to the crackdown on civil society in 2014, all these significantly undermined trust in the West. This decreased the power of popular mobilisation for the pro-Western agenda of the parties.
Thus, a pro-Western orientation is hardly an asset in the opposition's activities as compared to the incumbent case of the rentier state with domestic sources of rents, like Azerbaijan. Moreover, the 'pro-Western' energy policy of the incumbent even weakens the Western cooperation or its political relations with the liberal opposition in Azerbaijan. The quotation below describes well the logic behind such behaviour: While the mechanisms [of international influence by Levitsky and Way's linkage-and-leverage] are focused on cooperation and good will, Western leverage is low in oil rich states, as these states tend to be strong economically and because energy is one of the countervailing strategic interests of the West. Moreover in the case of Caspian states, Russia and China are counter hegemonic powers, which offer alternatives for cooperation without demands related to democracy promotion or anti-corruption issues. In fact, one can argue, that leverage sometimes go in the opposite direction. The West is more interested in the energy resources of the Caspian, than these states are-in cooperation with the West. (Pleines & Wostheinrich 2016, p. 295;emphasis added) This is in contrast with Belarus, where the opposition enjoyed wide support from Western institutions treating Belarus as the 'last dictatorship' in Europe.

Implications for the opposition: Belarus
As shown throughout the article, Belarus's political and economic system is largely centred around the president as the key distributor of the external (and internal rents) for the sake of maintaining the social contract. The welfare provided by the president pushed the opposition to the margins; opposing Lukashenka is equivalent to opposing the welfare status quo and threatening to ruin it for the sake of abstract European values.
Three main factors account for the weakness of the opposition in Belarus and the progressive decline of the liberal agenda across the country. First, political space is physically limited, forcing the opposition to favour online over offline activities (Pikulik 2010). The challenge for democratic forces trying to mobilise the public is to link activities in the virtual space with offline activities, using the two spaces to build on each other rather than allowing the virtual space to dominate and create a sense of activism that only exists virtually.
According to the qualitative media-monitoring conducted by the Belarusian Institute of Strategic Studies in 2012, the political forces are communicating to their own narrow audiences on issues of importance to experts rather than directly addressing the wider electorate on issues that matter to them (Pikulik et al. 2012).
The second factor is the dependency of the opposition on the West via complex donorimplementer networks. Pikulik and Bedford (2019) argued that there is a 'dependency' dilemma existing in the Belarus opposition. The political opposition, having no alternatives to donors/implementer resources (such as fundraising from business), has to rely on Western donors/implementers. It needs to comply with their expectations in order to survive and thrive, otherwise, it risks losing their support. The implementers make a bet on the opposition groups and have to account for the projects in their investment portfolios, namely those supported by the donor's organisations.
In other words, since Lukashenka successfully occupied the niche of the socially inclusive autocrat working for the fair distribution of resources, the opposition had to opt for regime change rather than trying to win support on the basis of welfare policies. This creates the paradox of 'the worse the better' associated with the foreign policy and donor-government nexus. Thus, in a period of confrontational politics (isolation and sanctions), the opposition is the only political partner of the West in Belarus. By contrast, during periods of cooperation, when Western governments see higher value in dialogue with Lukashenka (for example, the Minsk Process), the resources allocated to the domestic opposition shrink significantly (Pikulik & Bedford 2019). In other words, harsher repression against the opposition resulted in Western isolation of the incumbent and increased the flow of Western resources to the opposition. Less repression, paradoxically, led to the worsening of the opposition's situation, as the policy of direct Western engagement with the incumbent resulted in fewer or no resources being allocated by Western donors to the opposition (Pikulik & Bedford 2019).

Conclusion
This study analysed the 'converging divergence' between Azerbaijan and Belarus. The two countries combine autocratic institutions and Soviet legacies with a rentier economy but have different sources of rents-external in Belarus and internal in Azerbaijan. In this latter context, shaped by the driving forces of the oil-based economy and the leader's power base, oil rents were used to strengthen informal patronage networks and the repressive apparatus rather than distributed equitably through a welfare system providing health, education and social benefits, which remained one of the mostly poorly funded in Eurasia. Oil revenues have been invested largely in the industrial sector to support oil production at the expense of the non-oil sector. The national economy is highly dependent on oil with extremely centralised political and economic control and with an underdeveloped non-oil sector. This oil dependency prevented the creation of independent economic groups that would have contributed to political plurality by funding political parties. The domestic sources of funding were narrowing down due to the repressive measures against businessmen supporting the opposition, greater economic centralisation and general poverty that significantly reduced the income of parties in the form of membership fees and donations. The external financial support to the opposition has also been limited by the resistance of the parties to the idea of foreign funding and by the provisions of national legislation, while political support was shaped by the energy interests of external actors. The energy cooperation of established democracies with the autocratic leadership in Azerbaijan deprived the opposition of their major asset in relations with the West: their pro-Western and liberal course. In Belarus, the situation was reversed: the opposition enjoyed support from the West-around half a billion US dollars between 1998 and 2018 (Pikulik & Bedford 2019). At the same time, it had to fight an unusual enemy: a benevolent and inclusive autocracy that enjoyed public support.
Two types of rentier states-an exclusive, patronal rent-seeking regime in Azerbaijan and the inclusive, benevolent rent-seeking regime in Belarus with two other profound characteristics-sources of rents and position of the West-produced different challenges for their respective oppositions. In Azerbaijan, the dominant centre-right opposition parties had to address the transparency and fair distribution of resources, attract Western allies and fight for fair elections despite the corrupted electoral system. By contrast, in Belarus, the opposition became dependent on Western support and was inclined to offer society the sweat and tears of market reforms to replace the state paternalism sponsored by Russian rents.
Besides the structural effects of the rentier states, as expressed in the political economy of the two countries, the authoritarian conditions make the opposition's task even more difficult. While social injustice in Azerbaijan creates a strong potential power base for the opposition, its functioning is seriously impaired by repressive measures and restrictive legislation.
Thus, domestically the major objectives of the opposition in Azerbaijan should be the adjustment of their tactics and programme to attract the state-dependent groups, such as the health and education sectors, young professionals and the service sector, emphasising issues with cross-class relevance, such as corruption and equal opportunities. Externally, considering a narrow circle of allies of the opposition abroad, working for greater integration into the European regional networks of political parties and developing close relations with non-governmental international advocacy groups and with civil society overall might prove more effective than relying on political support from Western governments. The latter might be more responsive to the active promotion of the European integration agenda in substance and in the EU energy market. To date, the experience of Azerbaijani opposition in the multilateral European and trans-Atlantic institutions proved to be more effective when focusing on cooperation with the groups of countries, which are ardent supporters of the democratic institutions and do not have significant energy interests in the country.
Our comparative analysis proved that the rentier states are united by structural impediments to democracy through the effects which were described by Ross (2001) and Karl (2005). While we also showed that the structure is not solely responsible for the lack of democracy in rentier states, our analysis showed that in order for political development to occur, the foundations of the rentier economy must be changed. In the case of Azerbaijan, the beginning of the 'post-oil' era, signalled the need for the leadership to adapt to the increasing deficit in the revenues that sustained its network of patronage, which was supported both by the formal and informal economy. It climaxed at the end of February 2020, against the background of a sharp decline in the oil price coupled with the Covid-19 pandemic, when President Aliyev proposed a 'dialogue' with the opposition parties. 37 Younger generations, disillusioned by Western support for the regime and the increasing inequality and injustice of the patronage system, have turned, albeit in small numbers, to left-wing activism (Kucera 2020).
In the case of Belarus, the loss of the significant rents from Russia and Russia's persistence in pushing the integrational project created a trap for the incumbent regime. The loss of rents and financial weakening of the state was amplified by various factors that brought about a political crisis in August 2020. First, the government mismanaged Covid-19, concealing statistics and failing to take proper measures and inform society of the risks in spring 2020. Second, the independent candidates running for presidential office in summer 2020 managed to demonstrate clearly how the state violates its own rules/laws in hijacking the election process and stealing votes. As an example, they collected an extraordinary number of signatures for their candidates at the pre-election stage and made the state refuse these signatures as falsified proving to everyone who gave a legitimate signature that their rights were being violated. Another example is mobilising people to participate in the Electoral commissions and making them receive a 'no' due to some ridiculous reason from those authorising the lists. Third, the use of violence against protesters sparked further protests and further violence, creating a deepening cycle of state repression.
This study contributes to the debate on how the nature of rents, extracted from different sources, affects the political regime by extending it to conditions for the development of the opposition. It shows the effect of both structure and agency on how rents are distributed. In our two cases the relevant factors are the sources of rents (external in Belarus, internal in Azerbaijan) and the nature of power change (relatively fair elections in Belarus, unfair in Azerbaijan). Both factors create different challenges for the opposition. As the experience of Belarus proves, by inclusive rent distribution the leadership 'hijacks' the political agenda of the opposition, which could otherwise promote redistributive policies. In