Climate change prioritization in remedying energy supply deficits : the case of Bishkek

(1)

CLIMATE CHANGE PRIORITIZATION IN REMEDYING

ENERGY SUPPLY DEFICITS: THE CASE OF BISHKEK

Source: Anderson (2017)

Master Thesis Project Max Narinx [11342781] Supervisor: Eric K. Chu

Second reader: Yves van Leynseele Date of handing in: June 26, 2017 Word count: 28.766

(2)

(3)

ACKNOWLEDGEMENTS

I want to thank all the people who made it possible to write this thesis. Since it is not possible to mention all of you, I made a selection. First, I want to thank my supervisor Eric K. Chu for the guidance, fast responses to questions, feedback and inspiration. Furthermore, I am grateful to my hosts for letting me live with them as part of their family for 5 weeks in the city of Bishkek. Also, I want to especially thank my dear friends Nestana Rysbekova and Madina Ismailova for arranging my host-family, translating multiple interviews, joining me to conferences, picking me up from the airport, and most important companionship. Furthermore, I would like to thank Manas Toktobaev for translating one interview. I am really grateful to all of my Kyrgyz friends who supported me by helping me on several occasions to e.g. get into contact with interviewees, invited me for dinner, giving me company. Special thanks to my sister Renée Narinx and Aster Hoving for giving feedback on parts of this work. Finally, I want to thank all the interviewees for sharing their knowledge and experiences with me.

(4)

Abstract: This thesis work determined how the residential energy sector in the Bishkek (developing and transition economy) is governed in the light of infrastructure degradation, energy supply deficits and emerging climate change priorities. ‘Urban climate change experimentation’ and ‘low carbon energy transitions’ form the conceptual basis for this thesis. Combining these concepts implies that governing climate change goes beyond the institutional reading, since urban climate change experiments – all practices which purposively and strategically seek to direct, guide or control others or oneself to capture new forms of (socio-technical) learning or experience, whereof the purpose is to mitigate and/or adapt climate change in the imagined or existing urban community - are critical means through which governing takes place. The method used – 30 semi-structured interviews with government officials, energy consumers, academics, and energy company representatives have been conducted in April and May 2017. The remedies to the energy supply deficits (2007-2009 and 2012-2014) have been assessed on i.e. prioritization, climate change impact, and sector liberalization. Results show that remedies did not prioritize climate, liberalization and development of the RE sector, which indicates that Bishkek’s context does not facilitate a low carbon energy transition but actually facilitates an anti-transition towards a bigger carbon share. The main constraint in facilitating a low carbon energy transition is the national energy tariff policy, which cannot be adjusted due to socio-political tensions.

(5)

5 CONCLUSIONS... 73 5.1 Concluding findings……….……….……….……….……….……….. 73 5.2 Concluding statements……….……….……….……….……….……… 75 5.3 Discussion……….……….……….……….……….……….………. 75 5.4 Recommendations……….……….……….……….……….……….……… 76 BIBLIOGRAPHY……….. 77 APPENDICES……….. 82 1. List of Interpreters……….. 82

2. Interview guide/topic list………. 83

3. Fieldwork Protocol……….. 84

4. Timetable……….. 85

(7)

LIST OF TABLES

TABLE 1: Interview respondents per category……… 21

TABLE 2: Operationalization categories and indicators………….……….……….……….……. 24

TABLE 3: Power plants in the Kyrgyz Republic……….……….……….……….……… 31

TABLE 4: National Energy Tariffs for End Consumers……….……….……….……….…… 64

LIST OF FIGURES

FIGURE 1: Energy-related carbon-dioxide emissions by country……….……….……….. 11

FIGURE 2: Kyrgyz Primary Energy Consumption in 2014……….……….……….……….. 18

FIGURE 3: Operationalization scheme……….……….……….……….……….………. 23

FIGURE 4: The ‘electricity circle’ of Central Asia……….……….……….……….………. 30

FIGURE 5: District heating system map of Bishkek……….……….……….……….………… 32

FIGURE 6: Imported Gas in tera-joule (TJ) in the period 1990-2014……….……….………. 34

FIGURE 7: Post-reform energy production chain……….……….……….……….……… 36

FIGURE 8: Domestic transmission line construction……….……….……….……….……….49

FIGURE 9: Bishkek CHP and its smog in the city skyline……….……….……….………. 52

FIGURE 10: Percentage share in TPES per energy source……….……….……….……….. 54

LIST OF ABBREVIATIONS

GHG : greenhouse gas GHGs : greenhouse gases RE : renewable energy

RES : renewable energy sources GDP : gross domestic product TPES : total primary energy supply DH : district heating

NAP : national adaptation program HPS : hydropower station

HPSs : hydropower stations

CHP : central heat and power plant TJ : terajoule

UNDP : United Nations Development Program ECA : Europe and Central Asia Region ADB : Asian Development Bank

EBRD : European Bank for Reconstruction and Development JSC : joint-stock company

OJSC : open joint-stock company

(8)

1. INTRODUCTION

Contemporary energy use heavily relies on the combustion of fossil fuels and is the main source of greenhouse gas (GHG) emission and hence the main contributor to climate change – arguably the biggest threat to societies all over the world (Macay, 2008). In this light, the climate change problem can be regarded as an energy problem (MacKay, 2008). However, ‘low carbon energy transitions’ – in this thesis defined as: Changes in the energy system that ensure meeting the growing energy demand, with a focus on energy sector liberalization, while the carbon share reduces due to increased use of renewable energy sources (RES) – can reduce GHG emissions while meeting the demand for energy (Bridge et al., 2013). This implies that although the climate change problem is currently an energy problem, it does not have to be in the future, since RES can supply the energy demand without emitting GHGs (Twidell and Weir, 2015).

Research on climate change mitigation and adaptation has started to focus on governing climate change by means of experimentation at the urban level (Bulkeley et al., 2016). Urban climate change experiments – all practices which purposively and strategically seek to direct, guide or control others or oneself to capture new forms of (socio-technical) learning or experience, whereof the purpose is to mitigate and/or adapt climate change in the imagined or existing urban community (Bulkeley et al., 2016) – are essential, because experimentation is central in the exercise of government (Bulkeley et al., 2016; Bulkeley and Broto, 2013). The urban aspect is important because cities are the sites where climate change consequences are being felt, climate change is caused – two third of the global GHG emissions occur in cities (Droege, 2011: 55) – and solutions are being sought (Bulkeley et al., 2016; Wolfram and Frantzeskaki, 2016; Bulkeley and Broto, 2013; Hodson and Marvin, 2012; Dodeman, 2009; Bestill and Bulkeley, 2006). Governing through urban climate change experimentation can facilitate low carbon energy transitions, because one of the elements is to purposively try to adapt and/or mitigate climate change.

The city of Bishkek, capital of the Kyrgyz Republic, faced an energy crisis – a society-wide economic problem caused by a constricted supply of energy leading to diminished availability and increased price to consumers (Grossman, 2015) – in the period 2007-2009 and several periods of energy supply deficits that have been depriving a substantial part of the population from reliable energy access (Balabanyan et al., 2014; Liu, Pistorius, 2012; Jefferson Institute, 2009). During these episodes Bishkek could not reliably supply sufficient energy to its residents and commercial entities. Especially during the heating seasons, when the energy demand peaks, the energy systems failed on a regular basis (Balabanyan et al., 2015; Jefferson Institute, 2009). This implies that changes in the energy sector are needed/desired. I am interested in examining what actions were taken to improve the energy supply, and if those actions have initiated a low carbon energy transition, and if climate change experimentation occurred. In particular, I am interested in how the energy supply problems have influenced the city’s governance regarding climate change prioritizations and low carbon energy transitions. I choose to focus

(9)

on the residential energy use, because with approximately 60 percent of the total energy use, it is the biggest consumption sector (Energy Charter, 2011: 20).

This research is relevant because climate change impacts are being felt worldwide and this contribution to the knowledge on urban climate change experimentation could improve the general knowledge about GHGs strategies. Hence, the focus on low carbon energy transitions by means of climate change experimentation can contribute to the mitigation of climate change and climate change impacts, but also have additional benefits at local levels such as improving public health by reducing carbon rich air pollution. By assessing the energy sector in Bishkek, this research contributes to the knowledge and understanding of the city’s energy sector governance. This research could potentially contribute to a more sustainably performing energy sector. Scientifically, this research is significant as it contributes to the relatively new and growing field of research on urban climate change experimentation. More specific, by focusing on Bishkek – a city in a developing country – an underrepresented category in this field of research will be addressed. In addition, no urban climate change experimentation research has been performed on a Central-Asian city thus far. Because the Central-Asian countries are also transition economies – countries that have been changing from central planned to free markets systems such as former Soviet Union countries (Bouzarovski, 2009: 453) – an unrepresented context will start to unveil in the field of urban climate change experimentation.

To ascertain the above, three aspects of the city’s energy sector will be taken into account: How the residential energy sector of Bishkek is governed; what the municipality of Bishkek is doing to remedy the energy supply deficits; and how climate change is prioritized in the energy sector governance and remedies to the supply deficits. This leads to the following research question: How is Bishkek’s municipal residential energy system governed in light of energy supply deficits and emerging climate change priorities? Three sub-questions will help answering the research question. First, How is the residential energy sector governed in the municipality of Bishkek in terms of sociohistorical politics and finances? Second, What is the municipality of Bishkek doing to remedy the energy supply deficits? Third, What is the municipality of Bishkek doing to address transition priorities with regard to climate change? The objective of this research is to contribute to the knowledge and understanding of municipal residential energy sector governance in cities that faced episodes of energy supply deficits, situated in developing countries and transition economies, in the light of low carbon energy transitions and global climate change prioritization.

In this research, I will first discuss the literature on ‘low carbon energy transitions’ and ‘urban climate change experimentation’ as well as how they are interlinked. In addition, I will provide some contextual information on Bishkek’s current energy portfolio. Then I will explain that semi-structured interviews were used to gather data and I will explain how the research was executed. Afterwards, the results will be presented and finally this thesis will conclude by answering the research question.

(10)

2. THEORETICAL FRAMEWORK

In this section I will assess the literature on the concepts of ‘energy transitions’ and ‘urban climate change experimentation’, and how they are interlinked. First, I will discuss the how the concepts ‘energy’ and ‘development’ a r e related to climate change. Second, I will describe how energy transitions are a strategy for coping with climate change. Third, I will explain that energy transitions are determined by their geographical context or ‘geographies of energy transitions’. Fourth, I will assess why current research sees cities as the most significant arenas for governing climate change. Fifth, I will argue that the governing of energy transitions through the form of urban climate change experimentation is justified. Finally, I will assess the knowledge gaps in the literature, and explain how the case of Bishkek contributes to filling this gap.

2.1 Energy and Development: Geographical Perspectives

2.1.1 Energy problem

Energy is essential to human life, since “all production and many consumption activities involve energy as an essential input” (Paul and Bhattacharya, 2004). In this light, all societies require energy services to meet their basic human needs (IPCC, 2011: 2). However, production, consumption and distribution of energy also account for harmful environmental consequences on different levels, such as local air pollution and global climate change (MacKay, 2008; Stern; 2006). The burning of fossil fuels for energy accounts for 74 percent of globally emitted GHGs, while the main reason to burn fossil fuels is for energy (MacKay, 2008: 5). Therefore, the climate change problem can be seen as predominantly an energy problem (MacKay, 2008: 15).

2.1.2 Development problem

Besides perceiving climate change as an energy problem, it can also be seen as a development problem (Gupta, 2014). Development and climate change are intrinsically linked because economic growth generally leads to increased energy use; and as MacKay (2008) stated, energy production, distribution, and consumption account for t h e v a s t amounts of GHG emissions (Gupta, 2014: 22; Bilgen, 2014: 891; Stern, 2006). Climate change and development are linked in four specific ways.

First, most societies wish to develop; however, development will boost production, distribution and consumption which requires increased amounts of energy for which more GHGs will be emitted (Araújo, 2014; Gupta, 2014,). Among others, Bilgen (2014; 891) and the energy Information Administration (2016) state that globally speaking, societies are developing out of poverty. As a consequence, much of the world’s increase in energy production and consumption occurs in the developing countries (see Figure 1). Araújo (2014: 114) adds that population growth, which is much higher in developing countries, also accounts for the enhanced GHG emissions in many developing

(11)

(Energy Information Administration, 2016; Bilgen, 2014: 891; Khan et al., 2014: 337). The Energy Information Administration (2016) predicts the world consumption of marketed energy will increase by 48 percent between 2012 and 2040. Enerdata (2016) states that growth rates of energy production and consumption differ per region. For example, in 2014, the G20 countries, which account for about 80 percent of the global energy consumption, showed an almost stagnation in energy demand (+0,5 percent) and stagnation of CO2 emissions (-0,2 percent), despite an economic growth of 2,8 percent, while on the other hand, India’s energy consumption grew by 6 percent the same year (Enerdata, 2016).

FIGURE 1: Energy-related carbon-dioxide emissions by country. Source: Geels, 2014

Second, development can lead to enhanced adaptation and mitigation of GHG as developed societies are more likely to have knowledge and awareness about climate change, which increases climate related policy choices (Gupta, 2014). In addition, developed societies are more likely to dispose of technological innovations that help to decrease GHG emissions (Gupta, 2014). This implies that development is required in order to fully mitigate climate change. Furthermore, development generally also enhances climate change adaptation capabilities (Gupta, 2014).

Third, climate change mitigation measures can have additional benefits on societal developments (Gupta, 2014). For example, reducing the use of fossil fuel reduces GHG emissions could also results in co-benefits such as reduced air pollution, enhanced energy security and increased public health (Gupta, 2014: IPCC, 2011).

Fourth, enhancing carbon sinks can lead to a decline in deforestation rates, of which the particular ecosystem services benefit and hence the local communities as well (Gupta, 2014). However, whether deforestation enhances the development of societies remains arguable in literature (Gupta, 2014: Agrawal et al., 2011).

(12)

Although economic growth causes growing energy consumption, there is no consensus that increasing energy consumption causes economic growth. A general observation from energy consumption and economic growth studies is that a consensus on the relationship between energy consumption and economic growth lacks, due to differentiating geographical contexts (Ozturk, 2010: 340; Payne, 2010; 54; Soytas and Sari, 2009).

2.1.3 Renewable Energy Sources

To cope with climate change, it is possible to “adopt mitigation measures to reduce GHG emissions and enhance sinks that absorb GHG, adaptation measures that deal with the impacts of climate change, or adopt geo-engineering measures (large-scale interventions in the climate system)” (Gupta, 2014: 22). Assuming that development, energy use, and climate change are intrinsically linked, and large parts of the world are actually developing, the challenge is to acquire energy from sustainable renewable energy sources (RES) (Bridge et al., 2013); such as sunshine, wind, water and wave flow, and the cycles of ebb and tide (Droege, 2011: 3-4). In general, renewable energy is defined as “the energy obtained from naturally repetitive and persistent flows of energy occurring in the local environment” (Twidell and Weir, 2015: 3). Implementing the use of RES can lower GHG emissions from energy systems and therefore is a climate change mitigation strategy (IPCC, 2011: 3). Twidell and Weir (2015: 7) argue that techniques to gain energy from RES have become adequate to meet the energy demands. Among others, Bridge et al. (2013: 331) state that the energy challenge, and hence the climate change challenge, of the twenty-first century is to accomplish an energy transition towards a more sustainable energy system.

2.2 Energy Transitions

No universally accepted definition of ‘energy transition’ exists (Araújo, 2014: 112; Bridge et al., 2013). Bridge et al. (2013: 332) state that energy transitions imply “a change in conditions of the current energy system”. Araújo (2014: 112) defines energy transitions as “a shift in the nature or pattern of how energy is utilized within a system”. Before determining the definition, I will use in this thesis I will first discuss some examples of how energy systems differ per context and how that influences the energy transition itself.

2.2.1 Geographies of Energy Transitions

Bridge et al. (2013: 332) state that varying geographical contexts of energy systems are crucial for the goal of an energy transition. They argue that the ‘geographies of energy transitions’ – embedded components of energy systems in particular geographical settings – influence the path and form of transition. For example, in the global South, energy transitions usually point to a significant increase in the availability and affordability of energy services, which could result in an increase in the total carbon emissions (Bridge et al., 2013: 332-333). Droege (2011: 1) states that the challenge of developing countries is to “stabilize a growing demand for secure energy supplies without

(13)

establishing polluting industries and power supplies”. In the context of transition economy countries – countries that are changing from central planned to free market systems, for example former Soviet Union countries – energy transitions are framed as the ‘liberalization’ from a centrally planned energy system to a market-based energy sector (Bouzarovski, 2009: 453). For transition economies, it is claimed that “the liberalization and unbundling of energy activities can provide the most effective path for overcoming the economically and environmentally inefficient practices” Bouzarovski, 2009: 453). Bridge et al. (2013: 336-337) add that the degree to which power and authority over a network are centralized can be a constraint in reaching the goals of an energy transition, since research has shown that in centralized structures of energy systems constrain both the ability for decentralized renewable energy to develop as well as demand-side efficient use of energy. In the global North, energy transitions are usually defined as movements towards secure and low carbon futures, which is also called a ‘low carbon energy transition’ (Bridge et al., 2013: 333). For this research, I will combine aspects of above definitions, because the Kyrgyz Republic is first, a developing country which struggles to meet the growing energy demand, second, a transition economy and third, the Kyrgyz government committed to significantly decrease GHG emission (State Agency of Environmental Protection and Forestry, 2017). Therefore, for this research I will define ‘low carbon energy transition’ as: Changes in the energy system that ensure meeting the growing energy demand, with a focus on energy sector liberalization, while the carbon share reduces due to increased use of RES.

Hence energy transitions are depending on geographical contexts. To better understand how contextual characteristics of energy systems facilitate or constrain a low carbon energy transition Bridge et al. (2013: 334) provide a basic conceptual framework consisting of six spatial concepts. These concepts can be used to gain a more in-depth understanding of the form and pace of energy transitions (Bridge et al., 2013: 339). First, ‘location’. Location is both an absolute as well as a relative characteristic, which describes the ‘relational proximity’ of the system’s elements (Bridge et al., 2013: 334). For example, emerging railway-networks brought cities closer together in terms of relational proximity, but it also increased the relative distance between them and places that were not connected to the network (Bridge et al., 2013: 334). Hence, the de-carbonization of the energy sector may be determined by shifts in “ the nature and location of primary energy sources as well as the location of new build fossil fuel power plants” (Bridge et al., 2013: 334). Second, ‘landscape’ – “the assemblage of natural and cultural features across a broad space and the history of their production and interaction” (Bridge et al., 2013: 335) – focusses on how the location features described above interact with the cultural history. For example, if a society has been relying on a certain proximate energy source which for some reason is not proximate anymore, a transition has to be made to meet the energy demand. These choices depend on the historical cultural characteristics. For example, if a society heavily values air quality, it is less likely that they will transition to coal use. Third, ‘territoriality’ describes “how social and political power are organized and exercised over space” (Bridge et al., 2013: 336). Territoriality refers to the activities of partition and integration by political and economic actors. For example, all energy related infrastructure have been territorialized in a different way (Bridge et al., 2013: 336). For

(14)

example, energy systems have different degrees of centralization/decentralization and nationalization/privatization. In general, centralized energy governance and infrastructure are known to constrain decentralized RE development and energy end-consumer efficiency (Bridge et al., 2013: 336). Fourth, ‘spatial differentiation and uneven development’ implies that taken together, the locations, landscapes and territorialities, associated with a low-carbon energy transition could lead to uneven spatial development, which can trigger spatial convergence or divergence (Bridge et al., 2013: 337). Fifth, ‘scaling’ refers to “the material size and areal extent of the phenomena”, which implies that energy transitions have different sizes, depending on which different energy technologies can or will be implemented or are prioritized (Bridge et al., 2013: 337). For example, by governments energy is often perceived as a national issue, because the lack of energy could lead to socio-political instability, which governments do not want to risk (Bridge et al., 2013: 338). Sixth, ‘spatial embeddedness and path dependency’ refers to the embeddedness and lock-in of fossil fuels. For example, rural developing economies are generally less locked-in on fossil fuels than urban economies, because they often do not have an existing energy infrastructure. Hence, for these economies, it may be easier to move towards RES use (Bridge et al., 2013: 337).

2.3 Governing Energy Transitions in Cities

Although there are many different scales of action involved in low carbon energy transitions, current literature emphasizes that governing climate change at the urban level might be the most effective (Bulkeley et al., 2016; Wolfram and Frantzeskaki, 2016; Bulkeley and Broto, 2013; Hodson and Marvin, 2012; Dodeman, 2009; Bestill and Bulkeley, 2006). Below, I will first briefly explain what governance entails, followed by a discussion on the literature regarding climate change governance in cities.

2.3.1 Government and Governance

The term government refers to the formal institutions of the state and their monopoly of legitimate coercive power (Stoker, 1998: 17). Governance differs from government by considering that a strong state monopoly power does not apply (Paavlova, 2007: 94). Historically, governance has been used in a broad sense to refer to “any practice that more or less deliberately seeks to direct, guide or control others or oneself” (Dean 2007: 36). Hence, Stoker (1998: 18) states that the concept governance “recognizes the capacity to define and pursue collective goals which do not rest on the power of government to command or use its authority, it sees government as able to use new tools and techniques to steer and guide”. This implies that governance acknowledges the roles of supranational and subnational state and non-state actors, and the complex interactions between them, in the process of governing (Bestill and Bulkeley, 2006: 144).

(15)

The governing of climate change has increasingly been seen as a multi-level process operating at the local, national, international and intermediate levels simultaneously (Paavlova, 2007: 94; Bestill and Bulkeley, 2006). Climate change studies used to focus on the national level, but since the mid-1990s, an increasing number of scholars have argued that cities, rather than nations, may be the most significant sites for governing climate change (Bulkeley et al., 2016; Wolfram and Frantzeskaki, 2016; Bulkeley and Broto, 2013; Hodson and Marvin, 2012; Dodman, 2009; Bestill and Bulkeley, 2006). Hence, although the governing of climate change is a multi-level process, it is justified to focus on the city level, because cities are becoming the most relevant sites for coping with climate change. Cities are places with concentrations of the climate change causes, impacts, as well as responses to minimize these (Stern, 2006: 87-517). Araújo, (2014: 116) and Droege (2011: 55) state that ongoing urbanization resulted in cities being the places where most the people live as well as the places where two-thirds of current energy consumption occurs. These concentrations result in enhanced environmental impact, health impact and climate change impact, in relatively small areas (Araújo, 2014: 116; Droege, 2011). On the contrary, the concentrations of people and industries offer favorable circumstances for technological innovations (Dodman, 2009), also because cities are likely to be subjected to climate change-related consequences. This implies that new ideas and innovations can spread and adoption fast, both in technical and behavioral solutions (Dodeman, 2009). Wolfram and Frantzseskaki (2016) call this ‘qualitative momentum of networked places’, which implies that causes positive changes are more likely to occur in the networked places themselves, but also will have trickle-down effects on the rural areas. In addition, urban authorities and local governments can generally implement climate change measures effectively, because they hold responsibilities in relation to land use planning, local public transportation and the enforcement of industrial regulations (Dodeman, 2009). Hence, developing urban responses are significant as they can accelerate change towards both local and global sustainability (Wolfram and Frantzseskaki, 2016).

2.3.3 Urban Climate Change Experiments

Bulkeley and Broto (2013: 361) argue that when researching climate change governance, an approach is required that reaches beyond looking at government practices. They argue that government is accomplished through social and technical practices which should be considered as critical means. In other words, governance includes processes of social and technical learning which contribute to the changing of norms, values, goals, operational procedures, and actors that govern decision-making processes (Bos and Brown, 2012: 1341). Bulkeley et al. (2016) state that the social and technical learning or ‘will to improve’ results in interventions with the aim to “direct, conduct, or intervene in social processes to produce desired outcomes and avert undesired ones”. When these interventions are designed to contribute to mitigating and adapting of climate change in the city, they can be considered urban climate change experiments (Bulkeley and Broto, 2013: 362). More specifically, Broto and Bukeley (2013) state that an intervention is an urban climate change experiments when first, it is purposive and strategic but explicitly seeks to capture new forms of learning or experience;

(16)

second, the purpose is to reduce emissions of greenhouse gases (mitigation) and/or vulnerabilities to climate change impacts (adaptation); third, a climate change experiment is urban when it is delivered by or in the name of an existing or imagined urban community. Climate change experimentations are linked to urbanism, because in urban communities the nature of existing socio-technical systems provide the right circumstances for interventions (Bulkeley and Broto, 2013: 372). Bulkeley and Broto (2013: 362) emphasize the importance of trial and error aspect of experimentation, because it designates best practices that support climate change mitigation and adaptation and it challenges existing systems. Experimentation can be a significant tool in overcoming system lock-in and socio-technical standards (Bulkeley et al., 2016: 5; Bos and Brown, 2012: 1341;). Urban climate change experiments are socio-technical because “they challenge the technical basis of GHGs emissions, the social practices that produce them or both” (Broto and Bulkeley, 2013). Hence, Bulkeley and Broto (2013: 363) argue that urban climate change experiments are critical means through which governing is performed, because they are significant in both a social and material sense.

Empirically, a survey of urban climate change experiments in 100 cities around the world showed that urban climate change experiments are not confined to certain world regions, such as Europe and North America (Broto and Bulkeley, 2013; Joss, 2010: 242). Furthermore, many climate change experiments have a focus on energy: 78 percent of the infrastructure projects are related to energy and 45 percent of all experiments focus on energy (Bulkeley and Broto, 2013: 370). In addition, most climate change experiments take place in the urban infrastructure sector (31.1 percent), built environment sector (24.7 percent), and transport sector (18.8 percent) (Broto and Bulkeley, 2013).

2.4 Synthesizing Theories

Below, I will briefly discuss how the theories of energy transitions and urban climate change experimentation are linked, and how this leads to a theoretical gap in existing knowledge.

2.4.1 Urban Energy Transitions as Governance through Climate Change Experimentation Urban climate change experiments contribute to achieving a low carbon energy transition, because their purpose is to reduce GHG emissions, or to seek new forms of learning in doing so. Therefore, at the urban level, governing by climate change experimentation might be the most important form of urban climate change action and hence a main part of a low carbon energy transition at the urban level. Hence, urban interventions with the purpose of socio-technical learning and the mitigating/adapting climate change are significant for establishing a low carbon energy transition, which implies that energy transitions are potentially the results of experimentation.

2.4.2 Knowledge gap(s)

In the recent literature on urban climate change experimentation, there are multiple knowledge gaps that remain unaddressed. First, there is an underrepresentation of research on climate

(17)

change experiments in cities in developing countries (Broto and Bulkeley, 2013) – 2015 was the first year in which investments in RES (excluding large hydro) were higher in developing economies than in developed countries (UNEP, 2016: 11). This implies that there should be more research focusing on cities in developing countries. Second, according to (Broto and Bulkeley, 2013) the role of local authorities outside of the institutional reading is currently underrepresented in scientific literature. Therefore, researchers should try to implement a more varied range of actors/responses in future climate change experimentation work. More specific, more attention should be private sector climate change responses as it is not sufficient anymore to examine urban climate change experiments from a municipality perspective alone (Broto and Bulkeley, 2013). Third, in the case of interventions on energy systems there is still a separation between interventions seeking to reconfigure energy consumption and energy production, while Broto and Bulkeley (2013) argue that research should focus more on interventions leading towards systemic change (Broto and Bulkeley, 2013).

2.5 Problem Statement: Synthesizing the Gap in Knowledge

As literature on urban climate change experimentation mostly focusses on cities in developed countries. Cities in developing countries, as well as cities in transition economies remain underrepresented (Broto and Bulkeley, 2013). For example, no climate change experimentation research has focused on Bishkek or other cities in Central Asia, although researching cities in these countries is relevant, because most Central Asian countries are not only developing countries, but also transition economies, which means economies that are changing from centrally planned to free market systems (Myant and Drahokoupil, 2012; Bouzarovski, 2009). Researching urban climate change experimentation in this underrepresented context is relevant to determine whether certain contexts or contextual features are facilitating or constraining urban climate change experiments. In addition, Central Asian cities are particularly interesting research subjects as many of them have been facing energy shortages on a regular basis, while natural resources are abundant (Balabanyan et al., 2015; Daryl et al., 2013; Sadykov, 2013). It will be interesting to examine these cities in terms of: Why energy shortages occur, how they are governed in terms of remedies, and how they influence the prioritizing of climate change mitigation. Below I will present how the case of the energy system in Bishkek is relevant for filling this knowledge gap.

2.5.1 The Kyrgyz Republic and Bishkek

Bishkek is the capital of the Kyrgyz Republic, a developing country situated in Central Asia. It gaining independence from the Soviet Union in 1991, and hence is a transition economy (Myant and Drahokoupil, 2012; Bouzarovski, 2009). After the collapse of the Soviet Union, the Kyrgyz Republic experienced an economic downturn of which it is still recovering (Kronenbeg, 2014: 255). In general, transition economies suffered an economic downturn when the market system got introduced (Melo, Denizer and Gelb, 1996) and compared to other countries, transition economies have suffered a general

(18)

economic disadvantage (Myant and Drahokoupil, 2012). In 2015, the Kyrgyz Republic was ranked 183th of 229 countries in terms of gross domestic product (GDP) per capita, which averages 3400 USD (Central Intelligence Agency, 2016). Officially, Bishkek has 937.400 inhabitants (Chynybaeva, 2015), however Flynn and Kosmarskaya (2012) argue that this number is an underestimate since the ‘novostroiki’ (new settlements) are not included in the official numbers.

2.5.2 Kyrgyz energy portfolio

For its energy needs, the Kyrgyz Republic’s total primary energy supply (TPES) – the sum of production and imports subtracting exports and storage changes – depends for 30,1 percent on RES (see Figure 2), of which about 100 percent is hydro-generated (International Energy Agency, 2017; Meyer, 2015; Kalybekovichy and Djumabekovich, 2012). The remaining 69,9 percent of the energy production is fueled by fossil the fossil fuels coal (30,9 percent), oil (30,2 percent) and natural gas (5,96 percent) (International Energy Agency, 2017). The Kyrgyz Republic imports 80 percent of the fossil fuels it consumes (Meyer, 2015). In terms the domestic consumption of coal, 59 percent is imported and of gas and petroleum use, 98 percent is imported (Meyer, 2015). Hence, of the fossil fuels that it consumes, only of coal a substantial part is being exploited domestically (Jefferson Institute, 2009). In terms of electricity, the Kyrgyz Republic relies heavily on domestic hydropower stations (HPSs) – in 2010, 91 percent of the generated electricity was generated at HPSs (Asian Development Bank, 2013: 147). During the winter seasons, the Kyrgyz Republic often relies on imports of electricity from in particular neighboring countries Kazakhstan and; in summer it exports excess electricity generation to Kazakhstan, Uzbekistan, China and Tajikistan (Putz, 2015; Meyer, 2015; Asian Development Bank, 2013: 147).

FIGURE 2: Primary Energy Consumption in 2014. Source: International Energy Agency, 2017

Renewables 30% Coal 31% Oil 30% Natural Gas 6% Other 3%

(19)

2.5.3 Bishkek’s energy supply problems

Since independence, the Kyrgyz energy system has been slowly moving from a centrally planned system to a market-based system (Bouzarovski, 2009: 453), however the government still owns the vast amount of shares of energy companies (Balabanyan et al., 2015). The Kyrgyz energy sector has faced problems in supplying sufficient energy, leading to an energy crisis – “a society-wide economic problem caused by a constricted supply of energy leading to diminished availability and increased price to consumers” (Grossman, 2015) – in 2007-2009 (Jefferson Institute, 2009; Liu, Pistorius, 2012). In Bishkek, this unmet energy demand results in an inability to provide reliable heat, warm water and electricity to its residents and factories, especially in the heating season (Jefferson Institute, 2009). For example, the urban area of Bishkek-Tokmok suffers an unmet heat and warm water demand of around 20-25 percent and a six-fold in electricity outages compared to the early 90s (Balabanyan et al., 2015). The unmet energy occurs in electricity outages, failures in the district heating (DH) – system “composed by one or more thermal plants and a distribution network of insulated pipes where hot water, superheated water or steam flow, delivering heat to the users through heat exchangers” (Verda and Colella, 2011: 4278) – in government induced restrictions on energy consumption, and in increasing electricity tariffs (Balabanyan et al., 2015; Juraev, 2009).

From literature, three main causes for this energy supply deficits can be pointed out. First, Marat (2015: 48) states that the governing of the Kyrgyz energy sector is highly corrupt. For example, illegal selling of hydro-electricity to neighboring countries made the water levels in the hydropower reservoirs drop excessively, resulting in the inability to meet the domestic energy demand (Marat, 2015; Juraev, 2009). As a result, the Kyrgyz Republic had to purchase coal, gas and electricity from neighboring countries to meet domestic energy demands (Marat, 2015: 50). The second cause of energy supply deficits can be found in the degrading energy infrastructure (Balabanyan et al., 2015; Liu and Pistorius, 2012; Jefferson Institute, 2009; Juraev, 2009). The energy system in Bishkek and the Kyrgyz Republic as a whole originates from the Soviet Union era and since Kyrgyz independence the energy infrastructure lacked sufficient maintenance, investments, planning and resource availability have caused the energy systems to deteriorate (Balabanyan et al., 2015). The energy companies, which are mostly government owned, account for huge financial losses, because energy is sold below the cost of production (Balabanyan et al., 2015). Therefore, sufficient financial means for maintaining and expending the systems lacked. Third, environmental challenges, such as dry years and harsh winters, have decreased the energy supply and increased the demand (Kurmanbek, 2016: 4; Jefferson Institute, 2009: 2; Juraev, 2009).

(20)

3. METHODOLOGY

This section elaborates on how data was collected and analyzed. Below I will first address the used fieldwork method, followed by the used analytical method.

3.1 Fieldwork Method

3.1.1 Semi-structured interviews

Although the sub-questions for this thesis were formulated in an exploratory way, my goal was to also discover the underlying motives. Hence, this research was exploratory and descriptive, but also explanatory. Because the research is both exploratory and explanatory, I conducted semi-structured interviews for gathering data (see Appendix 2), as Bryman (2012: 469) states that interviews are a good way to receive in depth information about processes and policies. The structured part of the interviews was formed by operationalizing the theoretical concepts in the literature, as well as contextual literature that reports on the situation of Bishkek’s energy systems. Hence, the structured part of the interviews provided the basis for the interviews. The unstructured part offered room to ask follow-up questions and had a more exploratory function.

The strategy regarding conducting the interviews was to start with introducing the topic and asking the respondents for consent regarding the terms of the interview (see Appendix 3). When consent was granted, I started the interviews by asking a general historical question first. I let the interviewees describe the situations and their experiences and were necessary I ask them an elaborative or follow up question. Second, I asked questions to determine which actors have been doing what to remedy Bishkek’s degraded energy infrastructure and unreliable energy supply. Third, I asked questions to determine what the municipality of Bishkek has been doing to address climate change priorities.

3.1.2 Sampling method and size

For sampling purposes, I used a technique that included both purposive and snowball sampling. Relevant respondents were selected based on the following criteria. First, the respondent had to consider herself/himself knowledgeable in the field of Bishkek’s energy governance in terms of either the institutional aspects, the social/historical/cultural aspects, the technical aspects, or the financial aspects. Second, depending on the type of respondent I needed at a certain time – I tried to alternate interviews with interviewees of different groups to be able to ask more specific questions in the latter interviews – the respondent had to consider herself or himself as part of one of specified target groups (see Table 1). These target groups I constructed beforehand going to Bishkek and adjusted it when I was in the process of conducting the interviews. To be able to construct a sound representation of the situation I selected interviewees based on the target groups that emerged in the literature and policy documents beforehand, and adjusted it on the basis of the first few conducted interviews. These target groups were: Government officials or energy company representatives, residential end consumers, academics, NGO representatives,

(21)

private sector representatives, and energy/environment consultants (Balabanyan et al., 2014: 5). As the government manages the energy infrastructure and owns the energy companies, they had to be part of this research. In the end I conducted 6 interviews with government officials or energy company representatives. The residential end consumers were significant to include in this research, as they could describe their experiences in times of energy supply deficits. In addition, they provided a general understanding about what residential end consumers prioritize in terms of energy use. 7 academics from several universities were interviewed, because they could potentially reveal a more critical view of e.g. the government institution’s induced policies. Furthermore, I conducted 7 interviews with NGO representatives as they are linked to the institutional reading of governance, as they often work together with the government, but also have their own agenda. In addition, 2 interviews with people who identified themselves as private sector business representatives were conducted. Finally, 3 consultants in the field of energy and sustainability were conducted. A total of 30 interviews were conducted and In addition, 2 conferences on were attended (see Table 1; see Appendix 5). One conference was hosted by the UNDP and concerned the developing of the Kyrgyz National Adaptation Program (NAP), the other conference was on developing private sector involvement in environmental problems such as climate change and biodiversity losses.

TABLE 1: Interview respondents per category

Type of respondent Interviewees

Government officials and energy company representatives 6

Energy end consumers 5

Academics 7

Private sector representatives 2

NGO representatives 7

Consultants 3

Total 30

After every interview, I asked the respondent if she or he could bring me into contact with another relevant respondent. Depending on what kind of respondent I needed, I asked for a specific target group. I transcribed several interview when I was abroad, to bring myself closer the data. I also kept brief daily and weekly diaries to reflect on the gathered data in order to detect emerging patterns in the data, but also to expose on what data I was still missing out on. Finally I reached saturation.

From the total of 30 interviews, I recorded 22 interviews, transcribed 20 interviews, took field notes of 7 interviews. 2 interviews I did not fully transcribe as I believed they lacked sound data. Of the two conferences I took field notes.

(22)

I asked my respondents if I could record the interview, as recorded interviews can be transcribed, coded, and used for quotes. To ensure ethical principle were not being harmed, I asked the interviewees for consent (Bryman, 2012: 135). I let the respondent read the consent form (Appendix 3) and asked for oral informed consent.

In the consent from I first introduce myself and explain the research topic, the purpose of the research, and relevance of executing the research. Second, I explain how I will handle the data in terms of anonymity and confidentiality. For example, I explain that I am the only person with access to the recordings and that I will delete the recordings after transcribing them. As the topic was often perceived as sensitive, I guaranteed anonymity by not mentioning their names in my thesis, but a vague description of their profession. If respondents did not agree on this, I asked them how they preferred to be referred to. When interviewees did not give me consent to record, I made key word notes during the interview and directly after finishing the interview, I wrote down everything that came to mind regarding the interview. Finally, I explained that the interviewee could withdraw from the interview at any moment if they felt uncomfortable; which might have gained some interviewees’ trust (Jacob, 2012).

After the interview, I offered the respondents to send them the recording or the transcript. Only 1 interviewee requested the recording. Furthermore I thanked the respondents for their time, information and consent. Finally I give them my contact information in case they changed their mind about anything concerning the interview, or in case they wanted to give me with additional information.

3.1.4 Interview setting

For ensuring smooth interviews, I tried to conduct the interviews in a quiet setting. Most of the time, I went to the respondent’s office, house, or met them in a café or restaurant, which were all silent enough for the recording. Some interviews got interrupted because the respondent received incoming phone calls, however I believe that did not influence the quality of the interviews.

When making the interview appointments, I specifically asked if they preferred an interpreter to join the interview (see Appendix 1). Surprisingly, only 10 of the 30 interviewees requested an interpreter. When an interviewee requested an interpreter, I arranged one.

3.1.5 Moment of research

The data gathering took place this year (2017) from 3rd_{of April until the 8}th_{of May. There were no}

occasions i.e. holidays, elections, extreme weather events, etc. that obstructed the research.

3.2 Analytical method

The gathered data then needed to be analyzed in order to detect analytic paths (Bryman, 2012: 565). This section will describe how I handled that process.

(23)

The three main components of this thesis (which relate to the sub-questions) are first, ‘historical governance of the residential energy systems in Bishkek’, which I examined by looking at the current state of the energy system. Second, the ‘strategies to remedy the degraded unreliable infrastructure and energy supply deficits’, and third, ‘prioritization of climate change and the idea of transition’ (see Figure 3). These are operationalized into categories by means of indicators derived from the literature and interviews (see Table 2; see Appendix 2), hence, in the process, I added and removed some indicators. FIGURE 3: Operationalization scheme

In order to operationalize (see Table 2), I took the three independent variables and defined the categories and indicators by firs reviewing the literature. Hence, I researched the first independent variable – current state characteristics – by looking at which energy systems are related to residential use in Bishkek; how they are shaped in terms of infrastructure; how they came to be; which resources they use (and why). In addition, I examined the governance structures of these energy systems (e.g. the rate of decentralization, denationalization and inclusiveness). In addition, I assessed finances of the sector in terms of prices of energy production, consumption, and profitability of the energy companies. Furthermore I studies the landscape features which include the natural potential for generating energy from RES, and socio-cultural/historical features that influence the energy sector characteristics. Finally, I assessed the spatial embeddedness and path dependency by examining the carbon lock in rate and the financial/political lock-in rate.

I determined the energy supply remedies variable by looking which investments were made (and by who) to ensure sufficient energy supply. In addition I assessed the government strategies to ensure sufficient energy supply by examining if and why the following actions were taken: increase efficiency in energy production, liberalization and denationalization of the energy sector, energy pricing politics, corruption fighting politics and educational practices. In order to not only include the institutional reading, I also looked at what households, communities and the private sector did to remedy the energy supply deficits. Governance of Bishkek’s Residential Energy Sector Climate Change Prioritization Energy Supply Remedies Current Energy Sector Characteristics

(24)

To determine the idea of transitioning towards a low carbon future and climate change prioritization, I examined the mitigation and adaptation practices with a focus on the actual implementations, but also on the ‘will to improve’ among the different actors at different levels. From this I determined if climate change is governed by urban climate change experimentation in the residential energy sector.

TABLE 2: Operationalization categories and indicators

Independent variable Categories and indicators

1. Current energy sector

characteristic - Residential energy production, supply and consumption à Heating/water system à Electricity system

- Resource use and proximity

à RES share: hydro, wind, solar, geo-thermal à Fossil fuel share: coal, gas, oil

à Proximity of resources - Governance of energy system

à Top-down/bottom-up (Power relations between actors) à Scale and inclusiveness of energy system

- Financial

à Prices of energy production/consumption à Profitability of energy companies

- Landscape

à natural features: fossil fuels, RES potential à socio-cultural/historical features

- Spatial embeddedness and path dependency à Carbon lock-in

à Financial/Political lock-in 2. Remedy supply remedies - Investments

à RES investments

à investments in current system - Government strategies

à Efficient energy production

à Liberalization/denationalization of energy sector à Rise energy prices

à Corruption control à Education practices - Household strategies

à independent household energy system à Efficiency energy use

- Community strategies - Private strategies 3. Transition concerning climate

change prioritization

- Mitigation practices à willingness to mitigate - Adaptation practices

à willingness for adaptation

- Urban Climate Change Experiments when:

à purposively seeks to capture new forms of learning or experience

à purpose is: climate change mitigation, and/or adaptation. (social and/or technical) à conducted at the urban level (existing or imagined)

3.2.2 Transcribing

As mentioned, I fully transcribed 22 interviews. In interviews that were joined by an interpreter, I transcribed the English spoken dialogue. I transcribed the interviews myself, because I believed this would bring me closer to the interview data. In addition, I believed it would help me to identify key themes in an early stage.

(25)

3.2.3 Coding

I coded all the interview transcripts by using the computer software ATLAS.ti. The coding itself entailed “looking for repetitions, indigenous typologies or categories, metaphors, transitions in the topics similarities and differences between segments, missing data (silence, avoidance and gaps) and theory related material” (Bernard & Ryan, 2003: 88-93). I applied labels based on the operationalization table (see Table 2) to fragments of interview transcript. Afterwards I grouped all coded fragments based on which category and indicator I found they belonged to. Having the codes ordered gave me the opportunity to identify patterns and themes in the data (Bryman, 2012: 569). Finally, I identified and linked reoccurring themes and key messages in order to structure the findings. By doing this, the structure of the results section became clear.

3.2.4 Analyzing data and Presenting findings

I used a thematic analysis to construct an index of central themes and subthemes to order the codes (Bryman, 2012: 579). I presented the findings in a logically story that address all the above-mentioned indicators. In the conclusion section I reflected back on the three sub-question related independent variables and triangulate the findings. I supported the findings by presenting quotes from the interview transcripts. To maintain focus on the research question, I only presented and discussed the findings that relate to the research questions (Bryman, 2012: 689).

3.2.5 Reliability and validity

In qualitative research, it is not possible to statistically test the findings, however, to ensure reliability and validity, I used a strategy based on a design by Noble and Smith (2015). I reflected on the personal biases of the interviewees in daily and weekly diaries I kept during my stay in Bishkek. In these diaries I also reflect on my thought processes regarding I acknowledging biases in the data (both in data gathering and in the data analyzing) or my personal bias in analyzing and interpreting de data. Furthermore, by reflecting on the data on a weekly basis, I ensured no remaining gaps in knowledge were left open. An important strategy was to interview respondents from different groups, in order to receive information to generate a comprehensive view of the situation. By interviewing respondents from the groups alternately, I made sure that I could ask emerging questions to the right respondent. In addition, I stored and handled the data confidential to ensure anonymity and hence safety of the respondents. Finally, I triangulated the data and findings to answer the research question.

3.2.6 Limitations

One of the main limitations was that respondents in the group government officials and energy company representatives often did not allow me to record the interview. In general, it was very hard to find respondents of this group, so finding alternative respondents was no option due to limitations. Eventually, I succeeded to find a sufficient number of respondents in this group. Some of the interviews with respondents from this group were rather uncomfortable; I felt like they were suspicious about my

(26)

research, which could have affected the answers they gave. Furthermore, the language was a minor limitation. Most respondents chose to do the interview in English, but on some occasions, I got the feeling that they could not express themselves as good or as detailed as they wanted. The interviews with an interpreter were sometimes a bit difficult as it was not always clear when the interpreter could start translating. After the first interviews, I asked the interviewee to pause after every two or three sentences to minimize data getting lost in translation. Furthermore, my positionality might have been a constraint in approaching respondents or in getting answers during the interviews. As I mentioned above, government officials were hard to reach and often did not agree on participating in an interview.

(27)

4. RESULTS

4.1 Historical Dependence in the Residential Energy Sector Governance

In this chapter, I will explain the governance of the energy sector in Bishkek since the final years of the Soviet Union and hence how the current governance has been shaped by the historical socio-economic, political, and physical context of place. In order to understand how energy is currently governed in Bishkek, multiple respondents emphasized that one first needs to reflect on the city’s (historical) contexts. Therefore, I will first address how the energy governance arrangement in Bishkek is currently still dependent on the former Soviet Union’s governance structure and corresponding infrastructure. For example, when the Kyrgyz Republic gained independence in 1991, the socio-economic and political system changed drastically with an economic downturn in the following decade. Second, after discussing the historical socio-economic and political setting, I will explain the governance of the energy sector in Bishkek by addressing the current political structure, the legislations currently in force, the finances regarding the energy sector, and the abundant corruption practices. Finally, the overall approach to climate change governance in the Kyrgyz Republic will be clarified by looking at the global climate change commitments and domestic strategies to reach them.

4.1.1 Bishkek’s Soviet Union energy sector heritage

Until August 31, 1991, the area that forms the Kyrgyz Republic had been under the political and cultural control of the Soviet Union for about 60 years, after which the social, economic, and political ideology abruptly changed (Igmen, 2012: 3). In this section, I will explain how it came to be that Bishkek in many ways still depends on the energy governance arrangement and infrastructure which dates back to the time of the Soviet Union. First, I will assess the energy infrastructures and resource use, and then the centrally planned or ‘top-down’ way of governing the energy sector.

4.1.1.1 Soviet Union: Sharing of resources and infrastructure

In the Soviet era, the energy system and its resource systems were governed in a top down manner. An energy efficiency consultant described the Soviet energy sector governance as follows:

“In Soviet Times, everything was planned from the very central level, so the population had the right to receive proper heat, hot water, and electricity. Therefore, everything was central, production and the distribution” (Source: interview transcript 11).

Sehring (2009: 75) and Walters (2016: 82) state that energy governance in the Kyrgyz Republic inherited its structure from the Soviet Union, which can be described as a hierarchical model with strong centralization and a lack of horizontal coordination, dominance of the presidential administration, weak position of local authorities, a top-down command style, resistance to new management methods and

Climate change prioritization in remedying energy supply deficits : the case of Bishkek