Energy Research & Social Science 94 (2022) 102884 Contents lists available at ScienceDirect Energy Research & Social Science journal homepage: www.elsevier.com/locate/erss Original research article Solar energy at the peri-urban frontier: An energy justice study of urban peripheries from Burkina Faso and South Africa Roberto Cantoni a, *, Federico Caprotti b, Jiska de Groot c a Institute of Environmental Science and Technology, Universitat Autònoma de Barcelona, Edifici ICTA-ICP, Carrer de les Columnes s/n, Campus de la UAB, 08193 Cerdanyola del Vallès, Barcelona, Spain b College of Life and Environmental Sciences, Geography, University of Exeter, Mail Room - Old Library, Prince of Wales Road, Exeter EX4 4SB, United Kingdom c African Climate and Development Initiative (ACDI), University of Cape Town, Upper Campus, Geological Sciences Building, Level 6 - 13, Library Road, Rondebosch 7700, South Africa A R T I C L E I N F O A B S T R A C T Keywords: Most of the global population that lack access to electricity services live in sub-Saharan Africa. Peri-urban areas Solar energy of large African cities, often characterized by the presence of informal settlements, exist in a kind of ‘scalar Urban political geography limbo,’ unable to benefit from either access to the city grid or from programs aimed at the electrification of rural Energy justice areas. In addition, in those areas where lack of electricity access is common, energy poverty combined with South Africa Burkina Faso proximity to the grid leads to a greater likelihood of illegal energy supply arrangements. In this fieldwork-based Ouagadougou study, conducted through population surveys and interviews in the peripheries of Ouagadougou, Burkina Faso, Cape Town and Cape Town, South Africa, we employ a hybrid theoretical framework, based on work in urban political ecology and energy justice, to analyze the situation of electricity access in the two areas. We find that the planned scale, scope, and technological design of solar energy projects in peripheral areas are crucial in determining whether and how a project will be beneficial for local communities. This study provides guidance beyond academia to national and international policymakers and executives of renewable energy companies, as well as tools for a more in-depth assessment of energy justice issues. 1. Introduction prioritize or extend services to these areas due to a range of factors: the prospect of payment difficulties on the part of low-income users, high Sub-Saharan Africa is home to around three quarters of the world's costs of electricity connections discouraging potential customers, and population without access to electricity: according to a 2019 report by the modest amounts of energy consumed by households, resulting in the International Energy Agency (IEA), 580 million people out of the lower earnings. At the same time, while peri-urban areas may exhibit 770 million globally who were unable to use this vital commodity lived some characteristics typical of rural areas, they cannot be considered as in sub-Saharan Africa [1]. While this number has been decreasing since wholly rural, as their economies and population densities are different. 2013, mainly thanks to the extension of electricity grids and the spread In cases where electricity management is distributed among several of decentralized energy infrastructure, the COVID-19 pandemic reversed agencies depending on geographical scale (e.g., one agency being this trend [2]. Following the 2020 inception of the pandemic, and the competent for cities, another for rural areas), conflicts of competence consequent redirection of state budgets to tackle it, funding for elec­ may arise between these. Additionally, peri-urban areas may be treated tricity access projects has decreased. This especially affected non- as marginal areas when compared to more established urban centers on electrified rural and peri-urban areas. It is not uncommon for peri- which investments are concentrated. urban populations to live close to high-voltage lines but to be unable In peri-urban areas, energy poverty combined with proximity to the to connect (a condition known in the literature as being ‘under-the- grid leads to a greater likelihood of illegal energy supply arrangements, grid’) [3]. such as energy theft, illicit connections, and tampering with electricity Peri-urban areas are frequently characterized by a lower socio- meters [4–6]. The urgency of finding a solution to the twin, linked economic level than urban areas. Utilities are often reluctant to problems of illegal connections and energy poverty, makes it necessary * Corresponding author. E-mail addresses: roberto.cantoni@uab.cat (R. Cantoni), f.caprotti@exeter.ac.uk (F. Caprotti), jiska.degroot@uct.ac.za (J. de Groot). https://doi.org/10.1016/j.erss.2022.102884 Received 20 July 2022; Received in revised form 9 November 2022; Accepted 10 November 2022 Available online 17 November 2022 2214-6296/© 2022 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by- nc-nd/4.0/). R. Cantoni et al. Energy Research & Social Science 94 (2022) 102884 to study peri-urbanity from an energy justice perspective, and to ask underlining how heterogeneous infrastructural configurations [14] are how to improve access in these low-income areas. This paper analyzes part of the process of energy bricolage and fragmentation seen in the peri-urban areas in two large African cities (Cape Town, South Africa, peri-urban [15]. Thus, a UPE-informed perspective helps us to under­ and Ouagadougou, Burkina Faso). The paper establishes a dialogue stand how scalar power relations are also relational, shifting, and between research on energy justice and literatures from urban geogra­ intertwined with solar energy in peri-urban areas [16]. phy and urban political ecology (UPE) that have engaged with peri- At the same time, they enable a consideration of energy landscapes in urbanity and scale. The next section outlines our hybrid theoretical a spatial context that is rooted in place: the energy landscapes we framework. consider are the interfaces between energy practices at a variety of scales, and the local communities with which we carried out our 2. Scale, peri-urbanity, energy justice in African energy studies research. As Castán Broto and Sanzana Calvet [17, p. 281] show, the lens of landscape “shows that place-based experiences and discourses also 2.1. Scale and peri-urbanity shape the political ecology of energy.” Additionally, several aspects of the socio-material politics involved in the production of energy land­ Our focus is on the peri-urban as a key site of production and scapes (notably around distribution, policymaking procedures, and the negotiation of renewable energy landscapes. Much of the energy, urban reproduction of inequalities) [18] are linked to their energy justice studies and development literature continues to reproduce a binary characteristics, and it is this connection that justifies the use of an energy distinction between the city and rural areas in terms of the types of justice framework, coupled with UPE. It is to these facets that the paper projects and approaches in either category. Peri-urban areas escape now turns. facile categorizations as wholly urban or rural. In part, this is because they are rapidly changing “transition spaces” [7, p. 1], where a mix of 2.2. Crossing the peri-urban and energy justice aspects uses and economies blend the urban and the rural. Their fluidity means that they are not easily integrated within urban planning and develop­ Over the last two decades, scholars have critically analyzed the tools ment frameworks [8]. Finally, peri-urban areas are often the location of used by policymakers and energy analysts in assessing energy projects. urban informality, which is not easily classifiable as formally urban due This strand of research, broadly known as energy justice theory, has to infrastructural and socio-economic disconnection and segregation typically been formulated in two ways: through tenet- [19] or principle- from the formal city. based approaches [20,21]. The former identified the three tenets of In focusing on the peri-urban, we base our analysis on a dynamic and distributive, procedural, and recognition justice, depending on whether relational concept of scale. We argue that considering the nexus between the focus was on an energy system's distribution of costs and benefits, justice and energy in peripheral urban areas necessitates a close the decision-making processes involved, and which actors are recog­ engagement with the multiple scalar dimensions and processes that nized as legitimate stakeholders, respectively. The latter is based on the produce the peri-urban as a continuously shifting reality. A relational eight principles of availability, affordability, due process, transparency understanding of scale helps to understand sites of de- and re- and accountability, sustainability, intergenerational equity, intra­ territorialization within and outside the city [9]. Additionally, consid­ generational equity, and responsibility [21]. Due to the scope and ering the multiple ways in which scale is performed is key to under­ comparative approach of our paper, in our analysis we focus on the three standing the tensions and power relations that course through energy tenets approach. landscapes. For example, while public authorities and agencies may be As well as using an energy justice lens, we root our analysis in the seen as hierarchically organized on a spectrum from national to local, under-studied peri-urban context. While interest in energy justice in when considering energy in sub-Saharan Africa, scale can also be un­ renewable energy projects in Africa has been high over the last decade, derstood to be produced on a matrix between energy centralization and Africa's peri-urban areas have not been a common and explicit target of decentralization [10]. In both South Africa and Burkina Faso there re­ research in energy studies. This is probably a result of a ‘scalar limbo’: mains a key tension between centralization (which is involved in the that is, the difficulty of attributing purely urban or purely rural char­ national, modernist project of energy strategy and electricity planning), acteristics to peri-urban areas, and the consequential geographical and and decentralization (including through renewables, independent sociological complexity of these areas in terms of energy use patterns power producers, and off-grid energy projects). and of the choice of the energy system that could most adequately satisfy When considering the broad mix of urban processes and forms of their needs. This study attempts to partially fill this gap. tenure existing in peri-urban areas (from formal to informal urbaniza­ Existing studies on renewable energy and social sciences in peri- tion, as well as small-scale agriculture, animal husbandry, and light urban locations in Africa cover a broad range of topics, such as house­ industry), linked to overlapping formal (the grid as well as formal re­ hold willingness to pay for green electricity in Tigray, Ethiopia [22]; the newables) and informal energy infrastructures (including off-grid re­ potential of solar water heater program in South African townships [23]; newables but also other forms of energy), it can be seen that energy transitions in urban informal settlements in Kenya, Uganda, approaching the peri-urban energy landscape from the lens of energy South Africa, and Zimbabwe [24,25]; urban energy governance in South reveals these areas to be what Munro and Bartlett [11, p. 72] describe as Africa's Renewable Energy Procurement Program [26]; investment, a “post-modern energy landscape, one that is neither ‘traditional’ nor service reliability and social sustainability in electricity access in ‘modern,’ rather the fragmented product of different social and eco­ Mozambique [6]; and the rejection of energy technologies by low- nomic processes.” It is precisely because of the collage-like nature of the income urban dwellers in Cape Town, South Africa [27]. However, peri-urban that being sensitive to its scalar dimensions is necessary only a small subset of these studies includes aspects of energy justice: when considering the justice aspects of renewable energy-focused Samarakoon [28] analyzed energy injustices in Malawi's off-grid solar development in urban peripherality. market; Boamah et al. [4] studied institutionalized corruption in solar Drawing on UPE, the emergence of specific energy landscapes in PV systems in Ghana and Kenya; and Cantoni et al. [29] examined solar peri-urban areas can be seen as parts and parcel of a metabolic process in energy in Burkina Faso. which infrastructure is a central component [12], and which is charac­ Before explaining our methodology, we need to provide an epistemic terized by materialities that are expressed both temporally and spatially, warning. This study employs a Western theorization of the energy justice and that flow through institutions, regulations, technical systems and concept: this is a known problematic issue (see [30,31]). Work on the encounters with everyday lives [13]. Analyzing the production of energy application of non-Western theorizations of justice to energy-related landscapes in peri-urban areas also enables us to link the paper's focus on cases has recently been conducted: for example, Sanusi and Spahn energy justice to the dynamic contexts of Ouagadougou and Cape Town, [32], and Pellegrini-Masini et al. [33] have both worked on sub-Saharan 2 R. Cantoni et al. Energy Research & Social Science 94 (2022) 102884 Africa's Ubuntu philosophy. This is an area that deserves broader population, mostly located in rural areas and informal settlements in scholarly attention, a full exploration of which falls outside of the scope urban and peri-urban areas [41]. This is key, because lack of grid access of this paper. A second limit of this research is that the numerosity of our is often linked to significant inequalities: c. 47 % of households without surveys, and thus their analytical power, differs significantly (100 re­ electricity suffer from energy poverty countrywide [42], and there are spondents in the Burkinabe sample versus 223 respondents in the South knock-on effects on health, education, socio-economic outcomes and African one). However, a sample of 100 respondents is considered other aspects. Solar electricity has been identified as a potential short- to acceptable in literature for drawing statistically meaningful conclusions mid-term solution for electrifying households in rural and informal [34,35]. Finally, in terms of its applicability, we are aware that the areas. While South Africa has policies and strategies (such as the NDP) economic conditions of South Africa are rather unique, and that would aimed at increasing the number of grid-connected large-scale solar ar­ limit the applicability of this study; however, Burkina Faso's economy rays, it also has decentralized, embedded generation schemes aimed at has many points in common with the economies of surrounding coun­ consumption near the site of generation [41]. tries (West African countries in particular). Therefore, we believe that In South Africa, grid-connected solar generation is mostly aimed at the results of that case study can be applicable to other countries in the households in formal urban areas. In informal settlements, solar energy subregion with a similar degree of economic development. technologies are mostly standalone systems (or, in some areas, based In the next section, in order to situate readers, we provide back­ around mini-grids). The issue of solar energy provision to informal ground information on the national energy policies of South Africa and settlements needs to be seen within a complex web of social, economic, Burkina Faso, and outline the structure of their energy sectors. political, and financial interrelations. As Kovacic et al. [44] point out in their critique of energy transitions research in informal contexts, a key 3. National energy policies and energy sector structures obstacle is the definition of ‘problems’ from outside, and the reduction of these ‘problems’ to technical energy access issues. 3.1. South Africa 3.2. Burkina Faso South Africa's generation and distribution profile is dominated by Eskom, a state-owned utility which is responsible for producing 96 % of Burkina Faso is a landlocked country with limited access to energy its energy [36] and 90 % of its electricity. Municipalities, redistributors, resources. Biomass is the main source of energy consumption for private and private generators supply the remaining 10 % [37]. In turn, Eskom households (c. 90 % of the total). This is a major national difference with is regulated by the National Energy Regulator of South Africa (NERSA) South Africa, with high electrification rates and generation largely based [36]. Eskom transmits electricity to South Africa's main users (indus­ on electricity from cheap coal (the country is reliant on thermal fossil trial, mining, commercial, agricultural, residential, municipal). In areas fuels for about 70–80 % of total electricity generation) [45]. In Burkina that are serviced by Eskom, most often it is the municipalities that Faso, 62.3 % of the urban population and 4.7 % of the rural population redistribute electricity to households and businesses within the munic­ had access to electricity in 2018. The country's National Plan for Eco­ ipal area. This, however, is not the case of Qandu Qandu, which is nomic and Social Development for the period 2016–2020 set several directly supplied by Eskom (as also occurs in some other peri-urban quantitative targets for access to electricity and renewables by 2020, areas). including: i) an increase in the contribution of renewables to total en­ There have been calls for change in South Africa's energy generation ergy production (from 6.4 % in 2015 to 30 %); ii) an increase in installed and distribution structure, and of the balance of primary energy sources, national electricity capacity (from 300 MW in 2015 to 1 GW) and spanning at least two decades. The country's initial public electricity electrification rate (from 59.9 % in urban settings, 3.1 % in rural ones in strategy, published in 2011, was the Integrated Resource Plan (IRP). 2015 to 75 % and 19 %, respectively; and iii) the reduction of electricity This focused on transitioning away from coal, and institutional change, cost by 25 CFA Fr per kWh ([46]: 64–65; [47]). with a 2030 target [38]. The IRP planned for an additional 17.8 GW of For the longer time horizon to 2030, the National Action Plan for renewable power, as well as a smaller increase in coal generation (6.3 Renewable Energy (PANER) also set quantifiable targets for electrifi­ GW). Most of the projected increase in renewable energy was based on cation and renewables [48]: i) increase in energy access to 42.4 % of the solar and wind, as well as nuclear and imported hydroelectricity [39]. population; ii) a projected capacity of renewable power plants of 318 Updates to the IRP dampened the initially strong focus on renewables. MW; iii) increase in the share of renewables as part of total installed Another attempt to incentivize renewable generation in the country was capacity to 36 % (excluding grid-linked electricity imports, the target is the Renewable Energy Independent Power Producer Procurement Pro­ 27 %); iv) 10 % increase in installed capacity of off-grid renewables gram (REIPPPP). The initiative resulted in the award of 6327 MW of systems; and v) increase to 26.9 % of the rural population served by projects, for a total investment of US$ 20.5 billions. Solar PV accounted autonomous renewables-based systems. Furthermore, the Burkinabe for 34 % of this investment [40]. government recently developed an energy policy consisting of two pil­ In parallel with the IRP, the country's National Development Plan lars (transition towards renewables, and energy efficiency) aimed at (NDP) aims for a significant change in South Africa's energy generation increasing the energy supply and bridging the urban-rural gap in energy profile, with the aims being a reduction of dependency on coal, and access. This trend has been institutionalized through the creation of the more reliance on renewable energies and gas (NPC, 2019). More National Agency for Renewable Energy and Energy Efficiency (ANEREE) broadly, the NDP aims for 90 % grid access by 2030 [36], with 95 % of in late 2016. It is expected that the planned higher reliance on renewable the population having access to electricity (on-grid and off-grid com­ sources will be favored by Burkina Faso's considerable solar potential bined) by the same year [37]. Between 1999 and 2005, 3.2 million and solar market development. households were connected to the grid by Eskom, and 1.7 million by Legally, the energy sector in Burkina Faso is regulated through Law local government; in 1996–2011, the percentage of households using n. 014–2017/AN of April 20, 2017, whose main innovations with electricity for lighting rose from 58.2 to 85 % [41]. To enable the poor to respect to the existing situation can be summarized as: i) reaffirming the be connected to electricity, since 2003, households are entitled to 50 liberalization of the production and distribution segments (once kWh of Free Basic Electricity (FBE) per month. Notwithstanding these SONABEL's monopoly); ii) taking into account the Economic Commu­ gains, grid access is unequally spread, with 77.4 % of White households nity of West African States' regulations, aimed at building a sub-regional electrified, but only 54.4 % of Black African, 47.9 % of Colored, and electricity market; iii) introducing dispositions to promote renewable 59.3 % of Indian/Asian households electrified by 2015 [42], a remnant energy and energy efficiency. The law gave legally binding value to the of the Apartheid system. A significant number of households (3.5 m), National Action Plans 2015–2030 for Renewable Energy (the mentioned remain without access to the grid [43]: this represents c. 15 % of the PANER) and Energy Efficiency (PANEE), both adopted in 2015. In 3 R. Cantoni et al. Energy Research & Social Science 94 (2022) 102884 particular, with respect to under-the-grid settlements, PANER developed highway and managed separately. Only the latter area, closer to the a strategy consisting of: i) optimizing and increasing the use of low- solar power plant, has been the object of our analysis (Fig. 2). voltage grids; and ii) increasing the coverage rate by extending me­ In the spring of 2019, a survey of 100 households was conducted in dium- and low-voltage grids in already-electrified areas, but also by the southern part of Zagtouli. Aimed at assessing energy use patterns, it electrifying new rural areas. included both quantitative and qualitative questions. Distributional as­ pects were investigated through socio-demographic and economic in­ 4. Methods formation, as well as by collecting information on the energy supply and expenditure of each household. With regard to procedural energy justice We use a concurrent mixed methods approach to analyze two cases and as recognition, respondents were asked whether they had been in South Africa and Burkina Faso. For each area, we employed a aware of, or involved in, deliberative or legal procedures resulting from qualitative-quantitative population survey both to extract the main the allocation of land for construction for the plant. sociodemographic characteristics of the populations and to acquire data The survey was paralleled by a set of 45 interviews. Of these, 10 were on the uses of different forms of energy. In parallel, we conducted in­ shorter interviews (up to 1 h) conducted with residents of Zagtouli; 35 terviews with survey respondents as well as with other relevant actors were longer interviews (1 to 2 h), conducted with entrepreneurs in the detailed below. Our intention for this study was mainly to collect and field of renewable energy, and solar energy in particular (21), repre­ analyze opinions from both citizens living in the studied areas and ex­ sentatives from NGOs (4), utilities and national agencies' executives (5), perts from outside: that is why this part of analysis predominates in our consulting firms (2), and academics (3). The interviews were mostly Results and discussion section. The South African case was analyzed conducted in Ouagadougou and its province. The interviewed residents with a focus on discourses and policies around solar power specifically, were selected according to their willingness to take part in deeper ex­ and renewables more broadly, at various scales of governance [49]. To changes following the survey's administration. The 35 participants in the engage with governance approaches to solar power on a national, pro­ longer interviews were selected principally from a list of energy com­ vincial, and municipal scale, interviews were conducted with re­ panies operating in Burkina Faso, retrieved from the directory of com­ spondents from a range of government, energy authority, civil society, panies in Burkina Faso (https://www.goafricaonline.com/bf/annuaire), and academic respondents (more information on this below). In addi­ then expanded through snowball sampling. The purpose of the in­ tion, this paper is grounded in analysis of the development and terviews was: 1) to understand the legislative and institutional frame­ deployment of a solar mini-grid project in Qandu Qandu, an informal work governing the Burkinabe energy market and the main obstacles to settlement of c. 3500 households, founded in 2018 in the partially the development of the solar energy market; 2) to collect information informal Khayelitsha area of Cape Town (Fig. 1). Qandu Qandu is for the about projects operating in the area; and 3) to gain information on the most part not connected to the grid, but there are a range of energy state of the art of ongoing research in the renewable sector. The in­ infrastructures present in the settlement, including illegal and generally terviews were analyzed qualitatively using NVivo®. The survey data unstable grid connections (c. 10 % of households), multiple privately- were analyzed through SPSS Statistics® software. operated solar mini-grids not connected to the grid, and energy supply The energy justice analysis was conducted by distinguishing between and distribution networks focused on charcoal, paraffin and diesel used the three aspects of distributive, procedural and recognition aspects. within the confines of the shacks that make up the settlement. Qandu Each of these three aspects was evaluated qualitatively from interview Qandu itself is precariously located in an area prone to regular flooding. and survey responses; distributive and recognition aspects were also In total, 54 interviews were conducted. The categories of in­ assessed quantitatively through specific survey questions (e.g., whether terviewees were: a) governance actors (22 interviews total). These respondents were connected to the electrical grid; whether they could included those working in government departments (5), provincial collect/buy energy sources according to their needs; whether they were governments (3), municipal governments (10), and officials from aware of the nearby solar plant). Eskom, the public utility (4); b) energy and development-related civil In the next section, we report our most significant results and discuss society organizations (11 interviews); c) scholars (7 interviews); d) them. residents of Qandu Qandu (14 interviews). These residents were a sample of those contacted for research on a British Academy-funded 5. Results and discussion research project (UWB190088) in the Urban Infrastructures of Well­ being portfolio. The interviews with residents were conducted from 5.1. Case study: Qandu Qandu, Cape Town, South Africa February 2020 to June 2021, during the timeline of the project. The other interviews (with respondent categories a–c) were carried out in The historical and geographical specificities that have produced the 2017–18, as part of a UK ESRC/Newton Fund-National Research Foun­ Qandu Qandu settlement range from the historical trajectories of ur­ dation of South Africa project (ES/N014138/1). In addition, to provide banization processes in and around Cape Town,1 to tensions between community-level qualitative data on solar power in Qandu Qandu, a state mandates to provide energy and other infrastructures to citizens, to survey was used in addition to the above-mentioned interviews. The issues with land tenure and ownership that hamper this mandate. survey totaled 223 respondents and was conducted between November Nonetheless, focusing on the specific context of Qandu Qandu directs 2020 and March 2021: it was aimed at constructing a dataset of resi­ attention to two key aspects of UPE as it is expressed in the Southern dents' energy practices. Interviews were analyzed using NVivo, while for city, namely everyday practices, and diffuse forms of power [52]. By the purposes of this article, descriptive and contextual statistics from the briefly using the example of energy-related risks and hazards, we aim to survey were used. show how fragmented access to energy is key to the production of The Burkinabe case is based on fieldwork conducted in Zagtouli, an socionatural hazardscapes of informality, which are in turn linked to en­ area in the outskirts of Ouagadougou. Zagtouli is administered as a ergy justice. neighborhood of Ouagadougou; in the most recent census (2006) it had One of the points raised by multiple respondents centers around the around 27,000 inhabitants [50]. Zagtouli is home to West Africa's issues of energy, safety and risk connected with the use of dangerous largest on-grid solar power plant: a 33 MW facility owned and built by sources of energy. As one resident interviewed in March 2021 and one in the French electricity engineering firm Cegelec. The plant, financed by February 2021 put it: France's Development Agency and the European Union, has been oper­ ational since November 2017. It is scheduled to supply 5 % of the country's total electricity consumption, about 55 GWh per year [51]. 1 These are heavily influenced by South Africa's legacy of Apartheid, the Zagtouli is split into a northern and a southern area, separated by the N1 system of racial segregation that ended in 1994. 4 R. Cantoni et al. Energy Research & Social Science 94 (2022) 102884 Fig. 1. Map of Qandu Qandu, Cape Town (source: QGIS and Google Maps). “Few weeks back I think there were eight hokies [shacks] that were Town alone, for example, there were over 1195 shack fires that saw burnt because of paraffin and then three guys were found dead in emergency response by the city in January–September 2021 [53]. In those burnt hokies including a small child who was about nine to ten addition, the lack of reliable energy access means that there is little years old.” lighting in the community after dark, which increases the sense of fear and looming danger. This can be understood as part and parcel of the “You see these wires [informal grid connections]; many houses have process of “slow infrastructural violence” [54] recurring harms gradu­ burnt because of these wires. The illegal electricity is not safe.” ally accrue and concentrate along gendered, racialized or other lines. This shows how the density of the settlement, coupled with the use of In contrast, solar mini-grids were perceived as positive developments flammable materials and energy practices around combustible fuels and because of the potential for reducing energy-related risks affecting the informal grid connections, produce a highly relational risk landscape whole community. As one resident put it, the electricity delivered that links energy use by individuals or by a single household to diffuse, through the mini-grid: real risks that affect the whole community. This is part of a broader hazard landscape affecting informal settlements everywhere: in Cape 5 R. Cantoni et al. Energy Research & Social Science 94 (2022) 102884 Fig. 2. Map of Zagtouli (source: QGIS and Google Maps). “is unlike the electricity that we stole from those lines across above formal energy grid connections, but high-tension power lines exist near, our houses. There will be less people that will get killed by this and run through, the settlement. Thus, the Qandu Qandu community electricity and kids as well […]. Some of us have to rebuild our exists in an under-the-grid space. This is borne out by our survey: out of houses because of the electricity […] because we just do it yourself. 223 respondents, 36.3 % thought that lack of access to electricity was You see it is dangerous.” the biggest problem faced by the community. Only 6.7 % respondents had access to electricity at all, through means such as grid access (formal Thus, the political ecology of energy in Qandu Qandu reveals how or informal), car batteries, generators or solar home systems, and all energy and community are intertwined in a relational web. These re­ respondents relied on paraffin/kerosene and/or liquefied petroleum gas lations, in turn, include and incorporate different scales of governance, (LPG). This is also reflected in lighting in the home: out of all residents, as well as specific aspects of the ecology of risk, hazards and fear only 5.8 % used electric light bulbs, and 3.6 % used fluorescent strip [55,56]. lights, while the rest used paraffin/kerosene lamps and/or candles; 82.5 % of the surveyed residents reported that lighting was never adequate to 5.1.1. Energy justice analysis meet their needs. Likewise, most (93.7 %) residents reported dissatis­ Distributive justice is, in part, about access to energy services as well faction with the use of a mix of energy sources (paraffin/kerosene, LPG, as the siting of physical energy infrastructure [57]. Qandu Qandu has no 6 R. Cantoni et al. Energy Research & Social Science 94 (2022) 102884 and electric stoves) for cooking. This had knock-on effects on environ­ This highlights the sense of a lack of recognition not only of the mental quality and safety: only one respondent had a hood or extractor respondent as a citizen with rights, but of the respondent as a human fan for evacuating exhaust gases and vapor from cooking. Given this being. In this context, the isolation mentioned by the respondent is not general context, 93.3 % of the residents stated a desire to leave the simply the isolation of Qandu Qandu as an off-grid, informal commu­ settlement, a clear indication of a large deficit in terms of distributive nity, but isolation as non-recognition of the community's existence and justice related to energy and other essential services. its dignity within the wider Cape Town and South African polity. The introduction of solar mini-grids to the community had several The installation of solar mini-grids in Qandu Qandu occurred in a impacts that can be tied to distributional justice. Firstly, it was seen as a context whereby procedural justice is already sorely lacking, in that safe, stable and reliable source of energy that enabled not simply the residents have little access to decision-makers at different scales of consumption of electricity, but the addressing of multiple needs linked governance, and little to no access to mechanisms whereby their energy to electricity, as pointed to by a resident interviewed in February 2020: needs and rights can be expressed, represented and actioned. Addi­ tionally, the introduction of solar mini-grids, and pricing packages “If we can have electricity here in Qandu Qandu, I think our life can associated with these, into an already diverse economic mix around be better in many ways [...] we will have something to charge our energy stacking practices, introduces informational complexity that is cell phones. So, we will be in social media. We will know whatever difficult to assess: residents are effectively “being made responsible for that is happening around us. So, electricity is the best.” provisioning their electricity in the midst of precarity” but at the same time a procedural justice issue is that “the energy poor seldom have the At the same time, there are disadvantages associated with gaining requisite knowledge or support to make decisions that advance their access to solar-powered mini-grids. Apart from lesser capacity than that interests” [28, p. 8]. However, this is to some extent offset by the fact available through grid access, the mini-grids installed in Qandu Qandu that, in Qandu Qandu, the private sector provider opened a shack-based deliver direct current (DC), while most appliances available on the office within the community, providing an anchor point for the services market use alternating current (AC). This means that community it sells there, and providing a point of information for interested resi­ members either have to purchase and replace specific items of electrical dents. Thus, while residents have little to no access to adequate infor­ equipment, or rely on the DC-compatible ones supplied by the private mation at the scale of local government, they can access private sector- sector utility that runs the mini-grids. Thus, there is a distributional related solar mini-grid information with more ease, which is a proce­ justice deficit related to the additional costs accruing to community dural improvement. As a resident noted: members who need to adapt to connect to a DC system. This, in turn, is linked to a broader and multi-scalar issue of spatialized justice and eq­ “I was happy when I see it [...] two guys they say [...] come and see uity in service access and delivery: although some of the residents of this, they say you can rent this I said it is fine, I knew I was no longer Qandu Qandu can now access electricity by paying for services delivered going to pay 50 Rands or 200 for the guy you know who would come through solar mini-grids, most poor citizens connected to the grid in to collect the money for [...] that illegal electricity I knew I was going serviced formal or informal urban areas would otherwise have been to last the whole month.” eligible to a monthly 50 kWh of FBE. Thus, there remain critical issues The above quote also underlines one of the key procedural, around continued lack of access to benefits and entitlements that cannot information-based aspects of solar mini-grid provision: the enhance­ be realized by residents due to the off-grid nature of informal peri-urban ment of the ability to budget and forecast personal energy (and by settings. consequence, household economic) expenditure. This in turn reduces Injustices associated with recognition include not only mis­ continuous stress and instability-related aspects of energy precarity. recognition and inappropriate representation at different levels of pol­ Many of the aspects of energy justice aspects highlighted in the South itics and decision-making, but also cultural domination and disrespect African case are also reflected in the Burkinabe case, to which we now [19]. All of these are operational in Qandu Qandu and emerge as key turn. themes in community members' interactions with political structures. One of the ways in which misrecognition works in the Qandu Qandu 5.2. Case study: Zagtouli, Ouagadougou, Burkina Faso context is the way in which the community's needs only seem to be acknowledged by politicians around the time of elections (in order to Three areas were identified within Zagtouli Sud: western, central, stimulate voting), and then swiftly forgotten. This was mentioned and eastern. The central area (35 respondents) is characterized by a frequently by respondents. The pattern that is identified concerns visits higher economic level than the others, as was easily inferred during the by those seeking election or promoting voting for a specific party: access population survey by a visual analysis of the size and quality of housing, to services is offered as a vague promise in return for appropriate voting the presence of family-owned cars, watchdogs, families of servants often behavior. And yet, as one interviewed resident put it in March 2021: living in their landlords' households, better roads, and more modern “We vote, when we are done doing that they don't deliver on our amenities (TVs, fans, etc.). In the eastern and western areas (65 re­ wishes. So it feels like we are not being taken seriously. How to put it, spondents), the prevailing energy patterns are closer to other rural areas like they don't care about the issues of poor people should I say or the in the region: households are characterized by a general lack of elec­ wellbeing of poor people [...]. Their politics [laughing], when it gets tricity, or in rarer cases, by autonomous electricity production through to vote they say vote you'll get electricity will get water [...] when the solar home systems. time comes those don't happen.” While illegal connection to the grid, and the related possibility of fires, were not perceived as much of an issue in Zagtouli as in Qandu This underlines the mechanism of what respondents feel are Qandu, the main stated risks were those generated by the long-term use exploitative interactions, aimed at using the community (and the votes of biomass, and the economic constraints linked to gas use. With respect they represent) with no positive result, and indeed no acknowledgement to both firewood and charcoal, by far the most frequently highlighted between election times, representing non-recognition, as underlined by issue were cooking-related smoke and heat (30 %), and their negative another resident in February 2020: respiratory and visual health effects (23 %), mostly for women, who “We have to be like treated like human being, all in all as we are were the ones in charge of preparing meals in 96 % of all cases. Seasonal human being. [....]. We feel we have been isolated. In terms of san­ scarcity of wood (8 %) and charcoal (11 %), together with wood hu­ itations and all necessities that all the people need. […] We do not midity during the rainy season (7 %) affecting fuel quality, were also belong anywhere.” frequently mentioned issues. As for gas, problems included seasonal scarcity (28 %), affordability (12 %), possible leaks (9 %), and bottle 7 R. Cantoni et al. Energy Research & Social Science 94 (2022) 102884 filling scams (3 %). declared the plant had had no impact on their daily energy use patterns Not unexpectedly, the predominance of each of the problems and habits. That is telling, especially when we compare it to the case of mentioned depended on the sub-area where residents lived, and as such, Qandu Qandu, where the construction of mini-grids actually led to an on their socio-economic status. Residents of the central area, who mostly improvement in access to electricity. While it is true that the power plant make a limited use of firewood, mentioned negative consequences from was not planned to have a direct impact on the Zagtouli community (but its use less frequently than residents of the peripheral areas (χ2 (1, N = rather to act as a grid stabilizer), it is meaningful that, considering the 100) = 17.55, p < 0.001); on the contrary, they mentioned gas-related plant's international resonance (the French and Burkinabe Presidents issues more frequently (χ2 (1, N = 100) = 7.03, p < 0.001);. Thus, were present at its inauguration), so many residents did not even know especially for worse-off residents, “socionatural hazardscapes of infor­ about it. This reflects a lack of authorities' initiative to inform the pop­ mality” in Zagtouli materialize in health issues. This status disparity ulation about (let alone, involve them in) the development plans leads us to analyze energy justice aspects in Zagtouli, and the relevance envisaged in the area. of the solar power plant to those aspects. When asked about the possibility of grid connection in the future, residents of the peripheral areas showed disillusionment. A young 5.2.1. Energy justice analysis resident from the western area, who lives alone in an adobe and straw As can be inferred by the section above, access to energy is consid­ hut, stated: “SONABEL doesn't care about us: we are only a few, so it's erably varied within Zagtouli, with the central area's residents enjoying not economical for them to connect us. And anyway, connection costs the opportunity to benefit from a wider range of sources than residents would be too big to bear for us”. This resident owns a small solar panel, of the peripheral areas. An example of the range of possibilities offered which, he says, is barely sufficient to charge his mobile phone. That is to better-off residents, is the following statement by a resident of the the only electrical appliance he owns in his hut. The feeling of dis­ central area, head of a family of five and owner of a detached house with empowerment and abandonment from the public company that appears a small garden: from his words is not uncommon among his fellow residents. A young, married housewife, mother of three, living in the eastern area, talked “We built our home here and we bought a large solar panel. We along similar lines: “I know about the solar plant, but it's not for us. It's decided not to connect to SONABEL [i.e., the grid] because we don't for the city [Ouagadougou].” When asked if she had a solar panel in her need it. You see the electricity post there? We could have connected household, she mentioned that they used to have one to charge mobile but it was too expensive. And then you have brownouts all the time phones, but it got broken and it was not possible to repair it. with the grid”. The breakdown of the small solar panel points out a further issue, related to the low quality of solar equipment bought by lower-income This evidences both the opportunity to choose which source of en­ residents. This was also emphasized by several local entrepreneurs, ergy to use, and a preference for off-grid, a common finding among and may lead to residents' loss of confidence in renewable energy. There residents of the central area. It reflects the results of a study on Ghana by emerges from these accounts not only a lack of recognition [19], but a Boamah and Rothfuß [58]: the energy security resulting from owning general resignation to not benefiting from the power plant or, more solar PV systems–especially large ones–implies the possibility of enjoy­ generally, from the public electricity supply. ing the benefits of lighting even when the rest of the on-grid community Perhaps both a lack of knowledge about the ongoing solar plant- cannot: namely, during brownouts and blackouts. PV systems, therefore, related works and the above-mentioned feeling of disempowerment act as a proxy for high socioeconomic status. contributed to the fact that no major protest developed against the The situation appears quite different in the two other areas of Zag­ construction of the plant in Zagtouli. Of the 100 respondents inter­ touli, where worse-off residents do not benefit from a similar range of viewed for the survey, only one, whose house was in the western area of possibilities or hold any hope of grid connection in the near future. The Zagtouli (some 2.5 km northeast of the plant), complained that part of structural difference between the central and peripheral areas, and the the lands used for the plant were communal lands on which one of his condition that Cantoni et al. (2021) have referred to as the “unattainable cousins used to take his goats to pasture, and that he would have liked to proximity” of electrification [29] for the eastern and western areas, is a receive some compensation. Several other residents–especially those reflection of the Zagtouli's peri-urban location. Many residents of the with a higher educational level, and who used to get their information central area have bought land in Zagtouli because it was cheaper than in from the press or radio, expressed local pride: “Zagtouli has become a the capital city. They are typically higher-income professionals (medical unique location in Africa. We are at the forefront of solar energy in the doctors, engineers, agents of SONABEL, state officers), working in country. Of course we are proud!”, affirmed a high-school teacher and Ouagadougou and commuting daily between Zagtouli and the capital. Zagtouli resident. They enjoy salaries that are higher than those of the residents living in Nevertheless, focus group discussions with local authorities and the western and eastern areas, who work in Zagtouli in lower-paid jobs household representatives, conducted by a World Bank project team in (farmers, tailors, butchers, sellers, petty traders). 2019, revealed minor hostility from the local community towards con­ Thus, the very location of Zagtouli, at the border of the urban and the struction work (related to the transport of machinery during plant rural, contributes to the maintenance of a status disparity and to construction, which produced dust and intense noise, and led to public generate energy injustices. It also generates problems of territorial nuisance and frustration). In these discussions, residents argued that jurisdiction. The jurisdiction of the Burkinabe national electrification they should have been compensated for the damage suffered. Moreover, agencies means that Zagtouli is formally dependent, for its electrifica­ the constructing firms' failure to employ resident labor during con­ tion, on SONABEL. However, there are currently no plans to extend the struction created some tensions [29]. network to electrify the entire territory of Zagtouli Sud. In particular, the In the case of Zagtouli, the strategy pursued by the national gov­ lower-income areas would need electrification programs more akin to ernment involved the local authorities at the borough hall level, rural, off-grid projects managed by the Burkinabe Agency for Rural whereas, because of the misrecognition mentioned earlier, it is apparent Electrification (ABER). However, because Zagtouli is not classified as a that neither citizens nor NGOs were part of the decision-making process rural community, it does not fall under ABER's competence. Thus, an during the planning. While interviewed residents did not mention hav­ operational, scalar limbo is generated, which leads to the persistence of ing participated in public or private consultations, most did not see that an unsatisfying condition of unattainable proximity. as a problem: “No, I was not consulted. Why should I have been? I know One of the most significant results of the population survey con­ nothing about solar energy”, a local petty trader stated. “I'm happy that ducted in Zagtouli is that about a third of the respondents (32 %) were they [the constructors] are using those lands: there's nothing growing not aware of the existence of the solar power plant, while a further 62 % there, and if they manage to use them fruitfully, all the better”, remarked 8 R. Cantoni et al. Energy Research & Social Science 94 (2022) 102884 a local farmer. These statements reflect, first, a more general belief in injustices similar to those in the South African case, while, from a top-down decision-making by competent authorities, and second, a view distributional and procedural point of view, we note significant differ­ of local lands as a resource to be put to good use, devoid of any other (e. ences. Regarding the first point, the low local impact of the solar power g., symbolic) value. Unlike the Moroccan case studied by Rignall [59], plant is evident. The neighborhood/village remains, for the most part, in where it was the national authorities that tried to construct land planned a condition of unattainable proximity, not only for economic but also for for solar installation as marginal to facilitate investment, in Zagtouli it structural reasons (lack of a grid connection plan by the public utility), was the residents themselves that supported such a view. while wealthier households can choose to connect to the grid or use autonomous (solar) systems. As for procedural injustices, the perceived 6. Conclusion marginality of the area, combined with the lack of a public information/ consultation campaign, has led to socio-political invisibility of the power In this study, we have hybridized two strands of theory–urban po­ plant itself. litical ecology (UPE) and energy justice–to study the multi-scalar dy­ Our work highlights the intertwining of questions of UPE and energy namics of renewable energy projects in two African contexts that have justice: in particular, the uneven and inequitable landscapes UPE studies one geographical element in common: peri-urbanity. This element is are often also characterized by inequities in the distribution of energy, as significant not only as peri-urban territories are often considered polit­ well as by the marginality – better, marginalization – of the communities ically and socially marginal, but also because the phenomenon of diffuse that inhabits them. This latter aspect is evidence of procedural and peri-urbanization is an expanding feature, particularly in African met­ recognition injustices regarding who has and ought to have a say in ropolises. These show a recent historical trend of demographic growth, energy decision-making. In our analysis, we have shown how frag­ acting as magnets of attraction from rural areas. It is therefore urgent to mented access to energy is key to the production of socionatural haz­ analyze the dynamics of these transitional spaces at a time when inter­ ardscapes of informality, which are in turn linked to both energy justice, national climate policies are leading to an expansion of the renewable and the spatialization and ordinary everyday experiences of energy energy industry. landscapes, as seen through a UPE lens. Based on considerations from our UPE approach, we chose to analyze We believe that our work can provide useful information not only to energy systems comprised by different governance and technical con­ national policymakers to improve the energy justice aspects of future figurations, power, management modes and purpose, highlighting crit­ renewables projects, but also, more broadly, to international policy­ ical distributional, procedural and recognition aspects in terms of energy makers operating in countries with economic conditions similar to those justice. Drawing on UPE, our work has underlined how, on the one hand, studied. Specifically, the analysis of the Burkinabe case can improve energy landscapes in peri-urban areas are complex and relational, knowledge on the outcome of similar projects in other West African related to the processes, governance approaches and socio-technical countries; the South African case will contribute to a broader under­ systems through which they emerge, and at the same time are deeply standing of the criticalities of small-scale, decentralized projects. The spatial. The spatiality of the solar energy landscapes in the African peri- results of this study will also be of help for renewable energy companies, urban zone is a crucial topic for consideration, since these landscapes are which often dimension their plants on the basis of demographic and effectively rooted in place and in locality. Place is where the more economic considerations of the target areas, but with no prior analysis of diffuse and multi-scalar networks that make up energy landscapes are issues of distributive, procedural and recognition justice. interfaced with everyday lives, where tensions and opportunities around We call for further research linking political ecology and energy solar energy and its consequences are vocalized and operationalized. justice perspectives in the study of renewable energy landscapes in peri- Spatially and technically, as can be seen from the examination of urban settings. Based on our work, we underscore Jenkins et al.'s [57] South African and Burkinabe energy policies, a strong expansion of findings that procedural justice can be improved through promoting renewable infrastructures is expected: not only in rural areas, where three aspects of inclusion, namely, a) mobilizing local knowledge for just solar home systems are the most widespread technology, but also in the outcomes; b) expanded information disclosure, and c) enhanced repre­ peri-urban areas of large metropolises, where governments' technolog­ sentation at the level of different institutions and scales of governance. ical preferences and service delivery strategies are more complex and at This will enable scholars and policymakers to move past technical times unclear. However, as appears from the different choices made by ‘problem-solving’ approaches from ‘elsewhere’, and towards more the two national governments, approaches to the electrification of peri- contextually specific entry points that engage with peri-urban commu­ urban areas differed: in the South African case, a largely decentralized nities at different scales. approach was developed, with mini-grids as energy production units and operating almost wholly separately from, and in parallel to, existing CRediT authorship contribution statement energy networks and policies. In the Burkinabe case, the preference was to use the peri-urban territory for electricity production to stabilize the Roberto Cantoni: Conceptualisation, Methodology, Data curation, national grid, while the territory itself where the power plant was built, Investigation, Resources, Supervision, Writing - Original Draft, Writing - being under-the-grid, did not benefit. Review & Editing, Revising, Project administration, Funding We have highlighted how, in the case of Qandu Qandu, even though acquisition. the installed mini-grids were perceived as a positive novelty compared Federico Caprotti: Conceptualisation, Methodology, Data curation, to the existing reality of disconnection from the grid or illegal connec­ Investigation, Resources, Supervision, Writing - Original Draft, Writing - tions with fire risk, this did not prevent the simultaneous emergence of Review & Editing, Revising, Funding acquisition. energy injustices relating to various aspects: distributional ones (the Jiska de Groot: Resources, Funding acquisition, Review and Editing. incompatibility between the types of electricity used by mini-grids and appliances on the market); procedural ones (the limited knowledge that would prevent residents from taking part in decision-making processes); Declaration of competing interest and recognition ones (the perception of being politically manipulated in the run-up to elections, without this leading to effective consideration of None of the authors of the submitted article has any conflict of in­ their energy needs). Thus, the spatiality and place-based granularity of terest to disclose. the energy landscape becomes crucial when considering the facets of energy justice: it is here that approaches rooted in UPE and energy Data availability justice overlap and can produce useful and constructive analysis. As far as the Zagtouli case is concerned, we detected recognition Data will be made available on request. 9 R. Cantoni et al. Energy Research & Social Science 94 (2022) 102884 Acknowledgements Informal Settlements: case studies from South Africa and Zimbabwe, Energy Res. Soc. Sci. 56 (2019), 101202, https://doi.org/10.1016/j.erss.2019.05.012. [26] M. Davies, M. Swilling, H.L. Wlokas, Towards new configurations of urban energy The research presented in this paper was supported by: ERA4CS, an governance in South Africa’s renewable energy procurement programme, Energy ERA-NET Co-fund action initiated by JPI Climate, funded by BMBF (DE), Res. Soc. Sci. 36 (2018) 61–69, https://doi.org/10.1016/j.erss.2017.11.010. FORMAS (SE), BELSPO (BE), and IFD (DK) with co-funding by the Eu­ [27] A.N. Haque, C. Lemanski, J. de Groot, Why do low-income urban dwellers reject energy technologies? Exploring the socio-cultural acceptance of solar adoption in ropean Union’s Horizon 2020 Framework Program (Grant no. 690462); Mumbai and Cape Town, Energy Res. Soc. Sci. 74 (2021), 101954, https://doi.org/ the Economic and Social Research Council (Grant no. ES/N014138/2); 10.1016/j.erss.2021.101954. the British Academy (Grant no. UWB190088); Jiska de Groot’s Research [28] S. Samarakoon, The troubled path to ending darkness: energy injustice encounters in Malawi’s off-grid solar market, Energy Res. Soc. Sci. 69 (2020), 101712, https:// Leader Fellowship, funded by the AXA Research Fund. doi.org/10.1016/j.erss.2020.101712. [29] R. Cantoni, L. Skræp Svenningsen, S. Sanfo, Unattainable proximity: solar power References and peri-urbanity in Central Burkina Faso, Energy Policy 150 (2021), 112127, https://doi.org/10.1016/J.ENPOL.2020.112127. [30] B.K. Sovacool, M. Burke, L. Baker, C. Kumar Kotikalapudi, H. Wlokas, New [1] International Energy Agency (IEA), Africa Energy Outlook 2019, IEA, Paris, 2019. frontiers and conceptual frameworks for energy justice, Energy Policy 105 (2017) https://www.iea.org/reports/africa-energy-outlook-2019. 677–691, https://doi.org/10.1016/j.enpol.2017.03.005. [2] International Energy Agency (IEA), Energy access. https://www.iea.org/topics/ [31] G. Bombaerts, K. Jenkins, Y.A. Sanusi (Eds.), Energy Justice across Borders, Energy energy-access (accessed 15 September 2022). Justice across Borders, Springer, New York, 2020, https://doi.org/10.1007/978-3- [3] World Bank, Infrastructure, growth, and productivity in sub-Saharan Africa, in: 030-24021-9. Africa's Pulse 15, World Bank, 2017, https://doi.org/10.1596/26485. [32] Y.A. Sanusi, A. Spahn, Exploring marginalization and exclusion in renewable [4] F. Boamah, D.A. Williams, J. Afful, Justifiable energy injustices? Exploring energy development in Africa: a perspective from western individualism and institutionalised corruption and electricity sector “problem-solving” in Ghana and African Ubuntu philosophy, in: G. Bombaerts, K. Jenkins, Y.A. Sanusi (Eds.), Kenya, Energy Res. Soc. Sci. 73 (2021), 101914, https://doi.org/10.1016/j. Energy Justice Across Borders, Energy Justice Across Borders, Springer, New York, erss.2021.101914. 2020, https://doi.org/10.1007/978-3-030-24021-9_14. [5] M. Koepke, J. Monstadt, F. Pilò, K. Otsuki, Rethinking energy transitions in [33] G. Pellegrini-Masini, F. Corvino, L. Löfquist, Energy justice and intergenerational southern cities: urban and infrastructural heterogeneity in Dar Es Salaam, Energy ethics: theoretical perspectives and institutional designs, in: G. Bombaerts, Res. Soc. Sci. 74 (2021), 101937, https://doi.org/10.1016/j.erss.2021.101937. K. Jenkins, Y.A. Sanusi (Eds.), Energy Justice Across Borders, Energy Justice Across [6] D. Salite, et al., Electricity access in Mozambique: a critical policy analysis of Borders, Springer, New York, 2020, https://doi.org/10.1007/978-3-030-24021-9_ investment, service reliability and social sustainability, Energy Res. Soc. Sci. 78 13. (2021), https://doi.org/10.1016/j.erss.2021.102123. [34] J.F. Hair, W.C. Black, B.J. Babin, R.E. Anderson, Multivariate Data Analysis, 8th [7] A. Wandl, M. Magoni, Sustainable planning of peri-urban areas: introduction to the ed., Cengage Learning, United Kingdom, 2018. special issue, Planning Practice & Research 32 (1) (2017) 1–3, https://doi.org/ [35] J.E. Bartlett, J.W. Kotrlik, C.C. Higgins, Organizational research: determining 10.1080/02697459.2017.1264191. appropriate sample size in survey research, Inf. Technol. Learn. Perform. J. 19 [8] M.G. Mortoja, Y. Yigitcanlar, S. Mayere, What is the most suitable methodological (2001) 43–50. approach to demarcate peri-urban areas? A systematic review of the literature, [36] NPC, National Development Plan 2030. Our Future - Make it Work. Pretoria, Land Use Policy 95 (2020), 104601, https://doi.org/10.1016/j. National Planning Commission, Available at: https://www.gov.za/sites/default/f landusepol.2020.104601. iles/gcis_document/201409/ndp-2030-our-future-make-it-workr.pdf, 2019. [9] A. Paasi, Place and region: looking through the prism of scale, Prog. Hum. Geogr. [37] DoE SA, The South African Energy Sector Report 2019. Pretoria, Department of 28 (4) (2004) 536–546, https://doi.org/10.1191/0309132504ph502pr. Energy, Available at: http://www.energy.gov.za/files/media/explained/2019 [10] R.P. Thombs, When democracy meets energy transitions: a typology of social -South-African-Energy-Sector-Report.pdf, 2019. power and energy system scale, Energy Res. Soc. Sci. 52 (2019) 159–168, https:// [38] DoE SA, A Survey of Energy-Related Behaviour and Perceptions in South Africa. doi.org/10.1016/j.erss.2019.02.020. Pretoria, Department of Energy, Available at: http://www.energy.gov.za/files/me [11] P. Munro, A. Bartlett, Energy bricolage in northern Uganda: rethinking energy dia/Pub/Survey%20of%20Energy%20related%20behaviour%20and%20percept geographies in sub-Saharan Africa, Energy Res. Soc. Sci. 55 (2019) 71–81, https:// ion%20in%20SA%20-%20Residential%20Sector%20-%202012.pdf, 2012. doi.org/10.1016/j.erss.2019.04.016. [39] DoE SA, Electricity Regulations on the Integrated Resource Plan 2010-2030. [12] J. Silver, Disrupted infrastructures: an urban political ecology of interrupted http://www.energy.gov.za/IRP/2010/IRP_2010.pdf, 2010 (accessed 8 April electricity in Accra, Int. J. Urban Reg. Res. 39 (5) (2015) 984–1003, https://doi. 2021). org/10.1111/1468-2427.12317. [40] A. Ferreira Pinto, REIPPP: One of the World's Best Renewable Energy Tenders, but [13] G. Cederlöf, Out of steam: energy, materiality, and political ecology, Prog. Hum. There's Room for Improvement, PV Magazine, 2021, 30 September. Geogr. 45 (1) (2021) 70–87, https://doi.org/10.1177/030913251988462. [41] S. Essex, J. de Groot, Understanding energy transitions: the changing versions of [14] M. Lawhon, D. Nilsson, J. Silver, H. Ernstson, S. Lwasa, Thinking through the modern infrastructure ideal and the ‘energy underclass’ in South Africa, heterogeneous infrastructure configurations, Urban Stud. 55 (4) (2018) 720–732, 1860–2019, Energy Policy 133 (2019), 110937, https://doi.org/10.1016/j. https://doi.org/10.1177/0042098017720149. enpol.2019.110937. [15] M. Mouton, The Philippine electricity sector reform and the urban question: how [42] C.G. Monyei, A.O. Adewumi, K.E.H. Jenkins, Energy (in)justice in off-grid rural metro Manila's utility is tackling urban poverty, Energy Policy 78 (2015) 225–234, electrification policy: South Africa in focus, Energy Res. Soc. Sci. 44 (2018) https://doi.org/10.1016/j.enpol.2014.11.005. 152–171, https://doi.org/10.1016/j.erss.2018.05.002. [16] H. Ahlborg, A.J. Nightingale, Theorizing power in political ecology: the 'where' of [43] E.L. Meyer, O.K. Overen, Towards a sustainable rural electrification scheme in power in resource governance projects, J, Polit, Ecol, 25 (1) (2018) 381–401, South Africa: analysis of the status quo, Energy Rep. 7 (2021) 4273–4287, https:// https://doi.org/10.2458/v25i1.22804. doi.org/10.1016/j.egyr.2021.07.007. [17] V. Castán Broto, M. Sanzana Calvet, Sacrifice zones and the construction of urban [44] Z. Kovacic, J.K. Musango, K. Buyana, A. Ambole, S. Smit, B. Mwau, M. Ogot, energy landscapes in Concepción, Chile, J. Polit. Ecol. 27 (1) (2020) 279–299, S. Lwasa, A. Brent, Building capacity towards what? Proposing a framework for the https://doi.org/10.2458/v27i1.23059. analysis of energy transition governance in the context of urban informality in sub- [18] J.L. Rice, An urban political ecology of climate change governance, Geogr. Saharan Africa, Local Environ. 26 (3) (2020) 364–378, https://doi.org/10.1080/ Compass 8 (6) (2014) 381–394, https://doi.org/10.1111/gec3.12134. 13549839.2020.1849075. [19] D. McCauley, V. Ramasar, R.J. Heffron, B.K. Sovacool, D. Mebratu, L. Mundaca, [45] M. Moner-Girona, K. Bódis, T. Huld, I. Kougias, S. Szabó, Universal access to Energy justice in the transition to low carbon energy systems: exploring key themes electricity in Burkina Faso: scaling-up renewable energy technologies, Environ. in interdisciplinary research, Appl. Energy 223–234 (2019) 916–921, https://doi. Res. Lett. 11 (8) (2016) 84010, https://doi.org/10.1088/1748-9326/11/8/ org/10.1016/j.apenergy.2018.10.005. 084010. [20] B.K. Sovacool, Energy Ethics. Justice and the Global Energy Challenge, Palgrave [46] Finance and Developmentcollab Ministry of Economy, National plan for Macmillan, Basingstoke, 2013. economic and social development (PNDES) 2016-2020. http://extwprlegs1.fao. [21] B.K. Sovacool, M.H. Dworkin, Global Energy Justice. Problems, Principles, and org/docs/pdf/bkf166486.pdf, 2016. Practices, CUP, Cambridge, 2014. [47] Ministry of Energy, Plan d’action national de l’initiative energie durable pour tous [22] T. Arega, T. Tadesse, Household willingness to pay for green electricity in urban SE4ALL 2015-2020/2030. https://www.se4all-africa.org/fileadmin/uploads/se4a and peri-urban Tigray, northern Ethiopia: determinants and welfare effects, Energy ll/Documents/Country_AAs/Burkina_Faso_Agenda_d%E2%80%99Action_de_L% Policy 100 (2017) 292–300, https://doi.org/10.1016/j.enpol.2016.10.022. E2%80%99initiative_Energie_Durable_Pour_Tous.pdf, 2015. [23] C. Curry, J.A. Cherni, M. Mapako, The potential and reality of the solar water [48] Ministry of Energy, Plan d'action national des énergies renouvelables (PANER) heater programme in South African townships: lessons from the City of Tshwane, 2015-2020/2030. http://www.se4all.ecreee.org/sites/default/files/paner_bfa_ Energy Policy 106 (2017) 75–84, https://doi.org/10.1016/j.enpol.2017.03.028. juilllet_15_final.pdf, 2015. [24] A. Ambole, et al., Mediating household energy transitions through co-design in [49] F. Caprotti, S. Essex, J. Phillips, J. de Groot, L. Baker, Scales of governance: urban Kenya, Uganda and South Africa, Energy Res Soc. Sci. 55 (2019) 208–217, translating multiscalar transitional pathways in South Africa’s energy landscape, https://doi.org/10.1016/j.erss.2019.05.009. Energy Res. Soc. Sci. 70 (2020), 101700, https://doi.org/10.1016/j. [25] D. Conway, B. Robinson, P. Mudimu, T. Chitekwe, K. Koranteng, M. Swilling, erss.2020.101700. Exploring hybrid models for universal access to basic solar energy Services in 10 R. Cantoni et al. Energy Research & Social Science 94 (2022) 102884 [50] Humanitarian Data Exchange, Liste des villages du Burkina Faso – recensement [55] M. Davis, Ecology of Fear: Los Angeles and the Imagination of Disaster, 2006. https://data.humdata.org/dataset/liste-des-villages-du-burkina-faso-recense Metropolitan Books, New York, 1998. ment-2006, 2006. [56] M. Davis, Planet of Slums, Verso, London, 2006. [51] E. Bellini, Burkina Faso Commissions 33 MW PV Plant, PV Magazine, 2017, 28 [57] K. Jenkins, D. McCauley, R. Heffron, H. Stephan, R. Rehner, Energy justice: a November. conceptual review, Energy Res. Soc. Sci. 11 (2016) 174–182, https://doi.org/ [52] M. Lawhon, E. Ernstson, J. Silver, Provincializing urban political ecology: towards 10.1016/j.erss.2015.10.004. a situated UPE through African urbanism, Antipode 46 (2) (2014) 497–516, [58] F. Boamah, E. Rothfuß, From technical innovations towards social practices and https://doi.org/10.1111/anti.12051. socio-technical transition? Re-thinking the transition to decentralised solar PV [53] V. Lali, There Have Been Nearly 1,200 Shack Fires in Cape Town This Year, electrification in Africa, Energy Res. Soc. Sci. 42 (2018) 1–10, https://doi.org/ Claiming Over 80 Lives, GroundUp, 2021, 16 September. 10.1016/j.erss.2018.02.019. [54] Y. Truelove, H.A. Ruszczyk, Bodies as urban infrastructure: gender, intimate [59] K.E. Rignall, Solar power, state power, and the politics of energy transition in pre- infrastructures and slow infrastructural violence, Polit. Geogr. 92 (2022), 102492, Saharan Morocco, Environ Plan A 48 (3) (2016) 540–557, https://doi.org/ https://doi.org/10.1016/j.polgeo.2021.102492. 10.1177/0308518X15619176. 11