Keywords

1 Introduction

Access to clean, reliable, affordable and sustainable energy is one of the greatest sustainability challenges currently facing sub-Saharan African (SSA) countries and has been encapsulated in Sustainable Development Goal 7 (SDG 7) (IEA 2014) (Chaps. 1, 2, 7 Vol. 1). It has been estimated that up to 730 million people in SSA (or 80% of the population) have no access to electricity or clean cooking fuels, relying instead on traditional solid biomass fuels such as charcoal and fuelwood for their domestic needs (IEA 2014; Zulu and Richardson 2013) (Chaps. 1–2 Vol. 1). In fact biomass fuels have traditionally dominated household energy needs, especially for cooking (Karanja and Gasparatos 2019; IEA 2014) (Chaps. 2, 7 Vol. 1). However, the sourcing and use of conventional biomass fuels such as fuelwood and charcoal for cooking has various sustainability impacts as discussed below (Chaps. 2, 7 Vol. 1).

On the one hand, the use of traditional biomass fuels is associated with substantial negative health outcomes due to indoor air pollution from smoke and other pollutants produced during fuel combustion (Fullerton et al. 2008; Amegah and Jaakkola 2016). Furthermore, fuelwood collection and cooking in inefficient stoves can increase the risk of injuries (Das et al. 2017) and time diversion from educational and income-generating activities (Edelstein et al. 2008; Fullerton et al. 2008; Langbein 2017). Often these impacts are gender-differentiated as women and girls spend disproportionately higher amount of time collecting fuelwood and cooking in poorly ventilated areas (Ezzati and Kammen 2002; Karanja and Gasparatos 2019). Furthermore, reliance on traditional biomass puts a major recurring burden on household budgets, especially in areas experiencing fuelwood scarcity (e.g. from overexploitation or climate change) or where fuelwood collection is not possible (e.g. in cities) (Openshaw 2010; Karanja and Gasparatos 2019).

Furthermore, some of the prevailing fuelwood harvesting practices have been linked to negative environmental outcomes related to deforestation, forest degradation, carbon stock loss and biodiversity loss (Chidumayo and Gumbo 2013; Hosonuma et al. 2012; IPBES 2018; Karanja and Gasparatos 2019) (Chaps. 1–2, 7, 9 Vol. 1). Broader landscape degradation from unsustainable fuelwood harvesting can increase the vulnerability of the rural poor through the loss and degradation of ecosystem services related to livelihoods and food security,Footnote 1 creating thus a vicious cycle of biomass dependence and poverty (Cerutti et al. 2015; Chidumayo and Gumbo 2013; Mugo and Ong 2006; IPBES 2018). At the same time, charcoal production (in inefficient kilns) and use (in inefficient stoves) has been associated with significant greenhouse gas (GHG) emissions (Chidumayo and Gumbo 2013; Okoko et al. 2017) (Chap. 2 Vol. 1). Dependence on charcoal and fuelwood has been estimated as accounting for 1.9–2.3% of global GHG emissions (Bailis et al. 2015), which is comparable to global emissions from the aviation sector.

Conversely the biomass fuel sector (and especially the charcoal sector) is a vital source of livelihoods especially for many of the rural poor that produce and sell charcoal either as a primary or a secondary income-generating activity (Jones et al. 2016; Smith et al. 2017; Woollen et al. 2016). In this sense, fuelwood (and the derived charcoal) are important provisioning ecosystem services, catering for multiple human needs in urban and rural contexts of the continent (Woollen et al. 2016; Zorrilla-Miras et al. 2018). Charcoal production sometimes is also a means of sustainably managing and/or eradicating invasive tree species such as Prosopis julifloraFootnote 2 (FAO 2018).

Eradicating the energy poverty associated with high reliance on traditional biomass fuels is a key for attaining sustainable development in SSA. For example, it has been suggested that the large-scale promotion and uptake of clean cooking options (and the simultaneous phasing out of traditional biomass fuels) can have multiple sustainability benefits (IEA et al. 2019; Maes and Verbist 2012)Footnote 3 (Chaps. 1–2 Vol. 1). There have been many pilot and large-scale programmes during the past decades in SSA, promoting improved biomass stoves and stoves using electricity, liquefied petroleum gas (LPG), biogas, ethanol and briquettes (World Bank 2017). At the same time, there have been similar attempts to promote more efficient charcoal production technologies (Adam 2009; Schure et al. 2019). If successful, such initiatives and efforts can contribute substantially to multiple SDGs such as SDG1 (No Poverty), SDG2 (Zero Hunger), SDG3 (Good Health and Wellbeing), SDG5 (Gender Equality), SDG7 (Affordable and Clean Energy), SDG12 (Responsible Consumption and Production), SDG13 (Climate Action) and SDG15 (Life on Land) among others (Chap. 2 Vol. 1).

However, reducing household dependence on traditional cooking fuels through the promotion and uptake of clean cooking options, despite its high positive sustainability outcomes, remains a major sustainability challenge in SSA (IEA 2014; World Bank 2014). Besides factors related to costs (e.g. high upfront stove cost, recurring fuel costs, repair/change costs) that often hinder the switch from traditional to modern fuels/stoves (World Bank 2014), there are several other funding constraints complicating the effective implementation of clean cooking programmes (Karanja et al. 2020). In many SSA contexts, there is a lack of readily available modern cooking options, motivation, and incentives for fuel switch, as well as financing and other support measures (Zhou et al. 2011; World Bank 2017). Furthermore, some stove designs fail to consider the needs and cultural preferences of stove users (e.g. space for multiple pots, ability to cook local dishes) (Karanja and Gasparatos 2019; Jürisoo et al. 2018). All these factors have contributed to the lack of large-scale adoption and sustained use of efficient biomass cookstoves in parts of the continent, even if their initial adoption was successful (EkouevI and Tuntivate 2012; Ruiz-Mercado et al. 2011; Karanja et al. 2020; Debbi et al. 2014).

Ethanol fuel and stoves is one of the clean cooking options promoted in some SSA countries such Ethiopia, Kenya and Mozambique (Chap. 2 Vol. 1). However most of these efforts have been rather small-scale and have not moved beyond the pilot stage (World Bank 2017; Rogers et al. 2013). On the other hand, there is a long tradition of sugarcane ethanol production for transport biofuels in countries such as Malawi, Zimbabwe and South Africa (Gasparatos et al. 2015). For example, Malawi has been blending ethanol in conventional gasoline since the 1980s (often at blends as high as 20%) and has been considered as one of the most successful countries in this regard globally (Johnson and Silveira 2014). However, despite its wide availability, bioethanol has not been adopted in the residential sector in Malawi except for some pilot projects.

A growing body of literature has been exploring both the factors affecting the adoption of ethanol stoves in SSA and the impacts associated with sugarcane ethanol production. For example, studies have used consumer choice techniques to uncover the key attributes inherent in the selection of ethanol stoves (Ozier et al. 2018; Takama et al. 2012). Some studies have also explored user perceptions of ethanol stoves compared to other readily available options (Benka-Coker et al. 2018; Mudombi et al. 2018a) or the actual indoor air pollution/emissions (Pope et al. 2017) and associated health effects (Alexander et al. 2018). Some recent studies have discussed the possible environmental effects of sugarcane productionFootnote 4 on land use change (Beza and Assen 2017; Dlamini 2017; Romeu-Dalmau et al. 2018; Twongyirwe et al. 2018), biodiversity (Degefa and Saito 2017), water availability and quality (Hess et al. 2016; Ngcobo and Jewitt 2017; Kasambala Donga and Eklo 2018) and GHG emissions (Dunkelberg et al. 2014). Many studies have also explored impacts related to livelihoods/poverty (Manda et al. 2018; Mudombi et al. 2018b; Wendimu et al. 2016), food security (Dam Lam et al. 2017; Herrmann et al. 2018) and social conflicts (https://www.tandfonline.com/doi/full/10.1080/03057070.2016.1211401). As sugarcane can be viewed as a renewable resource in SSA, it could support energy access goals through the expansion of bioethanol use beyond the transport sector and into the residential sector (Johnson et al. 2017) (Chap. 2 Vol. 1).

The above suggests that when seeking to understand the potential and the sustainability of ethanol-based clean cooking options in SSA, it is important to understand broader factors affecting its production and adoption. This includes issues both related to the impacts of sugarcane production4 and factors affecting the promotion, adoption and sustained use of clean cookstoves.

The aim of this chapter is to explore some key aspects of the production and adoption of ethanol as a clean cooking option in SSA. However, there is currently no country in SSA with a significant track record of both domestic ethanol production and large-scale adoption of ethanol stoves (World Bank 2017; Gasparatos et al. 2015). As this prohibits a comprehensive analysis across the entire value chain of ethanol for cooking, this chapter synthesizes insights from two different sites, one related to ethanol adoption (Maputo, Mozambique) and one related to sugarcane production (Dwangwa, Malawi). Collectively these two sites offer unique and distinctive experiences in ethanol production and use in the SSA context. In particular, we explore some of the factors that led to the rapid penetration and adoption of ethanol as a clean cooking fuel in Maputo, as well as some of the impacts of ethanol feedstock production in Dwangwa on selected ecosystem services. In this way, we capture supply and demand issues on a somewhat comprehensive manner.

Section 5.2 outlines the characteristics of the study sites and the methodology for assessing the impacts of sugarcane production (Dwangwa, Malawi) and user perceptions of ethanol stoves (Maputo, Mozambique). Section 5.3 contains the main results related to the adoption of ethanol stoves in Maputo (Sect. 5.3.1) and the ecosystem services impacts associated with land use change in Dwangwa (Sect. 5.3.2). Section 5.4 synthesizes this evidence and outlines policy implications and recommendations for enhancing the sustainability and viability of ethanol-based clean cooking interventions in SSA.

2 Methodology

2.1 Study Sites

2.1.1 Charcoal Sector in Mozambique and Malawi

The two study sites discussed in this chapter include a sugarcane production area (Dwangwa, Malawi, Sect. 5.2.1.3) and an ethanol stove adoption and use area (Maputo, Mozambique, Sect. 5.2.1.2) (shown in Fig. 5.1). Both Malawi and Mozambique are least developed countries characterized by low levels of development in terms of gross domestic product (GDP) and the Human Development Index (HDI), ranking among the lowest in the world.

Fig. 5.1
figure 1

Location of study sites

Full size image

In both Mozambique and Malawi, traditional biomass fuels dominate the household cooking sector, with only 4% and 3% of the population, respectively, having access to clean cooking in 2016 (World Bank 2020). In Mozambique, traditional biomass fuels such as fuelwood and charcoal dominate the cooking fuel market, accounting for as much as 59% and 23% of cooking fuel demand, respectively, nationally (but with large variations between rural and urban areas) (EUEI 2012). Urban households in Mozambique predominately use charcoal for cooking regardless of income level, which suggests the strong position of charcoal in the national energy system (Castán Broto et al. 2020; EUEI 2012). Similarly, the urban population of Malawi is highly reliant on charcoal and fuelwood for cooking (Government of Malawi 2009; Republic of Malawi 2012; Zulu 2010; Makonese et al. 2018). Studies have found that low-income urban households depend almost completely on charcoal for cooking (91.2%), with this dependence being lower for middle- (28.9%) and high-income households (10.2%) (Practical Action 2017).

Overall, the charcoal sector plays a prominent role in the national economies of both countries. In Mozambique, it is estimated that the charcoal sector employs between 136,000 and 214,000 people on a full-time basis (EUEI 2012). Charcoal production accounts for a large fraction of rural livelihoods, offering valuable income diversification especially considering that in most rural areas of the country formal employment and income opportunities are very scarce and infrequent (Woollen et al. 2016; de Koning and Atanassov 2014; Jones et al. 2016). In Malawi, between 180,000 and 200,000 people are employed in the charcoal sector with an additional 928,000 people involved in the charcoal value chain (Kambewa et al. 2007). Similarly, the charcoal and fuelwood sectors are very important for rural livelihoods in Malawi, both in terms of absolute income and income diversification, considering the similar lack of formal income and employment options in most rural areas of the country (Smith et al. 2017; Kamanga et al. 2009).

At the same time, household expenditures on cooking fuel are quite high in both countries. In Mozambique, it is estimated that low-income households spend approximately 24% of their total income to purchase charcoal every month, while an average-income household spends about 15% of total income (Atanassov et al. 2012). Moreover, poor households essentially pay double the price for charcoal because they purchase it in small quantities, unlike high-income households that purchase in bulk and thus at lower prices (Atanassov et al. 2012). In Malawi, urban households also allocate a high proportion of their total income on energy, which ranges from 7.8% for high-income households to 13% for low-income households (Practical Action 2017).

Charcoal and fuelwood demand have been identified as a key driver of localized deforestation and ecosystem services degradation in both countries. Even though it is difficult to estimate exactly the actual deforestation rates in Mozambique (EUEI 2012), multiple studies have estimated the deforestation outcomes of charcoal production in many parts of Mozambique (Sedano et al. 2016, 2020; Silva et al. 2019). Similarly, charcoal and fuelwood production have been associated with significant deforestation in Malawi (Zulu 2010; Davies et al. 2010).

2.1.2 Maputo: Ethanol Stove Adoption Site

Maputo, the capital of Mozambique, has been the only major city in SSA to experience a large-scale commercialization of ethanol for cooking through a private sector initiative. In particular, CleanStar, a private company, promoted aggressively ethanol stoves and ethanol fuel as an alternative to charcoal, which dominates the cooking energy options in the city (Castán Broto et al. 2020; EUEI 2012) (Sect. 5.2.1.1). Even though the ethanol stove promotion built on a pre-existing initiative that commercialized ethanol for cooking in modest quantities, the large-scale promotion of ethanol effectively started in late 2012 (World Bank 2017). Ethanol stoves and fuel achieved rapid penetration shortly after the commencement of CleanStar activities through 160 ethanol distributors, reaching 34,000 consumers and a monthly consumption of 70,000–140,000 litres (Mudombi et al. 2018a). At the same time, CleanStar established an ethanol production facility in the city of Beira in central Mozambique using as feedstock cassava sourced from smallholder farmers from the north of the country (Bogdanski 2012; Costa 2019).

However, the ethanol production arm of CleanStar was discontinued in 2013 and its ethanol distribution component was transferred to a company called NDZiLO in June 2013 (Mudombi et al. 2018a). During CleanStar’s restructuring, cassava ethanol production effectively ceased, thus cutting ethanol supply to stove users in Maputo (Costa 2019). To meet this supply gap, the company started importing ethanol from South Africa in the last months before its collapse. However, this imported fuel was cheaper but of lower quality, often causing the underperformance and malfunction of the canisters and eventually influencing several users to switch back to charcoal or LPG (Mudombi et al. 2018a). Following the eventual collapse of CleanStar, NDZiLO started importing high-quality ethanol fuel from South Africa in an effort to revive the ethanol stove sector. At the time of fieldwork in July 2015 (see Sect. 5.2.2.1), there was still a strong ethanol demand in Maputo, amounting to approximately 80,000 litres per month from 10,000 consumers.

Despite these difficulties sustaining CleanStar’s activities, Maputo still arguably constitutes the largest consumer base for ethanol stoves in a major SSA city. In this respect Maputo offers a unique opportunity to explore the factors that constrained and facilitated the adoption of ethanol fuel and stoves, as an alternative to charcoal.

2.1.3 Dwangwa: Sugarcane Production Site

Malawi is the only SSA country to have fully integrated biofuels into its energy system (Sect. 5.1) (Chaps. 2–3 Vol. 1). Sugarcane ethanol has become a central element of the transport sector for more than 30 years (Johnson and Silveira 2014). In particular, sugarcane ethanol has been produced and blended with gasoline at proportions between 10 and 25% since the 1980s. This was essentially a response to the energy crises of the 1970s, which escalated the costs of importing refined oil products in this landlocked country. This has coincided with continuous efforts to integrate smallholders in the sugarcane value chain since the 1990s in order to enhance rural development (von Maltitz et al. 2019; Chinsinga 2017) (Chap. 3 Vol. 1). The sugarcane sector is also particularly important to the Malawian economy, accounting a substantial fraction of the national gross domestic product (GDP) (Chinangwa et al. 2017) (Chap. 3 Vol. 1).

Most of the sugarcane production is concentrated in the Dwangwa and Nchalo sugarcane belts (Chinangwa et al. 2017). The project site in Dwangwa contains a large-scale sugarcane plantation and a sugar mill operated by a multinational company (Illovo) since the late 1970s (Chinsinga 2017). Irrigated and rainfed sugarcane plots surround the core plantation and are managed by smallholders that sell sugarcane to the Illovo mill. The irrigated smallholders are part of the Dwangwa Cane Growers Limited (DCGL), while the rainfed smallholders are either independent or parts of smallholder associations (Chinsinga 2017; Gasparatos et al. 2018a). EthCo Malawi is a fully Malawi-owned company that operates an ethanol distillery, which uses the molasses by-product purchased from the Illovo mill. Due to the long history of sugarcane and ethanol production, Dwangwa offers a unique case study to explore some of the local impacts of bioethanol feedstock production in SSA.

2.2 Data Collection and Analysis

2.2.1 Maputo: Ethanol Stove Adoption Site

For the Maputo study, we use a qualitative research approach to elicit (a) user perceptions about ethanol stoves and fuel, compared to other options and (b) key lessons learned from the rapid and large-scale expansion of ethanol distribution, marketing and consumer use. For (a) we use a combination of a household survey with stove users and focus group discussions (FGDs) with ethanol users. For (b) we use in-depth interviews with an ethanol supplier (NDZiLO) and ethanol fuel and stove distributors.

For the household survey, we targeted 341 households that represent both users and non-users of ethanol stoves. The household survey was structured and mostly included closed-ended questions (with some open-ended questions). We selected respondents from neighbourhoods that had experienced a large-scale uptake of ethanol stoves and represented the predominant socio-economic background in Maputo (Mudombi et al. 2018a). We focused in the neighbourhoods of Benfica (N = 72), Chamanculo (N = 61), Hulene (N = 60), Mavalane (N = 69), Maxaquene (N = 65) and Urbanizacao (N = 14). For the household survey, we targeted the main female decision-maker within the household, which is usually the member most involved in food preparation and fuel/stove procurement (Gasparatos et al. 2018a). When the main female decision-maker was unavailable, we interviewed the spouse or another person that was involved in daily cooking activities (Gasparatos et al. 2018a). A full explanation of the survey protocol is included in Mudombi et al. (2018a).

Five FGDs were conducted with a total of 29 current and past ethanol users. The FGDs were mixed, with both men and women represented in each FGD. However, in contrast to the household survey, more men participated than women. This imbalance was due to the fact that the sampling list used to randomly recruit FGD participants was drawn from ethanol stove buyers from relevant shops. As men did the actual purchasing, it was challenging to invite the main female decision-maker from these randomly selected households. However, ground rules were laid out to ensure that all FGD participants contributed during the discussions.

Each FGD lasted between 60 and 90 min and contained open-ended questions about (a) participants’ socio-economic characteristics; (b) reasons influencing the decision to purchase ethanol stove for cooking; (c) frequency of stove use; (d) type of food cooked using the ethanol stove; (e) convenience of purchasing ethanol fuel; (f) ethanol expenditures compared to the previous main cooking fuel; (g) advantages and disadvantages using ethanol for cooking compared to the previous main cooking fuel; (h) reasons for stopping ethanol use for cooking; (i) willingness to resume ethanol use for cooking if addressing the reason influencing the decision to stop ethanol use; and (j) any other relevant comments.

Finally, we conducted expert interviews with 25 ethanol fuel and ethanol stove distributors and one ethanol supplier (Zoe Enterprises). These expert interviews were conducted in person with each individual respondent and lasted between 30 and 60 min. They contained open-ended questions that aimed to elicit information about (a) ethanol fuel and ethanol stoves as a business venture compared to previous business ventures; (b) livelihood benefits compared to previous ventures; (c) challenges experienced during involvement in the ethanol sector and potential solutions; and (d) any other relevant comments.

Household survey data was analysed in Microsoft Excel and R version 3.2.2 (R Core Team 2015) using descriptive statistics (e.g. proportions, means). The FGDs and expert interviews were transcribed and analysed through qualitative content analysis techniques. The transcribed text was coded manually and classified into relevant categories with similar meanings and themes (Hsieh and Shannon 2005).

2.2.2 Dwangwa: Sugarcane Production Site

For the Dwangwa study, we aimed at identifying the possible impacts of land use change for sugarcane production on ecosystem services. We use a combination of analytical techniques including geospatial analysis, ecological surveys, soil analysis and household surveys to identify the effects of land use change due to sugarcane production on carbon sequestration (regulating service), woodland products (provisioning services) and recreation and religious values (cultural ecosystem services) (Sect. 5.3.2).

First, we track the land use change in the study site through geospatial analysis. We conduct this analysis for two periods, i.e. the years 1975 during the early stages of sugarcane production and 2015 when the household survey was undertaken (see below). We track land use change both within the area of sugarcane production and in the surrounding areas. Understanding the type of land that was converted for sugarcane production offers important insights about the type of affected ecosystem services. We use a selection of satellite images from the Landsat image archive and correct them following different processes. The final classification includes the following land cover types: (a) sugarcane, (b) high-density forest, (c) low-density forest, (d) bare land, (e) agriculture other than sugarcane, (f) grassland, (g) water and (h) cloud or shadow. We edit manually these different classes using our knowledge of the study site and test the inherent robustness of the classifications. A full explanation of the analytical process is included in Romeu-Dalmau et al. (2018).

Second, we estimate changes in carbon stocks due to the land use change associated with sugarcane production. Carbon storage is a major regulating ecosystem service associated with biofuel feedstock production (Gasparatos et al. 2018b). In particular, we estimate carbon stock change in the above-ground biomass, below-ground biomass and soil organic carbon (SOC), before and after land conversion for sugarcane production. We perform this analysis for each of the land uses identified above and estimate carbon stock change over a 20-year cycle.Footnote 5 Carbon stocks are estimated using primary and secondary data collected through:

  • Standing biomass surveys in forest areas and soil sampling in forest, sugarcane and other agricultural areas (July 2015)

  • Literature review about the standing biomass of sugarcane and other common agricultural crops in the study site (mainly maize)

  • Literature review about allometric equations and other appropriate conversion factors

Subsequently, we use Monte Carlo simulation to assess the net changes in carbon stocks, as a means of incorporating the uncertainty associated with carbon stocks estimates. We follow the IPCC Guidelines for small data sets (IPCC 2006) and use R version 3.2.2 to perform the simulations (R Core Team 2015). A full explanation of the analytical process is included in Romeu-Dalmau et al. (2018).

Third, we use household surveys to identify which provisioning and cultural services local communities obtain from woodlands and their importance for the household. Following a literature review, we identify the main ecosystem services likely to be provided by the forest ecosystem in the study area (Gasparatos et al. 2018b). We then ask respondents two questions, with the first question eliciting whether respondents receive a given ecosystem service (Yes/No answer) and the second question eliciting the importance of this ecosystem service for the household. We survey households with differentiated levels of involvement in sugarcane production (i.e. plantation workers, sugarcane smallholders) and households not involved in sugarcane production (i.e. control groups). In particular, we sampled (a) formal plantation workers working for Illovo (N = 104); (b) irrigated sugarcane smallholders (N = 104); (c) rainfed sugarcane smallholders (N = 107); (d) a control group consisting of subsistence farmers living in the vicinity of the sugarcane-growing areas (N = 104); and (e) a control group consisting of subsistence farmers living approximately 50 km from the sugarcane belt (N = 99). Groups A–D live in the vicinity of the sugarcane plantation so it is safe to assume that they have been affected by the land use change, while Group E has not been affected by the land use change associated with sugarcane production. In order to ensure the effective randomization of respondents, we selected randomly the respondents of Groups A–C through lists obtained from Illovo, DCGL and the rainfed sugarcane grower associations (Sect. 5.2.1.3) and control groups through transect walks. Detailed information about the survey approach is included in Gasparatos et al. (2018a). The results are analysed through descriptive statistics (Sect. 5.3.2.2).

3 Results

3.1 Maputo: Ethanol Stove Adoption Site

3.1.1 Consumer Perceptions for Ethanol Fuel and Stoves

Out of the 341 surveyed households, 29% reported that they own (or previously owned) an ethanol stove. Of these, at the time of the survey, about 54% still used the ethanol stove regularly, 4% used it occasionally, and about 42% were no longer using it. Considering the entire household survey sample, the adoption profile is current users (17%), quitters (12%) and non-adopters (71%).

The different groups offered radically different reasons of why they adopted, not adopted and discontinued using the ethanol stoves (Table 5.1). The adopters usually cite the speed and convenience of the stove in terms of lighting and the lower smoke emissions. On the other hand, non-adopters and quitters mentioned the high costs (both for stoves and fuel), the poor design that is prone to malfunction or unable to meet family needs and the lack of access to fuel.

Table 5.1 Factors influencing the adoption and use of ethanol stoves and fuel according to the household survey
Full size table

FGD participants evoked similar reasons for adopting ethanol stoves and fuel and/or discontinuing use. In particular, former and current ethanol stove users pointed to the various factors influencing their decision to purchase ethanol stoves, with the most prominent being (a) the reasonable cost of the fuel and stove at the time of initial adoption, (b) convenience of use, (c) environmental friendliness, (d) lower smoke emissions, (e) higher safety, (f) cleanliness, (g) social exposure (i.e. influence from friends and neighbours) and (h) marketing strategies employed by the ethanol suppliers (Table 5.2). Most of the respondents indicated that the ethanol stove was easy to use, required little time to ignite and turn off and could be used both inside and outside the house all year long. Social exposure was also listed as an important factor, as some respondents bought ethanol after seeing their neighbours using it.

Table 5.2 Factors influencing the adoption and use of ethanol stoves and fuel according to the focus group discussions
Full size table

However, despite incurring the initial stove purchase costs, some respondents eventually reduced the frequency of using the ethanol stove (or discontinued altogether its use). The most commonly mentioned reason was the high recurring fuel costs, which were in fact increasing over time (Table 5.2, Sect. 5.2.1.2). Similar to the survey, other reasons included the problematic stove design that was prone to malfunction (i.e. fuel tanks that rust easily) and the occasional lack of fuel availability (Table 5.2) (Sect. 5.2.1.2).

3.1.2 Factors Facilitating Rapid Ethanol Penetration

Expert interviews with personnel from Zoe Enterprises, a business venture in Maputo, provide further insights about the factors contributing to the initial rapid penetration of ethanol for household use in the city. These include the (a) enabling policy and institutional environment, (b) effective utilization of pre-existing market channels, (c) extensive awareness-raising campaigns and capacity-building efforts and (d) effective post-acquisition customer services and support.

Firstly, respondents suggested that the government of Mozambique had created a favourable environment for both local and international investors in the sector through the formulation of appropriate policies and regulatory framework. Furthermore, the government provided incentives and subsidies to bioethanol producers, which made ethanol production price-competitive for the household market. Central to all these was the approval of the National Policy and Strategy for Biofuels (NPSB), which provided clear guidelines to both the public and private sector to enhance participation in biofuel activities. The NPSB was aimed at reducing the dependence on imported fossil fuels and essentially created a local biofuel market, including for the household sector. Even though at the time of the survey, the cooking ethanol was imported from South Africa as CleanStar (and its smallholder-based cassava production and ethanol distillation) had ceased operations (Sect. 5.2.1.2), the government still subsidized the cost, hence reducing the overall cost assumed by customers. For example, according to the respondent from Zoe Enterprises:

Initially, the government was not supportive as the only uses for ethanol they were familiar with, were as spirit in the health sector (which had no taxes) and as a beverage (which had 40% tax including import duty and VAT). However, after lobbying and explaining to the government that ethanol can be used for cooking in low-income households to substitute charcoal and firewood, we were able to convince them. Now we only pay 17% VAT, and no import duty. (Personal Communication, Manager, Zoe Enterprises, September 2015)

Secondly, it was also suggested that the effective capitalization of pre-existing market channels facilitated the rapid expansion of the ethanol sector in the city. Essentially, bioethanol was stored and distributed through a pre-existing network that commercialized imported ethanol gel fuel (Sect. 5.2.1.2), with the pre-existing consumers forming the initial consumer base of the ethanol. This allowed for the minimization of customer acquisition downtime. Furthermore, the ethanol fuel was retailed via existing outlets that were already known to the customer base, hence making easier the acquisition and improving the accessibility to the customers.Footnote 6 Moreover, selling both the stove and fuel in the same shop, and supply them through the same distributor reduced distribution costs and consumer effort in acquiring the stove and fuel. As was mentioned by an interviewee:

Previously, Zoe Enterprises was selling gel fuel. We had our own business license, the premises, the staff, our gel fuel client base and our distribution networks. When we were approached by CleanStar Ventures who produced the ethanol, and which was going to be cheaper than the gel fuel we were distributing, we agreed [to work together]. They worked under our kind of umbrella. When they got their license, we started operating as CleanStar, with Zoe Enterprises becoming the department of sales and marketing. Our role was to produce the NDZilo fuel [a mixture of ethanol and other additives], and perform marketing activities, since we had links with the gel communities and distributing to our retailing networks. (Personal communication, Sales and Marketing Lead, Zoe Enterprises, September 2015)

Thirdly, a series of capacity-building efforts and awareness-raising campaigns through different channels further accelerated ethanol penetration in Maputo. Information sharing about the importance of the ethanol fuel and especially ethanol’s economic, social and environmental benefits was a key element of these efforts. Relevant information about the ethanol stoves and fuel was disseminated through TV commercials, billboards and door-to-door visits by the sales team. These activities targeted a large segment of the local community and essentially familiarized many potential users with the new fuel. This ignited interest and catalysed the purchase of the stove and the fuel. As was mentioned by a respondent:

We had a big team, so we coordinated, supervised, and worked six to seven days a week. We were in the community, we were in meetings. We started by entering one neighborhood, then we worked first with the chief who is the community leader who then convened meetings with other smaller leaders. We had to do the demonstration, we had to explain to them and then we entered other neighborhoods and had our teams moving from place to place inside the community. We also did television advertisements and had billboards everywhere as we had more financial resources to support these activities. (Personal communication, Sales and Marketing Lead, Zoe Enterprises, September 2015)

Finally, the effective post-acquisition customer services and support was another major factor contributing to the quick uptake of ethanol fuel and stoves. By maintaining an updated record of their customers, ethanol stove retailers were able to follow up and receive user feedback. This included, for instance, reasons why ethanol purchase had declined over time. Oftentimes, as discussed above, this decline was influenced by poor stove design and fuel quality. This enabled the distributors to understand better the specific product-related problems and replace problematic stoves as needed. Such customer databases were important especially following collapse of the ethanol production (Sect. 5.2.1.2), as the distributors were able to contact directly ethanol stove users and explain the situation. Identifying and resolving consumer challenges early on helped in maintaining a demand for bioethanol despite the fuel supply challenges. This was described by an interviewee as follows:

We have a list of names and contact details of all the customers who have bought our ethanol stove. The information was mainly collected for the stove warranty and to get feedback from the customers. The information was used to contact the customers whose canister tanks had rusted for replacement and for informing them that NDZiLO was then available in the market. (Personal communication, Sales and Marketing Lead, Zoe Enterprises, September 2015)

3.2 Dwangwa: Sugarcane Production Site

3.2.1 Land Use Change

Irrigated sugarcane production in the Illovo plantation and the smallholders scheme (DCGL) have caused significant land use change (Fig. 5.2). In particular irrigated sugarcane cultivation led to the conversion of low-density forest, high-density forest and agricultural land dedicated to food crops and particularly maize, which is the staple crop in the area (Figs. 5.2 and 5.3a). Rainfed sugarcane production has also directly converted agricultural land, but it is difficult to quantify the actual magnitude of this land use change. The land use change observed in the surroundings area is minimal compared to the direct land use change occurring within the boundaries of irrigated sugarcane production (Fig. 5.3b). Although it is not possible to conclusively establish causality, the land use change observed in the surrounding area can be potentially linked to farm displacement elsewhere in the area and/or the attraction of population due to generation of direct and secondary employment opportunities.

Fig. 5.2
figure 2

Land use map of the Dwangwa area for 1975 and 2015. (Source: Romeu-Dalmau et al. 2018)

Full size image
Fig. 5.3
figure 3

Land use change within the irrigated sugarcane production area (a) and around it (b)

Note: Figure 5.3b illustrates the land use change that occurred in the surrounding areas of irrigated sugarcane production in a comparable area to Fig. 5.2b. (Source: Romeu-Dalmau et al. 2018)

Full size image

3.2.2 Ecosystem Services

Sugarcane areas store on average 65 tC/ha, which is the second highest carbon storage amount among the different land uses (Table 5.3). Soil organic carbon (SOC) constitutes the largest carbon stock in each land use. Net carbon storage over a 20-year period is higher for sugarcane compared to surrounding land uses (Fig. 5.4). This suggests that sugarcane production can generate carbon stock gains in the study site, offering thus an important regulating service. This is possibly due to the fact that the densely planted sugarcane crops have higher standing biomass compared to the low standing biomass of surrounding agricultural and woodlands that are already partly degraded from fuelwood extraction (Sect. 5.2.1.3).

Table 5.3 Carbon stocks in the different land uses (in tC/ha)
Full size table
Fig. 5.4
figure 4

Net carbon stock change due to sugarcane conversion

Note: The box represents the interquartile range (IQR; difference between the 25th and 75th percentiles). The thick black line represents the median, and the top and bottom whiskers indicate the highest values within the upper range (75th percentile +1.5 ∗ IQR) and the lowest values within the lower range (25th percentiles −1.5 ∗ IQR). Source (Romeu-Dalmau et al. 2018)

Full size image

While these results show clear carbon sequestration benefits, the conversion of woodlands has also most likely led to the loss of forest-related provisioning ecosystem services such as fuelwood, medicinal plants, wild food and fodder for livestock. Although it is not possible to quantify the actual loss of these ecosystem services, the fact remains that these ecosystem services are important for the livelihoods of the local communities.

Based on the household survey, the two main provisioning services collected from forest were non-timber forest products (NTFPs) and fuelwood, with 62% and 48% of the responding households involved in their collection (Fig. 5.5). Fuelwood was collected throughout the year, while NTFPs and wild food were collected just a few times within the year. Other major provisioning services derived from woodlands include indigenous vegetables (21%) and wild fruits (20%). Fuelwood was reportedly considered to be of high importance to 94% of the surveyed households, with fodder and non-timber products also reportedly to be of high importance by 93% and 85% of the respondents, respectively (Table 5.4). A small proportion of respondents do not consider medicinal plants and honey to be of high importance for their households. On the other hand, woodlands do not seem to provide significant cultural ecosystem related to recreation and religious values. Respectively, 87% and 80% of the respondents mentioned that they do not derive such services from the landscape.

Fig. 5.5
figure 5

Fraction of respondents obtaining provisioning ecosystem services from woodlands

Full size image
Table 5.4 Self-reported importance of woodland ecosystem services for surveyed households
Full size table

4 Discussion

4.1 Synthesis of Findings

Section 5.3 outlined some of the critical issues associated with the demand and adoption of ethanol as a clean cooking alternative in Maputo (Mozambique) and the impacts of sugarcane production in Dwangwa (Malawi). In particular Sect. 5.3 explored issues of bioethanol marketing and use in Maputo, as well as the factors that influenced its successful and rapid penetration as a clean cooking fuel. In Malawi the analysis focused on the possible impacts of sugarcane production on specific ecosystem services.

When it comes to the feedstock production side, cassava-based ethanol production in Mozambique following the CleanStar model was rather complex, and due to multiple institutional, cost-related and logistical reasons, it eventually collapsed (Sect. 5.2.1.2) (Mudombi et al. 2018a; Costa 2019). On the contrary, sugarcane and ethanol production in Malawi are quite optimized, with a long history of achieving high levels of output at cost-competitive prices (IRENA 2016; Mitchell 2011). However, due to the large-scale production model adopted (Chap. 3 Vol. 1), sugarcane production in Dwangwa has caused extensive land use and land cover change, which has had variable effects on the provision of different ecosystem services and the wellbeing of local communities (Chinsinga 2017; Kiezebrink et al. 2015).

Land use change, and especially the loss of crop land, possibly had negative effects on provisioning services related to food crops and woodland products, which are important for the livelihoods of local communities (Sect. 5.3.2.2). However, other studies in the same area have suggested that the actual effects of sugarcane production on food crop production might have been less pronounced for smallholders, as improved access to fertilizers and other agricultural resources enabled higher yields, enabling thus, to some extent, the compensation of cropland loss (Herrmann et al. 2018) (see also Chap. 3 Vol. 1). For regulating services, land conversion for sugarcane production led to gains in carbon stocks (Fig. 5.4), which is rather unusual for biofuel projects that convert woodland (Achten and Verchot 2011). Most likely these carbon stock gains were due to the fact that the converted woodland areas were already rather degraded from extensive fuelwood collection (Romeu-Dalmau et al. 2018), which is prevalent throughout rural Malawi (Sect. 5.2.1.1). Landscape conversion did not seem to have had an appreciable effect on the provision of cultural ecosystem services, as few respondents indicated that they derive such services from woodland areas (Sect. 5.3.2.2). However, it should be noted that the pathways linking bioenergy production and cultural ecosystem services are rather complicated and indirect (Gasparatos et al. 2018b), usually depending on the actual context of bioenergy production and use (e.g. Ahmed et al. 2019; de Hoop 2018).

When it comes to the ethanol adoption side, our study finds that high costs have been the main reasons for not adopting or discontinuing ethanol use for cooking (Tables 5.1 and 5.2). High costs have been identified as major constraints to stove adoption in many parts of SSA (Karanja et al. 2020; World Bank 2017; Rehfuess et al. 2014). Furthermore, users tend to prefer ethanol stoves for short cooking tasks such as boiling water for tea due to their convenience. Still, charcoal is preferred for longer and slower-cooking tasks, which further suggests that even for ethanol stove adopters, the comparatively higher operational cost factors limit its use to some degree (Mudombi et al. 2018a). Indeed many FGD participants suggested that the initial adoption of ethanol stoves was influenced by the reasonable initial costs, but this changed later on due to the collapse of domestic ethanol production, which in turn caused ethanol price hikes and decreases in quality (Table 5.2). Simple calculations based on the cost of energy supplied to the pot suggest that local ethanol prices need to drop to about 0.50 USD/L to make ethanol truly cost-competitive with charcoal, which is about half the current cost and rather close to production costs in major producing countries such as Brazil and the USA (Mudombi et al. 2018a).

As mentioned above the CleanStar ethanol production model was not optimized and solely geared towards the production of cheap ethanol, but attempted to incorporate many social aspects such as smallholder support during cassava farming (Costa 2019). Furthermore, it used a relative small-scale distillation facility that was located both far from the cassava production sites and the ethanol demand areas. The tendency to “weigh down” household fuel-switching efforts with multiple social benefits that may later complicate economic feasibility is among the reasons why some major cookstove programmes such as the Global Alliance for Clean Cookstoves tend to emphasize more commercial approaches involving building entrepreneurship and market demand (GACC 2014). However, such cost dynamics may change substantively if linked to large-scale production models such as the ones currently operational in Malawi (Puzzolo et al. 2019).

4.2 Policy Implications and Recommendations

The findings discussed in this chapter suggest major intersections between clean cooking and the SDGs. For example, the widespread adoption and sustained use of ethanol stoves and fuels essentially contributes directly to SDG7 on access to clean, affordable, reliable and sustainable energy (Sect. 5.1). However, many of the issues discussed in Sect. 5.3 associated with stove adoption and sugarcane production suggest strong intersections with multiple other SDGs such as SDG2 (Zero Hunger), SDG12 (Responsible Consumption and Production), SDG13 (Climate Action) and SDG15 (Life on Land), among others. This suggests that in order to maximize the positive impacts of ethanol stoves and fuel for contributing to progress to multiple SDGs, it would be important to both ensure the wide adoption and sustained use of ethanol stoves/fuel on the one hand, and the sustainable production of ethanol fuel on the other. Below we draw from the main results of this study some recommendations on how to enhance the potential of ethanol stoves and fuel to catalyse progress for meeting the SDGs.

First, to support the extensive development and widespread adoption of modern bioenergy options, SSA governments would need to design and implement comprehensive and cohesive policies that support various end uses, activities and actors along the bioethanol value chain (Chap. 2 Vol. 1). Comprehensive government support in terms of policies and institutional frameworks, subsidies and incentives are key for the establishment of biofuels investment and market (Jumbe et al. 2009; African Union 2013). This comprehensive support is clearly seen in Malawi through the long-term production of ethanol and its effective integration in the national energy system (Johnson and Silveira 2014). It was also partly seen in Mozambique in enabling both local and international actors to invest in biofuels sector (Schut et al. 2014). Such comprehensive support can in theory allow for cost reduction (see below) and the development of infrastructure for fuel/stove distribution, user education and awareness raising through diverse channels (Karanja et al. 2020). As discussed throughout this chapter, all these factors have played a major role in the long-term production of ethanol in Malawi and the rapid adoption of bioethanol in Maputo.

Second, it would be important to reduce the costs of ethanol production and provide incentives to consumers for stove uptake. It has been argued that in order to realize the wide commercialization and market penetration of clean fuels in SSA, it is crucial to understand the economics of biofuels industry (Amigun et al. 2008; Mitchell 2011). Regardless of whether ethanol is locally produced or imported, its costs as a household fuel must be competitive against other cooking fuels such as charcoal, LPG or electricity (World Bank 2017). However, the affordability of clean fuels such as bioethanol still remains a challenge, hindering their adoption and sustained use. This is further compounded by the fact that clean fuels such as ethanol are part of the formal sector and therefore regulated, compared to fuels such as charcoal and firewood, which are in the informal sector and hence unregulated (Smith et al. 2015; Ndegwa et al. 2016). A major issue directly related to the cost structure of ethanol is likely to be the taxes levied on ethanol. There are various convincing arguments that ethanol should be lightly taxed or not taxed at all when destined as a cooking fuel due to its health and environmental benefits compared to traditional cooking fuels (Sect. 5.1) (Chaps. 2, 7 Vol. 1) (Dalberg 2018; World Bank 2017). Governments can also support stove uptake by providing subsidies on stoves and fuels, therefore further enhancing its affordability to the consumers (Karanja et al. 2020).

Third, it would be important to develop and strengthen national and local bioethanol markets by putting in place institutional structures that provide a space for both the private and public actors to thrive along the bioethanol value chain (World Bank 2017; GACC 2014). The utilization and leveraging of existing market structures could be very important in fostering the scaling up of the adoption of a new product as the CleanStar experience has shown (Sect. 5.3.1). Governments can further support such efforts through research and development, building technical capacity in their respective ministries and raising awareness to potential users through targeted educational activities and campaigns (Karanja et al. 2020; Rehfuess et al. 2014). Further, and if appropriate, national government can facilitate the mandatory blending of ethanol into transportation fuels. This was done with success in Brazil and the USA and has not only created a viable biofuel market but has also driven down the cost of ethanol production (see comment above). This would simultaneously help reduce ethanol costs for cooking fuel and at the same time diversify the market options for producers, reducing thus their risks. However, embarking on a major transportation fuel ethanol programme would need sound research and justification. Furthermore, there would be a need for related guidelines and standards to ensure the effective consumer protection (Karanja et al. 2020).

Finally, depending on the location, scale of production and production practices, biofuel systems in SSA can have both positive and negative environmental, social and cultural impacts, including on ecosystem services (Gasparatos et al. 2015, 2018b). These impacts essentially influence the landscape sustainability and the wellbeing of local communities. It is thus important to adopt policies that acknowledge these aspects and possible trade-offs and provide guidelines on how to assess impacts and minimize negative trade-offs during the development of biofuel policies, programmes and projects. It should be noted that in SSA context, land use management strategies interact strongly with policies aimed at reducing dependence on traditional biomass and the significant GHG emissions associated with land use change (van de Ven et al. 2019). In this context, provisions that further support and strengthen the role of environmental impact assessments (EIAs) would be highly beneficial (OECD 2011), as EIAs are often the only avenue to hear the voices of local communities during the development of large agro-industrial projects in SSA (Ahmed et al. 2019).

5 Conclusions

Bioethanol, if priced correctly, has a great potential to substitute charcoal as a cooking fuel in SSA, therefore contributing directly to the achievement of SDG7. Its uptake can have multiple benefits related to health (SDG3), the stimulation of local industries and marketing chains (SDG9) and the reduction of GHG emissions (SDG 13) and deforestation and land degradation (SDG15). This is in addition to providing a clean fuel that appears to have a relatively high degree of user acceptance despite its comparatively higher cost. However, in order for ethanol cooking to evolve from a niche market within a fuel stacking usage context, there would be a need for substantive cost reductions to the point that it becomes comparable to charcoal. Alternatively (or better at the same time) the environmental and social costs of charcoal should be better reflected in local and national taxation regimes and energy and land use policies.

Considering the many possible environmental, social and economic benefits from moving away from charcoal to ethanol, there is a strong case to be made for national support for ethanol. Government support could be in the form of low taxation and the stimulus packages supporting low-cost ethanol production. The development of regional collaboration, trade and technology transfer could also support such markets, as exemplified by the fact that cost-effective supply was available in one country (Malawi), whereas robust demand existed in a neighbouring country (Mozambique).

Finally, there is real need to understand the pull factors that could accelerate the adoption and promote the sustained use of clean stoves and fuels in SSA. Addressing consumer affordability would be definitely critical towards this end. However, it would be equally important to understand some of the dynamics and trade-offs at both the fuel production and use levels and to identify ways to reduce the negative impacts and curb the barriers to adoption. Malawi and Mozambique are two countries that have complementary, albeit different, experiences in the production and adoption of ethanol fuel that can foster such learning for other SSA countries.