Household Energy Use Among Female-Headed Households in Urban Ethiopia: Key Issues for the Uplift of Women

Abstract

This article examines the linkages between cooking energy and gender based on a study conducted in Arba Minch town of southwestern Ethiopia. It primarily investigates cooking energy end-use within female-headed households (FHHs) residing in the town and specifically focuses on women who disproportionately suffer the impacts of energy accessibility. This study reveals that commercial cooking fuels are increasingly scarce and expensive, and the costs of modern cooking appliances are beyond the purchasing ability of most urban FHHs. Meeting the energy requirements in sustainable manner, thus, continues to be a major challenge that hinders FHHs’ efforts to improve their living situations. The results further show that though the income is an important factor, it is not the only powerful determinant in the consumption of cooking fuels within FHHs. Other socioeconomic characteristics were found to have a significant impact too. This study further examines the relevance of the “fuel stacking” model and finds that the majority of the FHHs, regardless of their economic status, depend on wood fuels as their primary cooking energy source. Even when family incomes rose, complete transition to cleaner fuels has not taken place; instead, it added fuels in the fuel stacking process.

Keywords

Biomass fuels clean energy cooking end-use energy accessibility energy ladder fuel stacking urban female-headed households

Introduction

Background of the Study

Energy is one of the most essential inputs for sustaining people’s livelihood, and without energy, modern life would generally cease to exist (Amoah, 2019; Bakhsh et al., 2020; Clancy et al., 2003). Niez (2010) suggested that lack of access to sufficient and sustainable supplies of energy affects as much as 90 per cent of the population of many developing countries. Cecelski (2004) pointed out that, despite many efforts, energy accessibility is widespread, and gender inequality exists at every level of the energy sector. As revealed by the United Nations Economic and Social Commission for Asia and the Pacific (ESCAP, 2008), the lack of access to energy services aggravates many social concerns, including poverty, ill-health, unemployment and inequity. According to Kohlin et al. (2011), due to the lack of access to electricity, the traditional unimproved cooking methods are though predominantly rural phenomena; there are also many urban households who cannot get modern energy but spend large fractions of their budgets on cooking fuels (Barnes et al., 2004).

Access to basic energy services to the urban female-headed households (FHHs) is an important issue that requires far more attention from policy makers in order to alleviate poverty. Cecelski (2004) stated that women and men have different roles in the energy system, that is, women bear the main burden of providing and using fuels for cooking. A report by the Ministry of Finance and Economic Development indicated that there has been a significant increase in the percentage of FHHs in urban Ethiopia (Woldehanna et al., 2008). At the same time, the incidence of households headed by women is very likely to grow in the study town of Arba Minch, and these FHHs are viewed as being at greater economic disadvantage than the male-headed households (MHHs) because they appear to have less income-earning capacity than their male counterparts. This study focuses on the views of women and gives due consideration to urban FHHs while addressing cooking energy consumption since they are primarily responsible for procuring and processing cooking fuels for their families and spending considerable time around cooking fires.

Statement of the Problem

The study area is one of the recently growing towns of the country and has experienced a significant flow of people from the surrounding rural areas. In spite of the improvement in access to clean fuels in the last few years, prices for commercial cooking fuels are already very high in the market for the majority of urban FHHs. A substantial portion of the urban FHHs continues to suffer as their incomes have not kept pace with the rising prices, consequently facing higher financial burden to meet their cooking demands as energy costs can form a significant part of household budgets with FHHs considered to be in a worse position than those of MHHs (Clancy, 2006; Gould et al., 2020; Parikh, 1995).

The increasing scarcity and costs of household energy have an adverse effect on the livelihood of FHHs who are limited in their ability to utilize energy for anything more than satisfying basic needs. Moreover, the high and direct dependence on biomass fuels coupled with low efficiencies in its end-use at household level, mainly for cooking purposes, is contributing to unnecessary high level of forest resource removal, which has resulted in serious shortage of biomass fuels and higher wood and charcoal prices, hitting adversely all urban households but most critically the FHHs. This hinders their efforts to move out of poverty and seriously constrains their ability to improve their living situations. Therefore, the livelihood of the urban FHHs will continue to be significantly impaired by energy accessibility that will seriously slow down their wellbeing, and the ultimate result will certainly affect the development of the study area.

In the consumption of household energy, Alam et al. (1998) indicated the hypothesis that households ascend an energy ladder and argued urban households shift from using biomass to modern fuels for cooking purposes. Barnes and Toman (2006) also confirmed the energy ladder hypothesis prescribing income to be the sole factor and pointed out that people with low incomes generally use traditional fuels as their main cooking fuels and people with higher incomes tend to use modern fuels. These researchers have developed the notion of a “fuel-income ladder” to explain the shift from traditional fuel use to modern fuels consumption with the rise of households’ income. However, wood fuel is still the choice for cooking among all FHHs regardless of their economic status. Yet, the majority of higher income FHHs do not currently substitute biomass fuel for other clean fuels in order to satisfy their cooking needs, instead they use a mix of commercial energy sources. It is actually a step towards “multiple fuel cooking” or “fuel stacking”.

Masera et al. (2000) looked into fuel stacking using data mainly from rural areas in Latin America, and a number of other studies did consider the urban energy supply and use pattern at a national level focusing on the major cities of the country. So far, no studies, to the best of my knowledge, have explored the multiple use of cooking fuels among well-off urban FHHs in the medium level towns of the country. The use of modern fuels is always accompanied by traditional fuels, and any conventional fuel does not completely substitute biomass. The objectives of this study were mainly to investigate the nexus of cooking energy use and gender in the urban area and identify the demographic and socioeconomic determinants in the overall domestic cooking behaviour of urban FHHs. The article aims to provide empirical evidence to answer the questions such as: What are the most important characteristics that determine the domestic multiple cooking fuel use? and What looks like the pattern of using a variety of cooking fuels among FHHs residing in the town?

Theoretical Perspectives on Cooking Fuels and Domestic Energy Use

In a household survey conducted by Clancy (2006), it was found that poor men and women do not necessarily become poor in the same ways, for example, a man might lose his job, and a woman, who has always depended on her husband for financial support, may become a widow, forcing her to start looking for a paying job later in her life, which she might be ill-equipped to do so. Men and women have different ways of adopting strategies for addressing their poverty; men are more easily able to migrate, while women stay for managing the household and creating informal sector business they can run from home. Therefore, the energy strategies that are intended to assist people to move out of poverty must take these gender aspects into account (Parikh, 1995).

Kohlin et al. (2011) pointed out that in sub-Saharan African countries, which include countries situated in the western, northern and eastern parts of Africa including Ethiopia (Ali, 2020), FHHs are more likely to be poor and thus less able to afford the up-front cost of new stoves or electricity connections, For example, in Kenya, ^´FHHs constitute a higher proportion of the poor both in the rural (54.1 per cent female vis-à-vis 52.5 per cent for male heads) and urban areas (63.0 per cent female vis-à-vis 45.9 per cent for male heads), and almost similar is the case for Ghana too (Twumasi et al., 2020). Several studies have revealed that the trend and extent of poverty is worse among women in association with the growth of female headship of household. Woldehanna et al. (2008) also has reported that, in urban Ethiopia, FHHs are characterized by low and insecure income as most of them are engaged in informal activities, and hence, the incidence of poverty is found to be higher in depth and severity in FHHs than those of their male counterparts.^^

Cecelski (2004) viewed that women use energy differently than men, because of their different household activities and because they make critical decisions about fuel substitution and the purchase of stoves and other appliances, based on their fuel preferences and budget constraints. Clancy et al. (2003) argued that women and men have different perceptions about the benefits of energy; men see the benefits of electricity in terms of leisure, quality of life and education for their children, while women see electricity as providing the means for reducing their workload, improving health and reducing expenditure. The study in Arusha of Tanzania by Meikle and North (2005) showed that FHHs make more use of electricity for household activities than the MHHs since the husband in MHHs is more concerned with expense rather than convenience. Nevertheless, women are most vulnerable groups to energy-related problems and are most likely to suffer the health effects of energy-inefficient appliances. Meikle and Bannister (2005) pointed out that burning of traditional biomass over open fires or inefficient stoves contribute to health-threatening indoor air pollution, which causes a variety of respiratory illnesses such as chronic obstructive pulmonary disease, asthma, bronchitis and pneumonia.

Several studies have stated the notion of an energy switching or the energy-ladder hypothesis as a model for households to substitute or to switch between available fuels (Alam et al., 1998; Barnes et al., 2004; Heltberg, 2004). These authors have argued that as the household income increases, household tends to shift from traditional fuels (in order: dung, crop residues, wood, charcoal and coal) to modern fuels (in order: kerosene, LPG and electricity), and the introduction of superior fuels will phase out traditional fuels as households will switch to the former. The energy-ladder concept is used to describe the way in which households climb the ladder with increase in economic status (Smith, 1987, 1990). The energy ladder model envisions a three-stage fuel switching process. The first stage is marked by universal reliance on biomass. In the second stage, households move to “transition” fuels such as kerosene, coal and charcoal in response to higher incomes and factors such as deforestation and urbanisation. In the third phase, households switch to LPG, natural gas or electricity. The main driver affecting the movement up the energy ladder is hypothesized to be income, and the shift moves up from traditional to commercial fuels along the gradient of income levels (Barnes et al., 2004).

However, the notion of an energy-ladder has been challenged because fuel preferences could be explained by many other factors rather than income (Heltberg, 2004; Masera et al., 2000; Ntobeg, 2007). These authors demonstrated the multiple fuel models rather than the energy switching scenario due to the result of complex interactions among economic factors, social factors and cultural factors. According to them, fuel wood is very seldom replaced entirely when families adopt LPG; none of the family ceased using fuel wood even in the households that have been using LPG for many years. Ntobeg (2007), for example, suggested that the coming of electricity is not as a replacement for the firewood as such, rather merely as supplement.

According to Meikle and Bannister (2005), household energy types and consumption levels are determined by income, fuel availability, fuel prices, cultural preferences, demographic distribution and physical environment. On the other hand, household energy use appeared to be most strongly related to sociodemographic variables (income, household size and age), whereas attitudinal variables and self-transcendence values (tradition/security and power/achievement) were important too (Abrahamse, 2011). Abebe and Koch (2011) suggested that household energy types and consumption levels are determined by a variety of household’s socioeconomic characteristics, such as income, age, education level, sex and occupation of the household head, type and ownership of improved biomass cook stoves, type and ownership of substitutable cook stoves, children and adults in the household, house ownership and characteristics of the house.

It has been argued that households with low levels of income rely on wood fuels, while those with higher incomes consume electricity and LPG that are cleaner and expensive. A recent study has shown that cleaner fuels are priced higher than the traditional ones (TERI, 2010). Nonetheless, not all modern fuels are more expensive than traditional fuels. Nowadays, the market prices of traditional fuels are getting higher when compared to the prices of some modern fuels such as electricity. It is found that most FHHs pay more for wood fuels (wood and charcoal) than they would for electricity for cooking. The majority urban FHHs are not looking the wood fuel as inferior, rather still they use it as the main cooking fuel. Many middle- and high-income FHHs satisfy their needs by using energy combinations rather than replacing wood fuels with conventional fuels. Instead of moving up the ladder step-by-step as income rises, most households tend to consume a combination of fuels for cooking purpose depending on many more factors (Amoah, 2019; Bakhsh et al., 2020).

Materials and Methodology

This study was conducted by the author during 2013 in Arba Minch town, which is one of the largest towns in southwestern Ethiopia. It is located about 505 km from the national capital Addis Ababa and 275 km from the regional capital of Hawasa. The town consists of two settlements, Secha (the uptown, where almost all administrative offices are located) and Sikela (the downtown, which is the business centre of the city) with a total of 12 kebeles (smaller administrative units in a town). Based on census conducted by the Central Statistical Agency (CSA, 2008), Arba Minch town with an area of 5,557 hectares (55.57 km²) has an estimated total population of 82,363, out of which 52.36 per cent are female.

Selection of Sample Households and Collection of Data

For the study, 272 urban FHHs were selected based on purposive and random sampling techniques. Diverse residential housing units, which represent differences in quality of housing and living, were considered—of the six sample kebeles (smallest administrative units) covered for the survey, three kebeles with more shanty houses and the rest three kebeles with more of better-off housing units were included. A questionnaire was developed with closed- and open-ended formats, and observations were collected directly interviewing the households’ heads. Data on the consumption of cooking energy resources were gathered in terms of monetary expenditures, which were also computed in terms of heat energy values. The research employed correlational quantitative design, in which a multivariate analysis was performed using cross-sectional data to explain the relationship to find out the most influential factors believed to influence an FHH’s expenditure on cooking fuels. Then, using households’ monthly income, sample FHHs were grouped under three income groups. FHHs with income less than Birr 1,200 per month (Birr is Ethiopian currency and 1 USD = 19 Birr in 2013) were considered as low-income FHHs, and middle-income FHHs are those falling in the range of Birr 1,201–2,800. Those with greater than Birr 2,801 were categorized into high-income FHHs.

Statistical Methods and Approach

The research employed correlational quantitative design, in which a multivariate analysis to explain the relationship and to analyse the significance of the factors believed to influence a FHH’s cooking fuel use in the study area. Coefficients were often converted into standardized regression coefficients (beta), which indicate the change in the dependent variable associated with changes in the independent variable. Analysis of variance (ANOVA) summary table was determined to test whether the model is statistically adequate. The significance of a multiple coefficient of correlation was assessed with the F ratio, which is used as a general indicator of the probability that any of the predictor variables contributes to the variance in the dependent variable within the population. This study tested the relevance of energy mix use that can be determined by many other factors, and association was drawn between energy consumption and other various factors by employing a multivariate analysis model in the analysis of cooking energy consumption pattern among urban FHHs. Here, cooking energy consump-tion is the response variable, and FHHs tend to use multiple fuels due to a number of factors, which are considered as predictors of the pattern of household cooking energy consumption. There are different factors that can be considered to be significant for determining the cooking energy use of the residential sector; the most important parameters selected in the current analysis as causal variables are age, occupation, education, marital status, income, family size, number and quality of cooking equipment being used, house ownership and quality of housing unit. Correlation coefficients were calculated to measure the strength of the relationship between these characteristics and the expenditure made on cooking fuels. The study also indicates the relative significance of the independent variable and identifies the different variables in order of their importance as predictors of urban domestic energy consumption with the help of beta coefficients. A multivariate analysis of the consumption pattern was used to reveal the probability of consuming different traditional as well as modern fuels in the town. For multivariate analysis, the model is Y = β₀ + β₁X₁ + β₂X₂ … β₉X₉ + ui, where Y = the value of the dependent variable (total household energy consumption). Β₀ = the constant, β₁, β₂ … β₉ = coefficients, ui = values of an unobserved error term.

Results and Discussion

Cooking Fuels and the End-Use Technologies

The analysis of the data on patterns of cooking energy consumption can be expressed either in terms of expenditure (Birr) made on it by an FHH or as the amount of energy consumed in terms of heat value of energy (mega-joules). Out of the surveyed FHHs, about 40.4 per cent of the respondents had their monthly income below Birr 1,200 and about 38.3 per cent between Birr 1,201 and 2,800 and about 21.3 per cent had greater than Birr 2,801 per month (see Table 1). The lowest monthly income for the sample FHHs was Birr 500, while the highest was as much as Birr 4,500 per month. Mean monthly income for households headed by females was Birr 1799.08. The average monthly cooking fuel expenditure for the sample FHHs, which is Birr 101.70, making up 5.65 per cent of the family mean monthly income. The lowest monthly expenditure for the sample FHHs was Birr 55, while the highest was Birr 169 per month. The disparity in expenditure among sample FHHs, which can be explained by coefficient of variation of 17.84 per cent, is smaller.

The average monthly income for low-income FHHs was Birr 823.04 and the mean monthly expenditure on cooking fuels was Birr 90.53; although the poor do have fewer energy expenditures than more wealthy FHHs, cooking energy as a per cent of total expenditure is often well above 10 per cent of their income (see Table 1). Low-income FHHs have to allocate 10.99 per cent of their income for purchasing energy, which creates a higher financial burden on their budgets. In the medium-income FHHs, the mean monthly income was Birr 1811.32 and the average expenditure on cooking fuels was Birr 102.90 per month, which constitutes 5.68 per cent of the average income of the group. In the high-income FHHs, the average monthly income was Birr 3334.38 and the average monthly expenditure on cooking fuels was Birr 117.53, which constitutes 3.52 per cent. Increasing income of the FHHs dictates more expenditure, and more cooking fuel is consumed with less significant strain on their budgets.

Table 1.

Monthly Income and Expenditure (in Birr^a) of Various FHHs Income Groups

Income Group	Income Range (Birr)	Sample FHHs	Monthly Income			Cooking Fuel Expenditure
Income Group	Income Range (Birr)	Sample FHHs	Mean	Sd	Cv (per cent)	Mean	Sd	Cv (per cent)
Low	<1200	102	823.04	165.75	20.14	90.53	13.44	14.85
Middle	1201–2800	106	1811.32	392.63	21.68	102.90	16.96	16.48
High	>2801	64	3334.38	411.82	12.35	117.53	18.64	15.86
Average		272	1799.08	1012.14	56.26	101.70	19.14	17.84

Source: The author.

Notes: ^aBirr = Ethiopian currency, Sd = Standard deviation, Cv = Coefficient of variation.

Biomass fuels account for 88.10 per cent of total energy consumption in terms of cooking energy expenditure, whereas modern fuels make up only 11.90 per cent (see Table 2). Out of Birr 24,307 spent on biomass cooking fuels per month by the sample households, Birr 16,005 (57.86 per cent of the total cooking fuel expenditure) was spent on wood constituting the highest share of the total expenditure for household energy. FHHs spent about Birr 7,835 (28.32 per cent of the total cooking fuel expenditure) per month for charcoal consumption followed by that of sawdust (Birr 530, 1.92 per cent only). Out of Birr 3,293 spent on modern cooking fuels, Birr 2,028 (7.33%) was spent on kerosene, while the share of electricity was about Birr 1,265 (4.57%).

Table 2.

Total and Per Capita Expenditures (in Birr) Made on Cooking Fuels by Various FHHs Income Groups

Income Group	Biomass Cooking Fuels						Modern Cooking Fuels				All Inclusive
	Wood		Charcoal		Sawdust		Kerosene		Electricity		All Inclusive
	Total	Per Capita	Total	Per Capita	Total	Per Capita	Total	Per Capita	Total	Per Capita	Grand Total	Per Capita
Low	5,570	9.49	2,945	5.02	125	0.21	594	1.01	nil	nil	9,234	15.73
Middle	6,590	14.84	3,045	6.86	185	0.42	512	1.15	575	1.30	10,907	24.56
High	3,845	16.86	1,845	8.09	220	0.96	922	4.04	690	3.03	7,522	32.99
Total/Average	16,005	12.71	7,835	6.22	530	0.42	2,028	1.61	1265	1.00	27,663	21.96

Source: The author.

The per capita monthly biomass cooking fuels consumption for the whole sample FHHs was Birr 19.35. The per capita monthly biomass cooking fuel use was Birr 14.72 for low-income FHHs. These figures rose to Birr 22.12 for the medium-income FHHs and Birr 25.91 for the high-income FHHs. The per capita monthly modern cooking fuel use for the whole sample FHHs was Birr 2.61. The per capita monthly modern cooking fuel use was only Birr 1.01 for low-income FHHs. The figure rose to Birr 2.45 for the medium-income FHHs and Birr 7.07 for the high-income FHHs. The total monthly consumption of cooking fuel for the whole sample FHHs was Birr 27,663. The expenditure varied from Birr 7,522 for the high-income FHHs, to Birr 10,907 for the medium-income FHHs. Monthly per capita energy expenditures for all sample FHHs was Birr 21.96, and it varied from Birr 15.73 for the low-income FHHs to as high as Birr 32.99 for the high-income FHHs. This indicates that per capita energy consumption and total expenditure made on cooking energy by a family significantly increased with a rise in a family income.

The expenditure made on cooking fuels was subsequently converted to calorific values which are measured by a basic unit of energy in the metric system known as Joule (J). Since a Joule is relatively a small amount, here, the household cooking energy consumption is discussed in terms of mega joule (MJ). Heat values of energy consumed per FHH are computed by using the standard energy content of all energy sources (Annex. 1). Wood consumed is purchased by FHHs from wood vendors who carry wood on their backs and heads and delivered directly to households. In the field work conducted, vendors serve almost all sample FHHs at an average price of Birr 2.30 for one kg of wood. It means an FHH bought 0.435 kg for 1 Birr. One kilogram of wood provides 15.072 MJ of energy (Annex. 1). Therefore, an FHH gets 6.556 MJ (15.072 MJ × 0.435) heat value of energy for 1 Birr (or from 0.435 kg of wood). Total monthly wood consumed by the whole sample FHHs was about 6962.18 kg (16,005 Birr × 0.435 kg), which provides a total heat value of 104928.86 MJ. The total monthly wood energy consumed varied from 25207.90 MJ for high-income FHHs to 43204.04 MJ for medium-income FHHs and 36516.92 MJ for low-income FHHs. Total and end-use monthly wood energy consumptions for the whole sample FHHs were 104928.86 and 10492.89 MJ, respectively (see Table 3).

Table 3.

Total, End-Use and Per Capita Cooking Energy Consumption by FHHs in the Town

Energy and its cost	Biomass Cooking Fuels				Modern Cooking Fuels			Grand Total
Energy and its cost	Wood	Charcoal	Sawdust	Total	Kerosene	Electricity	Total	Grand Total
Total energy y (kg, L, kWh)	6962.18 kg	2413.18 kg	530 kg	9905.36 kg	162.24 L	3238.4 kWh
Total energy (MJ)	104928.86	71745.10	8877.5	185551.46	5455.32	11663.30	17118.62	202670.08
End-use energy (MJ)	10492.89	14349.02	1420.40	26262.31	2727.66	8747.48	11475.14	37737.45
Per capita energy (MJ)	83.34	56.99	7.05	147.38	4.33	9.26	13.60	160.98
Total energy cost (Birr)	16,005	7,835	530	24,370	2,028	1,265	3,239	27,663
Per capita cost (Birr)	12.71	6.22	0.42	19.35	1.61	1.00	2.61	21.96

Source: The author.

The per capita energy was 62.21 MJ for low-income FHHs, and this figure rose to 97.31 MJ for the medium-income FHHs and 110.56 MJ for the high-income FHHs with respect to wood fuel (see Table 4). The two main sources of supply of charcoal in the town were private charcoal shops (retail sales) and charcoal vendors. Out of the total charcoal users, 87.56 per cent and 10.44 per cent of the FHHs obtained charcoal from charcoal vendors and the retail shops, respectively. The majority of FHHs purchase a sack of charcoal from charcoal vendors who deliver the fuel directly to them. The average price of a kg of charcoal was Birr 3.25; accordingly, an FHH bought 0.308 kg of charcoal for Birr one. One kilogram of charcoal provides heat value of 29.73 MJ of energy (Annex. 1). So, for Birr 1, an FHH could get heat value of 9.157 MJ (29.73 MJ × 0.308 kg). Total monthly charcoal consumed by the whole sample FHHs was about 2413.18 kg (7835 Birr × 0.308 kg), which provides total heat energy of 71745.10 MJ. The total monthly charcoal energy consumed varied from 16894.67 MJ for high-income FHHs to 27883.07 MJ for medium-income FHHs, and it was 26967.37 MJ for low-income FHHs. Total and end-use monthly charcoal consumption for the whole sample FHHs were 71745.10 and 14349.02 MJ, respectively (see Table 3). Monthly per capita charcoal energy consumption for the whole sample was 56.99 MJ. It was 45.94 MJ for low-income FHHs which rose to 62.80 MJ for the medium-income FHHs and 74.10 MJ for the high-income FHHs (see Table 4).

Sawdust was also other important source of energy utilized for coking purposes in the town. About 14.29 per cent of the sample FHHs were found to use this resource. The average price of sawdust is Birr 1 per kg, and an FHH bought 1 kg of sawdust for Birr 1. One kilogram of this fuel delivers 16.75 MJ of heat energy (Annex. 1), so an FHH gets heat value of 16.75 MJ (1 × 16.75 MJ). Total monthly sawdust consumed by the whole sample FHHs was about 530 kg (530 Birr × 1 kg), which provided total heat energy of 8877.5 MJ. Thus, the total monthly sawdust energy consumed varied from 3,685 MJ for high-income FHHs to 3098.75 MJ for medium-income FHHs, and it was 2093.75 MJ for low-income FHHs. Total and end-use monthly sawdust energy consumptions for the whole sample FHHs were 8877.5 and 1420.40 MJ, respectively (see Table 3). Monthly per capita sawdust energy consumption for the whole sample was 7.05 MJ. It was 3.57 MJ for low-income FHHs, which rose to 6.98 MJ for the medium-income FHHs and 16.16 MJ for the high-income FHHs (see Table 4).

Kerosene was usually bought from the nearby fuel stations of the town, with 91.8 per cent of households reportedly able to purchase this fuel within 1 km of their residence. The average price of 1 L was Birr 12.50. Therefore, 0.08 L of kerosene costed for Birr 1. One litre of kerosene delivers 33.63 MJ of heat value (Annex. 1). Therefore, 0.08 L of kerosene delivered 2.69 MJ (0.08 L × 33.63 MJ) of heat value. Total monthly kerosene consumed by the whole sample FHHs was about 162.24 L (2028 Birr × 0.08 L), which provided a total heat value of 5455.32 MJ. The total monthly kerosene energy consumed, therefore, varied from 2480.18 MJ for high-income FHHs to 1377.28 MJ for medium-income FHHs, and it was 1597.86 MJ for low-income FHHs. Total and end-use monthly kerosene energy consumption for the whole sample FHHs was 5455.32 and 2727.66 MJ, respectively (see Table 3). Monthly per capita kerosene energy consumption for the whole sample was 4.33 MJ. It was 2.72 MJ for low-income FHHs, which rose to 3.10 MJ for the medium-income FHHs and 10.88 MJ for the high-income FHHs (see Table 4).

With regard to electricity supply, the condition is better. Almost the entire sample dwelling units had access to electricity supply through either private connection (97.44%) or shared connection (2.56%). According to Ethiopian Electric Light and Power Authority, the price of electricity was based on fixed rate payment. The average price of 1 kWh of electricity was Birr 0.39. Since Birr 0.39 was equivalent to 1 kWh, therefore, Birr 1 was equivalent to 2.56 kWh. An FHH bought 2.56 kWh of electricity for Birr 1, and 1 kWh of electricity is equivalent to 3.6 MJ of energy (Annex. 1). Thus, for Birr 1, an FHH bought heat value of 9.22 MJ (2.56 kWh × 3.6 MJ). Total monthly electricity consumed by the whole sample FHHs was about 3238.40 kWh (1,265 Birr × 2.56 kWh), which provided a total heat value of 11663.30 MJ. The total monthly electricity energy consumed varied from 6361.8 MJ for high-income FHHs to 5301.50 MJ for medium-income FHHs, and it was nil for low-income FHHs. Total and end-use monthly electricity energy consumption for the whole sample FHHs was 11663.30 and 8747.48 MJ, respectively (see Table 3). Monthly per capita electricity energy consumption for the whole sample was 9.26 MJ, and it was nil for low-income FHHs. However, this figure rose to 11.94 MJ for the medium-income FHHs and 27.90 MJ for the high-income FHHs (see Table 4).

Table 4.

Monthly Total and Per Capita Cooking Energy (MJ) Use by Various FHHs Income Groups

Income Group	Biomass Cooking Fuels						Modern Cooking Fuels
	Wood		Charcoal		Sawdust		Kerosene		Electricity
	Total	Per Capita	Total	Per Capita	Total	Per Capita	Total	Per Capita	Total	Per Capita
Low	36516.92	62.21	26967.37	45.94	2093.75	3.57	1597.86	2.72	nil	nil
Middle	43204.04	97.31	27883.07	62.80	3098.75	6.98	1377.28	3.10	5301.5	11.94
High	25207.90	110.56	16894.67	74.10	3685	16.16	2480.18	10.88	6361.8	27.90
Total/average	104928.9	83.34	71745.11	56.99	8877.5	7.05	5455.32	4.33	11663.3	9.26

Source: The author.

As indicated in Table 3, on aggregate, the surveyed FHHs consumed a total cooking energy of 202670.08 MJ per month, of which biomass fuels constituted 185551.46 MJ (91.55%), while the remainder 17118.62 MJ (8.45%) was consumed in the form of modern fuels. Of the total biomass energy consumed, wood accounted for the highest share of 104928.86 MJ followed by charcoal (71745.10 MJ) and sawdust (8877.50 MJ). Of the total monthly heat energy values obtained from modern cooking fuels per FHH, electricity has a dominant share (11663.30 MJ) followed by kerosene (5455.32 MJ). Monthly per capita biomass energy consumption for the whole sample FHHs was 147.38 MJ. The total per capita monthly biomass energy consumption was 111.72 MJ for low-income FHHs, which rose to 167.09 MJ for the medium-income FHHs and 200.82 MJ for the high-income FHHs (see Table 4). The total per capita monthly modern energy consumption for the whole sample FHHs was 13.60 MJ. Thus, per capita monthly modern fuel consumption varied considerably from FHH to FHH according to their income status. The total per capita monthly modern energy consumption was 2.72 MJ for low-income FHHs, and this figure rose to 15.04 MJ for the medium-income FHHs and 38.78 MJ for the high-income FHHs (see Table 4).

Nevertheless, what becomes important to the consumers is not the gross energy consumed but how much end-use or useful energy did the FHHs consume. The amount of useful energy received by a family for a given gross energy input is determined by the efficiency level of energy resource utilisation. According to Clancy (2006), the total energy use is the amount of total energy that people use regardless of the efficiency of the appliances that they use, whereas the end-use energy is the energy that is adjusted for the efficiency of the appliance, technology and mode of use by the household. Thus, it stands for the amount of energy effectively utilized. As shown in Table 3, the efficiency level of the use of energy source depends on the type of energy resource. Wood fuels such as wood and charcoal are used to function at the efficiency level of 10 per cent and 20 per cent, respectively, whereas modern fuels such as kerosene and electricity are used to function at better efficiency levels of 50 per cent and 75 per cent, respectively (Annex. 1).

It is very difficult to separate the type of fuel and the end-use equipment used by households. The type of energy used depends on the availability and cost of the associated end-use device. For this reason, many urban FHHs of the study area are unable to change their cooking energy consumption from using woody biomass to electricity by installing electric mitads (stoves to bake injera using electricity). So far, the majority of FHHs are dependent upon the simplest and most common method of cooking, open hearth or three-stone fire, mainly to bake injera (flat pancake like bread commonly made of locally grown grain called Teff), which is by far the most important domestic function among FHHs.

As the data in Table 5 show, about 29.35 per cent of FHHs own traditional injera “mitad” (a clay-made circular pan used for baking “injera”), whereas the mirte (improved injera baking biomass stove), which is more efficient and allows proper burning of the wood, is used by 7.61 per cent urban FHHs and only about 4.21 per cent of FHHs own electric mitads (see Table 5). Once the FHHs obtain electric connection, they almost usually use it entirely for lighting. When asked about the reasons why electricity is not used for cooking purpose, almost all the non-adopters responded that they cannot afford to spend on electric mitads. There are two types of charcoal stoves used in the sample population, traditional iron charcoal stoves and lakech (“excellent”) charcoal stoves, which constitute 4.48 per cent and 31.79 per cent of the total energy cooking end-use appliances, respectively. The proportion of FHHs using lakech stove is higher and almost similar among the low and middle income groups, whereas the proportion of FHHs using kerosene stove (22.55%) is higher and almost similar among the middle and high-income groups.

Table 5.

Urban FHHs Owning Different Cooking Appliances

Income Group	Earthen Plate injera mitads	Traditional Iron Charcoal Stoves	Mirtemitads	Lakech Charcoal Stoves	Kerosene Stoves (Buta gas)	Electric injera mitads	Total Number of Cooking Appliances	Percentage
Low	102	28	5	82	34	nil	251	34.10
Middle	85	5	29	94	68	11	292	39.67
High	29	nil	22	58	64	20	193	26.23
Total	216	33	56	234	166	31	736	100
Percentage	29.35	4.48	7.61	31.79	22.55	4.21		100

Source: The author.

In addition to baking injera, wood is also used for cooking local foods, namely, kurkufa and fosossie (local foods prepared from maize dough), preparing wot, making coffee and tea as well as brewed alcoholic drinks among the lower income FHHs, while charcoal is most preferred to wood for cooking wot and for making coffee and tea in all income FHHs. The contribution of modern energy source to the similar activities was minimal except for kerosene, which was proportionately used less for preparing wot and making tea. The financial limitation of households to have electric stoves constrained most households to substitute electrical energy for fuel wood. Instead, households continue to depend on wood fuels to meet their energy needs for cooking.

Cooking Energy Determinants

The incidence of households headed by women is very likely to grow in the study area through various routes. One of the main routes is the termination of marriage, either through divorce or widowhood or separation, but the most common one is to never marry. The majorities (37.9 per cent) of the respondents are single, 22.1 per cent are widowed, 16.6 per cent are divorced, 14.3 per cent are separated, while only 9.1 per cent are married. The minimum and maximum family size was 1 and 11, respectively. The average family size in the FHH was 6. About 47 per cent of the respondents have family size of less than three members. The family size numbering 29.4 per cent have between four and seven family members, 23.5 per cent had between eight and above members. The total family members in the sample were 1,259, of which there were more female members (52.49 per cent) than the male ones (47.51 per cent). The family members in low-, medium-, and high-income groups were 587, 444, and 228, respectively.

The result shows the maximum age observed from the sample respondents was 66, while the minimum was 25. About 36.8 per cent of the respondents were between 31 and 40 years, 15.4 per cent in the range of 21–30 years, 33.8 per cent between 41–50 years, and 13.9 per cent were 51 and above. In terms of housing status of the respondents, it was found that 43 per cent FHHs live in their own houses, while 30.1 per cent were rented from kebeles and the rest 26.8 per cent were rented from private owners. Those lacking own houses are also living in an overcrowded rooms and poor housing conditions with a serious lack of basic facilities. The educational status of the surveyed FHHs shows that 34.9 per cent of the respondents had obtained degree and diploma, 27.6 per cent had completed secondary education, 13.2 per cent had primary education and 12.5 per cent of the respondents did not have any formal schooling at all. Respondents amounting to 27.6 per cent were government employees, most of urban FHHs in the low-income group are engaged in informal sector of the economy in order to sustain their life, about 20.2 per cent were found to be relying on low paying and insecure income sources such as daily labour and petty trades such as Areki, Tella, Chakka preparations (all are sorts of homemade/locally brewed alcoholic beverages and injera selling) to make their living and 14.3 per cent engaged in their own relatively better private jobs.

The Pearson coefficient of correlation was utilized to compute the relationship between the expenditure made on cooking fuels (dependant variable) and other independent variables. Multicollinearity could not make significant impact on the quality and stability of the fitted regression model. The coefficient correlation matrix of the model estimate was evaluated, and the exogenous variables are not highly correlated. The highest value recorded was 0.787, which was between age and family size of the FHH (Table 10). As indicated in Table 6, the Pearson’s coefficients of correlation computed were 0.603 for income, 0.418 for education, 0.362 for occupation, 0.325 for house ownership, 0.541 for number of cooking end-use equipment, 0.495 for quality of cooking end-use equipment and 0.423 for quality of housing unit, all suggesting the existence of positive association with the consumption of cooking fuels. However, the Pearson’s coefficients of correlation computed were negative for age (–0.136) and family size (–0.144) but not significant at 0.025 and 0.018 levels, respectively. Similarly, modern cooking fuels expenditure is positively and significantly related to all independent variables except age (–0.088) and family size (–0.057) (see Table 6). The data further reveal that the association between biomass cooking fuel consumption and occupation level of the household is statistically supported at 1 per cent probability level and other variables such as education (0.154), income of the FHH (0.151) and house ownership (0.127) are significant at p < 0.05. On the other hand, family size (–0.085), quality of cooking end-use equipment (–0.085), age (–0.51) and number of cooking end-use equipment (–0.104) were computed to have negative and insignificant effect on the likelihood of expending on wood fuels.

Table 6.

Association between Fuel Expenditure and Independent Variables

Type of Cooking Fuels Used	Family Income	Education	Occupation	Age	Family Size	House Owner-ship	Number of End- Use Equip.	Quality of End- Use Equip.	Quality of the House
Expenditures made on biomass cooking fuels	0.151^*	0.154^*	0.180^**	–0.051	–0.085	0.127^*	–.104	–0.085	–0.028
Expenditures made on biomass cooking fuels	0.013	0.011	0.003	0.407	0.161	0.036	0.087	0.162	0.102
Expenditures made on modern cooking fuels	0.469^**	0.271^**	0.183^**	–0.088	–0.057	0.203^**	0.685^**	0.616^**	0.623^**
Expenditures made on modern cooking fuels	0.000	0.000	0.002	0.149	0.346	0.001	0.000	0.000	0.000
Expenditures made on all cooking fuels	0.603^**	0.418^**	0.362^**	–0.136^*	–0.144^*	0.325^**	0.541^**	0.495^**	0.423^**
Expenditures made on all cooking fuels	0.000	0.000	0.000	0.025	0.018	0.000	0.000	0.000	0.000

Source: The author.

Notes: ^*Correlation significant at 0.05 level (two-tailed). ^**Correlation significant at 0.01 level (two-tailed).

FHHs, where the head has a higher level of income, education and occupation, are more likely to spend more on all sorts of cooking fuels. The amount of modern cooking energy consumed is positively and strongly correlated with the ownership and quality of the house, the number and quality of cooking appliances used. The positive signs of these coefficients more importantly signify that FHHs with higher income have greater capacity to pay and would choose the use of modern cooking fuel, and respondents with higher education and occupation levels would have greater awareness about the health effects of biomass fuels. FHHs with a head that had large number of end-use equipment had higher modern fuel adoption probability than the households with a head with lower number of equipment. The positive coefficient for quality of cooking appliances shows that the higher the consumption of modern cooking energy, the better was the efficiency in the utilisation of cooking energy resource. House ownership and quality of housing unit also considered to be determinant factors that significantly affect FHHs dependence on the use of modern fuels for household cooking. Modern cooking fuel use is strongly associated with house ownership status of the household. The data imply that the association between quality of housing unit and use of modern fuels is significant at 1 per cent probability level. FHHs residing in their own housing units tend to use more clean fuels than those of rented FHHs. One can infer from the results that residential quality is one of the prerequisites to consume modern cooking fuels. Obviously, FHHs living in better housing units are able to use the readily available utilities.

In the model summary, which gives the R-square statistic, the coefficient of multiple correlation computed was 0.695 and the coefficient of determination was 0.482. The value of 0.695 indicates that there is a very high statistical association between the exogenous variables chosen in the model and the endogenous variable of the main equation (cooking fuel consumption). The R-square and adjusted R-square of 0.482 and 0.465, respectively, confirmed that the model is well fitted and variables are appropriate. The coefficient of determination confirms that about 48 per cent variation in the dependent variable is explained by the variables, and it can be said that independent variables in the model are relevant and appropriate to explain dependent variable, that is, consumption of cooking energy. Besides R-squared (see Table 7), we can use ANOVA to check how well the model fits the data. To corroborate the above assertion, the ANOVA results of the model, with F-value of 27.129, estimated at 9 and 262 df (and a low standard error of 14.00), gave a ρ value of 0.000. F-test is used to test whether or not there is significant relationship between the price of the house and all other independent variables. The acceptability of the model is tested by using the ANOVA (see Table 8). The model is acceptable and is significant to explain the cooking fuel consumption because the significant value of F statistics is less than 0.05 per cent, which implies that the variation explained by the model is not due to chance. From Tables 7 and 8, we can find how well the model fits the data and ANOVA, which give us information about the model as a whole.

Table 7.

Model Summary

Model	R	R-Square	Adjusted R-Square	Std. Error of the Estimate
1	0.695	0.482	0.465	14.00346

Source: The author.

Table 8.

ANOVA

Model	Sum of Squares	df	Mean of Squares	F	Sig.
Regression	47879.476	9	5319.942	27.129	0.000
Residual	51377.403	262	196.097
Total	99256.879	271

Source: The author.

After checking for the model-fit and the evaluation of the F-value and R-square, it is important to find out the coefficients, which give results of the regression analysis and evaluate the standardized coefficients or betas, which are used to indicate the rate of change in independent variable when independent variable is changed by one unit. The “coefficients of correlations” (see Table 9) and “matrix” thereof (see Table 10) present the optimal weights in the regression model.

Table 9.

Beta Coefficients

Model	Unstandardized Coefficients		Standardized Coefficients	t Values	Sig.
Model	Beta	Std. error	Beta	t Values	Sig.
Constant	57.790	6.810		8.486	0.000
Family income	0.006	0.002	0.325	3.694	0.000
Education	0.209	0.710	0.019	0.295	0.768
Occupation	2.063	0.629	0.170	3.277	0.001
Age	–0.135	0.155	–0.063	–0.870	0.385
Family size	0.118	0.495	0.018	0.238	0.812
House ownership	2.645	1.361	0.117	1.944	0.053
Number of cooking end-use equipment	9.110	1.329	0.363	6.853	0.000
Quality of cooking end-use equipment	–0.460	0.742	–0.043	–0.621	0.535
Quality of house	–0.112	1.463	–0.005	–0.076	0.939

Source: The author.

Table 10.

Coefficient Correlation Matrix

	Cooking Fuel Expense	Family Income	Education	Occupation	Age	Fami-ly Size	House Owner-ship	No. of End- Use Equipment	Quality of End- Use Equipment	Quality of the House
Cooking fuel expense	1	0.603^**	0.418^**	0.362^**	–0.136^*	–0.144^*	0.325^**	0.541^**	0.495^**	0.423^**
Sig.		0.000	0.000	0.000	0.025	0.018	0.000	0.000	0.000	0.000
Family income	0.603^**	1	0.694^**	0.429^**	–0.170^**	–0.248^**	0.451^**	0.463^**	0.742^**	0.654^**
Sig.	0.000		0.000	0.000	0.005	0.000	0.000	0.000	0.000	0.000
Education	0.418^**	0.694^**	1	0.416^**	–0.115	–0.157^**	0.417^**	0.213^**	0.580^**	0.483^**
Sig.	0.000	0.000		0.000	0.058	0.009	0.000	0.000	0.000	0.000
Occupation	0.362^**	0.429^**	0.416^**	1	–0.154^*	–0.210^**	0.133^*	0.120^*	0.428^**	0.333^**
Sig.	0.000	0.000	0.000		0.011	0.000	0.028	0.047	0.000	0.000
Age	–0.136^*	0.170^**	–0.115	–0.154^*	1	0.787^**	0.191^**	–0.090	–0.160^**	–0.055
Sig.	0.025	0.005	0.058	0.011		0.000	0.002	0.139	0.008	0.363
Family size	–0.144^*	–0.248^**	–0.157^**	–0.210^**	0.787^**	1	0.163^**	–0.057	–0.201^**	–0.124^*
Sig.	0.018	0.000	0.009	0.000	0.000		0.007	0.349	0.001	0.041
House ownership	0.325^**	0.451^**	0.417^**	0.133^*	0.191^**	0.163^**	1	0.165^**	0.385^**	0.588^**
Sig.	0.000	0.000	0.000	0.028	0.002	0.007		0.006	0.000	0.000
Number of end-use equipment	0.541^**	0.463^**	0.213^**	0.120^*	–0.090	–0.057	0.165^**	1	0.454^**	0.280^**
Sig.	0.000	0.000	0.000	0.047	0.139	0.349	0.006		0.000	0.000
Quality of end–use equipment	0.495^**	0.742^**	0.580^**	0.428^**	–0.160^**	–0.201^**	0.385^**	0.454^**	1	0.511^**
Sig.	0.000	.000	0.000	0.000	0.008	0.001	0.000	0.000		0.000
Quality of the house	0.423^**	0.654^**	0.483^**	0.333^**	–0.055	–0.124^*	0.588^**	0.280^**	0.511^**	1
Sig.	0.000	0.000	0.000	0.000	0.363	0.041	0.000	0.000	0.000

Source: The author.

Notes: ^*Correlation significant at 0.05 level (two-tailed). ^**Correlation significant at 0.01 level (two-tailed).

Income of the FHH has a significant and positive effect on the total cooking fuel consumption. The beta coefficient 0.325 significant at 0.00 recorded for income implies that for every unit change in monthly income of the FHH, the monthly fuel expense increases by 33 per cent. Thus, the relationship or association between total energy consumption and monthly income of the household is significant at 0.000 probability level. The degree of association indicates that in urban areas where all energy sources are commercialized, access to energy is determined by the purchasing power of the families. Thus, households with better income level could have better access to all sorts of energy available in the market. Statistically, beta coefficient 0.363 significant at 0.00 shows that there is a corresponding increase in expenditure made on domestic cooking energy with an increase in the number of cooking end-use equipment as a whole; it, therefore, implies that the more end-use equipment an FHH had, the more the expenditure on total cooking fuel. A significant relationship was also found between occupation of the FHH and cooking energy expenditure. The probability of expending on total cooking fuels increases with increase in the level of occupation. The beta coefficient 0.170 at p < 0.05 significance level asserts that for every unit of occupation increased in the study area, the people will tend to increase their expense for cooking fuel by 17 per cent. Statistically presented beta coefficients for family size (0.018), for education (0.019) and for housing tenure of the FHH (0.117) that are non-significant at 0.812, 0.758 and 0.053, respectively, show positive influence but were not strong enough to significantly affect the monthly expenditure on cooking energy fuels. Other variables such as age (beta coefficient of –0.063), stove quality (beta coefficient of –0.043) and housing quality (beta coefficient of –0.005) turned out to be insignificant and negative.

Most FHHs cannot afford modern cooking fuels and proper appliances, that is, they have less access to appliances of higher quality because they are too expensive for them. Therefore, most urban FHHs cannot easily make a transition from biomass to electricity for cooking end-use since the high costs of modern cooking stoves are major constraints for them. This indicates that no substantial or complete switching from wood to electricity had occurred in household energy use. The majority households often lack the ability to optimize their consumption through improved technologies. A family has to purchase and improve domestic cooking appliances and engage in a better level of occupation in order to acquire adequate income for the healthier utilisation of the available modern cooking fuels. Though access to electricity in urban areas is not considered as a problem, yet only 4.21 per cent of the urban FHHs rely on electricity for cooking. When urban FHH gets connected to electricity they generally continue to use biomass fuels for cooking and the use of electricity was mainly for lighting and TV viewing and radio listening rather than cooking. The study results indicate that biomass is the most important source of energy and remain to be dominant source of energy in all income groups in urban FHHs, not only for the poor but also for non-poor FHHs. Most of the middle and higher income FHHs supplemented their use of wood fuels with conventional fuels for the purpose of cooking. Even these FHHs had not completely substituted their biomass fuels with electricity. They still spend more on wood and charcoal. This is an indication that there is a slow long-run energy transition prospect in the town.

In fact, Ethiopia has one of the lowest rates of access to modern energy services, whereby the energy supply is primarily based on biomass which in case of domestic use is usually stemming from unsustainable sources. With a share of 92.4 per cent of country’s energy supply, waste and biomass are the primary energy sources, followed by oil (5.7%) and hydropower (1.6%). According to estimates made by a recent study, at the national level, there appears to be a surplus of woody biomass supply. However, the same study revealed that there is a severe deficit of supply when the data are disaggregated to lower local levels. According to the same study, 307 Woredas (districts) out of the total number of 500 Woredas are consuming woody biomass in excess of sustainable yield (Energypedia, 2020). The results of the present study lend support to the findings reported at the national level regarding the pattern of domestic energy use which continues as in the past. The key question that arose here is: Why are the clean sources of fuel that are accessible to the FHHs in the study area not being used for cooking? Clearly, the increase in FHH’s income is too slow to permit FHHs to switch from wood fuels consumption to more technologically efficient sources of energy for the purpose of cooking.

Conclusion

Cooking takes up a big portion of urban FHHs end-use energy consumption. Disparities in cooking practices are not reflected across the different socioeconomic groups. The conditions of cooking in most FHHs are poor, with women generally cooking indoors in non-ventilated areas. Despite accessibility to electricity in urban areas, many FHHs still rely on biomass as their primary source of cooking energy. Even as income of FHH increases, the proportion of FHHs using wood for baking injera is not declining at large. Wood fuels (wood and charcoal) are by far the most used cooking fuels for a large majority of urban FHHs in spite of the growing scarcity and high prices of these resources. Even though it has irregular supply, the share of kerosene is also significant in the overall consumption of cooking energy. The use of kerosene is becoming more popular cooking fuel among middle- and high-income FHHs in the town.

Most FHHs still appear not to be benefiting significantly from modern fuel supply availability. Modern energy sources are either unavailable or limited to certain purposes other than cooking. LPG is not an alternative urban cooking fuel option; it is not commonly used due to lack of its general availability and much higher cost for household use. Biomass-based clean fuels, such as biogas, have not yet been commercialized; thus, the alternative wood fuels still remain viable solutions for FHHs with a significant health risk as a result of indoor pollution. The bulk of wood fuels is used for cooking in FHHs despite the fact that they have access to grid electricity. Only very small proportion of FFHs is found to use electricity for cooking purposes. Physical access alone does not ensure the households to benefit from the energy services. The real access to energy services can be limited by the purchasing power of the household and cost of energy-using equipment.

Most FHHs face high burden of energy accessibility or unable to use modern energy for cooking, as they have limited access to modern end-use technologies. The costs of the available better appliances are beyond the purchasing ability of most of urban FHHs. Most of the FHHs in the study area cannot afford electrical appliances to substitute for polluting and time-consuming traditional cooking technologies in order to meet their cooking demands. The provision and adoption of modern energy technologies (lakech and mirte) among FHHs has not been a great success in the town. Most end-use technologies used by most FHHs in the town are inefficient, and such energy inefficiency mode of utilization of traditional fuels leads to the massive waste of wood and contributes to deforestation. Such limited use of clean energy and reliance on burning traditional biomass has very limited opportunities for economic and social advancement.

It was observed that the problem of clean energy accessibility is acute since the majority of FHHs consumed less end-use cooking energy services due to their large dependence on traditional fuels that are used at very low efficiency. Their cash income is so low that even modest changes in energy expenditures can be a real hardship for them. In particular, lower income FHHs often end up spending a substantial proportion of their household income on energy when compared to the share spent by better-off groups. A family in lower economic status spends up to 11 per cent of its income on domestic energy, so fuel crisis affects families of the lower economic status adversely in comparison to those in the high-income levels. This becomes a large economic burden on their budget.

This study revealed that the shift to modern fuels with increasing income is not well manifested in the town and FHHs are not directly moving up the “energy-ladder” even as their incomes grow. Higher income FHHs move up the “energy-ladder” eventually switching to electricity for lighting, not for cooking. An increase in household income, however, does not necessarily mean a complete switching to clean cooking energy sources. FHHs even in the high-income group do not totally switch to modern cooking energy sources, and they do not directly replace their polluting and time-consuming traditional cooking technologies with the modern ones. It was found that while middle and higher income FHHs enjoy additional energy options, they were not predominantly reliant on modern energy sources but use a mix of commercial energy sources to meet their cooking energy needs. Despite the fact that the majority of middle and higher income FHHs combine the use of biomass fuels with other clean source of energy particularly kerosene to satisfy their cooking needs, wood fuels remain to be the dominant fuel among all lower income FHHs for cooking end-use. This implies that wood fuels (wood and charcoal) are still the choice of all FHHs, whether poor or better-off and FHHs seem inexorably linked. Thus, a transitional move with technological options for the majority is a challenge. Most FHHs lack access to modern energy technologies, which slow down the possibilities for using locally available clean sources of energy. The prices of wood fuels are still rising, and the high up-front costs for improved appliances prevent the majority of FHHs from changing their consumption patterns. This trend looks set to continue in the coming years, which means that unless swift and effective action is taken, the number of vulnerable energy consumers could also increase markedly.

The unsustainable use of wood fuel and the use of inefficient energy conversion technologies may result in serious adverse consequences for health, environment and economic development. Despite the fact that biomass fuels would remain important fuels for majority of the FHHs, no efforts were made to increase their supply. One great concern, however, is that the local authority does little to control access to the hinterland forests of the town from where wood fuel is extracted and supplied. An increase in household energy demand has led to massive deforestation on the outskirts of the town. This has resulted in serious shortage of wood fuels and higher prices. As women are the most intensive users of environmentally detrimental and unsafe fuels and are amongst the most vulnerable groups to energy-related problems, it is important that special attention be given to urban FHHs who deserve much closer attention from energy policy.

The energy policy must, therefore, support ways to use wood fuels more efficiently and sustainably through encouraging conservation of natural vegetation by growing trees as energy resources must be strengthened and intensified (Ponce et al., 2019). It is important to improve the efficiency of the traditional cooking stoves and change their cooking practices by employing cleaner alternative fuels and end-use technologies to improve energy efficiency so that the pressure on surrounding forests could be alleviated and household energy-related problems tackled. In addition to adopting improved stoves, women need to have well housing status. There is a need to scale up the kitchen and housing conditions for cooking process since many of the FHHs were found to be not only lacking their own houses but also living in an overcrowded rooms and poor housing conditions with a serious lack of basic facilities.

Footnotes

Declaration of Conflicting Interests

The author declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.

Funding

The author received no financial support for the research, authorship and/or publication of this article.

References

Abebe

, & Koch

S. F.

(2011). Clean fuel-saving technology adoption in urban Ethiopia (Working Papers 229). Economic Research Southern Africa. https://EconPapers.repec.org/RePEc:rza:wpaper:229

Abrahamse

(2011). Factors related to household energy use and intention to reduce it: The role of psychological and socio-demographic variables. Human Ecology Review, 18(1), 30–40.

Alam

, Sathaye

, & Barnes

(1998). Urban household energy use in India: Efficiency and policy implications. Energy Policy, 26, 885–891.

Ali

A. M.

(2020). Energy insecurity and alternative sources in Ethiopia: New energy policy is the need of the day. International Journal of Natural and Social Sciences, 7(3), 1–15.

Amoah

S. T.

(2019). Determinants of household’s choice of cooking energy in a global south city. Energy and Buildings, 196, 103–111.

Bakhsh

, Sadiqa

, Yasin

M. A.

, Haider

, & Ali

(2020). Exploring the nexus between households’ choice of cooking fuels, sanitation facilities and access to information in Pakistan. Journal of Cleaner Production, 257, 120621.

Barnes

, Krutilla

, & Hyde

(2004). The urban household energy transition: Social and environmental impacts in the developing world. Routledge.

Barnes

, & Toman

(2006). Energy, equity and economic development. In Lopez

& Toman

M.A.

(Eds.), Economic development and environmental sustainability: New policy options (pp. 245–272). Oxford University Press.

Cecelski

(2004). Re-thinking gender and energy: Old and new directions (Energia/EASE Conference, May 1–3). University of Twente, Twente, The Netherlands.

10.

Clancy

(2006). Urban poor livelihoods: Understanding the role of energy services. Best practices paper prepared for DFID KaR R8348. University of Twente, Twente, The Netherlands.

11.

Clancy

, Skutsch

, & Batchelor

(2003). The gender-energy-poverty nexus: Finding the energy to address gender concerns in development (Report to DfID). University of Twente, Twente, The Netherlands.

12.

CSA (2008). Population and housing census of Ethiopia: Result at country level. Analytical Report on Urban and Rural Population Size (Vol. II). Central Statistical Agency.

13.

Energypedia (2020). Ethiopia energy situation. https://energypedia.info/wiki/Ethiopia_Energy_Situation#Biomass.

14.

ESCAP (2008). Energy security and sustainable development in Asia and the Pacific. United Nations Economic and Social Commission for Asia and the Pacific. Bangkok, Thailand.

15.

Gould

C. F.

, Schlesinger

S. B.

, Molina

, Bejarano

M. L.

, Valarezo

, & Jack

D. W.

(2020). Household fuel mixes in peri-urban and rural Ecuador: Explaining the context of LPG, patterns of continued firewood use, and the challenges of induction cooking. Energy Policy, 136, 111053.

16.

Heltberg

(2004). Fuel switching: Evidence from eight developing countries. Energy Economics, 26(5), 869–887.

17.

Kohlin

, Sils

, Pattanayak

, & Wilfong

(2011). Energy, gender and development: What are the linkages? Where is the evidence? (Policy Research Working Paper 5800). The World Bank Sustainable Development Network Social Development Unit.

18.

Masera

, Saatkamp

, & Kammen

(2000). From linear fuel switching to multiple cooking strategies: A critique and alternative to the energy ladder model. World Development, 28(12), 2083–2103.

19.

Meikle

, & Bannister

(2005). Focus on the significance of energy for poor urban livelihoods: Its contribution to poverty reduction. Development Planning Unit, University College London. https://www.ucl.ac.uk/bartlett/development/sites/bartlett/files/migrated-files/44_0.pdf

20.

Meikle

, & North

(2005). The impact of energy use on poor urban livelihoods in Arusha, Tanzania. Development Planning Unit, University College, London. https://assets.publishing.service.gov.uk/media/57a08c60ed915d3cfd001352/r8321.pdf

21.

Niez

(2010). Comparative study on rural electrification policies in emerging countries (Information Paper). International Energy Agency.

22.

Ntobeg

(2007). Rural energy systems and the rural development process: A case study from Limpopo Province (Master of Arts thesis). University of South Africa, Pretoria, South Africa.

23.

Parikh

J. K.

(1995). Gender issues in energy policy. Energy Policy, 23(9), 745–754.

24.

Ponce

, Alvarado

, Ponce

, Alvarado

, Granda

, & Yaguana

(2019). Green returns of labor income and human capital: Empirical evidence of the environmental behavior of households in developing countries. Ecological economics, 160, 105–113.

25.

Smith

K. R.

(1987). The biofuel transition. Pacific and Asian Journal of Energy, 1(1), 13–32.

26.

Smith

K. R.

(1990). Risk transition and global warming. Journal of Energy Engineering-Asce 116(3), 178–188.

27.

TERI (2010). Alleviating energy poverty: Indian experience. The Energy and Resources Institute.

28.

Twumasi

M. A.

, Jiang

, Ameyaw

, Danquah

F. O.

, & Acheampong

M. O.

(2020). The impact of credit accessibility on rural households clean cooking energy consumption: The case of Ghana. Energy Reports, 6, 974–983.

29.

Woldehanna

, Hoddinott

, Ellis

, & Dercon

(2008). Dynamics of growth and poverty in Ethiopia (1995/ 96-2004/05). Development Planning and Research Department, Ministry of Finance and Economic Development.