Correlation of woodfuel production participation among rural households in the drylands of Ethiopia

Woodfuel production and consumption have been a concern for multiple stakeholders involved in household energy use, deforestation, and climate change. While research into the underlying decision-making process is growing, it remains insufficient. Such a study offers opportunities to develop policies that enable diversification of household energy consumption and livelihood options away from woodfuel use. Policymakers often lack an understanding of factors correlated with households' participation in woodfuel production. Therefore, this study examined the correlation between household participation in woodfuel production and factors that influence households' participation in woodfuel production in dryland areas of Ethiopia. Data were collected from 1,114 purposively selected woodfuel-producing and non-producing households through household surveys, key informant interviews, and focus group discussions. The sample included 775 participant households and 339 non-participant households. The collected data were analyzed using descriptive statistics and a binary logistic regression model. The results revealed that drought and related shocks are the main factors that forced households to participate in woodfuel production. The model results indicated that age, education, landholding, livestock holding, production asset value, ownership of improved cook stoves, number of years lived in the area, distance from the forest, access to forest extension, and institutional membership are statistically significant factors that negatively influence household participation in woodfuel production. On the other hand, household expenditure and drought occurrence positively and significantly influence the participation of households in woodfuel production. The findings of the study suggest that sustainable management and utilization of dryland forest resources require considering socioeconomic, demographic, institutional, and environmental factors correlated with households' decisions to participate in woodfuel production. This can be achieved through sound institutional setups and policy frameworks in the sector.

1 Introduction

Woodfuel (charcoal and firewood) offers various benefits such as providing energy for over 40 million people globally (FAO, 2017). Woodfuel serves a crucial energy source for the developing world (Njenga et al., 2023), accounting for 50%−90% of household energy in developing countries (Singh et al., 2018). Approximately 70% of the population in sub-Saharan Africa rely on woodfuel as their primary energy source (Sola et al., 2017). Similarly, approximately 92% of the population in Ethiopia depends on woodfuel (Mondal et al., 2018). In both rural and urban Ethiopia, woodfuel is the primary source of energy for cooking and heating, and it is expected to remain a significant energy source in the future, not only in Ethiopia but also in developing countries worldwide (Tucho and Nonhebel, 2017). Currently, woodfuel production is increasing due to the preferences and demands of the growing urban population (Bekele and Kemal, 2022; Yalew, 2022).

In addition to serving as an important energy source, woodfuel is also a vital means of livelihood for rural households and plays a role in improving the assets of developing countries such as Ethiopia (Sola et al., 2017; Girma et al., 2022). In Ethiopia, woodfuel is the most important forest product, with a total volume of 116 million cubic meters consumed in 2013 (MEFCC, 2017). Ethiopia's woodfuel production and consumption in 2015 was estimated at 108,173,872 and 108,171,205 m³ at the national level, respectively (UNEP, 2019). Woodfuel, specifically charcoal, provides regular household income and seasonal income during lean agricultural seasons. It serves as a safety net to mitigate economic shocks and contributes toward uplifting households from poverty (Obiri et al., 2014; Jones et al., 2016; Ndegwa et al., 2016; Smith et al., 2017; Brobbey et al., 2019). Woodfuel is essential for meeting a wide range of needs within rural household livelihoods, especially in dryland areas (Girma et al., 2022).

Despite its significant household economic and energy contributions, woodfuel has a negative impact owing to its correlation with adverse environmental effects. The over-dependency on woodfuel production and consumption from forests and woodlands leads to depletion of natural resources in developing countries, including Ethiopia (Yalew, 2022). Moreover, the exploitation of forest resources for woodfuel production is not properly managed and is unsustainable in Ethiopia (MEFCC, 2017; Tassie et al., 2021). It is estimated that 37 million cubic meters of woodfuel is produced through unsustainable extraction from natural forests. Unsustainable woodfuel production and consumption contribute to forest and land degradation as well as climate change impacts (Njenga et al., 2023). Charcoal production has a stronger negative correlation with forest and land degradation compared to firewood (Chidumayo and Gumbo, 2013; FAO, 2017). In Ethiopia, the annual forest degradation due to unsustainable woodfuel production is estimated to be between 0.58 and 0.87 million hectares (Yalew, 2022). According to Chidumayo and Gumbo (2013), greenhouse gas emissions from charcoal production in tropical ecosystems amounted to 71.2 million tons of carbon dioxide and 1.3 million tons of methane in 2009, which was equivalent to 7% of tropical deforestation. As a result of unsustainable woodfuel production and consumption, forest degradation accounts for approximately 46% of the total greenhouse gas emissions from the forestry sector (FDRE, 2011). Without new policy interventions, the number of people dependent on woodfuel production and consumption is expected to increase due to population growth and an increasing demand for biomass, exacerbating the negative impact (Rafaj et al., 2018).

Previous evidence has shown that socioeconomic, demographic, and institutional factors play a critical role in shaping the outcomes of forest ecosystems, largely through their impact on human behavior and incentives for woodfuel production (Mulenga et al., 2015; Brobbey et al., 2019; Tassie et al., 2021). Furthermore, different types and levels of institutional arrangements interact with socioeconomic and demographic attributes in various ways, resulting in specific outcomes for forest use and conditions for manageable woodfuel production (Mulenga et al., 2015; Vollmer et al., 2017). This is particularly essential in the design and implementation of policies and programs aimed at achieving environmentally sustainable woodfuel production in the drylands of Ethiopia. A rural household, pursuing feasible livelihood strategies, is the ultimate decision-maker regarding the use of natural resources to carry out activities for achieving desired objectives (Abera et al., 2021). When making production decisions, smallholder rural households aim to maximize their livelihood benefits over time based on their existing resource assets. These decisions are influenced by the prevailing socioeconomic and policy environment. For example, a household's decision to engage in woodfuel production in dryland woodlands may be affected by various factors such as socioeconomic, policy, and natural elements (Brobbey et al., 2019; Tassie et al., 2021). It is worth mentioning that household poverty, food security status, and natural disasters such as drought can motivate households to participate in such activities.

This study aims to understand the factors that influence households' decision-making process regarding whether to produce woodfuel or refrain from it. This understanding is crucial for the success of any policy intervention in the woodfuel sector, directing interventions toward livelihood improvement, poverty reduction, and environmental conservation, specifically targeting woodfuel and other natural resources. This study provides opportunities to develop and refine policies regarding woodfuel that encourage people to transition from woodfuel use to renewable energy resources. However, the identification and analysis of these factors as well as the necessary information for practical policy and development actions are limited and lack concrete evidence from the primary population directly involved in woodfuel production activities. Therefore, a detailed analysis of the factors determining woodfuel production in the dry woodland areas of the country is necessary for better targeting and designing intervention strategies.

This study was conducted in the drylands of Ethiopia to (1) assess the motivation of households engaged in woodfuel production, (2) characterize different shocks that push households toward woodfuel production, (3) understand the factors that influence and motivate households to participate in woodfuel production, and (4) identify policy options that align with the interest of smallholders and the forest ecosystem. The results of this study contribute to designing appropriate policies and interventions by the government and development practitioners to achieve sustainable woodfuel production and consumption.

2 Materials and methods

2.1 Study area

The study was conducted in dryland areas of Ethiopia, with sample districts drawn from various regions. Three livelihood groups of the target population were included in the study: farming, agro-pastoral, and pastoral communities. The target population comprised woodfuel producers and nonproducers found in the dryland woodland areas. The study was conducted in the eastern Ethiopia, and it covered the Somali region and rural areas of Dire Dawa Administration Council from eastern Ethiopia, the Amhara region from northwestern Ethiopia, the Afar and Oromia regions from central Ethiopia, and the SNNPR from southern Ethiopia. These areas have the typical woodfuel producers, and the study covered 11 dryland districts (Table 1).

Table 1

Table 1. List of the surveyed sample districts from dryland areas of Ethiopia.

2.2 Sampling method

The target population consists of communities in dry woodland areas of the country. These communities further comprise four groups of households based on their participation in woodfuel production: households that participate in charcoal production, firewood production, and both charcoal and firewood production and households that do not participate in woodfuel production. We selected sample households from each of these population strata. We used the snowball sampling method for participating households and the stratified random sampling procedure for non-participating households. As a result, we sampled 775 participating households and 339 non-participating households, obtaining a total of 1,114 households.

2.3 Data collection and variable selection

The data were collected from primary and secondary sources. Surveys were conducted with individuals who indicated that producing and selling woodfuel (charcoal or fuelwood) is one of their activities as well as with nonproducers. A mixed-method approach was used, combining focus group-based rural appraisal tools and semi-structured interviews.

Semistructured interviews were conducted to gather quantitative data from households that produce and do not produce woodfuel. Focus group discussions (FGDs) and key informant interviews (KIIs) were performed to collect qualitative data regarding the reason for household engagement in woodfuel production and the occurrence and severity of different shocks. One FGD per kebele was conducted with a diverse mix of men and women who were actively engaged and not engaged in woodfuel production. A total of 44 FGDs were held, with each group consisting of 8–12 woodfuel producers and non-producers of mixed genders. In addition, 33 KIIs were conducted in each district with experts, elders, and religious leaders. Initial participants were selected by district and kebele experts, and subsequent participants were selected through snowball sequential sampling. Different participants were used for each rural appraisal exercise. In addition to addressing a core set of questions, the research also followed the interests and experiences of specific groups, resulting in some data being collected only from certain groups, as noted in the results section. The expected explanatory variables used in the model were listed from related literature (Table 2).

Table 2

Table 2. Variables used in the binary logit model.

2.4 Data analysis and binary logit model specification

Before conducting data analysis, data management, data transformation, and diagnostic tests such as multicollinearity and correlation were performed. The collected dataset included both qualitative and quantitative data; descriptive statistical tools and qualitative narration were utilized for quantitative and qualitative data analysis, respectively. Quantitative data were analyzed using statistical tools such as frequency, percentage, mean, and standard deviation. Independent t-tests and chi-squared tests were used to analyze the correlation of continuous and dummy variables, respectively, with the households' decisions to produce woodfuel. Qualitative data collected through FGDs and KIIs were analyzed using text summaries and narratives and were subsequently triangulated with quantitative survey results.

Furthermore, the contributing factors of households participating in woodfuel production were analyzed using a binary logit model to identify and analyze the determinants of households' decisions to produce woodfuel. The selected model was suitable for the logistic nature of data distribution. For data analysis, STATA version 16 was used, and the results are presented in graphical, narrative, and tabular forms.

2.5 Binary logit model specification

The standard form of the logit model is given by Greene (2003). The logit formulas estimate the probability of Y to be 1 (decision to participate); the probability is referred to as follows (Equations 1 and 2).

$\begin{array}{ll}\text{Y}=0\text{is }1-{\text{q}}_{\text{it}}\text{(decision to not participate)}& (1)\end{array}$

$\begin{array}{ll}\text{P}\left({\text{Y}}_{\text{t}}=\frac{1}{{\text{X}}_{\text{t}}}\right)=\frac{{\text{e}}^{\text{β}\text{X}}}{{1+\text{e}}^{\text{β}X}}& (2)\end{array}$

An equivalent form can be stated as follows (Equation 3).

$\begin{array}{ll}\frac{e^{\text{β}X}}{{1+e}^{\text{β}X}}=\frac{1}{{1+e}^{\text{β}X}}& (3)\end{array}$

This can be described as :

$\begin{array}{ll}{q_{it}=\text{β}X}_{it}+u_{it}& (4)\end{array}$

where q_it = an unobservable latent variable for households participating in the production of woodfuel (Equation 4).

X_it = vector of explanatory variables,

β= vector of the parameter to be estimated,

μ_{it =} error term.

The observed binary (1, 0) indicates whether a household decides to participate in woodfuel production or not, as assumed in the usual logit model (Greene, 2004).

Now Y_it/(1– Y_it) is simply the odds ratio in favor of participating in woodfuel production, that is, the ratio of the probability that a household will participate to the probability that it will not participate. If we take the natural log of (4), we obtain as follows (Equation 4).

$\begin{array}{ll}Li=ln\frac{Y_t}{1-Y_t}=qit={\beta }_1+{\beta }_{2X_{it}}{+u}_{it}& (5)\end{array}$

That is, L, the log of the odds ratio, is not only linear in X_it but also (from the estimation viewpoint) linear in the parameters (Equation 5).

3 Results and discussion

3.1 Results on demographic, socioeconomic, institutional, and shock characteristics

This section presents descriptive statistics and a discussion based on the dataset. Table 3 presents a summary of the sample household characteristics, resources owned, natural factors, and statistical test results regarding the relationship of these factors in the sample groups—participants involved in woodfuel production and non-participants. The results indicate that proportionately more participant households were headed by male heads compared to the non-participant households (Table 3). This suggests that among the total sample observations, only 24% were women while the remaining 76% were men. Out of the total surveyed households, only 75.2% of men and 24.8% of women reported that they engaged in woodfuel production, whereas approximately 77.9% of men and 22.1% of women reported being non-participants in woodfuel production. Regarding the impact of education status, there is a statistically significant higher proportion of illiterate households among the woodfuel production participants compared to non-participants. The chi-squared test result indicates that households' decisions in woodfuel production participation had a statistically significant association with educational status at a 1% level of significance.

Table 3

Table 3. Chi-squared test of dummy variables correlated to woodfuel production.

The independent mean comparison results for the continuous explanatory variables between the two sample groups are presented in Table 4. The results indicate that the mean age of non-participants was 41.77 years, while that of participants was 39.5 years. Although the mean age of non-participants slightly exceeded the mean age of participants, the difference was statistically significant at a 1% significance level. Similarly, the mean length of stay of households in the area was 31.9 years for non-participants and 7.6 years for participants. The difference between the two groups was statistically significant at a 1% significance level.

Table 4

Table 4. t-test comparison of continuous variables correlated to woodfuel production.

On average, the annual income was ETB 41,304.88 (±44,675.76) for non-participant households and ETB 51,317.26 (±56,004.45) for participant households. Similarly, the average production asset value of non-participant households was estimated to be ETB 1,423.40 (±5,671.39) while that of participating households was ETB 3,073.87 (±8,244.64). The mean expenditure shows that a participant household spent ETB 24,921.31 (±46,192.14), while non-participant households spent ETB 46,675.03 (±59,376.45). In addition, the land and livestock holdings of non-participant households were estimated at ETB 4.0334 (±5.59726) and ETB 22.2096 (±30.33310), while those of participant households were ETB 1.5378 (±1.66186) and ETB 6.2106 (±6.44613), respectively. The independent sample t-test for the mean difference in annual income, production asset value, expenditure, land holdings, and livestock holding between the two sample groups showed a statistically significant difference at a 1% level of significance.

The study further assessed the institutional factors associated with households' participation decisions. The χ² statistical test results indicate statistically significant correlations between participation decisions and access to market information, institutional membership, and access to extension services for woodland forests at 10%, 1%, and 1% levels of significance, respectively. In addition, the study found that households were more motivated to participate in woodfuel production during times of stress, such as droughts and outbreaks of livestock diseases. Households' participation in woodfuel production was confirmed to be strongly associated with their exposure to drought and livestock disease, with these associations being statistically significant at less than a 1% level of significance (Table 4).

3.2 Motivations of households' participation in woodfuel production

Figure 1 presents the main factors that motivated households to engage in woodfuel production in the study area. The survey results indicate that the majority (37%) of the sampled households were motivated to produce woodfuel because selling charcoal was their main livelihood activity, while 30% were motivated for obtaining firewood. The seasonal gap-filling function of woodfuel contributed to ~17.7% of households engaging in firewood production and 15.6% of those engaging in charcoal production during times of food shortage. Similarly, woodfuel production was considered an alternative employment option due to the lack of other alternatives, as reported by 14.2% of charcoal producers and 10.9% of firewood producers. Approximately 16.3% of the respondents revealed that they engaged in woodfuel production, especially firewood collection, to cover unexpected expenses resulting from shocks such as crop failure and livestock disease.

Figure 1

Figure 1. Reasons for households' engagement in woodfuel production.

Woodfuel is one of the vital forest products that provide a safety net, gap-filling, and income-smoothing functions to meet households' daily and seasonal requirements (Schure et al., 2014; Smith et al., 2017). Specifically, woodfuel production is an activity that rural households engage in to generate supplementary income during periods of low agricultural and livestock activities and seasonal food shortages, as well as to generate cash income for household expenditures. Safety-net activities refer to activities performed by rural households to earn additional income when experiencing shocks such as asset loss or illness (Brobbey et al., 2019). Scholars have also shown that the absence of alternative sustainable livelihood activities forces households to engage in woodfuel production (Alhassan et al., 2022).

FGD participants also noted that individuals primarily engage in woodfuel production owing to unexpected expenses related to shocks such as loss of family members or livestock deaths, crop failures, and loss of employment; in response to high urban demand for woodfuel, especially charcoal; and for seasonal needs such as purchasing food during periods of scarcity in dry months.

Woodfuel production was reported to be the primary source of income for all producers. Participants of the FGD prioritized woodfuel production as a viable livelihood source, alongside alternative strategies such as crop production and livestock rearing. During the FGD, it was revealed that income was the main motivation for people to engage in woodfuel production. Another finding indicated that trading natural products helps rural households generate subsistence, if not all, of their cash income (Angelsen et al., 2014). The data from this study illustrate that dryland woodfuel production is a significant source of rural livelihoods, supported by the findings of Schure et al. (2014). A major issue in the dryland area is the lack of income-generating opportunities for rural communities (Abebaw et al., 2012). The FGD participants also mentioned that the start-up capital required for other income-generating activities is substantial and unaffordable for the community. In contrast, the start-up capital required for woodfuel production is minimal. Therefore, according to the FGD, woodfuel production is a much-needed opportunity for those with access to resources and markets to generate income in the drylands of Ethiopia.

3.3 Occurrence of different shocks and severity levels in woodfuel production areas

The results of KIIs, FGD, and household surveys demonstrate the presence of different shocks in the study areas (see Figure 2). The analysis of frequency of shock occurrences indicated that drought was the most common shock, occurring more frequently than other shocks (see Figure 2). Additionally, reduction in agricultural output prices, crop diseases, and family illnesses were identified as the three most important shocks that occurred frequently after drought. Similarly, livestock diseases, increase in agricultural input prices, floods, lack of sustainable household businesses, unemployment, and conflicts were also identified as problems related to shock occurrences, based on their frequency levels in the study areas (Figure 2).

Figure 2

Figure 2. Types of shocks in woodfuel production areas.

The results of the FGD revealed that various shocks occurring in the dryland areas influenced the livelihoods of households. FGD participants from woodfuel producer households reported that drought is the major problem affecting the different livelihood components of woodfuel producers. Consequently, they were forced to engage in woodfuel production to survive and as an alternative livelihood option. The KIIs also stated that households mainly engage in woodfuel production during times of shock occurrence, which affects their common livelihood resources, mainly livestock and crops. This finding is supported by previous studies. For example, Vollmer et al. (2017) demonstrated that charcoal income serves as a means to alleviate rural poverty. Charcoal income is positively correlated with valuable household assets and increases resistance to chronic shocks (Roussy, 2013; Vollmer et al., 2017). Furthermore, literature shows that woodfuel is used as a coping strategy for shocks that interrupt households' income components by affecting various valuable assets such as livestock (Schure et al., 2014; Jones et al., 2016; Brobbey et al., 2019).

Figure 3 presents the results of the household survey indicating the severity of different shocks. Approximately 42% of the households reported that drought was highly severe compared to other shocks in the study areas. In addition, increase in food prices, crop diseases, and livestock diseases were significant factors that adversely affected woodfuel producers (Figure 3). Other severe factors mentioned included food and agricultural input price increases.

Figure 3

Figure 3. Severity levels of different shocks in the study areas.

Findings from FGDs and KIIs revealed that there has been an increase in drought in dryland areas of the country, causing severe damage to household livelihoods. Drought had high and severe effects on livelihood resources, particularly due to water and fodder shortages, which are the main sources of cash income. The effects of drought have forced households to shift toward woodfuel production for their survival. Woodfuel production, especially charcoal, serves as a means to enhance households' ability to withstand various shocks, such as poverty (Zulu and Richardson, 2013; Vollmer et al., 2017).

3.4 Determinants of household participation in woodfuel production

In this section, we focus on understanding some of the major factors affecting household participation in woodfuel production using a binary logit model. Table 5 presents binary logit estimates of the factors influencing household participation in woodfuel production. The last column of Table 5 displays the marginal effect of each independent variable on a household's probability of participating in woodfuel production, holding all other factors constant. The results from the binary logit model show that out of the 17 explanatory variables entered in the model, 11 were significant at 1% and 5% significance levels.

Table 5

Table 5. Binary logit model estimation on factors affecting households' participation in woodfuel production.

Age of the household head was a significant factor affecting the participation decision at a 1% significance level. The negative sign for the age of the household head suggests that households with relatively older heads were less likely to participate in woodfuel production. The marginal effect indicates that as the age of the household head increases by 1 year, the probability of household participating in woodfuel production decreases by 0.5%, holding all other factors constant. This result agrees with the previous studies (e.g., Mulenga et al., 2015; Jones et al., 2016; Brobbey et al., 2019). A negative and significant relationship was found between education and the likelihood of households participating in woodfuel production at a 5% significance level. The marginal effect result shows that as households attain higher levels of education, the probability of participating in woodfuel production decreases by 8.2% compared to those with no education. This is largely because education provides access to a wider range of income-generating activities and opportunities. Education expands the potential for labor and employment. By contrast, households with lower levels of education may be more economically vulnerable and, therefore, more likely to depend on income from woodfuel (Mulenga et al., 2014, 2015). The result of this study is inconsistent with other findings; for example, Brobbey et al. (2019) found that educated households were more engaged in charcoal production.

Many studies on factors influencing household participation in woodfuel production have identified a lack of wealth or household resources as a major driving factor (e.g., Mwitwa and Makano, 2012; Mulenga et al., 2015). In this study, four wealth indicators, namely, land holding, livestock holding, total annual household income, and the value of production assets owned, were included in the model estimation. The model output confirms the findings in the literature, indicating that households with greater wealth are less likely to engage in woodfuel production. The households' decision to participate in woodfuel production and landholding had a negative and statistically significant relationship at a 5% significance level. The model results indicate that as the land holding size increases by 1 ha, the probability of household participation in woodfuel production decreases by 1.3%. This result aligns with the findings of previous research. For example, Mulenga et al. (2017) and Tassie et al. (2021)demonstrated that households with large landholdings are less likely to participate in woodfuel production.

The model results also indicate that livestock holding and the probability of household participation in woodfuel production are associated negatively and significantly at a 1% significance level. This implies that as livestock holding increases by one TLU, the probability of household participation reduces by 0.9%, while holding other factors constant. Similarly, the value of production assets was found to have a negative and significant relationship with the probability of household participation in woodfuel production at a 5% significance level. The result indicates that an increase of 1 birr in asset value reduces the probability of household participation in wood fuel production by 2.6%. This result is consistent with the findings of Mulenga et al. (2015), which demonstrated that as the asset value of households improves, the probability of their engagement in woodfuel production decreases.

Moreover, the ownership of improved cook stoves (ICS) is negatively associated with household participation in woodfuel production at a 1% significance level. The marginal effect also indicates that as households gain access to ICS, the probability of their participation in woodfuel production decreases by 19.1%. The result is in agreement with the findings of Sola et al. (2019), which revealed that ICS are a demand-side option for forest management by reducing the consumption of biomass in both rural and urban communities.

Household expenditure also showed a positive and significant relationship with household participation in woodfuel production at a 1% significance level. An increase of 1 birr in household expenditure increases the probability of household participation in wood fuel production by 6.7%, holding other factors constant. This implies that households with higher expenditures are more engaged in woodfuel production.

The distance from the household's residence to the forest and household participation in woodfuel production were found to be negatively and significantly associated at a 1% significance level. The marginal effect result shows that as the distance to the forest increases by 1 km, the probability of household participation in woodfuel production decreases by 1.3%, holding other factors constant. This result is likely because when households are located far from the forest and in a place where there is no road infrastructure, they will be discouraged from traveling long distances. Hence, their participation in woodfuel production activities will be obviously reduced.

The study results further indicate that the occurrence of drought and household participation in woodfuel production are positively and significantly associated at a 1% level of significance. The model also demonstrate that when the frequency of drought occurrence increases by one, the probability of household participation in wood fuel production increases by 3.3%, holding other factors constant. This result is consistent with the findings of Brobbey et al. (2019), which indicated that households that consider income from charcoal production to be seasonally important are more likely to engage in charcoal production. Furthermore, drought directly influences the quantity of charcoal produced as it was observed that during drought years, a larger quantity of charcoal is produced to cater to reduced agricultural and other income (Kiruki et al., 2020). This has also increased the dependency on woodfuel for livelihood. Other scholars demonstrated that the effects of natural hazards such as drought on common livelihood sources have forced households to engage in woodfuel production for the composition of their livelihood (Alhassan et al., 2022).

The length of time that households lived in the area was also considered in the analysis. The model result shows a negative association of this variable with the probability of household participation in woodfuel production, which was statistically significant at a 1% significance level. The marginal effect indicates that for every additional year that a household lives in the area, the probability of participating in woodfuel production decreases by 1.1%, holding other variables constant. This suggests that households that had lived in the area for a longer period are less likely to rely on woodfuel production compared to those who had lived in the area for a shorter time.

Institutional membership was expected to be an important factor in reducing the degree of household participation and dependency on woodfuel production. The results show that as households become members of different institutions, whether formal or informal, the probability of participating in woodfuel production decreases by 33%, while other variables are held constant.

4 Conclusion

This study aimed to understand the factors that influence the decision of rural households to engage in woodfuel production. Understanding these factors is crucial for identifying the key point for successful interventions in the utilization and management of dryland forests. The results revealed that the majority of households were motivated to produce woodfuel because charcoal and firewood were their primary sources of income. These resources were used to fill seasonal gaps, provide alternative employment options, and cover unexpected expenses resulting from shocks such as crop failure and livestock disease. Households primarily engaged in woodfuel production during times of shock that affected their main livelihood resources, such as livestock and crops. The results also showed that drought was the most common shock, followed by decreased agricultural output prices, crop diseases, and family illness, livestock diseases, increased agricultural input prices, floods, lack of sustainable household businesses, unemployment, and conflicts. These factors forced households to engage in woodfuel production as a means of survival and an alternative source of income.

The results of the binary logit model indicated that age, education, land holding, livestock holding, production asset value, ICS ownership, number of years lived in the area, distance from the forest, access to forest extension, and institutional membership all had a negative and statistically significant correlation with the participation of households in woodfuel production. These findings suggest that households' decision-making regarding woodfuel production is influenced by these variables. On the other hand, household expenditure and drought occurrence had a positive and statistically significant correlation with household participation in woodfuel production.

The study's results highlight that the importance of considering different socioeconomic factors, livelihood resources, and options, as well as addressing the occurrence of various shocks. Furthermore, improving the productivity of livestock, crops, and other alternative income sources is important for ensuring sustainable management of dryland forests and resource utilization. In addition, shifting from traditional energy consumption to renewable energy use is of paramount significance in reducing pressure on dryland forests. Overall, designing alternative energy consumption and livelihood options is crucial for creating a sustainable environment rather than relying on woodfuel.

Data availability statement

The original contributions presented in the study are included in the article, further inquiries can be directed to the corresponding author.

Ethics statement

Written informed consent was obtained from the individual(s) for the publication of any potentially identifiable images or data included in this article.

Author contributions

GG wrote the proposal, performed data collection, analyzed the data, and wrote, reviewed, and edited the manuscript. AS designed and wrote the proposal, designed the data collection tools, and performed data collection, reviewed, and edited the manuscript. TA wrote the proposal, performed data collection, and data entry. DB, AA, and AB performed data collection and data entry. All authors contributed to the article and approved the submitted version.

Acknowledgments

The research team acknowledges the support of the Ethiopian Forestry Development (EFD) for data collection. The team is also grateful to all participants involved in the household survey, key informant interviews, and focus group discussions as well as experts at the district and kebele levels for mobilizing the local communities, giving us their valuable time, and sharing useful information to the study.

Conflict of interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Publisher's note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

References

Abebaw, D., Kassa, H., Kassie, G. T., Lemenih, M., Campbell, B., and Teka, W. (2012). Dry forest-based livelihoods in resettlement areas of Northwestern Ethiopia. For. Policy Econ. 20, 72–77. doi: 10.1016/j.forpol.2012.02.002

Crossref Full Text | Google Scholar

Abera, A., Yirgu, T., and Uncha, A. (2021). Determinants of rural livelihood diversification strategies among Chewaka resettlers' communities of southwestern Ethiopia. Agricult. Food Secur. 10, 1–19. doi: 10.1186/s40066-021-00305-w

Crossref Full Text | Google Scholar

Alhassan, J., Ofosu, A., Iddrisu, S., and Kofi Garsonu, E. (2022). Wood fuel producers' insight on the environmental effects of their activities in Ghana. J. Sustain. Forest. 2020, 1–17. doi: 10.1080/10549811.2022.2053162

Crossref Full Text | Google Scholar

Angelsen, A., Jagger, P., Babigumira, R., Belcher, B., Hogarth, N. J., Bauch, S., et al. (2014). Environmental income and rural livelihoods: a global-comparative analysis. World Dev. 64, S12–S28. doi: 10.1016/j.worlddev.2014.03.006

PubMed Abstract | Crossref Full Text | Google Scholar

Bekele, B., and Kemal, A. W. (2022). Determents of sustainable charcoal production in AWI zone; the case of Fagita Lekoma district, Ethiopia. Heliyon 8:e11963. doi: 10.1016/j.heliyon.2022.e11963

PubMed Abstract | Crossref Full Text | Google Scholar

Brobbey, L. K., Hansen, C. P., Kyereh, B., and Pouliot, M. (2019). The economic importance of charcoal to rural livelihoods: Evidence from a key charcoal-producing area in Ghana. Forest Policy Econ. 101, 19–31. doi: 10.1016/j.forpol.2019.01.013

Crossref Full Text | Google Scholar

Chidumayo, E. N., and Gumbo, D. J. (2013). The environmental impacts of charcoal production in tropical ecosystems of the world: a synthesis. Ener. Sustain. Dev. 17, 86–94. doi: 10.1016/j.esd.2012.07.004

Crossref Full Text | Google Scholar

FAO (2017). The Charcoal Transition: Greening the Charcoal Value Chain to Mitigate Climate Chain and Improve Local Livelihoods. Rome: Food and Agriculture Organization of the United Nations.

Google Scholar

FDRE (2011). Ethiopia's Climate-Resilient Green Economy: Green Economy Strategy. Addis Ababa: Federal Democratic Republic of Ethiopia (FDRE).

Google Scholar

Girma, G., Shimeles, A., Abate, T., Seyoum, G., and Alemu, M. (2022). The urge for just transition: Evidence from understanding of wood fuel producers' livelihoods and vulnerability in the drylands of Ethiopia. Front. Sustain. Food Syst. 6:966137. doi: 10.3389/fsufs.2022.966137

Crossref Full Text | Google Scholar

Greene, W. (2003). Econometric Analysis. 5th Edn. Upper Saddle River, NJ: Prentice Hall.

Google Scholar

Greene, W. (2004). Fixed effects and bias due to the incidental parameters problem in the Tobit model. Economet. Rev. 23, 125–147. doi: 10.1081/ETC-120039606

Crossref Full Text | Google Scholar

Jones, D., Ryan, C. M., and Fisher, J. (2016). Charcoal as a diversification strategy: the flexible role of charcoal production in the livelihoods of smallholders in central Mozambique. Ener. Sustain. Dev. 32, 14–21. doi: 10.1016/j.esd.2016.02.009

Crossref Full Text | Google Scholar

Kiruki, H. M., van der Zanden, E. H., Kariuki, P., and Verburg, P. H. (2020). The contribution of charcoal production to rural livelihoods in a semi-arid area in Kenya. Environm. Dev. Sustainab. 22, 6931–6960. doi: 10.1007/s10668-019-00521-2

Crossref Full Text | Google Scholar

MEFCC (2017). Ethiopia's Forest Reference Level Submission to the UNFCCC (Revised). Addis Ababa: Ministry of Environment, Forestry, and Climate Change.

Google Scholar

Mondal, M. A. H., Bryan, E., Ringler, C., Mekonnen, D., and Rosegrant, M. (2018). Ethiopian energy status and demand scenarios: prospects to improve energy efficiency and mitigate GHG emissions. Energy 149, 161–172. doi: 10.1016/j.energy.2018.02.067

Crossref Full Text | Google Scholar

Mulenga, B., Nkonde, C., and Ngoma, H. (2015). Does Customary Land Tenure System Encourage Local Forestry Management in Zambia? A Focus on Woodfuel. Lusaka: Indaba Agricultural Policy Research Institute (IAPRI). doi: 10.22004/ag.econ.207021

Crossref Full Text | Google Scholar

Mulenga, B. P., Hadunka, P., and Richardson, R. B. (2017). Rural households' participation in charcoal production in Zambia: does agricultural productivity play a role? J. Forest Econ. 26, 56–62. doi: 10.1016/j.jfe.2017.01.001

Crossref Full Text | Google Scholar

Mulenga, B. P., Richardson, R. B., Tembo, G., and Mapemba, L. (2014). Rural household participation in markets for non-timber forest products in Zambia. Environm. Dev. Econ. 19, 487–504. doi: 10.1017/S1355770X13000569

Crossref Full Text | Google Scholar

Mwitwa, J., and Makano, A. (2012). Preliminary Charcoal Production Supply and Demand Assessment in Eastern and Lusaka Provinces. Lusaka, Zambia: United States Agency for International Development.

Google Scholar

Ndegwa, G., Anhuf, D., Nehren, U., Ghilardi, A., and Iiyama, M. (2016). Charcoal contribution to wealth accumulation at different scales of production among the rural population of Mutomo District in Kenya. Ener. Sustain. Dev.33, 167–175. doi: 10.1016/j.esd.2016.05.002

Crossref Full Text | Google Scholar

Njenga, M., Sears, R. R., and Mendum, R. (2023). Sustainable woodfuel systems: a theory of change for sub-Saharan Africa. Environm. Res. Commun. 5:051003. doi: 10.1088/2515-7620/acd0f3

Crossref Full Text | Google Scholar

Obiri, B. D., Nunoo, I., Obeng, E., Owusu, F. W., and Marfo, E. (2014). The Charcoal Industry in Ghana: An Alternative Livelihood Option for Displaced Illegal Chainsaw Lumber Producers. Wageningen: Tropenbos International.

Google Scholar

Rafaj, P., Kiesewetter, G., Gül, T., Schöpp, W., Cofala, J., Klimont, Z., et al. (2018). Outlook for clean air in the context of sustainable development goals. Global Environm. Change 53, 1–11. doi: 10.1016/j.gloenvcha.2018.08.008

Crossref Full Text | Google Scholar

Roussy, S. (2013). Enhancing Resilience to Shocks and Stresses ACF-International Briefing Paper.

Google Scholar

Schure, J., Levang, P., and Wiersum, K. F. (2014). Producing woodfuel for urban centers in the Democratic Republic of Congo: a path out of poverty for rural households? World Dev. 64, S80–S90. doi: 10.1016/j.worlddev.2014.03.013

Crossref Full Text | Google Scholar

Singh, D., Aung, T., and Zerrif, H. (2018). Energy for sustainable development resource collection polygons: a spatial analysis of woodfuel collection patterns? Energy Sustain. Dev. 45, 150–158. doi: 10.1016/j.esd.2018.06.003

Crossref Full Text | Google Scholar

Smith, H. E., Hudson, M. D., and Schreckenberg, K. (2017). Livelihood diversification: The role of charcoal production in southern Malawi. Energy Sustain. Dev. 36, 22–36. doi: 10.1016/j.esd.2016.10.001

Crossref Full Text | Google Scholar

Sola, P., Cerutti, P. O., Gautier, D., Iiyama, M., Schure, J., Chenevoy, A., et al. (2017). The environmental, socioeconomic, and health impacts of woodfuel value chains in Sub-Saharan Africa: a systematic map. Environm. Evid. 6, 1–16. doi: 10.1186/s13750-017-0082-2

Crossref Full Text | Google Scholar

Sola, P., Schure, J., Eba'a Atyi, R., Gumbo, D., Okeyo, I., and Awono, A. (2019). Woodfuel policies and practices in selected countries in Sub-Saharan Africa - a critical review. Bois et Forêts des Tropiques, 340: 27–41. doi: 10.19182/bft2019.340.a31690

Crossref Full Text | Google Scholar

Tassie, K., Misganaw, B., Addisu, S., and Tesfaye, E. (2021). Socioeconomic and environmental impacts of charcoal production activities of rural households in Mecha District, Ethiopia. Adv. Agricult. doi: 10.1155/2021/6612720

Crossref Full Text | Google Scholar

Tucho, G. T., and Nonhebel, S. (2017). Alternative energy supply system to a rural village in Ethiopia. Ener. Sustainab. Soc. 7, 1–14. doi: 10.1186/s13705-017-0136-x

Crossref Full Text | Google Scholar

UNEP (2019). Review of Woodfuel Biomass Production and Utilization in Africa A Desk Study. Nairobi.

Google Scholar

Vollmer, F., Zorrilla-Miras, P., Baumert, S., Luz, A. C., Woollen, E., Grundy, I., et al. (2017). Charcoal income as a means to a valuable end: Scope and limitations of income from rural charcoal production to alleviate acute multidimensional poverty in Mabalane district, southern Mozambique. World Dev. Persp. 7, 43–60. doi: 10.1016/j.wdp.2017.11.005

Crossref Full Text | Google Scholar

Yalew, A. W. (2022). Environmental and economic accounting for biomass energy in Ethiopia. Ener. Sustain. Soc. 12, 1–12. doi: 10.1186/s13705-022-00356-2

PubMed Abstract | Crossref Full Text | Google Scholar

Zulu, L. C., and Richardson, R. B. (2013). Charcoal, livelihoods, and poverty reduction: Evidence from sub-Saharan Africa. Ener. Sustain. Dev.17, 127–137. doi: 10.1016/j.esd.2012.07.007

Crossref Full Text | Google Scholar