SEARCH

SEARCH BY CITATION

Keywords:

  • Biogas energy;
  • Local Governance;
  • Ordinary Least Squares (OLS);
  • Households;
  • Behavioral model

Abstract

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Objectives of the Study
  5. 3. Materials and Methods
  6. 4. Results and Discussion
  7. 5. Conclusions and Policy Implications
  8. Acknowledgments
  9. References

In the context of climate change mitigation and poverty reduction, it has been argued that biogas energy is relevant, as it is economically and ecologically useful. In the 1980s, biogas use played an important role in the development of Burundi. Many schools and public institutions had implemented such installations. Unfortunately, many biogas infrastructures were destroyed in the civil war of the 1990s. This study analyzes what could be done, after a decade of crisis, to develop that sector. It aims to assess how and to what extent the inhabitants of villages are willing to contribute to the development of biogas technologies. We interviewed 150 farmers in order to assess their perception on the ecologic and economic features of biogas plants if implemented in their villages. The influence of socioeconomic, cultural, and demographic factors of households was assessed in this study. Results suggest that the maximum amount that a household is willing to pay each month for biogas use at a family level is positive for large-size households, households that are aware of climate change, consumers of candles, households with high income, households with an educated head, women, and breeders. However, the willingness decreases for households with older head of families. The study concludes that awareness campaigns on biogas benefits and financial and nonfinancial incentives are necessary. This policy should probably and primarily be oriented toward some more receptive categories of the population. Women should be fully involved, considering their positive motivation toward sustaining this sector.

1. Introduction

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Objectives of the Study
  5. 3. Materials and Methods
  6. 4. Results and Discussion
  7. 5. Conclusions and Policy Implications
  8. Acknowledgments
  9. References

Burundi is one of the countries with great lakes. More than 80% of the population lives in rural areas, with the agricultural sector as their main source of income. Since 2007, Burundi has become a member of the East African Community (EAC). Within the region, community members agreed to design and implement common policies related to the open market, security, and energy.

Burundi has several potential sources of energy, including hydro, biomass, and solar. Apart from peat energy that is exploited, fossil fuels (petroleum) are imported from abroad. Wind and geothermal energies are at an experimental stage. Despite the potential use of several energy sources, traditional biomass energy dominates the Burundian energy balance. Most of the biomass energy types such as firewood are not commercialized in a formal manner. The estimated annual amount of firewood consumed is 1,216,000 tons oil equivalent (TOE). Firewood ranks first within fuel wood. The latter is used for cooking and lighting in rural areas. Agricultural wastes come next, with 315,000 TOE followed by hydroelectricity that amounted to 45,580 TOE in 2008 (the values are shown in Table 1).

Table 1. Share of Energy Sources in the Burundian Energy Market (in Tonnes Oil Equivalent)
Energy SourcesQuantity AvailableShareRanking
Firewood1,216,00070%  1st
Agricultural residues  315,00018.13%2nd
Hydroelectricity   45,5802.6%4th
Charcoal  100,000 5.75%3rd
Bagasse   16,800 0.97%6th
Petroleum products   42,0002.4%5th
Peat    2,276 0.13%7th
Solar and biogas      174 0.01%8th
Total1,737,830100% 

These figures suggest that traditional biomass forms a large part of the national energy consumption. Seventy percent of the biomass sector consists of firewood, 5.82% of charcoal, 18.35% of agricultural residues, 0.97% of bagasse, and 0.01% of biogas and solar energy. Fossil energy amounts to 0.04% for peat and to 2.5% for petroleum products, whereas the hydroelectricity subsector represents 2.6%. These figures clearly indicate that fuel wood dominates the energy balance in Burundi. Firewood is used for cooking and/or for lighting during the night even though this form of energy consumption has negative impacts on small children and pregnant women, the vulnerable population [GTZ, 2007; UNCSD, 2011].

Besides health risks due to the indoor pollution, there is a risk of deforestation because of the overconsumption further reinforced by the population growth [IFDC, 2011].

A missing substitute for fuel wood is a big challenge in Burundi, especially as the population is growing.

The EAC and the IMF [2011a, p. 17] reported a natural population growth rate of 2.4% in Burundi. This population trend will have a serious impact on forest resources if no preventive policies are undertaken. According to the African Development Bank [2009, p. 140], the estimated consumption of fuel wood in rural areas is 3 kg per person per day.

Although traditional biomass energy remains an important energy source in Burundi, the demand for energy services is increasing. People aim to improve their welfare and well-being, which is often linked to higher energy consumption. The introduction of hydroelectricity has contributed to the welfare of people in the capital city and in urban areas. In rural areas where the majority of the population live, people do not have access to modern energy sources such as hydroelectricity.

Burundi is engaged in different international agreements, such as the Kyoto protocol (even though the country did not take up any commitments to GHG emission reductions), the UNFCC, and the realization of millennium development goals (MDGs). Encouraging renewable energies plays an essential role in all of these initiatives [MPDC and PNUD, 2011, pp. 61–62].

Against this background, we analyzed how and to what extent bioenergy technologies could be adopted in the region of Gihanga.

The literature shows that the adoption of new technologies in rural communities always faces challenges, especially when local participation is required.

Main barriers include economic, financial, sociocultural, demographic, and institutional factors [see Fenhann and Painuly, 2002, p. 15].

Mostly farmers are aware that climate change is a serious problem for humanity, as their crop yields are increasingly threatened by climate change. Although the rural population in developing countries does not emit much of GHG, their contribution to climate change mitigation is still necessary. Researchers agree that mitigation policies will enable them to contribute to the well-being of farmers without constraining future generations.

Key questions are about whether biogas technologies will be adopted by farmers in villages, whether sufficient feeding materials are available, and to what extent socioeconomic, demographic, and cultural factors affect the motivation of rural households to participate financially in biogas use. Evidence suggests that socioeconomic and demographic factors have a significant impact upon the households' willingness to participate in climate change mitigation through renewable energies [see Abdullah, 2009; Solomon and Johnson, 2009; Walekhwa et al., 2009].

In Uganda, for instance, Walekhwa et al. [2009] conclude that socioeconomic factors such as household income, fuel wood and kerosene cost, land ownership, livestock practice, and land size have a significant effect on the adoption of biogas technologies. Similar evidence is provided by Fenhann and Painuly [2002, p. 15].

According to these sources, there are technical and other conditions that are required for sustainable biogas development. Main factors include the availability of raw materials, economic and financial factors, the awareness and knowledge level of biogas advantages, technical factors, sociocultural factors, political will, and institutional aspects. These factors can be modeled as main barriers to the financial participation of households to invest in the biogas sector.

2. Objectives of the Study

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Objectives of the Study
  5. 3. Materials and Methods
  6. 4. Results and Discussion
  7. 5. Conclusions and Policy Implications
  8. Acknowledgments
  9. References

This study aims to contribute to climate change mitigation policy by pointing out factors that could impact on the feasibility of a biogas energy policy. Scholars have highlighted that biogas energy is relevant for both climate change mitigation and poverty alleviation, especially in rural areas where agriculture is the main source of income [Pacini et al., 2010].

Despite multiple benefits of the biogas sector, there is little information on how biogas technologies could be developed, especially in developing countries. There are many variables and structural problems that need to be tackled in these countries, mainly in the rural areas.

The adoption of biogas technologies and the engagement of rural farmers in biogas sector depend mainly on technical and socioeconomic factors. This study focuses on the district of Gihanga. This region has several advantages for biogas production compared to other regions in the country. These advantages relate to the availability of livestock manures and agricultural wastes, especially from rice production. In addition, the rural population of Gihanga relies heavily on firewood for cooking purposes from the National Park of Rusizi. Our results might be considered as a reference for other regions with a similar socioeconomic context. This paper aims to assess how and to what extent socioeconomic, cultural and demographic factors may foster the biogas sector in the district of Gihanga in Burundi.

3. Materials and Methods

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Objectives of the Study
  5. 3. Materials and Methods
  6. 4. Results and Discussion
  7. 5. Conclusions and Policy Implications
  8. Acknowledgments
  9. References

3.1. Data Collection

This study is based on a survey that was conducted in June, July, and August 2012. Before the formal survey, a baseline survey was carried out. This allowed the authors to schedule the formal survey, as key informants were identified. Before each interview and during the group discussion, study participants were informed and/or updated about the advantages of biogas usage at family and/or community level, and the requirements for its development. In the district of Gihanga, biogas technologies have not yet been developed and therefore some clarifications were necessary in order to minimize all bias-related reservations. The study participants were mainly farmers who were selected randomly from the six villages of the district. The gender aspect was taken into account insofar as the study included women that are household heads or women who could represent their husbands who were not available during the survey. The methodology applied in this study includes both qualitative and quantitative approaches. A quantitative approach was applied in collecting data from randomly selected households among the farmers of each village of the district. Primary data were then collected from 150 households that were categorized as farmers with livestock or without. The qualitative approach included three methods: focus group, observation, and discussion with local authorities and stakeholders. A focus group allows study participants to express their opinion on existing problems, and try to find solutions with the facilitation of a researcher. This method was chosen as it is useful to elaborate on appropriate solutions to real problems in the community [Abdullah, 2009, p. 23].

3.2. Empirical Model

The ordinary least squared (OLS) regression is used in the model to investigate the willingness to pay (WTP) for climate change mitigation through the development of family-sized biogas digesters. We adopted the OLS model in order to estimate the parameters (βi) of regression. These parameters indicate the influence level of variables on the willingness to invest in the biogas sector in villages.

The OLS is not the only method to estimate (βi). However, we chose this method because of the following advantages [Hewson and Whalley, 2009, p. 13]:

  1. If errors are independent and identically (and normally) distributed, the least squares estimate is also a maximum likelihood estimate.
  2. The Gauss-Markov theorem states that inline image estimated through OLS are the best linear unbiased estimators (BLUE).
  3. The OLS has been applied, as the WTP is the maximum value that households are willing to pay each month over the year [see Meeks, 2012, p. 69].

In this study, the monthly participation or the WTP variable is a dependent variable, and other variables are considered as influential.

This monthly participation is represented by yi and it is the maximum amount that a household is willing to pay each month.

The assumption has been formalized in the following way:

  • display math(1)

xi are variables that may affect the decision of financial participation.

The modeling of the assumption above is performed by the following expression:

  • display math

This expression can be written as

  • display math(2)

yi = the monthly payment for the biogas sustainability

  • display math

In the OLS model, βi explains the variation in the monthly participation caused by the variation of one unit of Xi (explanatory variables).

  • display math

If βi > 0, the variation of Xi thus has a positive effect on the monthly participation. In this case, the WTP is positive for this variable. The variable therefore presents an opportunity to invest in the biogas sector in the village.

If βi < 0, the variation of Xi has a negative influence on the WTP.

The technology will be adopted only if a specific policy is developed in order to trigger behavior change and not by focusing on this factor alone.

  • display math

In order to estimate inline image or β we minimize the square of the error term that is the residual sum of squares (RSS). The main problem is now to minimize the RSS.

It is worth noting that the normal equation is:

  • display math(3)

By solving the mathematical equation inline image we see that the residual sum is given by (1) − (3)

  • display math(4)

This equation assumes that the residuals are uncorrelated and have equal variance.

It means that var(ε) = σ2. Under this condition, a least squares regression will perform the optimal properties [see Hewson and Whalley, 2009].

The logic underlying the OLS approach is to minimize the residual sum of squares that is explained by inline image

By minimizing this expression, one is able to find the estimators β0 and βi that are BLUE. The expression is presented in the following:

  • display math(5)

In this expression, β is estimated by solving the mathematical equation (5); it is suggested to check whether the parameters are equal to zero. The validity of the model has to be checked by applying a statistical test. At this step, a null hypothesis is proposed where we assume that β is equal to 0 and an alternative hypothesis where β is different to 0. If the null hypothesis is true, one is dealing with a parameter omitted from the model.

  • display math

By applying a statistical test, we are able to reject or accept the hypothesis at a given significance level. The validity test of the regression is decided by looking at the t-statistic, its corresponding significance level and coefficients of multiple determinations (R2). Based on other accepted methods such as contingent valuation related to the WTP, a significance level of 10% has been considered in this study [see Solomon and Johnson, 2009 and Meeks, 2012].

Furthermore, a regression with an R2 that is around 10% can be considered as statistically valid [see Meeks, 2012].

4. Results and Discussion

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Objectives of the Study
  5. 3. Materials and Methods
  6. 4. Results and Discussion
  7. 5. Conclusions and Policy Implications
  8. Acknowledgments
  9. References

We analyzed how and to what extent the socioeconomic and cultural variables could impact on the willingness to invest in family-sized biogas digesters in the district of Gihanga. We assume that, all other things being equal, variables such as household income, gender of household head, household size, age of household head, belief in climate change, household head education, firewood expenditure, candle expenditure, and livestock practice have a significant impact on the WTP in the biogas sector.

4.1. Model Specification

In the previous section, we explained that the WTP is considered as the maximum amount that each household is willing to pay monthly. On the basis of the monthly candle expenditures, firewood expenditures and other energy sources that are regularly consumed in the household, we estimated the maximum amount, as we realized that people have some reservations about the exact amount due to cultural aspects. During the survey, the participants were asked to name the maximum amount of money that they would be willing to invest in a biogas digester.

In order to test for correlations between WTP and explanatory variables described above we followed the Meeks [2012]'s model. From an economic point of view, each economic agent is willing to invest depending on the expected profit that can be earned [Andersen, 2008].

In this study, expected interests are the access to energy and the contribution to climate change mitigation, as we assumed that people were aware that biogas use will reduce the overconsumption of forest resources. Biogas technologies will thus contribute to mitigate anthropogenic climate change.

WTP as a dependent variable is a function of the variables:

Household income (I), gender of household head (S), household size (Z), livestock practice (L), candle expenditures (C), firewood expenditures (F), household head age (A), the belief in climate change (B), and household head education (E).

The equation of the WTP in the biogas sector can be written as follows:

  • display math

Equation (1) implies that the WTP is a function of the change of explanatory variables that are involved in the model. As explained previously, the WTP is expressed by the maximum values that a household is willing to pay, and the OLS method of regression was used in the study.

  • display math

εi is the error, and w0 the constant in the model.

The above equation means that the change in each explanatory variable will affect the change in the WTP variable. In the following, we check for the statistical significance of the estimator (β).

4.2. Explanatory Variables of the WTP in the Biogas Sector

4.2.1. Description of Study Variables

This study aims to highlight the main factors that are expected to be more influential on the investment in biogas digesters in the Gihanga district. The explanatory variables are gender of household head, household income, education of household head, belief in climate change, family size, age of household head, livestock practice, and household expenditures for other sources of energy. The dependent variable is the maximum monthly amount that is to be spent. In Table 2, we distinguish categorical, continuous, and binary variables. For precise valuation, a categorical variable is a factor that contains information about membership in one or several possible options. People can belong to one group or many defined groups whereas for a continuous variable there is no restriction to particular values; individuals can belong to any possible group [Pasta, 2009].

Table 2. Types of Model Variables
VariablesExpected SignVariablesExpected Sign
  1. Source: Authors.

Sex of household head (F)PositiveLivestock practicePositive
Household incomePositiveBelief in climate changePositive
Family sizePositiveAge of household headPositive
Candle expendituresPositiveCharcoal expendituresPositive
Education of household headPositive  

The decision of investing in the biogas sector can be influenced by different variables. Humans are influenced both by visible and invisible factors. This study assumes that visible factors are household income, sex, age of household head, belief in climate change, and household head education. Invisible factors in this study are those individual characters that relate to genetics; psychology and/or to knowledge or experiences that have been accumulated since the childhood. The latter are also considered as endogenous factors that influence human behavior in decision making for a given purpose [see Johnson et al., 2013].

4.2.2. Expected Sign of Explanatory Variables

The analysis focuses on the willingness of households to invest in the biogas sector as a means to climate change mitigation. The explanatory variables of this investment decision could be both considered as barriers or opportunities for the development of the sector, depending on the sign and the statistical significance of related coefficients. In this case, we assume that all variables are positive and statistically significant.

For positive sign, a change in each explanatory variable will affect a positive variation of the dependent variable. It means that the variation of both the explanatory variable and the WTP for the biogas use will be in the same direction, and for the negative signs it is inverse. Table 3 details the variables and their expected sign in the model. In the discussion below we explain the findings and the feature of each variable that is involved in the model.

Table 3. Expected Sign of Explanatory Variables
VariablesTypeDescription
  1. Source: Authors.

Monthly paymentCategoricalA maximum amount that a household is willing to pay per month
Sex of household headBinaryGender of household head; a proxy variable for gender relations (1 = male; 0 = female)
Household incomeCategoricalEach household has a minimum revenue, especially from agricultural activities
Family sizeCategoricalNumber of persons living in the household
Candle expendituresContinuousExpenditures for candles that are used for lighting purposes per month
Charcoal expendituresContinuousExpenditures for charcoal that are used for cooking purposes per month
Livestock practiceBinaryThe practice of livestock; a proxy variable (1 = yes; 0 = no)
Belief in climate changeBinaryThe belief in climate change; a proxy variable (1 = yes; 0 = no)
Education of household headCategoricalNumber of school years of household head
Age of household headContinuousThe age of the household head

4.3. Econometric Issues

Before running the OLS regression, we controlled for multicollinearity issues in order to distinguish mutual influences between variables and their specific influences on the dependent variable. Scholars suggest that the variance inflation factors (VIF) can be applied in order to check for the multicollinearity problem. The rule of thumb is that if VIF for each variable in the model (VIFi) is ≥ 10, there is a problem with multicollinearity, and therefore adjustment methods need to be applied. We state that VIFi = (1/(1 − Ri2)), where R2. Based on the rule of thumb, there is no multicollinearity problem. As it is presented in Table 4, we found that for all model variables, the VIF is < 10.

Table 4. Results of VIF for the Model Variables
VariableVIF1/VIF
  1. Source: Authors upon basis of the survey of 2012.

Family size1.160.862961
Charcoal expenditures1.150.868970
Age of household head1.080.926503
Education of household head1.080.928269
Livestock practice1.070.937123
Candle expenditures1.060.947447
Belief in climate change1.050.950185
Household income1.050.954357
Mean VIF1.09 

This means that the correlation between the explanatory variables was not strong. This led the authors to conclude that the independent and error terms of the model are uncorrelated. It was therefore possible to analyze the specific impact of the explanatory variables. But it was also necessary before the regression to check whether the model variables are normally distributed, since there can be outliers that can bias the model estimators. For variables that are not normally distributed, we applied the logarithmic function, especially for variables that are in the form of maximum (point) values. We checked the normal distribution for household income, monthly payment, household age, candles expenditures, years of studies, and firewood expenditures.

According to the World Bank, donors, and other international agencies, the level of poverty, in terms of monetary measures, is below 1.25 USD PPP (per person per day) [Loewe and Rippin, 2012]. Considering this poverty line, it becomes apparent that the population of Gihanga is suffering extreme poverty, as 2 USD is the daily revenue of the whole family.

As it is more detailed in Table 5, the findings indicate that 64 of 139 interviewed households are ready to pay less than 5 USD, and more than 50% of households surveyed are ready to pay at least 6 USD; 27 of 139 of interviewed households can achieve 6 USD, 28 are ready to pay more than 8 USD, 18 can pay at least 12 USD, and 2 are ready to pay 15 USD per month. This figure shows clearly that people are willing to invest in the bioenergy sector depending on their monthly income and/or their habitual energy consumption.

Table 5. Distribution of Household Income and the Amount That a Household Is Willing to Pay Monthly in USD
What Is Your Monthly Income From Agricultural Activities? (in USD)How Much Are You Willing to Pay Monthly for Biogas Development (in USD)?
24.368.311.715Total Number of HouseholdsNumber of Households
  1. Source: Our Survey of 2012.

015.8%9%  3.7%0 5.5%0 9 6.5%
6.736.8%26.7%33.3%14.3%11%03424.5%
66.731.6%37.8%29.6%39.3%50%100%5338% 
133.310.5 15.5%22.2%14.3%17%02215.8%
2005.32%  7.4%17.8%00 9 6.5%
266.702%  3.7% 3.6% 5.5%0 4 2.9%
316.706.8%010.7%11%0 8 5.8%
Total19452728182139 100%

This argument is econometrically tested in the following subchapter.

Before performing the OLS regression, we checked for the normal distribution, firewood expenditure, household head age, family size, and education, by applying the log function.

For the distributions where outliers have been suspected, it was necessary to solve this problem in order to improve the quality of the model estimators. We controlled for these issues and then run the OLS (Figure 1).

image

Figure 1. Test for normal distribution of the error term for the OLS model. Source: Our computation using STATA.12 software upon basis of the results in Table 6.

Download figure to PowerPoint

4.4. Benchmark Results

The main aim of this chapter is to present the econometric results of the OLS regression. It allows insights into the questions how and to what extents the explanatory variables involved in the model could impact on the WTP for investments in the biogas sector. The rationale is that both private and public actors in the biogas sector are willing to inject their capital in an environment where socioeconomic and technical aspects could foster the uptake of biogas technologies. As one can observe in Table 6, the findings show that the maximum amount that a household is willing to pay (represented by the “WTP” variable) is affected by all model variables and statistically significant (F = 0.00).

Table 6. Ordinary Least Squared Modelinga
Explanatory VariablesWTP = Monthly ParticipationStandard Errors in Parentheses
***p < 0.01, **p < 0.05, *p < 0.1
  1. a

    In the above table, the influential variables are ranked on the basis of their statistical power in terms of p-value to influence the WTP a maximum amount for biogas development. In fact, we considered that a significance level of 10% is not powerful statistically than 5% and 1%. The last variable (Firewood expenditures) is not significant statistically, and therefore it is difficult to provide our opinion on their influence level.

  2. Source: Authors computation using STATA.12 software.

Household head age−0.299** (0.127) 
Household income0.000543** (0.000239)
Household head gender(W)0.200** (0.0961)
Belief in climate change0.465** (0.225)
Livestock practice0.233* (0.118)
Education0.0225* (0.0119)
Candle expenditures1.50e−05* (7.97e−06)
Family size0.0344* (0.0201)R2 = 0.2539
Firewood expenditures0.00110 (0.00147)Prob > F = 0.0000
Observations: 150

We assessed the specific impact of each explanatory variable on the WTP, and conclude that the change of one unit in the belief on climate change implies a positive variation of 0.465 in the WTP for biogas use at family level. The change of one unit of household income causes a positive variation of 0.000543 in WTP for family-sized biogas digester.

We also found a significant influence of the livestock practice on the willingness to invest. The average effect (AE) of the livestock practice on the WTP is 23%, and statistically significant at a 5% level. We also found that the gender issue is relevant in the WTP investment decision in the biogas sector.

The AE of the household head female sex is positive (20%), and statistically significant at a 5% level. Regarding the family size, we found that the change of one unit in family size provokes a positive variation of 3.4% on the willingness to invest in the family-sized biogas sector. For the household head age, a change of one unit in age causes a negative variation of 29% in the WTP. This is statistically significant at a 5% level. The variation of one unit in candle expenditures causes a positive change of 0.0015% in the WTP for a family-sized biogas plant. For the household head education, the findings suggest that the variation of one unit in education level provokes a positive change of 2% on the WTP for biogas use. The estimator for the firewood expenditures is not statistically significant based on considered significance levels in this study. We also checked whether the error terms of the OLS regression are normally distributed in order to confirm their independence in the model, and found that they are independent (see the representative graphic). These findings allow us to confirm the validity of the model. We are thus able to reject the null hypothesis that states the insignificant influence of household income, belief in climate change, household head sex, education, household head age, candle expenditures, and livestock practice. We conclude that the WTP in the biogas sector is significantly influenced by household income, belief in climate change, household head sex, household head age, family size, household head education, candle expenditures, and livestock practice.

5. Conclusions and Policy Implications

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Objectives of the Study
  5. 3. Materials and Methods
  6. 4. Results and Discussion
  7. 5. Conclusions and Policy Implications
  8. Acknowledgments
  9. References

The study aims to evaluate main barriers and opportunities for investing in family-sized biogas digesters in the district of Gihanga. We evaluated the socio economic and technical factors that could significantly impact on the willingness to invest in the biogas sector. In order to facilitate investors' decisions to provide capital for the biogas sector, it was necessary to analyze how and to what extent these barriers could impact on the WTP for investments in family-sized biogas digesters.

The findings show that socioeconomic and technical factors, especially household income, household head age, sex of household head, family size, education, and livestock practice are main factors for the biogas development in the district of Gihanga.

We learned that the belief in climate change significantly affects the WTP for family-sized biogas digester. Farmers are aware that biogas use in cooking could contribute to protect forest resources and therefore, the climate especially the rainfall period will be stable and predictable. Moreover, they argued that agricultural activities will be more productive to reduce the hunger in their families. For households where women were interviewed, we found that the average effect on the WTP is positive.

During the field research we experienced that even though decisions regarding household income are the responsibility of men. Women are motivated to invest the household income in biogas technologies. They revealed that this kind of energy, if developed in their village, will reduce several stresses that they face while gathering firewood for cooking, and getting budget for candle expenditures. Another finding is that the variation of household head age negatively affects the WTP for the biogas use. We conclude that elderly people are not willing to pay a monthly sum for an investment in biogas use. We found that old people in the district are not willing to adopt biogas technologies. Most of them declared that they have never seen these technologies; and that they are not motivated to apply them.

Biogas technologies offer potential advantages such as clean energy for cooking and lighting. Others concern the sanitation and effluents that are relevant for agricultural fertilization.

All biogas advantages combined contribute to human well-being, jobs, capabilities, and sustainable development, especially in the rural area. Unfortunately, most of the rural population is not aware of these advantages.

There are many barriers that need to be tackled in order to foster the development and uptake of the biogas sector in Burundi. Based on our findings and experiences, we suggest the following policy elements:

  • 1.
    Awareness campaigns in the region and in the whole country about biogas benefits are necessary and Women, particularly, should be involved in the development process of biogas technologies, since they are more motivated to participate for their sustainability.

Other findings argued, women in rural area are more affected by climate change consequences [see International Institute for Environment and Development, 2008, p. 25].

  • 2.
    Financial and nonfinancial incentives are necessary for households.

By deciding which households are subject to support, it is important to prioritize farmers who have cattle, large families, and families who have a certain level of education.

  • 3.
    A national biogas program should particularly focus on policies that foster access to water resources, and the support of farmers by providing financial and nonfinancial incentives.

Acknowledgments

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Objectives of the Study
  5. 3. Materials and Methods
  6. 4. Results and Discussion
  7. 5. Conclusions and Policy Implications
  8. Acknowledgments
  9. References

This paper was presented at the 36th IAEE conference in Daegu, Korea, 16–20 June 2013. The Centre of Studies on Sustainable Development (CSSD) of Free University of Brussels (ULB) financed this trip. E. Pirgmaer contributed to improve the quality of the English of this article. Y. De Smet, N. Nimenya provided useful comments in order to improve its quality. We thank J. Nzomoi and R. Luckach for their useful advices during the research. Lastly, we thank farmers of the District of Gihanga for their cooperation during the survey. Thank you all for your invaluable contribution!

References

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Objectives of the Study
  5. 3. Materials and Methods
  6. 4. Results and Discussion
  7. 5. Conclusions and Policy Implications
  8. Acknowledgments
  9. References
  • Abdullah, S. M. (2009), Willingness to pay for renewable energy options in developing countries: The case of Kenya, PhD thesis, pp. 1–337, Univ. of Bath, Bath, U. K.
  • ADFV (2010), Acting on climate change for sustainable development in Africa, in Climate Change, Economic Growth and Poverty Reduction in Africa, p. 10, United Nations Conference Centre, Addis Ababa, Ethiopia.
  • AFD (2011), The Paris-Nairobi climate initiative: Access to clean energy for all in Africa and countries vulnerable to climate change, in Background Document to Clean Energy for All Conference, p. 2, Paris, France.
  • AFDB (2009), Un plan d'action pour les infrastructures au Burundi, accélérer l'intégration régionale, p. 235.
  • Andersen, V. (2008), Models of Investment Decision Making in Finance, in Encyclopedia of Complexity and Systems Sciences, p. 22, Univ. of Nice-Sophia Antipolis, Nice, France.
  • APESA (2007), Méthanisation et production de Biogaz: Etat de l'art, p. 37.
  • ARE (2011a), Hybrid Mini-Grid for Rural Electrification: Lessons Learned, pp. 172 , ARE, Brussels, Belgium.
  • ARE (2011b), Rural Electrification with Renewable Technologies, Quality Standards and Business Models, pp. 156, ARE, Brussels, Belgium.
  • BEKE, E. T. (2011), Institutional constraints and adoption of improved rice varieties: Econometric evidence from Ivory Coast, Rev. Agric. Environ. Stud., 92(2), 117141.
  • Bellemare, C., A. Sebald, and M. Strobel (2010), Measuring the willingness to pay to avoid guilt: Estimation using equilibrium and stated belief models, IZA DP No. 4803, pp. 1–31, IZA, Bonn, Germany.
  • Curry T. E., S. Ansolabehere, and H. Herzog (2007), A survey of public attitudes towards climate change and climate change mitigation technologies in the United States: Analyses of 2006, Results, N°.LFEE, 2007-01 WP, 48 pp., MIT Press, Cambridge, Mass.
  • Day, D. L., T. H. Chen, J. C. Anderson, and M. P. Steinberg (1990), Biogas plants for small farms in Kenya, Biomass, 21(1990), 8399.
  • De-Graft Acquah, H., and E. E. Onumah (2011), Farmers perception and adaptation to climate change: An estimation of willingness to pay for climate change mitigation in Ghana, Agris, III(4), 3139.
  • Dekelver, G. and S. Ruzigama (2005), Report on the feasibility study for a biogas support programme in the Republic of Rwanda, p. 60, SNV, The Hague, Netherlands.
  • Derrick, A. (2000), Institutional barriers in developing renewable energy projects and markets, CTI Joint Industry Seminar on Technology Transfer, pp. 1–4, San Salvador, El Salvador.
  • Dessus, B. (2008), Le méthane d'origine Agricole, Cible à Privilégier dans la lutte contre le Changement Climatique, Revue Durable, No. 29, 16.
  • Fenhann, J., and P. Painuly (2002), Implementation of Renewable Energy, Technologies–Opportunities and Barriers, Summary of Country Studies, UNEP Collaborating Centre on Energy and Environment RisQ, pp. 143, National Laboratory, Roskilde, Denmark.
  • GTZ (2007), Feasibility study for a national domestic biogas program in Burkina Faso, 98 pp., Bonn, Germany.
  • GTZ (2009), Renewable Energies in East Africa, Country Chapter/Burundi, 2009, 19 pp., Bonn, Germany.
  • Hewson, P., and B. Whalley (2009), Introduction to Ordinary Least Squares Regression, 56 pp. [Available at http://users.aims.ac.za/∼ben/ccOLSnew.pdf.]
  • Humbad, A., S. Kumar, and B. V. Babu (2009), Carbon Credits for energy self-sufficiency in rural India: A case study in rural India, Energy Educ. Sci. Technol. A, 22(2), 187197.
  • IFDC (2011), Enquête sur le flux d'approvisionnement en Bois énergie à Bujumbura, 1–42.
  • IMF (2011a), Burundi: Document de stratégie pour la Réduction de la pauvreté, Rapport du FMI n° 11/53, p. 119, Fonds Monétaire International, Washington, D. C.
  • IMF (2011b), IMF Country Report No. 11/104, p. 57, IMF, Washington, D. C.
  • International Institute for Environment and Development (IIED), 2008, Climate change impacts on the progress towards the sustainability of MDG achievement across Africa, in 10th Meeting of the African Partnership Forum, pp. 1–40, Kyoto, Japan.
  • International Renewable Energy Agency- IRENA (2012), Renewable Energy Jobs & Access, p. 80.
  • ISAT & GTZ. 1997, Biogas Digest: Biogas Basics, vol. 1, 46 pp., Bonn, Germany.
  • Johnson, D.D.P., Michael E. Price, and Mark Van Vugt (2013), Darwin's invisible hand: Market competition, evolution and the firm, J. Econ. Behav. Organ., 90S(2013), S128S140.
  • Karekezi, S. and W Kithyoma (2003), Renewable energy in Africa: Prospects and limits, pp. 1–30, African Energy Policy Research Network, Nairobi, Kenya.
  • Kostka, G., U. Moslener and J.G. Andreas (2011), Barriers to energy efficiency improvement: Empirical evidence from small- and medium-sized enterprises in China, pp. 1–13, Frankfurt School, Frankfurt, Germany.
  • Loewe, M. and N. Rippin (2012), Changing Global Patterns of Poverty, pp. 14, 1615–5483. German Development Institute, Bonn, Germany.
  • Magnani, R. (2001), Guide d'échantillonnage, Food and Nutrition Technical Assistance Project, pp. 1–5, Academy for Educational Development, Washington, D. C.
  • Marcato, C. (2007), Origine, devenir et impact du Cuivre et du zinc des lisiers porcins. Rôle de la Digestion Anaérobie, Thèse de Doctorat, pp. 1–199, Institut national Polytechnique de Toulouse, Toulouse, France.
  • Meeks, J. V. (2012), Willingness-to-pay for maintenance and improvements to existing sanitation infrastructure: Assessing community-led total sanitation in Mopti, Mali, Graduate School Theses and Dissertations, pp. 1–173, Univ. of South Florida, Tampa, Fla.
  • Ministère de l'Eau, de l'Environnement, de l'Aménagement du Territoire et de l'Urbanisme and UNDP (2010), Deuxième Communication Nationale sur les Changements Climatiques au Burundi, 150 pp.
  • MPDC et PNUD (2011), Vision Burundi 2025, 104 pp. [Available at www.bi.undp.org.]
  • OECD/IEA (2006), Renewable energy: RD&D priorities, Paris, France.
  • OECD/IEA (2011), Energy for all: Financing access for the poor, Oslo, Norway.
  • Pacini, H., D. Khatiwada, and T. Lönnqvist (2010), Tailor-made solutions: Small-scale biofuels and trade, Bridges Trade BioRes Rev., 4(4), 116.
  • Pandey, B., P. Subedi, M. Sengendo, and I. Monroe (2007), Report on the Feasibility for a National Household Biogas Commercialization Program in Uganda, 122 pp., Winrock International, Brussels, Belgium.
  • Pasta, D. J. (2009), Learning when to be discrete: Continuous vs. categorical predictors, in SAS Global Forum 2009, pp. 1–10, ICON Clinical Research, San Francisco, Calif.
  • RECORD (2009), Techniques de production d'électricité à partir de biogaz et de gaz de Synthèse, n°07-0226/1A, 253 pp.
  • Solomon, B. D., and N. H. Johnson (2009), Valuing Climate Protection through Willingness to pay for Biomass Ethanol, Ecol. Econ., 68, 21372144.
  • Ulimwengu, J. and P. Sanyal (2011), Joint estimation of farmers' stated willingness to pay for agricultural services, IFPRI Discussion Paper 01070, 28 pp. [Available at www.ifpri.org.]
  • United Nation Conference on Sustainable Development (2011), Promotion of new and renewable sources of energy, in United Nation Conference on Sustainable Development, Sixty-Sixth Sessions, 24 pp.
  • Venkatachalam, L. (2004), The contingent valuation method: A review, Environ. Impact Assess. Rev., 24(2004), 89124.
  • Walekhwa, P., J. Mugisha, and L. Drake (2009), Biogas energy from family-sized digesters in Uganda: Critical factors and policy implications, Energy Policy, 37, 27542762.