Empirical study on influencing factors of biogas technology adoption in Khyber Pakhtunkhwa,Pakistan

Abstract

Climate change caused by global warming, and the growing scarcity of nonrenewable energy sources, have driven Pakistan to shift from a traditional energy consumption pattern to a renewable-energy-use pattern. The per capita energy consumption in rural Pakistan is very low, especially in rural areas heavily relying on traditional energy sources. This paper presents the extent of biogas technology adoption by Pakistani rural households and the factors affecting their decision to adopt the technology in three selected districts of Khyber Pakhtunkhwa province. The data were collected by interviewing 480 respondents by using a pretested and designed questionnaire. The results show that the household adoption rate of biogas technologies is low. The factors affecting the adoption decision of households included household income, access to credit, cultivated land area, the number of cattle in the household, education, and family size. The study also found fundamental barriers to the household adoption of biogas technologies, such as a lack of proper technical services by implementing organizations and insufficient governmental support. The authors make recommendations based on the findings to increase the adoption rate of biogas technologies in rural Pakistan.

Keywords

Adoption decision binary logistic regression biogas plant biogas technology Pakistan

Introduction

Energy access and availability are fundamental for individual consumers’ quality of life, economic growth, and employment opportunities. Increased energy supply and consumption lead to improved economic development and higher living standards.¹ The human development index of a country is directly proportional to its energy prosperity.² A shortage in the energy supply is a barrier to the development, economic growth, and prosperity of a country and adversely affects its environment, water availability, agricultural productivity, and human health.³ However, the increasing consumption of nonrenewable fossil fuels also negatively affects a country’s development because of exposure of its population to the adverse impacts of climate change. According to the International Energy Agency, the per capita electricity consumption in developed countries is markedly higher than that in other countries, resulting in the release of greater amounts of greenhouse gases such as methane, carbon dioxide, and sulfur dioxide into the atmosphere. This, in turn, results in environmental problems such as air pollution, deforestation, global warming, and climate change and also inflicts several respiratory diseases.⁴

Despite these serious environmental threats, the world still consumes nonrenewable energy sources on a large scale. In this context, awareness of climate change and resource depletion has been raised considerably by international treaties and protocols like the Kyoto Protocol and the Copenhagen Summit, which encourage the world to gradually shift to renewable energy sources to reduce greenhouse gas emissions.⁵ Today, many countries exploit such sustainable and environment-friendly renewable energy sources such as biogas and solar and wind energy systems.⁶

In the last few decades, Pakistan has consistently incurred severe energy crises with its high reliance on nonrenewable fossil fuels for electricity production, which costs almost US$7 billion annually.⁷ Despite higher investment in the energy sector, the energy supply still cannot meet the demand, although the per capita energy consumption in Pakistan in 2014–2015 (6187 kWh) was far less than that of the United States (108,424 kWh) and the United Kingdom (44,245 kWh).⁸ Also, the natural gas grid has not extended to rural areas, which makes the electricity shortage more severe than in urban areas, which use natural gas for electricity production. Consequently, rural communities rely heavily on traditional fuels such as petroleum, fuel-wood, animal dung cakes,^a crop residues, and tree residues. Use of these fuels adversely affects the health of rural residents, who suffer from an increased incidence of respiratory diseases and eye infections.⁹ Among these fuels, the supply of fuel-wood is decreasing due to higher rates of deforestation.⁷

On the other hand, both rural and urban areas have abundant waste by-products, such as crop residues, animal wastes, and other organic wastes. The accumulation of these wastes and lack of safe waste handling practices have also contributed to environmental and health problems. In this context, the production of biogas by using these organic wastes can potentially reduce the volume of these generated wastes in rural vicinities.^10,11

Pakistan, with a total of 170 million head of available cattle, has great potential to produce biogas as a major renewable energy source.¹² Among those animals, the number of cows and buffaloes is 72 million, and their dung may be used as feedstock for biogas production.¹³ It is estimated that on average, a medium size cow, bull, or buffalo produces 10 kg of dung, and 72 million animals could provide 720 kg of dung per day.⁷ According to research, it is estimated that at a temperature of 15°C, a dry mass of 1 kg of animal dung yields 0.19 m³ of biogas, and this conversion rate doubles by increasing the temperature to 27°C.¹⁴ Based on this calculation, all the collected animals’ dung would probably produce more biogas than is currently being produced.

On the other hand, forests provide 2,63,000 m³ of firewood, and most of it is used as fuel in the rural sector.¹⁵ The consistent collection of firewood from forests reduces carbon sequestration, and firewood burning by rural households increases carbon emissions, which might consequently lead to regional climate change. The current forest assets of the country are not enough to fulfill the growing demand for wood for various purposes.¹⁶

Biogas is produced from biomass through the process of anaerobic digestion,¹⁷ and it is the fourth largest energy source on a global scale providing nearly 14% of primary energy.^18,19 Several initiatives have been started for the promotion of biogas technology in Pakistan by various organizations such as Pakistan Council of Appropriate Technology, Pakistan Council for Renewable Energy Technologies (PCRET), Alternative Energy Development Board (AEDB), and Pakistan Renewable Energy Society. The PCRET was established in 2001 with the objective to disseminate biogas technology in the country including the research activities in this sector. Similarly, AEDB, established in 2003, is also playing its role in the promotion of renewable energies in the country in order to reduce emissions of greenhouse gases. Of these organizations, the major actor in dissemination of biogas technology is PCRET with an existing 2483 installed household biogas plants under Public Sector Development Program across the whole country.²⁰ Literature study shows that the number of existing biogas plants in Pakistan is 5357 with a potential of biogas production up to 12–16 million m³/day indicating the sufficient amount of alternate fuel for cooking purpose in rural areas.^21,22

Such initiatives are also taken by several industries for their facilities. For example, four industries (J.K. farms, Al-Hamad Exports, Tahir dairy farms, and Ashraf Zia textile industries) in Faisalabad, Punjab have their own large-scale biogas plants. Similarly, several NGOs are also involved in biogas technology promotion such as Revgreen Pakistan, Reon Energy Limited, National Rural Support Program, Koshish, Integrated Rural Support Program, Foundation for Integrated Development Action (FIDA), The Green Circle, etc. After the technology got much familiarity, many private firms initiated business of installing biogas technology, and now, the public is supposed to have sufficient awareness about installation of this technology.²³

Several studies have indicated institutional barriers, demographic factors, environmental factors, and household socioeconomic characteristics as factors influencing the adoption of biogas technology.^24–28 Mwirigi et al.²⁶ did research in sub-Saharan African countries focusing on the socioeconomic barriers to adoption of biogas technology as well as the factors enhancing such adoption. They found that financing, institutional reforms, research and development as well as support in transfer of knowledge are necessary for widespread adoption of biogas technology.²⁶ Bensah et al.²⁹ assessed the status of biogas industry and the barriers in developing sustainable marketing for biogas plants in Ghana. Their findings revealed that biogas companies have enough potential to expand their services by developing better public awareness strategies, standardized biogas plants’ models, and better financing schemes.²⁹ Another related study done by Qu et al.²⁷ investigated the obstacles that farmers faced in adopting biogas technology in China. They found that age, income, number of family members at home, and government support markedly affected the adoption of biogas technology.²⁷ Furthermore, Kabir et al.²⁵ did an empirical study on determinant factors influencing the adoption of biogas technology in Bangladesh by applying binary logistic regression. Their results from the regression model indicated that educational level, income level, cattle size, and gender of the household heads significantly influence biogas technology adoption.²⁵ Jan and Akram¹² predicted the household’s willingness to adopt a biogas system in Khyber Pakhtunkhwa (KP) province in Pakistan. They found that female drudgery, awareness of the advantages and disadvantages of using biogas, educational level, the daily electricity shortage, and the effect of the electricity supply shortage on children’s education were statistically significant factors.¹² The study of Berhe et al.³⁰ aimed to find the factors influencing the choices of households to use different types of fuels. The study showed that access to electricity, working age, access to credit services, gender, the size of cattle holding, and livestock mobility significantly influences household energy choices.³⁰ Wahyudi³¹ determined the factors affecting biogas technology adoption by cattle farmers in Indonesia. He conducted a cross-sectional survey and found that the households with higher social status, income, and education adopt biogas technology more likely than those having lower social status, income, and education.³¹ Looking at these studies, it is still difficult to specify certain factors affecting biogas technology adoption globally and within a country. This is because different agro-ecological regions have diverse socioeconomic characteristics.³²

Renewable energy is strategically important for Pakistan because it is an energy-deficient country. It is previously mentioned that various government organizations and nongovernmental organizations (NGOs) have started implementing biogas technology widely across the country. So far, however, biogas technology adoption has not extended as far as expected and requires further empirical research. In this context, the present study focused on determining the extent of biogas technology adoption and the factors affecting the decision to adopt biogas technologies at the household level. This study could be useful for governments and NGOs to improve policy instruments, project planning, and technical services to increase the adoption of biogas technology. This could lead to the formulation of a sustainable energy development strategy and the reduction of deforestation and greenhouse gas emissions.

Materials and methods

Sampling, data collection, and analysis

This study was done in rural areas of the KP province in Pakistan. KP is a farming-based province with low energy consumption. A multistage sampling technique was used to select study sites at various levels (district, union council, and village) in KP by using a purposive sampling method based on a relatively higher number of biogas plants. Using this procedure, we selected three districts, Dera Ismail Khan, Mardan, and Abbottabad, out of the total 24 districts, and then selected three union councils^b (UCs) (Kech, Saribahlol, and Lora, respectively) from the three districts (Figure 1). We then selected nine villages out of the total 25 villages in the selected UCs, as shown in Figures 2 to 4. We also used a purposive sampling method to select all 195 biogas users and a simple random sampling method to select 285 nonusers of biogas (out of the total 2286 households) from the study villages. We got the list of households from their local government departments. We used a purposive sampling technique for biogas users because they were far fewer than nonusers, which could otherwise have made a highly disproportionate sample size. Therefore, the results should be considered as self-illustrative rather than representative.

Figure 1.

Map of Pakistan showing KP Province and location of selected districts in KP Province.

Figure 2.

Location map of selected villages in district Mardan.

Figure 3.

Location map of selected villages in district Abbottabad.

Figure 4.

Location map of selected villages in district Dera Ismail Khan.

A semistructured questionnaire was administered to collect data, and variables were selected and designed by prestudying the resource endowment and socioeconomic features of the selected research areas and by studying the available literature. The questionnaire was pretested in the study sites and was updated based on the initial findings. The field survey in the selected areas was carried out from June to September 2016. A binary logistic regression model was applied to measure the relationship between the factors and the biogas adoption. The collected data were processed with the help of Statistical Package for Social Sciences (SPSS 16, IBM, USA). The quantitative findings were also verified by local key informants who were knowledgeable about the socioeconomic problems in the study sites.

The study team also verified the adequacy of the sample size in this study for using logistic regression. Therefore, the minimum sample size required in this study was calculated based on the following guideline³³

N=10 k/p

(1)

where p is the smallest of the proportions of negative or positive cases in the population and k is the number of covariates (independent variables). By following this guideline, the calculated number of cases was 325. This implies that the calculated sample size of 480 for this study was more than adequate for the logistic regression.

Description of biogas technology in study sites

The study focused on observing two types of biogas plants in the study sites: fixed dome and floating drum as shown in Figure 5. Generally, different organic materials can be used for biogas production.³⁴

Figure 5.

Biogas plants in three selected districts (Dera Ismail Khan, Abbottabad, and Mardan).

In study sites, animal dung (mainly cows and buffalo) was used as a feedstock for biogas production. The production capacity of these two types of biogas plants in all districts was different because of the different plant sizes: 5, 8, 10, 15, and 20 m³.

The fixed dome type of biogas plant was common in the study areas. It was made of bricks or concrete, and its construction required a high level of technical skills to avoid gas leakage.³⁵ Those plants consisted of an inlet, an airtight underground digester, and an outlet tank. In the inlet tank, water and the animal dung were mixed in equal amounts. From there, it went to the underground digester for biogas production through anaerobic digestion. The produced gas accumulated in the dome of the digester and put pressure on the substrate, pushing it out of the digester through the outlet tank.

The other type of biogas plant in the study areas was the floating drum type, which exists throughout the world.³⁵ It is constructed of bricks and a concrete pit partially sunk in the ground. Under the liquid of the digester, a steel dome is buoyed by the gradual production of gas and serves as a gas collector.³⁶ Generally, the amount of biogas produced is affected mainly by the design of the digester, the organic load rate, the hydro retention time, and the temperature.³⁷

For the selected study districts, biogas technology had been developed both by government organizations, such as the PCRET and the Pakistan Dairy Development Company (PDDC), and NGOs such as FIDA. Both the government organizations and the NGOs applied similar criteria and conditions in selecting beneficiary households; namely, a sufficient number of animals (cows or buffalo), enough land to build the biogas plant, and availability of water for biogas production. Those organizations also provided a financial subsidy for the construction of biogas plants.

Application of the binary logistic regression model

Binary logistic regression was used to determine the factors affecting the adoption of biogas technology. In this study, the adoption was the dependent variable, defined as the decision by a household to install a biogas plant. Both logit and probit are well-known approaches to adoption studies.³⁸ A decision as to whether to use a probit or a logit model is based on its computational convenience.³⁹ Logistic regression is used when the dependent variable is a dichotomy, and the independent variables are of any type. It applies maximum likelihood estimation after transforming the dependent variable into a logit variable.⁴⁰ It estimates the probability of a certain event occurring. The dependent variable is a logit, which is the natural log of the probabilities; that is

In (\frac{P}{1 - P}) = a + bX

(2)

By extracting P, the equation can be rewritten as

P = \frac{e^{a + bX}}{1 + e^{a + bX}}

(3)

where P is the probability of the event occurring, X are independent variables, e is the base of the natural logarithm, and a and b are the parameters to be estimated by the model. The empirical form of the model used in the study was

PrY = \frac{1}{1 + e^{- (a + bX)}}

(4)

where Y is the logit for the dependent variable.

The logistic prediction equation for the study was

Y = b_{0} + B_{1} X_{1} + B_{2} X_{2} + B_{3} X_{3} + \dots B_{n} X_{n}

(5)

where

b_{0}

is a constant term;

X_{1}, X_{2}, \dots, X_{n}

are independent variables likely to affect the probability of adopting the biogas technology; and

b_{1}, \dots b_{n}

are the coefficients to be estimated. The dependent variable was modelled as Y = adoption of biogas technology = P(Y) = 1 if the household chooses to adopt biogas technology and 0 otherwise.

Selection of the variables explaining the adoption of biogas technology

No rigid theoretical model has universally defined factors affecting biogas technology adoption. This is because rural households consider factors other than socioeconomic incentives, including noneconomic factors.²⁴ In this study, literature and field experiences were used to select the prospective variables that could affect the households’ decision to adopt biogas. A list of prospective variables of biogas adoption is shown in Table 1.

Table 1.

Definition of explanatory variables for the logistic regression model.

Variable	Type	Description
GenderHHH	Binary	Sex of the head of household (0 = Female, 1 = Male)
AgeHHH	Continuous	Age of the head of household in years
EduHHH	Continuous	Number of years spent in schooling by the head of household
SizeHH	Continuous	Number of persons in the household
HeadsCatt	Continuous	Number of cattle in the household
Income	Binary	Household categories based on level of monthly income (0 = lower income: up to Pakistani rupee (PKR) 15,000; 1 = higher income: PKR 15,001–33,000)
CultLand	Continuous	Cultivated land area in acres at the household level
AccCredit	Binary	Access to credit service by the head of household (1 = Yes, 0 = Otherwise)

Source: Authors’ analysis.

Independent variables in the model include gender, age, education, household size, number of cattle owned, income, cultivated land area, and access to credit. In a second stage, specific assumptions were made regarding each variable, as listed in Table 2.

Table 2.

Explanatory variables with priory signs for the logit model.

Description	Expected sign
Gender of the head of household (0 = Female, 1 = Male)	±
Age of the head of household in years	±
Number of years spent in schooling by the head of household	+
Total number of persons in the household	±
Number of cattle in the household	+
Household categories based on monthly income (0 = lower income: up to Pakistani Rupee (PKR) 15,000; 1 = higher income: PKR 15,001–33,000)	+
Cultivated land area in acres at the household level	+
Access to credit service by the head of household (1 = Yes, 0 = Otherwise)	+

Source: Authors’ analysis.

Gender, household size, and age were assumed to have mixed influences; that is, either positive or negative on the adoption of biogas technology. The rest of the variables were assumed to have a positive influence on the adoption of biogas technology.

Results and discussion

General characteristics of the households

It is important in social science research to determine the socioeconomic characteristics of the persons being studied. Table 3 summarizes the socioeconomic characteristics of the 480 sampled households in the study areas. These characteristics included age, cultivated land area, household size, head of cattle, years of education, and income.

Table 3.

General characteristics of the respondents.

Variable	Mean	Standard deviation	Minimum	Maximum
AgeHHH (years)	49.6	7.6	30	71
CultLand (acres)	1.75	1.0	0	5
SizeHH	9.79	4.7	3	21
HeadsCatt	6.39	3.9	1	20
EduHHH (years)	2.54	2.8	0	10
Income (PKR)	20,000	7144	7000	33,000

Source: Field survey, 2016.

Age was a variable considered important for the adoption of innovations. Results show that the mean age of the respondents was 49.6 years, which revealed a possible labor source for biogas-related activities. However, the main reason for such a high mean value of age was that the respondents for this study were heads of households. Age influences biogas technology adoption in a complex form because older individuals with more experience and higher economic status are expected to be more likely to adopt it than would younger individuals.⁴¹ However, the older farmers may also consider it a risk and therefore may have a lower level of adoption than younger ones.²⁴

The area of land owned is globally considered to be positively related to the likelihood of adoption of biogas technology.⁴² Our results show that on average, a studied person had 1.75 acres of cultivated land, which was considered enough to designate him or her as a farmer.

Household size determines the labor force available to feed the biogas plant daily, and therefore, it is considered an important factor affecting biogas technology adoption.⁴³ On average, the respondents in the study area had up to 10 family members, which indicated the availability of enough labor for carrying out biogas-related operations.

To install a biogas plant, it is mandatory to have enough cattle. We found that the mean value of the number of cattle was 6.39, indicating the possibility of enough feedstock for running a biogas plant (with a minimum value of 1 and a maximum value of 20). According to the aforementioned estimation by Amjid et al.,⁷ one head of cattle produces 10 kg of dung per day. Usually, a small-scale biogas plant requires a feedstock of four head of cattle (40 kg) per day.^44,45 Therefore, the number of cattle is considered a basic factor for adoption of biogas technology.⁴⁶ According to our findings in Table 3, the average number of cattle was 6.39, which probably produced almost 64 kg of dung per day.

Education is another important variable when dealing with technology adoption. Table 3 indicates that the respondents had completed an average of 2.54 years of education. We also found an average of 3.3 and 2 years for biogas users and nonusers, respectively. This result is similar to that of Mengistu et al.,²⁸ who found that the respondents’ mean number of years of education was 2.85 and that of biogas users and nonusers was 3.7 and 2, respectively. A study done by Kelebe et al.⁴¹ also reported similar findings. This implies that the inclusion of this variable in the model seemed logical. It indicates that a large portion of the respondents could at least read and write, implying that the households were able to be trained to use biogas technology. Household heads having more education are supposed to be more knowledgeable, informed, and aware of the adverse effects of fossil fuels on the environment.²⁴ They consider clean energy sources (such as biogas) more environment-friendly as compared with those having no education.

Results also show that the respondents’ average income per month was Pakistani Rupee (PKR) 20,000 (with minimum and maximum values of 7000 and 33,000), indicating the availability of income to afford the costs associated with the construction of a biogas plant.

Results on the extent of biogas technology adoption

Biogas technology provides several environmental and social benefits, but still it is adopted at a lower rate when transferred to rural households in developing countries.⁴⁷ The data presented in Table 4 show a union-council-wise comparison regarding the number of households, the total number of installed biogas plants, and the extent of biogas technology adoption. The project-operating organizations did not have a predefined number of biogas plants to provide in each district, UC, or village. However, they motivated the existing beneficiaries to spread the information about the running project. However, most of the households had still not accessed such information that was available on the websites of those organizations. Therefore, the households had to access such information through person-to-person communication before they were selected as beneficiaries, based on the required terms and conditions. However, the Dera Ismail Khan district had an effective way of disseminating project information at a household level; they used village organizations. Those village organizations were actually made by the project-operating organizations.

Table 4.

Extent of biogas technology adoption in the study sites.

UC	Total households	Biogas plants	Extent of biogas technology adoption
Saribahlol	481	15	0.031
Kech	1300	150	0.115
Lora	505	30	0.059
All	2286	195	0.085

Source: Field survey, 2016.

As shown in Table 4, the number of installed biogas plants in the area was 195, and the number of households was 2286. The extent of biogas adoption was determined by dividing the available number of biogas plants by the number of households. This extent was found to be low (0.085) for the study sites, with only 195 biogas users. We also found this figure low when compared with that found by a researcher for Kongwa and Bahi districts in semi-arid areas of Tanzania.³⁶ This low level of biogas adoption is particularly noticeable in rural areas, where it is more feasible.

Among the three UCs, the greatest extent of adoption was found for UC Kech (0.115) of the Dera Ismail Khan district, followed by UC Lora (0.059) of Abbottabad and UC Saribahlol (0.031) of Mardan. The reason for the highest figure for Dera Ismail Khan was that the project-operating organization (FIDA) provided a greater financial subsidy for the construction of biogas plants in this district than was provided by other organizations (PCRET and PDDC) in Abbottabad and Mardan respectively.

Results based on the factors affecting biogas technology adoption

Mitigation of some sustainability issues in both rural and urban settings in developing countries is partly possible with the adoption of biogas technology.⁴⁸ According to Table 5, the logistic regression model was statistically significant (χ² (8) = 103.84 and P = 0.000). The model correctly predicted 72.70% of the sample households, which was greater than the proportion by chance accuracy rate (64.75%), thereby satisfying the criteria for classification accuracy. The “goodness-of-fit” of the model was validated by looking at the nonsignificant value (P = 0.327 > 0.05) of the Hosmer and Lemeshow test, although the pseudo R² was relatively low (26.30%) (Table 5). This low coefficient of variation does not affect the quality of the model, because the pseudo R² in logistic regression does not have a meaning equivalent to that of the ordinary least squares or R² value because it comes from binary variables.²⁵ The independent variables in the analysis did not have a standard error greater than 2, indicating that there were no numerical problems among the independent variables. Multicollinearity among the independent variables was also checked to confirm the statistical power of the model. Usually, statisticians consider the signs of multicollinearity in the case of a higher bivariate correlation among independent variables (greater than +0.7 in the case of a positive correlation, or less than −0.7 in the case of a negative correlation), a tolerance value less than 0.2, and a variance inflation factor value greater than 10. We checked our results by following this guideline and found no signs of multicollinearity. According to the Wald χ² test, six of the eight independent variables were found to influence the adoption of biogas at a significance level of 0.05 (Table 5). These include the cultivated land area, education, head of cattle, family size, access to credit, and income. The variables determined in this analysis were different from those reported by Jan and Akram,¹² who followed the probit model on the willingness to adopt biogas technology in Pakistan.

Table 5.

Factors affecting the adoption of biogas technology.

Variable	Coefficient (1)	Standard error	Wald	Sig	Odds ratio
Constant	−3.231	0.814	15.735	0.000	0.040
GenderHHH (male)	0.121	0.356	0.115	0.734	1.129
AgeHHH	−0.017	0.014	1.556	0.212	0.983
EduHHH	0.147	0.038	15.361	0.000	1.159
SizeHH	0.094	0.022	18.037	0.000	1.098
HeadsCatt	0.173	0.030	34.346	0.000	1.189
Income (higher)	0.764	0.223	11.790	0.001	2.148
CultLand	0.316	0.103	9.442	0.002	1.372
AccCredit (Yes)	0.541	0.260	4.338	0.037	1.718
Number of observations: 480

−2 Log likelihood = 544.60; omnibus tests of model coefficients (χ² = 103.84, df = 8, and P = 0.000); Hosmer and Lemeshow test (χ² = 9.181, df = 8, and P = 0.327); pseudo R² = 26.30%; and overall percentage of correctly predicted sample households = 72.70.

Source: Field survey, 2016.

Gender

In Pakistan, ownership of assets and their control affect the decision as to whether to adopt an innovation. The role of gender in biogas adoption is very interesting and considered important. The gender of a household’s head has been assumed to have either a positive or negative effect on the adoption of biogas technology.^25,28,41 In this study, of the 480 respondents, 90% were males and 10% females. Our results in Table 5 indicate that the gender of the household head had a positive and nonsignificant (5% significance level) relation with biogas technology adoption. This means that male-headed households had a higher probability of adoption than female-headed households. In Pakistan, the management of fuel consumption is the responsibility of females, but males dominate the decision-making and the ownership of resources at the household level. These results look similar to those of Mengistu et al.²⁸ and Mwirigi et al.,²⁶ who found that male-headed households were more likely to adopt biogas than female-headed households. However, an opposite finding was reported by Kabir et al.,²⁵ who found that female-headed households were more likely to adopt this technology than were male-headed households.

Age

We found the relation of age of the household’s head to the probability of biogas adoption to be negative, implying that younger household heads were more likely to adopt it than were older household heads. However, this relation was found nonsignificant (P = 0.212). These results tally with those of Walekhwa et al.,²⁴ who discovered that the age of farmers and the likelihood of adoption of biogas technology were negatively correlated. This indicates that older people did not take more risks related to the adoption of innovations and were therefore less willing to make such decisions. On the other hand, our results contradict the findings of Kelebe et al.,⁴¹ who reported a positive relation between the age of the respondents and the probability of biogas adoption.

Years of education

Lack of education is one of the most prominent barriers to the dissemination of biogas technology in economically less-developed countries.⁴⁹ Our results show that the number of years spent in schooling and the likelihood of biogas technology adoption had a positive and highly significant relationship (P = 0.000), implying that more years of formal education of the household’s head increased the likelihood of adoption of biogas by a factor of 1.159. These results are similar to those of Mengistu et al.²⁸ and Jan and Akram,¹² who reported that more-educated heads of households were more likely to adopt biogas technology than those who had only a few years of schooling.

Family size

The proper operation and maintenance of biogas plants require labor.⁵⁰ Therefore, households with larger family sizes are expected to adopt biogas technology more than those with a smaller family size.⁵¹ On the other hand, a high number of family members may put pressure on the financial assets of the family, thereby negatively influencing biogas adoption.²⁸ Our results indicate that family size in the study area had a positive and significant effect (P = 0.000) on the households’ decision to adopt biogas, with a logit coefficient of 0.094 and an odds ratio of 1.098 (Table 5). These results are in agreement with those of Kelebe et al.,⁴¹ but contradict those of Abbas et al.,⁴² who discovered a negative relation between family size and biogas technology adoption.

Head of cattle

Livestock (mainly cows and buffalo) provide feedstock for biogas production in Pakistan. However, due to inadequate technical support, other sources of feedstock such as kitchen waste, industrial waste, and agricultural waste are not used. The selection of the size of a biogas plant depends on the optimum number of livestock for enough biogas production.⁵² We found that the assumption that there is a positive relation between the number of cattle and the likelihood of adoption of biogas technology was proved true. Results indicate that increasing the number of cattle increased the likelihood of adoption of biogas by a factor of 1.189 (Table 5). Similar findings were reported by Kabir et al.²⁵ and Christiaensen and Heltberg⁵³ in Bangladesh and China, respectively.

Income

We assumed that income was an explanatory variable affecting biogas technology adoption based on previous research studies^25,28,42 and field experience during a pilot survey in which we observed that rural households preferred biogas technology over other fuels because of fuel- and fertilizer-related (bio-slurry^c) benefits. Therefore, even with a higher income they were more interested in biogas compared with other cleaner energy sources like electricity. We grouped the respondents into two categories based on their monthly income: those at a lower income level (PKR up to 15,000) and those at a higher income level (PKR 15,001 to 33,000). Results for this variable were found to be significant at a significance level of 0.05, indicating that, as compared with lower-income households (the reference category), higher-income households were more likely to adopt biogas technology (with an odds ratio of 2.148). This indicates that besides subsidy provision, households were still markedly dependent on income. These results tally with the findings of Kambele⁵⁴ and Abbas et al.,⁴² who also found a positive relation between income and the likelihood of adopting biogas.

Cultivated land area

The cultivated land area is an important factor in the adoption of biogas technology because it is a source of fodder for livestock, which in turn provides feedstock for biogas production. Previous research shows a relation between farm size and biogas technology adoption.^24,25,28 The theoretical justification for including the cultivated land area instead of the total land area was that during the pilot study, it was noticed that the study area had sharecroppers as well who had much cultivated land. Also, households with a larger cultivated land area were supposed to have a higher income and more animal fodder. Therefore, we considered this variable very important to include in the model. According to Table 5, cultivated land area had a positive and significant relationship (P = 0.002) with the likelihood of a household adopting biogas, with an odds ratio of 1.372. These results are consistent with the findings obtained by Abbas et al.,⁴² who found that farm size was positively related to biogas technology adoption.

Access to credit

Adopting biogas technology leads to contribute to the conservation of the environment.⁵⁵ According to Putra et al.,⁵⁶ the consumption of firewood declines with the adoption of the biogas technology in Indonesia. In Pakistan, banks do not provide credit for construction of biogas plants in particular, but they provide funds in general to the households at the prevalent interest rate. Moreover, an informal credit service exists among individuals in time of need, but it lacks a legal payback structure. Keeping these facts in mind, we added access to credit as an explanatory variable affecting biogas technology adoption. This variable was also included by Kelebe et al.⁴¹ and Mengistu et al.²⁸ in their studies. We found that the project-operating organizations provided a financial subsidy for the construction of biogas plants in the study sites. However, subsidizing the construction of a biogas plant is a temporary solution, and so access to credit is important to raise the adoption of biogas technology exclusively.⁵⁷ The construction costs of a biogas plant are high; therefore, the beneficiaries were expected to spend a certain amount of their money for plant construction, implying that there might be a need for credit. The effect of access to credit on biogas adoption was proved to be positive and significant at a 5% significance level with an odds ratio of 1.718. These results are consistent with those of Mengistu et al.,²⁸ Gwavuya et al.,⁵⁸ and Kelebe et al.,⁴¹ who found access to credit as a positive influencing factor for biogas technology adoption.

Validity of findings

The above-mentioned quantitative results were verified by local key informants, who were knowledgeable about the socioeconomic problems in the study sites. The key informants included officials of biogas projects, members of the elected local government, and highly qualified persons at the local level. After verifying the results, they mentioned other barriers as well, such as promotional factors and institutional factors.

Among the promotional factors, they mentioned that although rural people could either watch television or listen to the radio, they could not use other information and communication technologies such as the Internet. Therefore, they did not have information about project-operating organizations (government or private). Information such as advertisements about biogas installation was usually available on websites but was not broadcast on television or radio.

Among the institutional factors, they mentioned that the government had no clear policy for biogas technology dissemination in the study areas. Only two governmental organizations, PCRET and PDDC, along with NGOs, implemented biogas projects for some time period. They also mentioned that the PCRET-made biogas plants gradually became inactive due to poor construction. The agricultural extension department of the government also did not play a role in the dissemination of biogas technology. The banks provided credit to households at a high enough interest rate that poor rural households could not afford credit, thereby impeding the adoption of these innovations. Also, despite subsidy provision, the poor households could not afford the high installation costs of biogas technology.

Conclusions and policy recommendations

The main purpose of this study was to find out the extent of adoption of biogas technology and its influencing factors in rural areas of KP in Pakistan. A logistic regression model was used to analyze the factors affecting the adoption of biogas technology. Results reveal that the extent of biogas adoption is very low, and several factors are involved in households’ decision to adopt biogas technology. Based on the significant results and higher odds ratios, the level of income of households was found to have a key role in biogas technology adoption. The second most prominent factor was the households’ access to credit. The next-important factors included cultivated land area, head of cattle owned, education, and family size. Lack of proper promotion of biogas technology by project-operating organizations and the insufficient role of the government in providing support services were also found as major barriers to biogas adoption.

Concerned government organizations and NGOs should encourage donor agencies to support biogas projects and should take proper steps to disseminate information on biogas technology widely in the country. Such organizations should place advertisements in the media to increase the awareness of people about biogas technology. Furthermore, the government should promote credit schemes and subsidy programs for keeping livestock, which is the source of feedstock for biogas production. This would not only benefit the farming community but would also bring improvements in the national economy through carbon trading and organic farming and would help mitigate global warming.

Footnotes

Declaration of conflicting interests

The authors declare no potential conflicts of interest with respect to the research, authorship, and/or submission of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: For the conduct of this study, the authors received financial support from the China Scholarship Council under the Ph.D. research studies scheme.

Notes

Syed M Amir has done PhD in Rural Development & Management at College of Humanities & Development Studies, China Agricultural University, Beijing, China. Currently, he is working as Assistant Professor at Department of Development Studies, COMSATS University Islamabad, Abbottabad campus. His research interests include rural education, renewable energy adoption and its impacts, rural development, land rights, disaster risk reduction, and vulnerability and resilience in disasters.

Yonggong Liu is a Professor at Department of Development Studies, College of Humanities and Development (COHD), China Agricultural University (CAU) in Beijing. Since 1991 he delivers seminars and lectures on development program planning and management, rural community development planning and management, agricultural innovation management and international development cooperation. His research areas cover agricultural and pastoral risk management, development program planning and impact monitoring and evaluation, natural resource management, climate change and impacts on agriculture and rural development, institutional and policy studies for agricultural and rural development. In the past 28 years, besides the academic activities within the university, he also serves as senior national and international consultant for designing, monitoring and evaluating agricultural and rural development programs implemented in China and other developing countries respectively financed by Chinese governmental line agencies and international development organizations.

Ashfaq A Shah received his Ph.D. degree from the College of Humanities and Development Studies, China Agricultural University Beijing, on July 2018. He is currently working as an Assistant Professor at the Department of Development Studies, School of Social Sciences and Humanities (S3H), National University of Science and Technology (NUST), Islamabad. His research interests include climate change adaptation, community based disaster risk reduction and management, vulnerability and resilience, risk perception, local Institutions preparedness in dealing with DRM/DRR, public health in emergencies.

Umer Khayyam received his Ph.D. degree in the subject of development studies from University of Münster, Germany, 2016. He is currently working as an Assistant Professor at the Department of Development Studies, School of Social Sciences and Humanities (S3H), National University of Science and Technology (NUST), Islamabad. He is working on different research projects and has got many publications in peer reviewed journals.

Zafar Mahmood has done PhD in subject of statistics from University of Agriculture Peshawar, Pakistan. Currently, he works as an Associate Professor at Department of Statistics, Mathematics and Computer Science in the same university. His research interests include applied statistics, computational statistics, statistical modeling, data simulation, data collection strategies in biological sciences, and research methodologies in social and agricultural sciences.

References

Hall

Introduction to special issue on new studies in EROI (energy return on investment). Sustainability 2011; 3: 1773–1777.

Asif

Energy crisis in Pakistan: origins, challenges, and sustainable solutions. Oxford University Press Catalogue, No. 9780195478761, 2012.

Amigun

Sigamoney

von Blottnitz

Commercialization of bio fuel industry in Africa: a review. Renew Sust Energ Rev 2008; 12: 690–711.

Organisation for Economic Co-operation and Development. Energy efficiency indicators: essentials for policy making. Paris: OECD Publishing, 2014.

Linares

Santos

Ventosa

Coordination of carbon reduction and renewable energy support policies. Clin Policy 2008; 8: 377–394.

Iwaro

Mwasha

Towards energy sustainability in the world: the implications of energy subsidy for developing countries. Int J Energy Environ 2010; 1: 705–714.

Amjid

Bilal

Nazir

, et al. Biogas, renewable energy resource for Pakistan. Renew Sust Energ Rev 2011; 15: 2833–2837.

National Electric Power Regulatory Authority (NEPRA). State of industry report 2015. Islamabad, Pakistan.

Jan

Ullah

Akram

, et al. Adoption of improved cook stoves in Pakistan: a logit analysis. Biomass Bioenerg 2017; 103: 55–62.

10.

Chun

Yang

Yoshikawa

Hydrogen generation from biogas reforming using a gliding arc plasma-catalyst reformer. Catal Today 2009; 148: 283–289.

11.

Zaidi

ZI.

Pakistan renewable energy report. United Nations Economic and Social Commission for Asia and the Pacific, 2011.

12.

Jan

Akram

Willingness of rural communities to adopt biogas systems in Pakistan: Critical factors and policy implications. Renew Sust Energ Rev 2018; 81: 3178–3185.

13.

GOP. Pakistan economic survey 2012–13. Islamabad: Economic Advisers Wing, Ministry of Finance, Government of Pakistan.

14.

Reddy

BS.

Rural energy systems for the year 2000 A. D. J

Energ Manag 1989; 13: 25–29.

15.

Government of Pakistan (GOP) Pakistan Economic Survey 2009–10. Economic Advisers Wing, Ministry of Finance, Pakistan.

16.

Asian Development Bank. Islamic republic of Pakistan: country environment analysis. Philippines: Asian Development Bank, 2008.

17.

Cvetkovic

Radoicic

Vukadinovic

, et al. Potentials and status of biogas as energy source in the Republic of Serbia. Renew Sust Energ Rev 2014; 31: 407–416.

18.

Karki

AB.

Biogas as renewable energy from organic waste. In: Doelle HW, Rokem JS, Berovic M (eds) Biotechnology. United Kingdom: EOLSS Publications, 2009, Vol 10, pp.262–292.

19.

Myles

Environmental, socioeconomic, health, and other positive impacts of building household biogas plants in rural India. In: Second World renewable energy forum: renewing civilization by renewable energy, Bonn, Germany, 29–31 May 2004.

20.

Rafique

Rehman

National energy scenario of Pakistan – current status, future alternatives, and institutional infrastructure: an overview. Renew Sust Energ Rev 2017; 69: 156–167.

21.

Akhter

Country presentations: promotion of renewable energy, energy efficiency and greenhouses gas abatement (PREGA). In: Second regional training and planning workshop on renew energy, ADB Headquarters, Manila, Philippines, 2004.

22.

Islamic

ADB.

Asian development bank. Republic of Pakistan: Country Environment Analysis, 2008.

23.

Kamran

Current status and future success of renewable energy in Pakistan. Renew Sust Energ Rev 2018; 82: 609–617.

24.

Walekhwa

Mugisha

Drake

Biogas energy from family-sized digesters in Uganda: critical factors and policy implications. Energ Policy 2009; 37: 2754–2762.

25.

Kabir

Yegbemey

Bauer

Factors determinant of biogas adoption in Bangladesh. Renew Sust Energ Rev 2013; 28: 881–889.

26.

Mwirigi

Balana

Mugisha

, et al. Socio-economic hurdles to widespread adoption of small-scale biogas digesters in Sub-Saharan Africa: a review. Biomass Bioenerg 2014; 70: 17–25.

27.

Bluemling

Which factors are effective for farmers’ biogas use? – Evidence from a large-scale survey in China. Energ Policy 2013; 63: 26–33.

28.

Mengistu

Simane

Eshete

, et al. Factors affecting households' decisions in biogas technology adoption, the case of Ofla and Mecha Districts, northern Ethiopia. Renew Energ 2016; 93: 215–227.

29.

Bensah

Mensah

Antwi

Status and prospects for household biogas plants in Ghana–lessons, barriers, potential, and way forward. Int J Energy Environ 2011; 2: 887–898.

30.

Berhe

Hoag

Tesfay

, et al. Factors influencing the adoption of biogas digesters in rural Ethiopia. Energy Sustain Soc 2017; 7: 1–11.

31.

Wahyudi

The determinant factors of biogas technology adoption in cattle farming: evidences from Pati, Indonesia. Int J Renew Energ Dev 2017; 6: 235–240.

32.

Bekele

Drake

Soil and water conservation decision behaviour of subsistence farmers in the Eastern Highlands of Ethiopia: a case study of the Hunde-Lafto area. Ecol Econ 2003; 46: 437–451.

33.

Peduzzi

Concato

Kemper

, et al. A simulation study of the number of events per variable in logistic regression analysis. J Clin Epidemiol 1996; 49: 1373–1379.

34.

Afazeli

Jafari

Rafiee

, et al. An investigation of biogas production potential from livestock and slaughter house wastes. Renew Sust Energ Rev 2014; 34: 380–386.

35.

Kauzeni

Masao

Sawe

, et al. Bio energy options: multidisciplinary participatory method for assessing bio energy options for rural villages in Tanzania. Stockholm Environment Institute, 1998.

36.

Wawa

AI.

The challenges of promoting and adopting biogas technology as alternative energy source in semi-arid areas of Tanzania: the case of Kongwa and Bahi districts of Dodoma region. PhD Thesis, the Open University of Tanzania, 2012.

37.

Sathish

Vivekanandan

Parametric optimization for floating drum anaerobic bio-digester using response surface methodology and Artificial Neural Network. Alexandria Eng J 2016; 55: 3297–3307.

38.

Burton

Rigby Dand Young

Analysis of the determinants of adoption of organic horticultural techniques in the UK. J Agric Econ 1999; 50: 47–63.

39.

Greene

WH.

Econometric analysis. 3rd ed. Upper Saddle River, NJ: Prentice Hall, 1997.

40.

Garson

GD.

Logistic regression: stat notes from North Carolina State University, www2.chass.ncsu.edu/garson/PA765/logistic.htm (2012, accessed 12 October 2015).

41.

Kelebe

Ayimut

Berhe

, et al. Determinants for adoption decision of small scale biogas technology by rural households in Tigray, Ethiopia. Energy Econ 2017; 66: 272–278.

42.

Abbas

Ali

Adil

, et al. Economic analysis of biogas adoption technology by rural farmers: the case of Faisalabad district in Pakistan. Renew Energ 2017; 107: 431–439.

43.

Eshete

Sonder

Heegde

The feasibility study of a national programme for domestic biogas in Ethiopia. Report for SNV/Ethiopia, 2006.

44.

Adeoti

Ilori

Oyebisi

, et al. Engineering design and economic evaluation of a family-sized biogas project in Nigeria. Technovation 2000; 20: 103–108.

45.

Boers

Workneh

Esthete

National biogas programme Ethiopia: programme implementation document. Ethiopia Rural Energy Development and Promotion Centre, Addis Ababa, Ethiopia, 2008.

46.

Gautam

Baral

Herat

Biogas as a sustainable energy source in Nepal: present status and future challenges. Renew Sust Energ Rev 2009; 13: 248–252.

47.

Tigabu

Berkhout

van Beukering

Technology innovation systems and technology diffusion: adoption of bio-digestion in an emerging innovation system in Rwanda. Technol Forecast Soc Change 2015; 90: 318–330.

48.

Juárez-Hernández

Castro-González

Assessing the impact of biogas on the energy sustainability of an urban restaurant in Mexico. Ingen Invest Tecnol 2016; 17: 61–71.

49.

Surendra

Takara

Hashimoto

, et al. Biogas as a sustainable energy source for developing countries: opportunities and challenges. Renew Sust Energ Rev 2014; 31: 846–859.

50.

Momanyi

Benards

AH.

Social-economic factors influencing biogas technology adoption among households in Kilifi county-Kenya. Environments 2016; 6: 20–33.

51.

Nyns

EJ.

New concept for the evaluation of rural biogas management in developing countries. Energ Convers Manage 1996; 37: 1525–1534.

52.

Singh

Sooch

SS.

Comparative study of economics of different models of family size biogas plants for state of Punjab, India. Energ Convers Manage 2004; 45: 1329–1341.

53.

Christiaensen

Heltberg

Greening China’s rural energy: new insights on the potential of smallholder biogas. Environ Dev Econ 2014; 19: 8–29.

54.

Kambele

Factors affecting the adoption of biogas technology as an alternative source of energy in rural Tanzania: a case study of Arumeru district, Arusha Region. PhD Thesis, University of Dar Es Salaam, Tanzania, 2003.

55.

Tarus

RC.

Attitudes influencing adoption of biogas fuel among workers and learners in selected Christian based training institutions in Nandi County , Kenya. Master Thesis, Kenyatta University, Kenya, 2016.

56.

Putra

Liu

Lund

The impact of biogas technology adoption for farm households–Empirical evidence from mixed crop and livestock farming systems in Indonesia. Renew Sust Energ Rev 2017; 74: 1371–1378.

57.

Ghimire

PC.

SNV supported domestic biogas programmes in Asia and Africa. Renew Energ 2013; 49: 90–94.

58.

Gwavuya

Abele

Barfuss

, et al. Household energy economics in rural Ethiopia: a cost-benefit analysis of biogas energy. Renew Energ 2012; 48: 202–209.