Energy Policy 63 (2013) 26–33

Contents lists available at ScienceDirect

Energy Policy
journal homepage: www.elsevier.com/locate/enpol

Which factors are effective for farmers’ biogas use?–Evidence

from a large-scale survey in China
Wei Qu a,e, Qin Tu b,d,n, Bettina Bluemling c
a
School of Economics and Business Administration, Beijing Normal University, Beijing, China
b
Emerging Markets Institute, Beijing Normal University, Beijing, China
c
Environment Policy Group, Wageningen University, LX Wageningen, The Netherlands
d
Institute of World Economics and Politics, Chinese Academy of Social Sciences, Beijing 100732, China
e
Institute of Rural Development, Gansu Academy of Social Sciences, Lanzhou, Gansu, China

H I G H L I G H T S

A binary Probit Model is used to test important factors of household biogas implementation in rural China.

The analysis is based on a survey of 1227 households from four provinces.

The promotion of government has very big and significant effects on household′s decision-making.

The agro-climatic conditions reveal to be decisive.

Many household characteristics including the subjective discount rate are significant factors.

ar t ic l e i nf o a b s t r a c t

Article history: The Chinese government since 1995 has carried out programs for the construction of household biogas
Received 15 May 2013 digesters in the Chinese countryside. Despite the large governmental spending in the building of
Accepted 6 July 2013 household biogas digesters, only 12.16% of the households suitable to produce and use biogas, have built
Available online 30 July 2013
a digester (Li, 2009). This article asks which factors on the household level may be important for the
Keywords: decision whether or not to build a biogas digester. Based on a survey with 1227 households from
Biogas Guangxi, Hubei, Shandong and Gansu provinces, results of a binary Probit Model show that the
Rural China governmental promotion of biogas has a significant effect on households’ decision. The question arises
Household behavior which households may, if governmental programs were to be running out, be most likely to construct a
household biogas digester? The household head′s age, the number of household members staying at
home, the total household income and the subjective discount rate of the respondents are significant
factors in a farm household′s decision to build a biogas digester. However, also agro-climatic conditions
reveal to be decisive, which is why technical solutions for tackling the low productivity of biogas
digesters in cold regions may need to be further considered.
& 2013 Elsevier Ltd. All rights reserved.

1. Introduction government increased considerably (Hao, 2011). During the Ninth

In China, programs for renewable energy such as biogas have a
rather long history, going back to as early as the 1960s. At that
time rural renewable energy programs were generally not moti-
vated by a shortage in energy supply or by environmental reasons,
more often they were part of a political program (Ma et al., 2010).
Since 1995, as a result of a series of guiding documents aiming at
the promotion of rural biogas, capital input from the central
n
Corresponding author at: Emerging Markets Institute, Beijing Normal Univer-
sity, Beijing, China. Tel.: +0086 10 588 042 58.
E-mail addresses: tuqin@cass.org.cn, tuqin@bnu.edu.cn (Q. Tu).
(1996–2000), Tenth (2001–2005) and Eleventh (2006–2010) Five-
year Plan periods, 55.4 million Yuan, 3.5 billion Yuan and 21.2
billion Yuan were respectively invested in rural biogas develop-
ment. Since the end of 2008, biogas received a wider role than just
for energy provision. With its incorporation in a governmental
program to counter the effects of the international financial crisis,
biogas was made a further means of expanding domestic demand
and revitalizing the rural economy. By the end of 2009, 35 million
household biogas digesters have been built in the frame of 56,500
biogas projects (Hao, 2011).
Despite the large governmental spending in building biogas
infrastructure, in 2006, only 0.2% of total primary energy con-
sumption in China was covered by biogas (Zheng et al., 2006).

0301-4215/$ - see front matter & 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.enpol.2013.07.019
 W. Qu et al. / Energy Policy 63 (2013) 26–33 27

Furthermore, only 12.16% of the households suitable to produce
and use household biogas have adopted the technology (Li, 2009).
According to Chen and colleagues, only about 20% of the biogas
potential is utilized in rural China (Chen et al., 2010). After such a
long time of governmental investment in biogas development, it is
hence timely to ask whether the large amount of monetary input
could reach its objective to provide momentum to the develop-
ment of biogas in the rural areas of China.
We can assume that the government only reached out to
certain households within the Chinese countryside. In this article,
the characteristics of these households will be investigated, as well
as what may have motivated them to build biogas digesters. From
these results, we can gain insight into which factors will in future
be important for the further diffusion of household biogas diges-
ters in rural China.
In the analysis of factors, we distinguish four categories, i.e.
agro-climatic conditions, socio-economic conditions of the house-
hold, some measures of the interviewees’ foresightedness, as well
as the influence of the government on households’ decision to
build a biogas digester. Each of these factors requires different
conditions for the further diffusion of biogas. For example, if our
research shows a large influence of agro-climatic conditions on the
likelihood to build a biogas digester, this would mean that biogas
may foremost further diffuse in certain regions of China.
If, however, the government proves to be influential for the
decision to build a biogas digester, the further diffusion of biogas
digesters may fall short as soon as the government ends the biogas
subsidy programs. By inquiring which factors on the household
level are important for the building of biogas digesters, we can
hence provide different policy recommendations and outlooks for
the future success of biogas development in rural China.
This article focuses on household biogas digesters that are
usually of the size of 8–10 m3 and where feedstock is provided
foremost by animal manure, and partly by agricultural residues.
These systems have been extensively described by literature (Chen
et al., 2010; Feng et al., 2009), which is why they will not be
further treated in this study.
In the next section, further reasoning is provided for the
selected parameters. In the subsequent section, the survey area
and methodology are presented. Results in Section 4 show that the
influence of the government on the decision whether to build a
biogas digester should indeed not be underestimated. We discuss
these results in Section 5 in the light of the other factors’ influence.
In the end, we conclude with some recommendations on how to
further develop the momentum of biogas use in rural China.
2. Household variables that may influence the building of
biogas digesters
From literature, four different groups of conditions can be
identified which are assumed to be influential for the decision
whether or not to build a household biogas digester (Li, 2009; Fan
et al., 2011; Groenendaal and Wang, 2010; Sidh and Basu, 2011): the
agro-climatic conditions, such as temperature, humidity and pre-
cipitation; the general socio-economic conditions of the household,
such as income, age-structure, education; related to these, the
foresightedness of those who are at home, i.e. the interviewees;
and the influence of the Chinese government as the promoter of
biogas.
An overview of the variables is given in Table 1. The variables in
the last column are the names of the independent variables which
are assumed to influence a household′s decision of whether to
build biogas infrastructure. In the following, these parameters are
further explained.
2.1. Agro-climatic conditions
For biogas, agro-climatic conditions are crucial determinants
for biogas yield, which is why they here are assumed to also play
an important role in farm households’ decision whether to build a
biogas plant next to their house. While Gansu Province in the
north-west of China, one of our survey locations, is a semi-arid to
arid area with continental climate, including warm to hot sum-
mers and (very) cold winters, Guangxi Province in the south of
China is located in a subtropical climate with hot and long
summers and high annual rainfall. Hubei Province, a further
survey site, also lies in a subtropical climate, however, the survey
site is located in the mountainous areas with cooler summers and
colder winters than in Guangxi Province. Shandong Province, the
fourth survey site, is located in the east coast, has a temperate
climate, in between a humid continental and humid subtropical
climate, with winter temperatures below zero degree. These
conditions are assumed to affect the general tendency of farm
households to invest in biogas digesters. While a considerable
amount of studies have been carried out on the implementation of
household biogas digesters within a particular region (e.g. Zheng
et al., 2006; Cheng et al., 2011; Tu et al., 2011), to the knowledge of
the authors, no study has so far compared the effect of agro-
climatic conditions on the decision whether to construct biogas
infrastructure, in comparison to other, e.g. socio-economic factors.

Table 1
The variables used in the analysis.

Variable categories Explanation of independent variables Variable names

Agro-climatic conditions Province dummy Provn

Socio-economic conditions Total income of household (Yuan) Inc
Total area of agricultural land of household (mu) Land
Percentage off-farm income in total income (%) Offinc
Economic gains Gains
Number of members staying at home Stay
Is there a skilled person in the family? (1 ¼ yes, 0¼ no) Skilled
Gender of household head (0 ¼ female 1¼male) Sex
Years of education of the household head Edu
Age of household head Age
Foresightedness Risk aversion RiskA
Interest in new technology (1-10 scale: 1 ¼ dislike 10¼ like a lot) Newtech
Discount rate Discountrate
Governmental influence Knowledge source for biogas? (1 ¼from government, 0¼ others) Knowsource
Knowledge of the government′s aim to promote biogas (1 ¼yes 0¼ no) Knowaim
Household head the leader of village (1 ¼yes, 0 ¼no) Cadre

1 mu = 1
15 ha.
n
Prov: We have four provinces in total, so we can use three province dummy variables, Prov1, Prov2, Prov3.
28 W. Qu et al. / Energy Policy 63 (2013) 26–33
2.2. Socio-economic conditions
Socio-economic conditions have been pinpointed as crucial for
biogas development (Li, 2009; Fan et al., 2011; Groenendaal and
Wang, 2010). Li (2009) found that villagers’ low income made it
difficult for them to invest in the building of biogas infrastructure.
In this case study, most biogas users belonged to upper- and
medium-income groups. Even if government subsidies covered
one third of the total investment, only these upper and middle
income groups can afford biogas infrastructure construction. Also
Groenendaal and Wang (2010) found that the villages with biogas
digesters are comparatively rich. Access to commercial loans in
rural China is very limited (Cheng et al., 2011), which is why we
may expect that, since the government only subsidizes part of the
investment, low income households may have neither the capital,
nor access to loans for investing in biogas infrastructure. We will
hence examine in how far the total income of a household, as well
as its total agricultural area (as a proxy for agricultural capital) will
be predictive for a household to build a biogas digester.
Apart from the income situation, the expected benefits from
building biogas infrastructure may be of importance for the
decision whether to build biogas infrastructure. These expected
benefits are, e.g. saving on cooking material and energy costs, as
well as receiving governmental subsidies for construction.
We constructed a dummy variable for considering these expected
“economic gains”. If the respondents answered that they had
considered those gains when they made the decision, they are
assumed to more likely build a biogas digester.
A further factor that will influence biogas infrastructure con-
struction is labor shortage. A large number of especially young and
educated rural inhabitants leave for the cities to seek employment
that will often yield much higher incomes than agriculture. If only
a few family members remain in the household, building a biogas
digester may not pay off. This is why the number of household
members staying at home will be considered in the analysis. We
assume that the more members stay at home, the more likely a
biogas digester will be constructed.
Due to migration, agriculture has in some cases lost its
importance as a major source of income. The role of agriculture
for the household may hence be predictive for biogas infrastruc-
ture construction. We assume that a household with larger farm-
land or a bigger share of farming income in total income may have
a higher likelihood to invest in a biogas digester, because the fuel
for biogas in rural China consists mainly of pig manure and other
agricultural remainders, which are provided by agricultural activ-
ities. We take both variables of the percentage of income coming
from off-farm income in total household income and total area of
household farmland as independent variables.
We further may assume that the gender of the household head
will play a role in the decision whether to build a biogas digester.
Since women are in general responsible for cooking, they will
benefit more from biogas since air quality in the kitchen will
significantly increase in comparison to the use of firewood or coal.
Furthermore, by using biogas, women will save time since they do
not have to collect firewood and stalks anymore (Sidh and Basu,
2011). If a household is headed by a woman, we hence assume a
greater likelihood to build a biogas digester.
Among common socio-economic factors for investment deci-
sions in new infrastructure or technology, the age of the house-
hold head and his or her years of education are often used.
However, Li (2009) found that education has not proven to be
important for the building of biogas infrastructure. This is the
reason why we will, apart from education, also consider another
factor expressing the general experience of the household head to
deal with other than agricultural activities. If a household head has
experiences beyond agriculture, we assume him or her to also be
supportive to increasing the household′s living standard by
investing in biogas.
2.3. Foresightedness
This category of parameters comprises some characteristics of
the interviewee in regard to his or her degree of risk aversion,
interest in new technology and time preference (discount rate).
It should show whether biogas distribution depends on innova-
tive, less risk-averse and forward looking households. Three
variables are employed (see also Table 1).
Interest in new technology is assumed to be supportive for the
decision of whether to build a biogas digester as it reduces
potential barriers towards technology application. Furthermore,
attitude towards risk (risk aversion) is assumed to affect the
likelihood to invest in the biogas infrastructure. It was measured
by asking interviewees’ decision for a simple portfolio: there are
two hypothetical choices, one is a riskless choice, i.e. an inter-
viewee could decide to get 100 Yuan1 for sure without any risk;
the other choice bears risk, because there is a 50% chance that the
interviewee may get nothing, and the other 50% chance that the
interviewee gets more than 100 Yuan (the amount varied between
150 and 2000). Without loss of generality, we can assume a simple
form of utility function—constant relative risk aversion (CRRA),
therefore, we can easily compute the coefficient (Ingersoll, 1987).
The general attitude towards long-term investment was inves-
tigated by using a subjective discount rate. A question in our
survey was: “Suppose that you have the choice between 1000
Yuan cash bonus today and a different, higher cash bonus next
year. Given your present situation, what is the smallest additional
amount X you would take for waiting a year?” The subjective
discount rate is computed as X/(1000+X). A high discount rate will
make it less likely that a household will invest in biogas digesters,
as a household is impatient and prefers short term benefits, but
the benefits of biogas mainly come in the long-term.
2.4. Governmental promotion
The government will play an important role in the construction
of biogas infrastructure in China. This may not only relate to
subsidies, but furthermore to the way how biogas is promoted.
The local state, i.e. the governmental levels below and including
the county level, has been crucial for rural economic development
in China (Edin, 2003), and as mentioned in the introduction,
biogas has recently been seen as a measure to stimulate the rural
economic and social development. However, if households were
motivated to construct biogas infrastructure foremost because
they follow a governmental initiative, then the further diffusion
of the technology is likely to fail as soon as the government ends
the program.
We hence included in our survey, three questions in regard to
this role of the government. On the one hand, we asked from
whom the household got to know about biogas. If there is a close
relation between the government as a source of information and
the building of a biogas digester, this will show that the future of
biogas may be uncertain when governmental programs stop. On
the other hand, we assume that the knowledge of governmental
goals with the distribution of biogas will influence the decision of
farm household to build a biogas digester. The government has
been widely promoting its multiple goals with the distribution of
biogas, e.g. improving the sanitation situation both in households
and the village, supporting rural development, ameliorating clean
energy supply to households, or protecting the forest and
1
1 US $≈6.3 Yuan.
 W. Qu et al. / Energy Policy 63 (2013) 26–33 29

Fig. 1. Location of the survey provinces.

improving the ecological environment of rural communities (see e. Table 2

g. Gosens et al. 2013). We can expect that if households know Percentage of survey households with or without biogas.

about these multiple benefits, they may be more likely build

Province %HH built %HH not built
biogas infrastructure. A further variable that was taken into
consideration was whether within a household, a member may Guangxi 77.9 21.1
be a village leader. Since the distribution of biogas infrastructure Hubei 85.4 14.6
Shandong 80.4 19.6
falls under governmental programs, the presence of a village
Gansu 68.5 31.5
leader in the household was assumed to have an effect on
increasing the likelihood of a household to build the biogas Total 78.1 21.9

infrastructure.

100
3. Study area and method
90
80
3.1. Study area 70
60
In order to get a national representative sample of rural 50
household biogas use, we did a household survey in four pro- 40
vinces, including Guangxi Zhuang Autonomous Region (at the 30
same administrative level as a province in China), and Hubei, 20
Shandong and Gansu Provinces (see Fig. 1). The reasons to choose 10
these four provinces are as follows: 0
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
(1) As mentioned under Section 2.1., agro-climatic conditions
are very important determinants for biogas use. Therefore, pro- Total Guangxi Hubei Gansu Shandong
vinces with different agro-climatic conditions were selected, from Fig. 2. Accumulated built biogas digester (%) in each year from 1996 to 2010.
the south of China to the north/northwest, with both decreasing Data source: Our household survey.
temperature and precipitation. Guangxi Zhuang Autonomous
Region is located in the very south of China. It has a subtropical
climate with warm and short winters, hot and long summers and with continental climate, including warm to hot summers and
high annual rainfall. Biogas digesters can be easily used for a (very) cold winters. Biogas digesters work well during the sum-
whole year without any provisions for keeping the digester warm, mers in Shandong and Gansu Province, but do almost not work in
although the biogas productivity of digesters in the winter is about the winters if there are no provisions to keep the digester warm.
half compared to the summer. Hubei Province is located in central (2) As outlined in Section 2.2, the economic situation is also one
and southern China, it also has a subtropical climate, with similar of the most important determinants for biogas use. As a conse-
temperatures in summers (hot), but slightly longer and colder quence, provinces with different economic conditions in their
winters compared to Guangxi. Shandong and Gansu Province are rural areas would have to be selected. Shandong is one of the
further north of Hubei, but Shandong is located in the east coast richest provinces in China. According to the China Statistical
area, with a climate in between humid continental and humid Yearbook 2010, published by the National Bureau of Statistics of
subtropical, and has a moderate temperate climate and annual China, Shandong Province is ranked no. 8 in terms of per capita
rainfall, with winter temperatures below zero degree. Gansu annual net income of rural households in 2009, among the 31
Province in the north-west of China is a semi-arid to arid area provincial level areas in China, while Gansu Province is the poorest
30 W. Qu et al. / Energy Policy 63 (2013) 26–33
Table 3
Household background.
Household (HH) average
No. HH members No. staying at homen Years of schooling
Guangxi 5.3 3.2 7.6
Hubei 4.2 2.7 6.9
Shandong 3.5 2.6 7.6
Gansu 4.6 3.4 6.1
Average 4.3 3.0 7.0
n
Number of members staying at home means the members who stayed at home more than 6 months in 2009.
one. Hubei and Guangxi are in the middle range, ranked no. 14 and
24, respectively.
The sample comprises 1227 households within 32 villages in
8 counties of the 4 above mentioned provinces. Surveys were
carried out from May to August 2010, in two counties in each
province. The government implemented the biogas program only
in those villages where at least more than a certain amount of
households (for example, more than 30) want to build a household
biogas digester. According to local officials we interviewed, around
30% to 60% of all villages in each county participated in the biogas
program. We randomly chose 4–5 villages from those who
participated in the biogas program in each county, and 30–40
households were randomly selected in each village. Accordingly,
each county has a sample size of about 150 households. As survey
villages participated in the biogas program, they are to a limited
extent representative for the average Chinese village when it
comes to the biogas use rate (see Table 2).
As we can see clearly from Fig. 2, the number of built biogas
digesters increased rapidly from 2000 to 2010. Our data shows:
rural areas in south China with more rainfall and warm weather,
like Guangxi and Hubei, which have more suitable agro-climatic
conditions for biogas use, started to build biogas on a large scale
since 2000, earlier than provinces in other parts of the country.
After having good experiences with the first pilot projects and new
models of biogas systems for cold weather conditions, biogas
digesters could be built under either a small greenhouse or a
warm livestock shed, to keep the digester warm enough and to
keep it still working during the cold winter in north or north-west
China. Rural areas in north China, like Gansu and Shandong,
started to build biogas on a big scale since 2005, almost 5 years
after those provinces in the south of China. Our data is consistent
with the findings of other studies (Hao, 2011; Cheng et al., 2011) in
terms of the fast development periods of rural biogas digesters.
3.2. Descriptive statistics
Our data is not a real random sample of all rural households,
because the government implemented the biogas program only in
those villages where at least more than a certain amount of
households wants to build a household biogas digester. In order
to study which factors affect households’ decision-making when
they have the chance to build a biogas digester, we can only
choose from villages which implemented a biogas program. Not
surprisingly, 78.1% of the households in our survey sample have
built biogas digesters. This construction rate differs by far from the
figure of Li (2009) who claims that only 12.6% of the households
suitable to produce and use biogas have adopted it. However, even
if the sample may not be representative in its distribution of
adopters and non-adopters, it will help us to gain insight into the
motivation of those having built biogas digesters.
In regard to other criteria, our sample is fairly representative.
As we can see from Table 3, in terms of income, we find that the
average number of members in a farm household is 4.3, with 4383
Head of HH average
Age Income per person in 2009 (yuan) % of male Years of schooling Age
32.8 4522 92.5 7.8 47.5
36.6 4264 88.9 7.4 46.7
36.6 5891 87.1 7.7 48.2
35.4 2984 94.4 6.4 48.4
35.1 4383 90.0 7.3 47.3
Yuan annual income per person in 2009, which is lower than 5153
Yuan, the national average income per capita for rural households
in 20092. The per capita income in our sample with 5891 RMB is
highest in Shandong Province; in Gansu Province, per capita
income is lowest with only 2984 RMB. The average number of
household members is highest in the province Guangxi with
5.3 persons in a household; with 3.5 persons, Shandong province
has the smallest farm households in the sample; Hubei and Gansu
are in the middle, with 4.2 and 4.6 persons, respectively.
In our sample, on average, 3.0 members are staying at home in
each household, which means 70% of the household members
were not migrating. This number changes in different areas, with
the lowest rate (60%) in Guangxi, the highest (74%) in Shandong
and Gansu. These numbers again can be considered a rather low
percentage of household members migrating to the cities in
comparison to national trends. The average years of schooling in
the whole sample is 7.0 years, of which the average years of
schooling is highest in Guangxi and Shandong with 7.6 years, and
lowest in Gansu with 6.1 years.
Among the interviewed households, 90% were headed by man.
The average age of the household head is 47 years with an average
of 7.3 years of schooling.
In conclusion, the descriptive data shows that the sample, in
comparison to the national average, has a lower average income
level, and is composed of households with a low rate of out-
migration, and with senior, foremost male household heads.
For the analysis of these data, we used a binary Probit model to
discuss the correlation between the independent variables of
Table 1 and the decision of whether or not to build a biogas
digester.
4. Results
Results of the binary Probit Model are shown in Table 4. We can
see that indeed the agro-climatic conditions (dummy variables of
two provinces, Guangxi and Hubei, in the south, highly significant
at 1% level) have the highest predictability for whether a biogas
digester is built or not. The comparison across provinces shows
that farm households in Guangxi and Hubei are much more likely
to build biogas digesters when compared to Gansu. The probabil-
ities are 18.6% and 19.7% higher than for Gansu. Guangxi and
Hubei are located in the south of China where the annual average
temperature and the amount of rainfall is high. Their natural
environments hence are more suitable for the use of biogas.
However, the natural conditions do not seem to be as significant
for all provinces. For Shandong Province, which is in a slightly
more favorable location than Gansu in terms of biogas
2
China Statistical Yearbook 2010, the per capita income of rural households in
2009 are 2980, 6119, 5035 and 3980 Yuan in Gansu, Shandong, Hubei and Guangxi,
respectively.
 W. Qu et al. / Energy Policy 63 (2013) 26–33 31

Table 4 When households decide whether to build a biogas digester,

Results of the Probit model Dependent variable: built biogas digester or not? the expected benefits prove to be important: 15.2% of the farm
0¼ no, 1¼ yes.
households consider whether, for the construction of a digester,
Variable Parameter Std. Err. Z p4z Marginal effects they may get governmental subsidies; 23.6% of the households
take into consideration whether, by using biogas, they can reduce
Sex 0.0438 0.180 0.24 0.808 0.0114 energy costs. Our results show that the dummy is significant (at 5%
Agennn 0.103 0.0330 3.13 0.002nnn 0.0264 level). Economic gains can stimulate the building of biogas
Age2nnn  0.00114 0.000339  3.36 0.001nnn  0.000291
Edu 0.0201 0.0168 1.20 0.230 0.00513
digesters for about 8.9%. As a consequence, more rural households
Skilled  0.0220 0.114  0.19 0.848  0.00559 may start to use biogas if the government increases subsidies or
Staynnn 0.128 0.0419 3.07 0.002nnn 0.0327 energy costs increase.
Cadre  0.0622 0.281  0.22 0.825  0.0163 Also the number of household members staying at home has a
Logincnn  0.0302 0.0135  2.24 0.025nn  0.00770
significant effect (at 1% level). More household members at home
Loginc2nn 0.00617 0.00264 2.33 0.020nn 0.00157
Land 0.00259 0.00995 0.26 0.795 0.000660 means higher energy costs for cooking and more available labor at
Offinc 0.0339 0.0275 1.23 0.219 0.00863 home who can take care of the biogas digester. Therefore a larger
Gainsnn 0.361 0.146 2.48 0.013nn 0.0888 family is more likely to use biogas. Each additional member
RiskA 0.129 0.0879 1.46 0.144 0.0328 staying at home will increase the probability of building a biogas
Newtech  0.00260 0.0155  0.17 0.867  0.000664
Discountratenn  0.667 0.281  2.37 0.018nn  0.170
digester by 3.3%.
Knowsourcennn 0.751 0.130 5.79 0.000nnn 0.229 Other socio-economic factors are not significant. Based on our
Knowaimnnn 0.371 0.133 2.79 0.005nnn 0.0985 results, the gender of the household head does not seem to make a
Guangxinnn 0.955 0.201 4.75 0.000nnn 0.186 difference. In line with another study (Li, 2009), the education
Hubeinnn 1.02 0.201 5.08 0.000nnn 0.197
level of the household head does not play a significant role for the
Shandong 0.246 0.152 1.62 0.106 0.0589
Constantnnn  3.41 0.887  3.85 0.000nnn decision whether to build a biogas digester. Furthermore, other
Log likelihood  379.4 variables like the experience of the household head in the non-
Pseudo R2 0.141 agricultural sector, as well as whether or not to have a cadre as a
Observations 867 household member, do not prove to be predictive.
Standard errors in parentheses, np o0.1.
Except for agro-climatic and socio-economic conditions, vari-
nn
p o0.05.
ables of foresightedness and governmental promotion are also used
nnn
p o 0.01. to predict households′ decision on building a biogas digester.
These variables reveal a mixed picture. Risk aversion and the
interest in new technologies do not play a significant role.
However, the subjective discount rate has a significant negative
implementation, the dummy variable is not significant. Other effect (at 5% level) on the decision of whether to build a biogas
factors will play a role here. digester. Farmers with high discount rates are less likely to build a
Among the socio-economic variables, we find that the house- biogas digester. Biogas is a clean energy mainly for cooking, it
hold head′s age (Age & Age2, both at 1% level), the number of reduces air pollution when replacing coal, and in this way can
household members staying at home (Stay, at 1% level), and the improve health conditions. Due to the reduction in coal or other
logarithm of household total income (Loginc & Loginc2 3, both at fuels, biogas has also long-term economic benefits. Our results
5% level) are significant to predict whether a household would show that the marginal effect is  0.17, which is consistent with
build a biogas digester. classical economic theory, since a biogas digester is a long term
Our results indicate a clear age structure: the variable of the investment, and farmers with low discount rates may be more
household head′s age has an inverse-U shape: both young and old patient and also more likely to make a long term investment—
household heads are less likely to build a biogas digester. The building a biogas digester.
probability of constructing a digester is increasing with the age of The Chinese government seems to be rather influential for farm
the household head, but reaches a peak at an age of 45 years, then, households’ decision whether to construct a biogas digester. If
the probability is decreasing with the further increase of a house- farmers got to know about biogas digesters from the government
hold head′s age. The marginal effect of a household head′s age is (rather than from other resources like friends, family, relatives or
hence not linear, it is changing with age. media (TV, newspapers and magazines)), they are much more
The total household income also has a quadratic form, but a U likely to build biogas infrastructure (variable Knowsource, highly
shape. The lowest point is at 2.45, which means that the total significant at 1% level), the probability is 22.9%, more than 1/5. The
annual income of the family is equal to 11.6 Yuan, which is much effect of the variable Knowsource is even bigger than agro-climatic
lower than the lowest income of our sample. Therefore, in terms of conditions, where the difference in the probability to implement
the total household income, all households in our data are located biogas, between a favorable (like Guangxi or Hubei) and an
at the right side of the lowest point (11.6 Yuan), which means that unfavorable (like Gansu) province, is less than 1/5. Also, when
the probability to build a digester is increasing with the household farm households know governmental objectives behind the pro-
income: rich households are more likely to build a biogas digester. motion of biogas, they are more likely to build biogas infrastruc-
While the total income has a significant effect on the likelihood ture (variable Knowaim, at 1% level); and it makes the probability
to build a biogas digester, our indicators for revenues from of building a digester 9.9% higher.
agriculture do not play a significant role: both the total area of
land as well as the share of off-farm income in total household
income, do not prove to be predictive. Results do not support the 5. Discussion
assumption that households with a higher orientation towards
agriculture would be more likely to invest in biogas. Results of our large-scale survey show that agro-climatic
conditions, the age of the household head, the number of house-
hold members staying at home, total income, expected economic
gains, the interviewees’ subjective discount rate, and government
3
Loginc ¼ ln(inc) Loginc2 ¼ (ln (inc))2. promotion all have significant effects on a household′s decision
32 W. Qu et al. / Energy Policy 63 (2013) 26–33
whether to build a biogas digester. The age distribution curve of
the household head gives a particularly interesting result, i.e. an
inverse U-shape. The result is in line with results from other
studies (Li, 2009). Young household heads are more likely to find
off-farm jobs in cities, therefore, they are less likely to live on crop
cultivation and to raise some pigs, as it is time-consuming and less
profitable. Whereas young household heads will not fear the new
technology introduced with biogas, elder household heads will
find it more difficult to learn how to manage a biogas digester, and
may therefore also be less likely to build a biogas digester. Middle
aged household heads are more likely to do farming at home
because they will, compared to younger people, face more diffi-
culties in finding good jobs in cities, and they are also more likely
to have elderly parents to take care of in their home. They have
both the ability to learn how to manage a biogas digester, as well
as enough motivation to raise pigs, which provide them with fuel
for the digester. This is how we may explain why both young and
elder household heads are less likely to build a biogas digester
than a middle aged household head. As a consequence, there is an
inverse-U shape for age structure.
Also the significant positive impact of the number of members
staying at home on the likelihood of building a biogas digester is
supported by the argument of another study (Tu et al., 2011),
which finds that due to migration, some households are too small
and therefore not likely to build biogas infrastructure. The findings
of the positive effect of higher incomes on the likelihood to build a
biogas digester are also in line with the study of Tu et al. (2011)
which shows that investment costs are still high for some rural
households despite governmental subsidies.
Our results also show that climate conditions, represented by
the four different provinces, are crucial for the further diffusion of
biogas digesters. Guangxi and Hubei Province in the south of China
are the places where we may most expect a further diffusion of
biogas digester construction. There is however also a risk in taking
a Province dummy representing climate conditions. In China,
provinces have a lot of leeway in implementing national programs,
which is why it is not easy to distinguish whether the effects in
our data may also be explained by province level policies.
This point particularly needs to be emphasized since our results
also show that the government does play an important role in
farm households’ decision whether to construct a biogas digester.
The government′s influence is not only expressed in terms of the
role of subsidies (captured in the variable Gains), but also in the
fact that knowing about biogas from the government will make
farm households much more likely to construct a biogas digester
(captured in the variable Knowsource, 22.9%). The momentum of
the government hence may fade away exactly since it has
invested a lot.
Furthermore, since we found that the individual subjective
discount rate also has a significant impact on the likelihood to
build a biogas digester, the future of further biogas development
may become even more uncertain under the conditions of a fast
changing economy. Results for foresightedness need however to
be treated with care as only 65.2% of the interviewees were
household heads. It is hence not clear in how far their risk aversion
and discount rate may have influenced the decision to build a
biogas digester.
A somewhat surprising factor that proved not to have a
significant effect on the decision to build a household digester, is
the agricultural income, and herewith the orientation towards
agriculture. In China, the most popular fuel used for a household
biogas digester is pig manure. Therefore, raising pigs or other
livestock may be much more related to the implementation of a
household biogas digester than the agricultural income, or the
orientation towards agriculture as a whole. Rather, knowing the
multiple goals of the government with the promotion of biogas
(variable Knowaim, see Table 4), and hence the multiple benefits
from biogas, seems to be predictive for implementing biogas.
Biogas seems to be a measure to improve living conditions, not
an agricultural infrastructure. Extension hence needs to point at
the multiple benefits of biogas to further support the momentum
of biogas digester construction.
All in all, results hence support a rather skeptical view on the
future of household biogas construction after the ending of
governmental programs. However, based on our insights, some
recommendations can be provided.
6. Conclusions
The Chinese government for a long time has promoted the
construction of household biogas digesters in the Chinese country-
side. By the end of 2009, 35 million household biogas digesters
have been constructed in the frame of national programmes (Hao,
2011). This article asks which factors on the household level may
be important for the decision whether or not to build a biogas
digester. It finds the governmental promotion of biogas to be
having a significant effect on households’ decision. The question
arises which households may, if governmental programmes were
to be running out, be most likely to construct a household biogas
digester?
We find that households where a large share of household
members stay at home, with a household head in his middle age
(around 45 years old), with a low subjective discount rate, and
which are not poor but have at least a medium income, are most
likely to build a biogas digester. These would be the households to
best target further distribution of household biogas digesters.
The agricultural orientation of these households (in terms of land
and a high share of agricultural income in the total income) does
not play a significant role. Extension hence should rather point at
the multiple benefits of biogas digesters as a means of improving
livelihood through improved sanitation both at home and the
village, reduced workload of cooking at home, improved indoor air
quality, or improved eco-environmental conditions of local com-
munities (see also Gosens et al. 2013).
On a regional scale, we find that in provinces where climatic
conditions are rather suitable for biogas, i.e. Guangxi and Hubei,
households have a stronger desire to build biogas digesters, while
under the conditions in Gansu and Shandong (low temperatures in
winter, with resulting low productivity of biogas digesters), farm
households’ interest in building biogas digesters is relatively weak.
Therefore, it will be of considerable help for the further promotion
of biogas in the north part of China, to consider technical solutions
for tackling the low productivity of biogas digesters and therewith
potentially reduce barriers to investment.
Acknowledgements
This research was carried out within the research project SURE
(SUstainable Natural REsource Use in Rural China, http://sure.
ernasia.org/), which is funded by the Royal Netherlands Academy
of Arts and Sciences (KNAW), grant 08-PSA-E-02, and the Chinese
Ministry of Science and Technology (MoST), grant 2008DFA90630,
as part of the Programme Strategic Scientific Alliances, as well as
in the frame of the project RenErGo (Renewable Energy Govern-
ance in China and the EU, CO-REACH64-090).
References
Chen, Y., Yang, G.H., Sweeney, S., Feng, Y.Z., 2010. Household biogas use in rural
China: a study of opportunities and constraints. Renewable and Sustainable
Energy Reviews 14 (1), 545–549.
 W. Qu et al. / Energy Policy 63 (2013) 26–33
Cheng, S.K., Li, Z.F., Shih, J., Du, X.L., Xing, J.J., 2011. A field study on acceptability of
4-in-1 biogas systems in Liaoning Province, China. Energy Procedia 5,
1382–1387.
Edin, M., 2003. Local state corporatism and private business. Journal of Peasant
Studies 30 (3–4), 278–295.
Fan, J., Liang, Y.T., Tao, A.J., Sheng, K.R., Ma, H.L., Xu, Y., Wang, C.S., Sun, W., 2011.
Energy policies for sustainable livelihoods and sustainable development of
poor areas in China. Energy Policy 39 (3), 1200–1212.
Feng, T., Cheng, S., Min, Q., Li, W., 2009. Productive use of bioenergy for rural
household in ecological fragile area, Panam County, Tibet in China: the case of
the residential biogas model. Renewable and Sustainable Energy Reviews 13,
2070–2078.
Gosens, J., Lu, Y., He, G., Bluemling, B., Beckers, T.A.M., 2013. Sustainability effects of
household-scale biogas in rural China. Energy Policy 54, 273–287.
Groenendaal, W.van, Wang, G.H., 2010. Microanalysis of the benefits of China′s
family-size bio-digesters. Energy 35 (11), 4457–4466.
Hao, X.R., 2011. Present status and prospect of rural energy development in China.
In: Presentation for the RenErGo Conference 2011.
33
Ingersoll, J., 1987. Theory of Financial Decision Making. Rowman & Littlefield,
Totowa, NJ.
Li, J., 2009. Socioeconomic barriers to biogas development in rural Southwest
China: an ethnographic case study. Human Organization 68 (4), 415–430.
Ma, H.Y., Oxley, L., Gibson, J., Li, W., 2010. A survey of China′s renewable energy
economy. Renewable and Sustainable Energy Reviews 14 (1), 438–445.
Sidh, S.N., Basu, S., 2011. Women′s contribution to household food and economic
security: a study in the Garhwal Himalayas, India. Mountain Research and
Development 31 (2), 102–111.
Tu, W.B., Zhang, L.X., Zhou, Z.R., Liu, X., Fu, Z.T., 2011. The development of renewable
energy in resource-rich region: a case in China. Renewable and Sustainable
Energy Reviews 15 (1), 856–860.
Zheng, Y.H., Li, Z.F., Feng, S.F., Lucas, M., Wu, G.L., Li, Y., Li, C.H., Jiang, G.M., 2006.
Biomass energy utilization in rural areas may contribute to alleviating energy
crisis and global warming: a case study in a typical agro-village of Shandong,
China. Renewable and Sustainable Energy Reviews 14 (9), 3132–3139.