To grow or not to grow? Factors influencing the adoption of and continuation with Jatropha in North East India

Renewable Energy 81 (2015) 627e638

Contents lists available at ScienceDirect

Renewable Energy journal homepage: www.elsevier.com/locate/renene

To grow or not to grow? Factors influencing the adoption of and continuation with Jatropha in North East India Kishor Goswami a, *, Hari K. Choudhury b, 1 a b

Department of Humanities & Social Sciences, Indian Institute of Technology, Kharagpur, India Indian Institute of Information Technology, Guwahati, India

a r t i c l e i n f o

a b s t r a c t

Article history: Received 2 January 2014 Accepted 28 March 2015 Available online

The study examines the personal, physical, economic, institutional, and risk and uncertainty factors that determine the adoption of and continuation with Jatropha plantations in North East India. The study is based on a sample of 144 present-growers, 137 past-growers, and 145 non-growers of Jatropha in the states of Assam and Arunachal Pradesh. Farmer characteristic such as their willingness to take risks and whether they have land that is not in use in agriculture play an important role. Two institutional factors are critical - availability of credit and extension services. If farmers are able to grow Jatropha in land that is really not suitable for agriculture, they are more likely to stick with the plantation. Non-farm labor availability and travel and time related to transportation of labor and seedlings also matter. The study recommends the extension of government credit facilities to the farmers as the opportunity cost of labor and land, the initial low returns, and the approximately 5-year payback period from Jatropha cultivation exert a negative and significant influence on farmers' desire to continue with Jatropha cultivation. The study also recommends that the government consider increasing the market price of Jatropha to ensure sustainability of the industry in the region. © 2015 Elsevier Ltd. All rights reserved.

Keywords: Jatropha Adoption and continuation behavior Past, present and non-growers Biodiesel industry North East India

1. Introduction Biofuel has been gaining in popularity in recent years largely due to global political, economic, and environmental events. The potential of biofuel as a clean energy which is a substitute for fossil fuel, and a stimulus solution for climate change and rural development have motivated many countries to consider the adoption of biofuel seriously [1e4]. Another reason for prioritizing energy generation from renewable sources by many countries is the need to comply with emission reduction commitments as per the provisions of the Kyoto Protocol [5] that has led many countries to amend their domestic laws accordingly. Many countries are therefore moving towards the partial and gradual replacement of fossil fuels with biofuel such as ethanol and biodiesel. The governments of several countries are using a wide variety of measures

* Corresponding author. Department of Humanities & Social Sciences, Indian Institute of Technology Kharagpur, West Bengal 721 302, India. Tel.: þ91 3222 281770 (office), þ91 3222 281771 (residence); fax: þ91 3222 282270, þ91 3222 255303. E-mail addresses: [email protected], [email protected] (K. Goswami), [email protected] (H.K. Choudhury). 1 Indian Institute of Information Technology Guwahati, India. http://dx.doi.org/10.1016/j.renene.2015.03.074 0960-1481/© 2015 Elsevier Ltd. All rights reserved.

such as mandatory blending of biofuel with standard gasoline or diesel and the granting of tax advantages or other industrial promotion measures such as tax exemptions and subsidies, supply and demand stimulation, and formal targets for biofuel usage in order to promote the biodiesel industry [6]. In India, too, the rationale for biofuel development has become stronger arising from the fact that India ranks sixth in the world in terms of total energy demand [7] with the demand for energy growing at a rate of 4.8 percent per annum [8]. In 2013, India's total supply of oil was 982.2 thousand barrels per day which was 1.1 percent of the world supply, while its consumption was 3509 thousand barrels per day which was 3.9 percent of the world consumption [9]. Thus, India's domestic production of oil satisfied only 28 percent of its consumption in 2013. In the meantime, India's oil import expenditure has grown nearly three folds since 2004-05 due to the combined effects of increasing global oil prices and the sustained growth in domestic consumption [10]. The oil import bill has increased from USD 87.1 billion in 2009e10 to USD 167.6 billion in 2013e14 [11]. Moreover, according to the Energy Information Administration (2010), India will become the fourth largest net importer of oil in the world by 2025, behind the United States, China and Japan [12]. This implies that a large sum of foreign currency reserves is out flowing in meeting the oil import bill.

628

K. Goswami, H.K. Choudhury / Renewable Energy 81 (2015) 627e638

Additionally, any instability in the price and supply of crude oil adversely affects the domestic economy as experienced in the 1970s and in the 2007e08 [13]. All of the above make it abundantly clear that energy security is a vital issue for the country as higher economic growth in the country will only further increase the demand for energy in the coming years, making the use of domestically produced biodiesel all the more vital in achieving energy security. 1.1. Statement of the problem The problems relating to the biodiesel industry in India are mostly on the supply side. According to the mandates of India's biofuel policy, it would create a 16.72 million ton demand for biodiesel by 2017 [14]. The Energy and Resources Institute (2005) estimated that, in order to meet the 20 percent blending of biodiesel with petro diesel by 2030, about 38 million hectares of wastelands have to be brought under Jatropha plantation with a potential yield of 5 tons per hectare [7]. This is a major challenge for the National Biofuel Policy, as Jatropha is a new perennial crop for the farmers with considerable risk and uncertainty entailed in its production, cost, price, profitability, and employment generation. There are reports that farmers who had taken up Jatropha in certain regions are abandoning their Jatropha plantations and switching over to some other activities whereas a few others are not taking care of their plantation [15]. Thus, the challenge for the policy makers is to get the rural people involved in Jatropha plantations since the objective as stated in the national biofuel policy of India is to meet the blending target along with the generation of substantial employment opportunities for rural people. Thus, there is a need to study the farmers' adoption behavior with regard to Jatropha plantations. Given this background, attempt is made in the present study to identify and analyze the factors that influence farmers' adoption of and continuation with Jatropha plantations in North East India.

require more open spacing. Manual harvesting, which is applicable to smaller plantations, is considered as best method for sustainable production and generating jobs in developing countries. According to an estimate of Kureel et al. (2007), one person can collect and decorticate 25e30 kg of seed per day [16]. In almost all the tropical and subtropical countries, Jatropha is used as hedging materials to protect crops from grazing animals, as it is not eaten by animals. It is also planted as a means to control soil erosion. It is also possible to use Jatropha hedges against bush-fire and/or soil improvement [21]. In different areas, people use latex of Jatropha plant for gum care [22], whereas Jatropha oil is used for lighting and cooking purposes [17]. However, Jatropha's potential as a petroleum fuel substitute has long been recognized. It was used during the Second World War as a diesel substitute in Madagascar, Benin, and Cape Verde, while its glycerin by-product was valuable for the manufacture of nitro-glycerin [17]. Moreover, Jatropha is also used for soap making purpose. Soap is made up by adding a solution of sodium hydroxide (caustic soda) to Jatropha oil. This simple technology has turned soap making into a viable small-scale rural enterprise appropriate to many rural areas of developing countries [17]. In England, the oil of Jatropha is used in wool spinning whereas in China for manufacturing of non or semi drying alkaloids [16]. The Jatropha industry is in its nascent stage, covering a global area estimated at about 900,000 ha [23]. According to the study of Gexsi (2008), more than 85 percent of Jatropha plantations are in Asian countries such as those in Myanmar, India, China, and Indonesia. Africa accounts for around 12 percent or approximately 120,000 ha, mostly in Madagascar, Zambia, Tanzania, and Mozambique. Latin America has approximately 20,000 ha of Jatropha, mostly in Brazil [23]. The area under Jatropha was projected to grow to 12.8 million hectares by 2015. Indonesia is expected to be the largest producer in Asia with 5.2 million hectares, whereas Ghana and Madagascar together will have the largest area in Africa with 1.1 million hectares, and Brazil is projected to be the largest producer in Latin America with 1.3 million hectares [23].

1.2. Background of Jatropha and its use Jatropha (Jatropha curcas L.), belongs to Euphorbiaceae family, is a fast growing shrub. It is native to South America and has a long history of its propagation by the Portuguese in Africa and Asia [16]. In natural condition the plant may be 3e5 m in height. However, in controlled commercial plantation the plant height is usually maintained within 2 m. The life span of Jatropha is 30e50 years [17,18]. A Jatropha plant can give its best seed yield only from sixth year onwards [19]. Under rainfed condition, the average seed yield of Jatropha from fifth year (excluding the year of establishment) onwards varies from 2.5 tons to 4 tons per hectare depending upon whether the soil is poor or average [19]. The findings of Nahar and Hampton [20] shows that the seeds become mature when the capsule changes from green to yellow after 2e4 months. The seeds contain 21 percent saturated fatty acids and 79 percent unsaturated fatty acids. There are several propagation methods for Jatropha, such as direct seeding, transplanting, planting (cutting), or tissue culture. Direct planting by cutting decreases the time of production as compared to direct seeding or transplanting. Propagation through seed (sexual propagation) leads to genetic variability in terms of growth, biomass, seed yield, and oil content. In India, Jatropha flowers during MarcheApril and SeptembereDecember. The fruiting usually takes place between September to December. Both manual and mechanical harvesting of Jatropha seed is possible. Normally, plant spacing of 8.8  8.8 feet (2.7  2.7 m) is considered more desirable for commercial cultivation [16]. However, spacing depends on the harvest method; mechanical harvest in large-scale commercial plantations may

2. Background of farmers' adoption behavior: theory and empirical models 2.1. Theory The theoretical model for the purpose of testing farmers' adoption behavior with regard to Jatropha is mostly based on the household production theory of Singh et al. [24] and Strauss [25] and the agroforestry adoption model of Mercer and Pattanayak [26]. Similar to agroforestry adoption [26], the choice facing household ‘i’ was considered, when deciding whether to adopt Jatropha. The utility maximizing household compares its expected net utility (EU * )2 with and without adoption. A reduced form of the net utility can be expressed as:

EU * ¼ Xi b þ εi

(1)

where, Xi is the vector of explanatory variables and b is the vector of coefficients to be estimated. Since the true net utility function is not know, the estimated function is considered as random by including the error term εi. A farmer adopts Jatropha, if he/she finds that expected utility (EU * ) of adoption of Jatropha is greater than nonadoption.

2 The derivation of the theory is available at Choudhury HK, Goswami K. Determinants of expansion of area under Jatropha plantation in North East India: A Tobit analysis. Forest Policy and Economics 13 (2013) 46e52.

K. Goswami, H.K. Choudhury / Renewable Energy 81 (2015) 627e638

2.2. Empirical models The adoption of a technology or a new crop or a management practice can be represented by a discrete decision to use or not to use (a variable defined as either one or zero), a proportional indicator (such as a share of the land or share of plants), an index of scale or extent of use (such as the number of hectares or the number of plants), the level of choices or intensity of use (per hectare or per plant), or frequency of use (such as the number of seasons or the number of applications per season). However, the use of the said technology, crop or practice at any one point in time is also not a robust indicator of adoption [27]. Thus, in order to study the farmers' adoption behavior, it is necessary to define the term “adoption”. In this context, the present study considers only those farmers who are continuing with a Jatropha plantation as an adopter of Jatropha. Following Jabbar et al. [28], and Wendland and Sills [29], the farmers' adoption of Jatropha were divided into two stages. In the first stage, whether a farmer adopts Jatropha or not was analyzed. If a farmer adopts Jatropha, whether he/ she would continue or abandon it was analyzed in the second stage. In both the cases, the dependent variable is binary in nature. Those farmers who discontinue their plantations are considered as pastgrowers for the purposes of the study. On the other hand, those farmers who are yet to adopt Jatropha are defined as non-growers. The different studies on adoption behavior suggest that the decision of farmers to adopt Jatropha is influenced by a host of factors [30,31]. In the present study, it is conceptualized that farmers in the first instance will decide whether to continue with Jatropha or abandon it on the basis of their experience after their initial adoption experience. This decision will be influenced again by a combination of economic, personal, institutional, physical, and risk and uncertainty factors. Fig. 1 gives the conceptual framework of farmers' adoption of Jatropha in the form of a flow chart in two sets of equations comprising the adoption of Jatropha and the decision to continue with or abandon the plantation. The literature shows that risk and uncertainty remain among the most under-studied aspects of adoption behavior. Studies are required that directly measure risk preference, and relate them to the adoption decision process [32]. Following Mercer's suggestion, the present study, in measuring risk and uncertainty, analyzes

629

farmers' risk preferences through an experiment and integrates it to the adoption decision process based on the NeumannMorgenstern expected utility concept [33]. In the experiment, the farmer was given the option to choose from two alternatives: (A) either the farmer could have INR 10 now for sure or (B) a coin would be flipped and, if the outcome is heads, he/she would receive INR 30 now and if it is tails, he/she would receive no money. A farmer was categorized as a risk averter if he/she chooses option A and a risk taker if he/she chooses option B. Though a third option (C) was available for those who were indifferent, in the present study, no respondent was found to choose this option. Accordingly, the third option was not considered anymore in the study. The incentive payoffs were kept at a lower level as the literature shows that nearly all individuals, regardless of personal characteristics, are moderately risk averse at higher payoff levels [34]. The use of appropriate quantitative tools is vital in analyzing farmers' adoption behavior. Feder et al. [35] have stressed that those using econometric tools should pay more attention to the quantitative importance of the explanatory variables rather than to their directional impacts. Numerous studies in recent years have used different models such Logit and Probit models, Multinomial Logit model, and Tobit model to analyze the influence of different factors on decision variables, with a few such studies conducted on adoption-related decisions [30,36e38]. The available literature make it evident that farmers' adoption behavior is influenced by many factors and that different researchers have resorted to different proxies to measure the impact of these factors. Moreover, the use of the Logit and Probit models are more frequent in analyzing the farmers' adoption behavior. Farmers' adoption of Jatropha was based on the theoretical model developed by Mercer and Pattanayak [26], where the ith farmer adopts Jatropha (JA* ) if the expected utility/benefit of adoption is greater than that without adoption. Thus the model is specified as: where, JA* is not directly observable. The researcher can observe a farmer's adoption decision of Jatropha such that JA is equal to 1 if

JA* ¼ F0 þ F1 Age þ F2 Age square þ F3 Education þ F4 Primary occupation þ F5 Farming experience þ F6 Cultivable land þ F7 Distance to nearest market þ F8 Availability of unemployed family member þ F9 Labor shortage þ F10 Location þ F11 Rainfallþ F12 Non  farm employment opportunity þ F13 Credit availability þ F14 Extension service þ F15 Risk behavior þ ei

Fig. 1. Conceptual framework related to farmers' adoption and continuation with Jatropha plantation.

(2)

630

K. Goswami, H.K. Choudhury / Renewable Energy 81 (2015) 627e638

the farmer adopts Jatropha, and 0 if he does not, which is observable, i.e.,

 JA ¼

1 if J*A > 0 0 if J*A  0

Whether a farmer continues with Jatropha or abandons it is also specified similar to the model of the farmers who adopt Jatropha. The ith farmer continues with Jatropha (JC* ) if the expected utility/ benefit of continuation is greater than that from its abandonment. Thus, the model is specified as: where, JC* is not directly observable. The researcher can observe a farmer's continuation decision of Jatropha such that JC is equal to 1 if the farmer continues with Jatropha, and 0 if he does not, which is

wasteland of the region in 2008e09. Moreover, underutilized/ degraded forest area (scrub dominated) constituted 8.2 percent (or 3634.35 sq. km) of the total wasteland of the region. These three categories of land can easily be used for Jatropha plantation and may work as a green cover to reduce soil erosion and landslides in the region. The initiative with regard to Jatropha plantation in NE India mostly remains in the hands of private organizations such as D1 Williamson Magor Bio Fuels Ltd. (D1WMBFLtd), and Sun Plant Agro Ltd. D1WMBFLtd, for instance, took a major initiative towards the plantation of Jatropha at the farmers' level in 2007e08, with the total area brought under Jatropha plantation estimated at 39,400 ha and 46,020 ha in 2007 and 2008 respectively (D1WMBFLtd, n.d.).3 Of the various North Eastern states, Assam has the largest amount

JC* ¼ m0 þ m1 Age þ m2 Age square þ m3 Education þ m4 Knowledge about Jatropha þ m5 Availability of unemployed family member þ m6 Labor shortage þ m7 Distance to nearest market þ m8 Slope of land under Jatropha þ m9 Average time to reach plantation site þ m10 Location þ m11 Rainfall þ m12 Non  farm employment opportunityþ m13 Alternative use of the land þ m14 Minimum expected income to continue with Jatropha þ m15 Expected market price þ m16 Expected payback period þ m17 Access to credit þ m18 Technical help in planting and pruning þ m19 Extension service þ m20 Risk behavior þ m21 Ant and pests attack þ yI

observable, i.e.,

 JC ¼

1 if J*C > 0 0 if J*C  0

With this background, the primary objective of the present study is to answer how do personel, physical, economic, institutional, and risk and uncertainty factors influence farmers' decision making with regard to adoption of and continuation with Jatropha plantation. 3. Study area and sampling strategy 3.1. Study area The study is confined to the Indian states of Assam and Arunachal Pradesh. Assam was selected because it is different from the other North Eastern states in terms of altitude and topography. In contrast, Arunachal Pradesh is meant to represent the other North Eastern states such as Manipur, Nagaland, and Tripura as it is similar in topography to these states. The primary data were collected from 22 villages in five districts in Assam, namely, Cachar, Karimganj, Karbi Anglong, North Lakhimpur, and Dhemaji, and 6 villages in the Papumpare district of Arunachal Pradesh (see Map 1). 3.1.1. Jatropha plantation in North East India One of the major concerns of the biofuel policy in India is how to make suitable land available for Jatropha plantation without exerting a negative impact on food production. However, this problem is not acute in North East (NE) India as the total wasteland in NE India is 44,315.06 sq. km in 2008e09, which is about 9.5 percent of the total wasteland of India [39]. Out of this 44,315.06 sq. km wasteland, 45.87 percent (20,325 sq. km) is land with or without scrub. Shifting cultivation, a pernicious form of cultivation resulting in soil erosion and degradation of forest, is practiced in NE India and constituted 17 percent (or 7568.70 sq. km) of the total

(3)

of area under Jatropha plantation (at 33,900 ha) followed by Tripura (at 26,000 ha). However, since 2010e11, farmers have started switching from Jatropha plantation to other activities in many places. The field survey of the present study shows that the majority of the farmers cultivated Jatropha in less than 1 ha of the area, with the average area brought under Jatropha for past-growers and present-growers remaining at 0.94 ha and 1.09 ha respectively. However, the independent sample test shows a significant difference (at the 10 percent level of significance) in the average area brought under Jatropha plantation between the two categories of Jatropha growers, i.e., present-growers and past-growers. A Jatropha plant can give its seed yield only from the sixth year onwards [40]. As the oldest plantations were around 4e5 years old at the time of the survey, only a few farmers reported the economic yield of the Jatropha seed. Almost all the respondents reported that the seedling provided by D1WMBFLtd was used for Jatropha plantation. However, the majority of the growers (both past and present) cited additional sources of income generation as the first reason for opting for Jatropha cultivation followed by financial support from government. 3.2. Sampling strategy The study is based on both primary and secondary data. The primary data were collected from 426 respondents using the multistage random sampling technique. In the first stage, purposive sampling was used based on the fair availability of present-growers, past-growers, and non-growers of Jatropha in the villages. After selection of the sample villages, the respondents were selected randomly. The present-growers, pastgrowers, and non-growers of Jatropha were sampled from the same places to ensure that the respondents faced a similar

3 Information about area under Jatropha plantation in NE is collected from the unpublished documents of D1 Williamson Magor Bio Fuel Ltd., Kolkata, 2010 through personal communication with the Managing Director of the company.

K. Goswami, H.K. Choudhury / Renewable Energy 81 (2015) 627e638

631

Map 1. District map of Arunachal Pradesh and Assam Note: i) The map is modified from www.mapsofindia.com ii) C indicates sample districts.

Environment and Forest (MoEF), Ministry of Development of North Eastern Region (DoNER), and Department of Economics and Statistics of Assam and Arunachal Pradesh. Moreover, relevant secondary data were also collected from companies/organizations directly involved in the industry such as D1 Williamson Magor Bio Fuels Ltd (D1WMBFLtd) and the North Eastern Development Finance Corporation Ltd (NEDFi).

environment. The selection of areas for primary data collection was made at household level based on personal visits, discussions with block development officers (BDOs), and agricultural extension officials. A total of 426 respondents were interviewed, consisting of 144 present-growers, 137 past-growers, and 145 nongrowers of Jatropha from different blocks in the districts through a pre-tested questionnaire during November 2011 to March 2012. A total of 11 focus group discussions (FGDs) were also organized, one in each block, to collect and cross-verify information related to Jatropha. The secondary data were also collected from the North Eastern Council (NEC), Ministry of

Table 1a Descriptive statistics of the factors used in the Probit models. Factors

Unit of measurement

Present-grower

Past-grower

Mean (Std. Dev.) Min. Personal factors Age Education Primary occupation Knowledge about Jatropha Physical factors Cultivable land Availability of unemployed family member Labor shortage Distance to nearest market Slope of land under Jatropha Average time to reach the plantation site Rainfall

Year 41.19 (12.39) Year 7.44 (4.92) 1 for farmer and 0 for otherwise 0.61 (0.49) 1 for yes and 0 for no 0.68 (0.47) Hectare 1 for available and 0 for not available 1 for yes and 0 for no Kilometer Degree Minute 1000 milimeter

3.09 (2.94) 0.42 (0.50) 0.53 12.83 7.78 14.12

(0.50) (9.07) (8.47) (11.26)

2.84 (0.58)

18 0 0 0

Non-grower

Expected sign

Max. Mean (Std. Dev.) Min Max. Mean (Std. Dev.) Min. Max. 74 17 1 1

41.39 (12.06) 7.50 (5.10) 0.50 (0.50) 0.26 (0.44)

18 0 0 0

75 17 1 1

37.46 (11.33) 7.290 (5.03) 0.55 (0.50) 0.25 (0.43)

18 0 0 0

66 18 1 1

Positive Positive Positive Positive

0.6 0

22.67 NA 1 0.23 (0.43)

NA 0

NA 1

1.99 (2.32) 0.37 (0.48)

0 0

16 1

Positive Positive

0 1.5 0 0

1 28 50 60

0 1.5 0 2

1 28 45 90

0.49 (0.50) 12.72 (9.04) NA NA

0 1.5 NA NA

1 28 NA NA

Negative Negative Negative Negative

2.18

0.82 (0.38) 12.33 (9.16) 8.45 (8.71) 26.51 (17.62)

3.51 2.90 (0.58)

2.18 3.51 2.87 (0.58)

2.18 3.51 Negative

632

K. Goswami, H.K. Choudhury / Renewable Energy 81 (2015) 627e638

Table 1b Descriptive statistics of the factors used in the Probit models. Factors

Unit of measurement Present-grower

Past-grower

Non-grower

Mean (Std.Dev.) Min. Max. Mean (Std. Dev.) Min. Economic factors Non-farm employment opportunity Alternative use of land under Jatropha Minimum expected income to continue with Jatropha Expected price of Jatropha seed Expected payback period Institutional factors Availability of/access to credit for Jatropha plantation Technical help in pruning and planting Extension Service Risk and uncertainty factors Risk Pests and disease attack

1 for yes and 0 for no 0.47 (0.501) 1 for yes and 0 for no 0.61 (0.49) Rs. 000 per hectare 14.36 (8.14) Rupees Year

0 0 1

1 0.75 (0.43) 1 0.85 (0.35) 48.2 20.49 (9.84)

Max.

Mean (Std. Dev) Min. Max.

0 1 0.64 (0.48) 0 1 NA 3.75 56.25 NA

Expected sign

0 NA NA

1 NA NA

Negative Negative Negative

20.94 (7.74) 15.68 (6.98)

12 5

50 50

26.69 (9.88) 24.43 (10.66)

10 7

55 60

NA NA

NA NA

NA NA

Negative Negative

1 for yes and 0 for no

0.38 (0.49)

0

1

0.34 (0.47)

0

1

NA

NA

NA

Positive

1 for yes and 0 for no 1 for yes and 0 for no

0.55 (0.50) 0.68 (0.47)

0 0

1 1

0.23 (0.42) 0.60 (0.49)

0 0

1 1

NA 0.60 (0.49)

NA 0

NA 1

Positive Positive

1 for risk taker and 0 for risk averter 1 for yes and 0 for no

0.58 (0.50)

0

1

0.28 (0.45)

0

1

0.43 (0.50)

0

1

Positive

0.51(0.50)

0

1

0.56(0.45)

0

1

NA

NA

NA

Negative

Note: i) Figures in parentheses show standard deviation; ii) NA ¼ not applicable.

4. Results and discussion -vis jJatropha 4.1. Farmers' adoption behavior vis-a Probit models were estimated, with and without village fixed effects, to fulfill our objective of analyzing the influence of personal, physical, economic, institutional, and risk and uncertainty factors on farmers' adoption as well as continuation decisions. Tables 1a and 1b portray the summary statistics of the factors used. With regard to the present-growers and non-growers, significant differences were observed for the variables such as age, primary occupation, cultivable land, availability of unemployed family

member, and availability of credit and extension services (see Table 2). On the other hand, in the case of present-growers and past-growers, significant differences were also observed for the availability of unemployed family member, labor shortage, average time to reach the plantation site, non-farm employment opportunity, alternative use of the land, minimum expected income to continue with Jatropha, expected market price, expected payback period, access to credit, technical help in planting and pruning, extension service, and risk behavior. Tables 3a and 3b presents the regression results of the Probit model and specification test of adoption of Jatropha with and without village fixed effects. The results indicate that cultivable

Table 2 Independent sample mean test and McNemar's chi2 test across categories of farmers. Factors

Personal factors Age Education Primary occupation Knowledge about Jatropha Farming experience Physical factors Cultivable land Availability of unemployed family member Labor shortage Distance to nearest market Slope of Land under Jatropha Average time to reach plantation Site Economic factors Non-farm employment opportunity Alternative use of land under Jatropha Minimum expected income to continue with Jatropha Expected price of Jatropha seed Expected payback period Institutional factors Availability of/access to credit for Jatropha plantation Technical help in pruning and planting Extension service Risk and uncertainty Risk Pests and disease attack

Independent sample mean test significance level

McNemar's chi2 test

Present vs. Non-grower

Present vs. Past-grower

Present vs. Non-grower

Present vs. Past-grower

Prob > chi2

Exact McNemar sig prob.

Prob > chi2

Exact McNemar sig prob.

0.008 0.801 e e 0.292

0.895 0.922 e e e

e e 0.0478 0.270 e

e e 0.058 0.3203 e

e e e 0.269 e

e e e 0.320 e

0.000 e e 0.921 e e

e e e 0.645 0.520 0.000

e 0.010 0.734 e e e

e 0.013 0.799 e e e

e 0.000 0.001 e e e

e 0.000 0.001 e

e e e e e

e e 0.000 0.000 0.000

0.220 e e e e

0.250 e e e e

0.053 0.000 e e e

0.062 0.000 e e e

e e e

e e e

0.000 e 0.000

0.000 e 0.001

0.000 0.001 0.002

0.000 0.001 0.002

e e

e e

0.787 e

0.857 e

0.035 0.622

0.044 0.681

e

K. Goswami, H.K. Choudhury / Renewable Energy 81 (2015) 627e638

633

Table 3a Results of the Probit model of adoption of Jatropha plantation. Without fixed effect (first model)

Factors

Personal factors Age Age square Education Primary occupation Farming experience Physical factors Cultivable land Distance to nearest market Availability of unemployed family member Labor shortage Location Rainfall Economic factors Non-farm employment opportunity Institutional factors Credit availability Extension service Risk and uncertainty factors Risk behavior Constant Number of observations ¼ 289 Wald Chi Square (15) ¼ 54.52 Prob > Chi-Square < 0.001 Pearson Chi-Square (273) ¼ 286.50 Prob > Chi-Square ¼ 0.275 Pseudo R2 ¼ 0.150 Log Pseudo likelihood ¼ -170.355 Area under ROC Curve ¼ 0.744

With fixed effect (second model)

Coefficient

Robust std error

Marginal effects

Coefficient

Robust std error

Marginal effects

0.0512 0.0003 0.0001 0.1563 0.0153

0.0423 0.0005 0.0201 0.2133 0.0096

0.020 0.000 0.000 0.062 0.006

0.045 0.000 0.012 0.345 0.012

0.043 0.001 0.024 0.246 0.010

0.018 0.000 0.005 0.137 0.005

0.1050** 0.0005 0.1062 0.0388 0.8716*** 0.8287**

0.0527 0.0170 0.1696 0.1624 0.3324 0.3321

0.042 0.000 0.042 0.015 0.326 0.331

0.114** 0.038 0.117 0.124 0.270 5.717***

0.057 0.043 0.185 0.181 0.790 0.351

0.046 0.015 0.047 0.049 0.107 2.276

0.6191***

0.2150

0.243

0.817***

0.236

0.317

1.3331*** 0.3743*

0.2928 0.2206

0.486 0.148

2.273*** 1.255**

0.477 0.575

0.723 0.458

0.3684** 1.0544

0.1698 1.3923

0.146 e

0.370** 0.181 16.639 e Number of observations ¼ 289 LR Chi Square (39) ¼ 79.58 Prob > Chi-Square < 0.001 Pearson Chi-Square (249) ¼ 276.86 Prob > Chi-Square < 0.109 Pseudo R2 ¼ 0.199 Log Pseudo Likelihood ¼ 160.526 Area under ROC Curve ¼ 0.777

0.146 e

Note: *** ¼ 1 percent, ** ¼ 5 percent, and * ¼ 10 percent level of significance are considered. dy/dx is for discrete change of dummy variable from 0 to 1.

Table 3b Specification test of the Probit model of adoption of Jatropha plantation. Factors

Without fixed effect (first model) Coefficient

_hat 1.054 _hatsq 0.211 Constant 0.060 No. of observations ¼ 289 LR Chi2(2) ¼ 61.96 Prob > Chi-Square < 0.001 Pseudo R2 ¼ 0.155 Log Pseudo likelihood ¼ 169.339

With fixed effect (second model)

Standard error

z

P>z

Coefficient

0.159 0.145 0.089

6.620 1.450 0.670

0.000 0.146 0.500

1.023 0.148 0.109 0.136 0.035 0.092 No. of observations ¼ 289 LR Chi2(2) ¼ 80.22 Prob > Chi-Square < 0.001 Pseudo R2 ¼ 0.200 Log Pseudo likelihood ¼ 160.210

land, availability of credit, extension service, and risk behavior have a statistically significant and positive influence on adoption in both models (see Tables 3a and 3b). Non-farm employment opportunity and rainfall have a negative influence on the adoption of Jatropha plantations. The estimates of the marginal effects calculated at sample mean indicate that holding other factors constant, a) an increase in the amount of cultivable land by 1 ha increases the probability of adoption by about 4 percent (in both models); b) an increase in rainfall by 1000 milimeter reduces the probability of adoption of Jatropha by 33e228 percent; c) availability of non-farm employment opportunity reduces the probability of adoption by 24e32 percent; d) credit availability for Jatropha increases the likelihood of adoption by 49e72 percent; e) the availability of extension services increases the probability of adoption by 15e46 percent; and f) being a risk lover or risk taker increases the probability of adoption of Jatropha by around 15 percent (see Table 3a). Scarcity of suitable land is one of the major constraints facing biofuel development in developing countries [7,41]. The present

Standard error

z

P>z

6.890 0.800 0.380

0.000 0.422 0.701

study confirms that availability of cultivable land has a positive impact on the adoption of Jatropha, the reason being that farmers with a sizeable amount of cultivable land adopt Jatropha as a means of diversifying their cropping pattern. Among the respondents in the study, the average cultivable land with adopting farmers was about 50 percent more than that among the non-adopting farmers, the variation ranging from 2 ha to 3 ha. Moreover, the adoption of Jatropha varies across states, which is also an indication of the role that availability of land plays in adoption. In comparison to Assam, people of Arunachal Pradesh have more access to cultivable land due to the low population density of the latter state. According to the estimates of Census 2011, Arunachal Pradesh has the lowest density of population, i.e., 17 per sq. km in comparison with other Indian states such as Assam where it is 397 per sq. km [42]. Rainfall has a negative influence on the adoption of Jatropha, which may be due to the fact that higher rainfall causes severe damage at flowering time to Jatropha and may, in turn, lead to both low production and discouragement of farmers. Higher rainfall also causes fungal attack and restricts root growth [17]. The negative

634

K. Goswami, H.K. Choudhury / Renewable Energy 81 (2015) 627e638

influence of non-farm employment opportunity on the adoption of Jatropha is attributable to the fact that Jatropha is a new crop with a 5e6 year gestation period. Coupled with the element of uncertainty emanating from the adoption of a new crop, it is therefore natural for farmers to consider Jatropha plantation as a secondary option with regard to income generation. This leads in turn to a trade-off in the allocation of labor between Jatropha plantation and other primary crops or activities. In the present study 60 percent (or 57 percent in the case of non-growers) respondents had an annual per capita household income of less than INR 10,000, of which 65 percent (or 59 percent in the case of non-growers) were farmers. This fact indicates the importance of non-farm employment to fulfill farmers' livelihood needs. Analyzing the situation in the context of adoption of new crop like switchgrass, Hipple and Duffy (2002) also found that farming activities and practices that create conflicts between on-farm management and off-farm employment discourage adoption of alternatives [43]. The availability of institutional help in the process of adoption also plays a key role in farmers' decisions on adoption. The study shows that the availability of credit for Jatropha plantation and extension services increases the likelihood of adoption of Jatropha with extension services playing an important role in the adoption of Jatropha plantation in North East India. This finding is in the line with the findings of Mujeyi (2009) in the context of decision to adopt commercial utilization of Jatropha [44], Mohamed and Temu (2008) in the context of impact of credit on adoption of agricultural  and Yang (2007) in the context of insurance, technologies [45], Gine credit, and technology adoption in Malawi [46]. Since Jatropha is a new perennial crop for the farmers, awareness about the different uses of Jatropha is very low among the farmers. Therefore, the impact of extension services is positive as predicted. The literature on farmers' adoption behavior shows that risk is an important factor in explaining the adoption of any new crop or innovation or technology, with the likelihood of adoption of new crops or innovation higher for persons who are risk takers. Those who adopt a wait-and-see attitude are however less likely to be interested in growing a new crop as Qualls et al. [47] show for switchgrass in the South Eastern United States. The present study confirms this finding in that the likelihood of adoption of Jatropha plantation is higher among risk takers since there is considerable risk and uncertainty in Jatropha plantation. The goodness-of-fit of the adoption model was examined using the Wald Chi-Square, LR Chi-Square, Pearson Chi-Square, and ROC Curve values (Table 3a). Link test was used to test the model specification presented in Table 3b. In order to avoid the problem of heteroscedasticity, robust standard error was used to test the significance of the coefficients. The results of these tests show that the estimated Probit model of adoption of the Jatropha plantation satisfies the goodness-of-fit criteria. The study however fails to reject the null hypothesis of a few factors, namely, age, education, primary occupation, farming experience, distance to the nearest market, availability of unemployed family member and labor shortage due to lack of sufficient evidence. Thus, the study finds the influence of these factors on the adoption decision to be statistically not significant. 4.2. Farmers' continuation with Jatropha plantation The study also analyzed the factors influencing continuation with Jatropha plantation by farmers in order to throw some light on the post-adoption scenario. Continuation of Jatropha was considered as an activity where farmers continued with same Jatropha plantation for at least 3 years. Knowledge about Jatropha, distance to nearest market, access to credit, and risk behavior have a positive and significant influence on

continuation, whereas labor shortage, average time to reach the plantation site, rainfall, non-farm employment opportunity, alternative uses of the land, and expected payback period have a negative influence in the second model with village fixed effect (see Tables 4a and 4b). Other factors remaining constant, the estimates of significant marginal effects calculated at sample mean show that, in the two models (see Tables 4a and 4b): a. Some knowledge about Jatropha increases the probability of continuation by 30 and 36 percent; b. Labor shortage reduces the likelihood of continuation by 28 percent and 21 percent; c. A one kilometer increase in the distance to nearest market increases the probability of continuation by 36 percent in the second model; d. Every 1000 milimeters increase in the average rainfall reduces the probability of continuation by 148 percent in the second model; e. A 1 min increase in the average time to reach the plantation site reduces the probability of continuation by about 2 percent in both the models; f. The availability of non-farm employment opportunities reduces the probability of continuation by 38 percent in both the models; g. The availability of alternative uses for the land under Jatropha reduces the probability of continuation by 36 percent and 40 percent; h. The probability of continuation decreases by 3 percent and 4 percent with every one year increase in the expected payback period. Since the usual payback period for Jatropha is about 5 years [48], initial financial support for low income growers becomes vital for continuation. i. Access to credit increases the probability of continuation by 29 percent and 44 percent. j. Being a risk taker increases the probability of continuation with Jatropha by 45 percent in the first model and 50 percent in the second model (see Table 4a). The positive influence of knowledge about Jatropha on continuation with Jatropha plantation indicates that lack of knowledge leads to inadequate plantation practices on the part of the growers, which results in inadequate maintenance, poor growth, and higher mortality of plants, with farmers consequently losing interest and abandoning their plantations. Moreover, knowledge enables farmers to anticipate correctly the prospects of and benefits from Jatropha, thus motivating them to adopt Jatropha plantation. Similar to our findings, analyzing the adoption of biofuels technologies by smallholder farmers in Zimbabwe, NyamwenaMukonza (n.d.) also found that perception towards Jatropha play a greater role in its adoption [49]. Maingi (2010) also found that the likelihood of adoption of Jatropha would be higher under farmer's knowledge of the benefits from the plant [50]. The negative impact of the labor shortage problem on continuation with Jatropha arises from the fact that, when faced with labor shortages, farmers prioritize agricultural activities to ensure food security. Further, Jatropha plantations suffer when farmers face labor shortage as this reduces the time spent on proper maintenance of plantation sites. In contrast, increases in the distance to nearest market increases the probability of continuation with Jatropha. This is because access for selling of non-plantation crops on a regular basis is limited. Since Jatropha is not harvested frequently, harvested seeds can be stored for a longer duration unlike many other crops, and requirement of production inputs are limited, growers may prefer to continue with this crop instead of shifting to other crops.

K. Goswami, H.K. Choudhury / Renewable Energy 81 (2015) 627e638

635

Table 4a Results of the Probit model of continuation with Jatropha plantation. Without fixed effect (first model)

Factors

Personal factors Age Age square Education Knowledge about Jatropha Physical factors Availability of unemployed family labor Labor shortage Distance to nearest market Slope of land under Jatropha Average time to reach plantation site Location Rainfall Economic factors Non-farm employment opportunity Alternative use of the land Minimum expected income to continue with Jatropha Expected market price Expected payback period Institutional factors Access to credit Technical help in planting and pruning Extension services Risk and uncertainty factors Risk Behavior Ant and pests attack Constant Number of observations ¼ 281 Wald Chi Square (21) ¼ 115.97 Prob > Chi-Square < 0.001 Pearson Chi-Square (259) ¼ 196.31 Prob > Chi-Square ¼ 0.998 Pseudo R2 ¼ 0.563 Log Pseudo likelihood ¼ 85.054 Area under ROC curver ¼ 0.943

With fixed effect (second model)

Coefficient

Robust std error

Marginal effects

Coefficient

Robust std error

Marginal effects

0.017 0.000 0.058** 0.776***

0.061 0.001 0.029 0.276

0.007 0.000 0.023 0.302

0.007 0.000 0.023 0.952***

0.062 0.001 0.035 0.295

0.003 0.000 0.009 0.364

0.199 0.731*** 0.030 0.012 0.045*** 0.112 0.208

0.244 0.243 0.021 0.022 0.009 0.490 0.457

0.079 0.284 0.012 0.005 0.018 0.044 0.083

0.121 0.524* 0.924*** 0.024 0.046*** 0.531 3.741***

0.264 0.276 0.062 0.023 0.010 0.825 0.447

0.048 0.206 0.365 0.009 0.018 0.209 1.479

0.874*** 0.957*** 0.027* 0.007 0.085***

0.270 0.291 0.015 0.015 0.019

0.337 0.362 0.011 0.003 0.034

1.000*** 1.040*** 0.025 0.010 0.102***

0.309 0.294 0.017 0.015 0.026

0.383 0.395 0.010 0.004 0.040

0.735* 0.235 0.234

0.405 0.279 0.317

0.286 0.094 0.093

1.179*** 0.233 0.127

0.445 0.299 0.744

0.444 0.092 0.050

1.188*** 0.159 4.405

0.238 0.255 e

0.447 0.063 e

1.367*** 0.286 0.505 0.078 0.299 0.031 13.477 e e Number of observations ¼ 281 LR Chi Square (46) ¼ 240.11 Prob > Chi-Square < 0.001 Pearson Chi-Square (234) ¼ 191.28 Prob > Chi-Square ¼ 0.957 Pseudo R2 ¼ 0.617 Log Pseudo Likelihood ¼ 74.594 Area under ROC Curver ¼ 0.957

Note: *** ¼ 1 percent, ** ¼ 5 percent, and * ¼ 10 percent level of significance are considered. dy/dx is for discrete change of dummy variable from 0 to 1.

Table 4b Specification test of the Probit model of continuation with Jatropha plantation. Factors

Without fixed effect (first model) Coef.

_hat 1.007 _hatsq 0.049 Constant 0.051 No. of observations ¼ 281 LR Chi2(2) ¼ 219.63 Prob > Chi-Square < 0.001 Pseudo R2 ¼ 0.564 Log Pseudo likelihood ¼ 84.873

With fixed effect (second model)

Standard error

z

P>z

Coef.

0.109 0.079 0.137

9.230 0.620 0.370

0.000 0.534 0.712

1.017 0.114 0.058 0.024 0.064 0.120 No. of observations ¼ 281 LR Chi2(2) ¼ 241.06 Prob > Chi-Square < 0.001 Pseudo R2 ¼ 0.619 Log Pseudo likelihood ¼ 74.156

The time required to reach the plantation site is positively related to an increase in the transportation cost. At a constant price of Jatropha seed, the increase in transportation cost reduces profit, thus adversely affecting the interest of farmers in Jatropha plantations, which would lead eventually to the abandonment of the plantation. Analyzing the factors causing the growers to discontinue protected vegetable production after few seasons in Sri Lanka, Wijerathna, Weerakkody, and Kirindigoda (2014) found that about 60 percent of the discontinued group claimed the higher cost of transport which resulted in their discontinuation [51]. Non-farm employment opportunities and alternative uses for the land under Jatropha represent an opportunity cost for farmers of Jatropha plantations. Since the plantation of Jatropha involves a long-term commitment from farmers along with the opportunity cost of land and labor, the higher opportunity cost induces the

Standard error

z

P>z

8.940 2.430 0.540

0.000 0.015 0.592

farmers to either switch land or labor from Jatropha plantation to some other use, particularly in the absence of a reasonable return from the plantation. Comparing the returns from Jatropha with Groundnut in the context of viability, livelihood trade-offs, and latent conflict in Jatropha plantations for biodiesel in Tamil Nadu, India, Montobio and Lele (2010) found that, as the Jatropha cultivation is not exceeding the opportunity costs of groundnut cultivation for rainfed farmers, 30 percent of the plantations had been removed and 50 percent were kept without maintenance [15]. The negative effects of the expected payback period from the Jatropha plantation on the decision to continue with Jatropha are explicable on the grounds that Jatropha plantations entail a certain amount of up-front investment. An increase in the expected payback period would discourage investors from continuing with that project. Analyzing the economic benefits and costs of Jatropha

636

K. Goswami, H.K. Choudhury / Renewable Energy 81 (2015) 627e638

plantation in North East India, Goswami, Saikia, and Choudhury (2011) urged for initial state funding particularly for operation and maintenance of Jatropha plantation as the plant is having about 5 years of payback period [48]. Similarly, Gunatilake, Pohit, and Sugiyarto (2011) also stated that a longer payback period of Jatropha creates additional problems for smallholders in biodiesel production [52]. Utilizing credit facilities for activities connected with Jatropha plantation increases the probability of continuation with Jatropha since access to credit enables the farmers to meet the expenses of the establishment, and the operation and maintenance costs of the Jatropha plantation. This is more so for a crop like Jatropha with longer gestation period. Analyzing women's access to entrepreneurial resources in the context of textile trading in informal economy, Olarenwaju and Olabisi (2012) also mentioned that access to credit is crucial in continuation of any ongoing activity [53]. Moreover, farmers would want to delay the burden of loan repayment by showing that they are still continuing with Jatropha plantation. The positive relations between continuation with Jatropha and farmers' risk-taking behavior implies that the risk-loving farmers would be happy to take the risk of plantation and wait longer than risk-averse farmers in anticipation of positive results. Therefore, the probability of continuation with Jatropha increases with risk-taking behavior. As with the adoption model, Wald Chi-Square, Pearson ChiSquare, and ROC Curve values were used to examine the goodness-of-fit of the farmers' continuation with the Jatropha model. The results of these tests show that the estimated Probit model of continuation with Jatropha plantation satisfies the goodness-of-fit criteria. Moreover, the specification test presented in Table 4b shows that there is no omitted variable bias problem in the models. 4.3. Discussion The objective of the present study was to analyze the factors those are crucial in adoption and continuation of Jatropha plantation. The econometric analyses suggests that several factors (both economic and non-economic) play an important role in the process of adoption and continuation of Jatropha plantation in North East India. Availability of land and labor for Jatropha plantation are significant constraints. Although, at governmental level, policies are designed to stop conversion of normal cultivable land for Jatropha, present study suggests that these policies are not fully implemented at the ground level. About 19 percent of Jatropha growers used their regular agricultural land for Jatropha plantations. Land conversion from agriculture to Jatropha, interestingly was higher for past-growers (26 percent) relative to present-growers (13 percent). Moreover, about 73 percent of the amount of land brought under Jatropha plantation can be used for other agricultural activities, but limited markets make this difficult. In general, however, since there are alternative uses of land, this makes the adoption and continuation decisions tenuous. Similarly, availability of non-farm employment opportunity works against the continued production of Jatropha. Thus, Jatropha production, like many other agricultural decisions, are driven by the local demand and supply of labor and land. Several institutional factors need further discussion. To the extent, that farmers are able to grow Jatropha in land that is not economically viable for alternate uses, the more likely that they will continue with these plantions. Extension services can play an important role in educating farmers in making these land use decisions. As the econometric analyses suggest, farmers with better

knowledge are more likely to continue with Jatropha plantations. Similarly, availability of credit is an important variable. This is more so during the gestation period as financial support in the initial stage is crucial to continue with the crop. However, proper use of credit is also equally important. The background research in this context suggests that credit is available only in Cachar, Lakhimpur, and Papumpare districts. Of the total sampled Jatropha growers (both present and past growers), about 40 percent received loans for Jatropha plantation. Out of those who received loans, only 42 percent used their loan for the given purpose with misuse of loans being significantly higher for the past-growers (93%) than the present-growers (28%). A simple cost-benefit analysis was also undertaken to better understand the net returns to a hectare of Jatropha plantation. This simple analysis shows that labor and seedling costs are the major establishment cost component of a plantation. Costs incurred on cleaning and weeding are the primary operational and maintenance cost components [48]. However, it was noted that in North East India, seedlings were distributed for free so it was only the labor cost that matter. The econometric analyses reinforces this understanding, as it is found in Table 4a that labor shortage problem reduces the likelihood of continuation by 21 percent to 28 percent. Different studies indicate that Jatropha plantations are economically viable, although not highly profitable [48,54]. Preliminary analysis of the study shows that plantations are not viable at the current market price of Jatropha seeds of INR 5 and a low harvest of 1.5 tons of seeds per hectare. However, the expected harvest from plantations in NE India is 2.5 tons [48]. Thus, price of seeds and production level play crucial role. However, the profitability of Jatropha cannot be compared with other perennial crops (like Rubber and Tea4) as Jatropha are grown in areas where the opportunities of growing other crops are limited. Moreover, the harvests of these plantation crops require nearby processing units, which is not the case with Jatropha seeds. 5. Conclusion and recommendations India's biofuel policy seeks to increase demand for biodiesel and encourages blending of 20 percent biodiesel with other fuels. This is a major challenge as Jatropha is a new perennial crop for the farmers, who see considerable risk and uncertainty in its production, profitability and employment generation. The study was able to quantify the impact and to show the direction of impact of the factors on farmer's decision to adopt and continue with Jatropha plantation. The findings of the study reveal that, of all the factors, farmers' characteristics, institutional factors, and structural issues are found to play the most significant role in the farmers' decision to adopt and continue with Jatropha plantation. Farmers' characteristics such as their willingness to take risks and whether they have land that is not in use in agriculture play an important role. The institutional factors which play an equally important role are availability of credit and extension services. The first of these helps to reduce the short term pinch imposed by growing a perennial crop, while the second leads to better knowledge and land use decisions. If farmers are able to grow Jatropha in land that is really not suitable for agriculture, they are more likely to stick with the plantation. Structural factors such as non-farm labor availability and travel and time related to transportation of labor and

4 Initial investments in Rubber and Tea are INR 38,070 (http://planning.up.nic.in/ innovations/inno3/ph/rubber.htm) and INR 136,900 (http://planning.up.nic.in/ innovations/inno3/ph/tea.htm) respectively against Jatropha where it is about INR 13,945 [48].

K. Goswami, H.K. Choudhury / Renewable Energy 81 (2015) 627e638

seedlings also matter. Thus, to the extent that markets and transport infrastructure improves, this will aid Jatropha production. The findings of the study have important implications for policy measures to expand the biodiesel industry in the region. Since institutional factors, such as the access to credit, help in the adoption and continuation of Jatropha plantations, government institutions ought to extend such facilities to farmers where they are currently not available. Moreover, credit provision needs to be coupled with proper utilization of approved credit and here the role of extension services comes in. Apparently, there have already been improvements in the use of credit, possibly because of improved monitoring, but this aspect needs to be further strengthened. As the opportunity cost of labor and land have a negative and significant influence on continuation with Jatropha plantations, the government could think about increasing the market price for Jatropha. The biodiesel price in India is artificially low and is less than half of that of subsidized fossil diesel (INR 26.5 per liter,56 of biodiesel against INR 53.1 of fossil diesel in Delhi7). Low biodiesel price make Jatropha less attractive and can demotivate farmers. Therefore, for achieving sustainability in production of Jatropha seeds, it is necessary to increase the market price of biodiesel which will ultimately minimize the gap between the expected and actual income from Jatropha. Jatropha production is a small but essential part of India's strategy for increasing renewable energy sources in the country. The study shows that there are serious bottlenecks to increasing Jatropha production but these problems can be remedied with some important institutional interventions. Acknowledgments This paper is the outcome of a project entitled Sustainability of Biodiesel Industry in North East India, which is financially supported by SANDEE (SANDEE/July 2010/002). We gratefully acknowledge this support. We wish to thank our SANDEE advisor Subhrendu K. Pattanayak for helpful suggestions and comments. We also thank Priya Shyamsundar, Mani Nepal and the SANDEE Secretariat for all the support they have provided towards the execution of the project. We found the comments and suggestions from J. M. Baland, M. N. Murty, and E. Somanathan helpful. We also acknowledge the assistance from Bhabesh Hazarika, Atanu Hazarika, and from organizations like D1 Williamson Magor Bio Fuels Limited, and North Eastern Development Finance Corporation Ltd in carring out this study. References [1] Tapah BF, Santos RCD, Leeke GA. Processing of glycerol under sub and supercritical water conditions. Renew Energy 2014;62:353e61. [2] Abbaspour M, Ghazi S. An alternative approach for the prevention of deforestation using renewable energies as substitute. Renew Energy 2013;49:77e9. [3] Jumbe CBL, Mkondiwa M. Comparative analysis of biofuels policy development in Sub-Saharan Africa: the place of private and public sectors. Renew Energy 2013;50:614e20. [4] Kumar A, Sharma S. An evaluation of multipurpose oil seed crop for industrial uses (Jatropha curcas L.): a review. Industrial Crops Prod 2008;28:1e10. [5] Chandak A, Somani S. Renewable fuel saving obligation: innovative mechanism for emission reduction. J Petroleum Technol Altern Fuels 2011;2:21e4.

5 http://www.biodieselmagazine.com/articles/7508/study-recommends-raisingbiodiesel-price-in-india. 6 http://www.google.co.in/url?sa¼t&rct¼j&q¼&esrc¼s&source¼web&cd¼ 4&ved¼0CEgQFjAD&url¼http%3A%2F%2Farticles.timesofindia.indiatimes.com% 2F2011-01-02%2Findia-business%2F28351977_1_Jatropha-bio-diesel-recordfluctuation-average-procurement-price&ei¼6oGVUp6PO4iRrQeMjoGADw& usg¼AFQjCNGOWEj1fSK-b46vPHNLrBBcuwkwgQ&bvm¼bv.57155469,d. bmk&cad¼rja. 7 http://www.mypetrolprice.com/2/Diesel-price-in-Delhi? FuelType¼1&LocationId¼2.

637

[6] Rajagopal D, Zilberman D. Review of environmental, economic and policy aspects of biofuels. World Bank Policy Research Working Paper 4341. Washington, DC: The World Bank; 2007. Available at:, http://www.wilsoncenter. org/news/docs/worldbankSept.2007.pdf [accessed 12.02.2011]. [7] The Energy Resources Institute. Liquid biofuels for transportation: India country study on potential and implications for sustainable agriculture and energy. German Technical Cooperation; 2005. Available at: http://www.gtz. de/de/dokumente/en-biofuels-for-transportation-in-india-2005.pdf [accessed 05.07.2009]. [8] Agoramoorthy G, Hsu MJ, Chaudhary S, Shieh P. Can biofuel crops alleviate tribal poverty in Inida's drylands? Appl Energy 2009;86:S118e24. [9] Energy Information administration. International energy statistics. 2014. Available at: http://www.eia.gov/cfapps/ipdbproject/iedindex3.cfm? tid¼5&pid¼53&aid¼1&cid¼ww,IN,&syid¼2009&eyid¼2013&unit¼TBPD [accessed 18.11.2014]. [10] Singh S. India: biofuels annual 2009. GAIN Report No. IN9080. 2009. Available at:, http://gain.fas.usda.gov/Recent%20GAIN%20Publications/General% 20Report_New%20Delhi_India_6-12-2009.pdf [accessed 19.10 2009]. [11] Chandrasekhar CP, Gosh J. India's external resilience. BusinessLine, The Hindu; 2014. Available at: http://www.thehindubusinessline.com/opinion/columns/ c-p-chandrasekhar/indias-external-resilience/article6098035.ece [accessed 27-12-2014]. [12] Energy Information Administration. Country analysis briefs: India. 2010. Available at: http://www.eia.doe.gov/cabs/India/pdf.pdf [accessed 1909.2010]. [13] Hamilton JD. Causes and consequences of the oil shock of 2007e08. Brookings Pap Econ Activity 2009;40:215e61. [14] Planning Commission. Report of the Committee on development of Biofuel. Government of India; 2003. Available at: http://planningcommission.nic.in/ reports/genrep/cmtt_bio.pdf [accessed 05.07.2009]. [15] Montobbio PA, Lele S. Jatropha plantations for biodiesel in Tamil Nadu, India: viability, livelihood trade-offs, and latent conflict. Ecol Econ 2010;70:189e95. [16] Kureel RS, Singh CB, Gupta AK, Pandey A. Jatropha: an alternative source for biodiesel. National Oilseed and Vegetable Oil Development Board, Ministry of Agriculture, Government of India; 2007. [17] Brittaine R, Lutadalio N. Jatropha: a smallholder bioenergy crop e the potential of pro-poor development. Integr Crop Manag 2010;8. FAO, Rome. Available at: http://www.google.co.in/url?sa¼t&rct¼j&q¼&esrc¼s&source¼ web&cd¼1&cad¼rja&uact¼8&ved¼0CCIQFjAA&url¼http%3A%2F%2Fwww. fao.org%2Fdocrep%2F012%2Fi1219e%2Fi1219e.pdf&;ei¼3Wm-VLXFLMLn 8AWa-YGYDA&usg¼AFQjCNG5oGCTUGU3jYnMmXSSB9nEIRSaVg&bvm¼bv. 83829542,d.dGc [accessed 26.11.2014]. [18] Sunil N, Kumar V, Varaprasad KS. Origin, domestication, distribution and diversity of Jatropha curcus L. In: Bahadur B, Sujatha M, Carels N, editors. Jatropha, challenges for a new energy crop. Genetic improvement and biotechnology, vol. 2. New York: Springer; 2013. p. 137e52. [19] Punia MS. Cultivation and use of Jatropha for bio-diesel production in India. National Oilseed and Vegetable Oil Development Board, Ministry of Agriculture, Government of India; 2010. Available at: http://www.cpamn.embrapa. br/agrobioenergia/palestras/CULTIVATION%20AND%20USE%20OF% 20JATROPHA%20FOR%20BIO-DIESEL%20PRODUCTION%20IN%20INDIA.PDF [accessed 26.11.2010]. [20] Nahar K, Hampton MO. Jatropha: an alternative substitute to Fossil fuel. IFAS Publication Number HS1193. Gainesville: University of Florida, Institute of Food and Agricultural Sciences; 2011. Available at:, http://edis.ifas.ufl.edu/ HS1193 [accessed 02.01.2015]. [21] Henning RK. The Jatropha system: integrated rural development by utilization of Jatropha curcas L. as raw material and as renewable energy. 2004. Available at: http://www.Jatropha.de/Journal/The%20Jatropha%20paper%20June% 202004.pdf [accessed 02.09.2009]. [22] Oduola T, Popoola GB, Avwioro OG, Oduola TA, Ademosun AA, Lawal MO. Use of Jatropha gossypifolia stem latex as a haemostiatic agent: how safe is it? J Med Plants Res 2007;1:14e7. [23] Gexsi. Global market study on Jatropha. London: World Wide Fund for Nature; 2008. Available at: http://www.Jatropha-alliance.org/fileadmin/documents/ GEXSI_Global-Jatropha-Study_FULL-REPORT.pdf (accessed 27.05.2013). [24] Sing I, Squire L, Strauss J. The basic model: theory, empirical results, and policy conclusions. In: Singh I, Squire L, Strauss J, editors. Agricultural household models: extensions, applications, and policy. Baltimore: Johns Hopkins University Press; 1986. p. 39e69. [25] Strauss J. The theory and comparative statics of agricultural household models: a general approach. In: Singh I, Squire L, Strauss J, editors. Agricultural household models: extensions, applications, and policy. Baltimore: Johns Hopkins University Press; 1986. p. 71e91. [26] Mercer DE, Pattanayak SK. Agroforestry adoption by smallholders. In: Sills E, Abt K, editors. Forests in a market economy. Forestry sciences series, 72; 2003. p. 283e99. [27] Smale M, Edmeades S, Wood S, Kalyebara R. Elements of the conceptual framework and sample survey design. In: Smale M, Tushemereirewe WK, editors. An economic assessment of banana genetic improvement and innovation in the Lake Victoria region of Uganda and Tanzania. Washington DC: International Food Policy Research Institute; 2007. p. 12e24. [28] Jabbar MA, Beyene H, Saleem MA, Gebreselassie S. Adoption pathways for new agricultural technologies: an approach and an application to vertisol management technology in Ethiopia. Socioeconomic and Policy Research

638

[29]

[30]

[31] [32] [33] [34] [35] [36]

[37]

[38]

[39]

[40] [41] [42]

[43]

[44]

K. Goswami, H.K. Choudhury / Renewable Energy 81 (2015) 627e638 Working Paper no. 23. Ethiopia: International Livestock Research Institute; 1998. Wendland KJ, Sills EO. Dissemination of food crops with nutritional benefits: adoption and disadoption of soybeans in Togo and Benin. Nat Resour Forum 2008;32:39e52. Choudhury HK, Goswami K. Determinants of expansion of area under Jatropha plantation in North East India: a tobit analysis. For Policy Econ 2013;13: 46e52. Jayson B, Xiarchos IM. Why are Californian farmers adopting more (and larger) renewable energy operations? Renew Energy 2013;55:322e30. Mercer DE. Adoption of agroforestry innovations in the tropics: a review. Agrofor Syst 2004;61e62:311e28. Sen A. Microeconomics: theory and applications. Oxford: Oxford University Press; 2007. Binswanger HP. Attitudes toward risk: experimental measurement in rural India. Am J Agric Econ 1980;62:395e407. Feder G, Just RE, Zilberman D. Adoption of agricultural innovations in developing countries: a survey. Econ Dev Cult Change 1985;33:255e98. Goswami K, Choudhury HK, Saikia J. Factors influencing farmers' adoption of slash and burn agriculture in North East India. For Policy Econ 2012;15: 146e51. Adesina AA, Mbila D, Nkamleu GB, Endamana D. Econometric analysis of the determinants of adoption of alley farming by farmers in the forest zone of southwest Cameroon. Agric Ecosyst Environ 2000;80:255e65. Pattanayak S, Mercer DE. Valuing soil conservation benefits of agroforestry: contour hedgerows in the Eastern Visayas, Philippines. Agric Econ 1998;18: 31e46. Wastelands Atlas of India. Categorywise and districtwise wasteland area. Ministry of Rural Development and National Remote Sensing Agency, Government of India; 2011. Available at: http://www.dolr.nic.in/ WastelandsAtlas2011/Wastelands_Atlas_2011.pdf [accessed 28.09.2012]. NABARD Consultancy Services. Jatropha plantation in Tripura: techno feasibility report. Agartala: Tripura Regional Office; 2007. Peters J, Thielmann S. Promoting biofuels: implications for developing countries. Energy Policy 2008;36:1538e44. Census of India. Density of population. Government of India; 2011. Available at: http://www.censusindia.gov.in/2011-prov-results/data_files/india/Final_ PPT_2011chapter7.pdf [accessed 06.04.2013]. Hipple PC, Duffy MD. Farmers' motivations for adoption of switchgrass. In: Janick J, Whipkey A, editors. Trends in new crops and new uses. Alexandria, VA: ASHS Press; 2002. p. 252e66. Available at: http://www.hort.purdue.edu/ newcrop/ncnu02/v5-252.html [accessed 28.12.2014]. Mujeyi K. Socio-economic of commercial utilization of Jatropha (Jatropha Curcas) in Mutoko district, Zimbabwe. J Sustain Dev Afr 2009;11:36e53.

[45] Mohamed KS, Temu AE. Access to credit and its effect on the adoption of agricultural technologies: che case of Zanzibar. Afr Rev Money Finance Bank 2008:45e89. Available at: http://aisberg.unibg.it/bitstream/10446/27425/1/ MOHAMED-TEMU%20suppl.%202008.pdf [accessed 28.12.2014].  X, Yang D. Insurance, credit, and technology adoption: field experimental [46] Gine evidence from Malawi. Policy Research Working Paper 4425. The World Bank; 2007. Available at:, http://elibrary.worldbank.org/doi/pdf/10.1596/18139450-4425 [accessed 28.12.2014]. [47] Qualls DJ, Jensen KL, English BC, Larson JA, Clark CD. Analysis of factors affecting farmers' willingness to adopt switchgrass production. In: Paper presented at the Southern agricultural economics association annual meeting, Corpus Christi, Texas, February 5e8; 2011. Available at: http://ageconsearch. umn.edu/bitstream/98625/2/SAEA2011Qualls.pdf [accessed 20.03.2011]. [48] Goswami K, Saikia J, Choudhury HK. Economic benefits and costs of Jatropha plantation in North-East India. Agric Econ Res Rev 2011;24:99e108. [49] Nyamwena-Mukonza C. Adoption of biofuels technologies by smallholder farmers in Zimbabwe. Available from: www.redesist.ie.ufrj.br/ga2012/paper/ ChipoMukonza.pdf; [accessed 28.12.2014]. [50] Maingi RN. The potential role of Jatropha Curcas L. for environmental management and sustainable livelihoods in Kibwezi, Kenya. Thesis submitted in partial fulfilment for the degree of Master of Environmental Studies (Community Development) in the School of Environmental Studies of Kenyatta University. 2010. Available at:, http://www.google.co.in/url? sa¼t&rct¼j&q¼&esrc¼s&source¼web&cd¼4&ved¼0CDUQFjAD&url¼http% 3A%2F%2Fir-library.ku.ac.ke%2Fbitstream%2Fhandle%2F123456789%2F6886% 2FRegina%2520Maingi.pdf%3Fsequence%3D1&ei¼ZHyuVN_vCcq_ uASxkYD4Bw&usg¼AFQjCNHOf1qjkJI-PTTN8pvY0RtaLSCkuA&bvm¼bv. 83339334,d.c2E&cad¼rja [accessed 28.12.2014]. [51] Wijerathna M, Weerakkody WAP, Kirindigoda S. Factors affecting the discontinuation of protected agriculture enterprises in Sri Lanka. J Agric Sci 2014;9:78e87. [52] Gunatilake H, Pohit S, Sugiyarto G. Economy-wide impacts of biodiesel production and use in India: a computable general equilibrium model Assessment. 2011. p. 1e16. Asian Development Bank South Asia Working Paper Series 4. [53] Olarenwaju OA, Olabisi YS. Women's access to entrepreneurial resources in informal economy: a qualitative analysis of Yoruba women textile traders' access to entrepreneurial resources at Balogun market, Lagos, South-West Nigeria. Econ Insights e Trends Challenges 2012;64:30e43. [54] Shinoj P, Raju SS, Kumar P, Msangi S, Yadav P, Thorat VS, et al. An economic assessment along the Jatropha-based biodiesel value chain in India. Agric Econ Res Rev 2010;23:393e404.