Estimating farmers' willingness to pay for climate

Environ Monit Assess (2015) 187: 38 DOI 10.1007/s10661-014-4254-z

Estimating farmers’ willingness to pay for climate change adaptation: the case of the Malaysian agricultural sector Muhammad Mehedi Masud & Ha Junsheng & Rulia Akhtar & Abul Quasem Al-Amin & Fatimah Binti Kari

Received: 1 July 2014 / Accepted: 29 December 2014 / Published online: 30 January 2015 # Springer International Publishing Switzerland 2015

Abstract This paper estimates Malaysian farmers’ willingness to pay (WTP) for a planned adaptation programme for addressing climate issues in the Malaysian agricultural sector. We used the contingent valuation method (CVM) for a monetary valuation of farmers’ preferences for a planned adaptation programme by ascertaining the value attached to address climatic issues in the Malaysian agricultural sector. Structured questionnaires were distributed among the sampled farmers. The study found that 74 % of respondents were willing to pay for a planned adaptation programme and that several socioeconomic and motivation factors have greater influence on their WTP. This paper clearly specifies the steps needed for all institutional bodies to better address issues in climate change. The outcomes of this paper will support policy makers to better design an efficient adaptation framework for adapting to the adverse impacts of climate change.

M. M. Masud (*) : H. Junsheng : F. B. Kari Faculty of Economics and Administration, University of Malaya, 50603 Kuala Lumpur, Malaysia e-mail: [email protected] R. Akhtar Faculty of Economics and Management Sciences, International Islamic University, Kuala Lumpur, Malaysia A. Q. Al-Amin International Business School (IBS), Universiti Teknologi Malaysia, Kuala Lumpur, Malaysia

Keywords Adaptation . Agriculture . Climate change . Willingness to pay (WTP) . Contingent valuation method (CVM)

Introduction Among the consequences of climate change are extreme weather and unexpected temperature and rainfall fluctuations which present significant risks to the agroeconomy (Georgescu et al. 2011; Lobell et al. 2011; Fischer et al. 2005). Lobell et al. (2008) showed that climate change in Southern Africa may result in a drop of maize of up to 30 %. Main crops such as grain and maize can drop by up to 10 % in South East Asia by 2030. Similar findings were reached by IPCC’s recent publications and reports (IPCC 2007). Developing countries are facing the biggest challenges in their drive to become developed countries due to climate vulnerability. Poor countries are more vulnerable due to their present exposure and economic and social sensitivity to climate change (Stern et al. 2006). Over the past few decades, Malaysia has experienced a growing warming trend. In the southern areas of peninsular Malaysia, the frequency of long dry periods tended to be higher with a significant increase in the mean and variability of the length of the dry spells. Furthermore, all the indices of wet spells in these areas show a decreasing trend (Jamaludin et al. 2008). Increasing temperatures result in extreme weather and climate variability. In Malaysia, temperature and rainfall are projected to increase between +0.6 and 3.4 °C and

38 Page 2 of 12

−1 and +32 % in 60 years, respectively (INC 2000). The rise in sea level is about 13–94 cm in 100 years (INC 2000). Like other countries, Malaysia is experiencing the adverse effects of climate change on key economic sectors such as energy, industry, transport, forestry, agriculture, water and coastal resources, public health and the waste sector (INC 2000). The Malaysian Agricultural Research Institute (MARDI) has estimated that a 1 °C increase in daily average temperature reduces 10 % of the rice yield in peninsular Malaysia (Abdullah 2007). There is a growing fear that climate change could significantly affect the productivity of the rice crop (Tao et al. 2008). It is clear that Malaysian agricultural productivity is decreasing due to climate vulnerability and other climate change-related issues (Siwar et al. 2009). Due to weak institutional capacity, limited engagement in environmental and adaptation issues and a lack of validation of local knowledge (SPORE 2008; BNRCC 2008; Royal Society 2005; Adams et al. 1988), farmers are little concerned with climate change. In view of the significance of climate change in rice production, there is a need to determine farmers’ perceptions and attitudes towards climate change in order to offer an efficient adaptation framework. The adaptation framework might help famers to adapt to the adverse effect of climate change. Accordingly, there is a need to analyze the implications of the climate change adaptation policies to ascertain an efficient and effective adaptation framework. From the above circumstances, it is clear that a new economic model is necessary for Malaysia to progress and join the league of highincome nations. To do this, it should not neglect environmental issues. It is necessary to value environmental endowments for sustainable development. Climate change directly or indirectly affects the social and economic sustainability of farmers. Climate change causes crop failures, low productivity and high production costs and results in loss of income for farmers, while also raising the seasonal unemployment rate (Siwar et al. 2009; Alam et al. 2011). Since farmers are dependent on agriculture, when productivity and agricultural production decrease, this will certainly affect their revenue. Under such circumstances, adaptation strategies become necessary for farmers because failure to adapt could lead to ‘serious disadvantages, social dislocation and displacement of people, even morbidity and mortality’ (Downing et al. 1997). According to Stern (2007), it is high time to take steps to tackle the upcoming risk of climate change to

Environ Monit Assess (2015) 187: 38

society. Planned adaptation is urgently needed. Adaptation strategies must be implemented through planned adaptation programmes. These strategies do not only assist in combating the adverse effects of climate change, but also encompass a big number of technical, social, economic and environmental challenges (Iglesias et al. 2007; Olesen and Bindi 2002). The development of adaptation methods must take into account future socioeconomic and climate change scenarios. Practitioners need to understand the relevance of a future climate to a future society, rather than its relevance to current social norms (Iglesias et al. 2011). Identification of relevant adaptation strategies for the greatest susceptible population is a formidable challenge. As such, decision makers must be aware that adaptation strategies to climate change cannot provide equal benefits to all regions and ethnic groups. Adaptation is an important strategy in reducing shortrun and long-run vulnerability to climate change (IPCC 2007). Farmers, particularly those in developing countries like Malaysia, must adapt to climate change to reduce its negative impact and exploit the benefits of adaptation. Farmers in Africa and Asia have been adjusting to climate change by employing autonomous adaptation strategies given the current vulnerabilities and expected magnitude of the climate change impact; these autonomous adaptation efforts will most likely fall short of the necessary adaptation (Stern et al. 2006). Therefore, planned or policy-guided adaptation is important. Adaptation to climate change has the potential to substantially reduce many of the adverse impacts of climate change, reduce vulnerabilities and promote sustainable development through enhancing the welfare of the poorest members of society (Smit and Pilifosova 2003) by improving food security, facilitating access to safe water and shelter, increasing income and improving sustainability of existing resources for example. In effect, adaptation is a way of reducing vulnerability, increasing resilience, moderating the risk of climate impact on lives and livelihoods and taking advantage of opportunities posed by actual or expected climate change. Against this background, the current study explores farmers’ perceptions and adaptation strategies to climate change. Specifically, the paper seeks to (1) estimate the WTP of the farmers for a planned adaptation programme to climate change in North West Selangor, Malaysia and (2) determine the factors that influence their willingness


to pay (WTP). This study will help policymakers and government to better address the challenge of climate change and meet expected growth targets by providing them with data pertaining to the awareness levels of farmers towards climate change and their WTP for a planned adaptation programme. To this end, the study of the WTP of farmers is necessary, especially given the fact that, to date, no such research has been conducted in the Malaysian context. There is therefore a pressing need to survey the behaviour and WTP of Malaysian farmers for a planned adaptation programme to climate change in North West Selangor, Malaysia. In order to develop a clear picture and sound understanding of WTP for a climate change adaptation programme and to socioeconomic and motivation factors that influence farmers’ WTP, we test several hypotheses. In particular, we aim to determine WTP for climate change adaptation programmes, as no such study has yet been undertaken in Malaysia. Thus, we propose the following hypotheses: H1: Farmers’ socioeconomic factors have a positive direct relation with WTP for planned climate adaptation programmes. H2: Motivation factors have a positive and significant relation with WTP for planned climate adaptation programmes.

Contingent valuation method A contingent valuation (CV) method was employed to estimate the WTP of farmers as applied to environmental valuation (Carson 2012). Many studies also employed CVM to quantify the benefits of nonmarketed environmental goods and attributes in such a way that they can enter directly into cost-benefit calculations predominantly in western, North American and East Asian advanced and developed countries (Carson et al. 2010). The CVM methodology highlights diverse issues like improvements in water quality and sanitation (Howard et al. 2010; Vörösmarty, et al. 2010; Orgill et al. 2013), valuing forestry (Canadell and Raupach 2008; Gelo and Koch 2012; Mason et al. 2013), exposure to flood risk (Lantz et al. 2012; Kellens et al. 2013), wetland conservation (Yoon 2009; Kaffashi et al. 2013; Turner 2013), offsetting carbon emissions and groundwater contamination, health economics (Georgiou and Turner 2012; Del Borghi et al. 2013; Everard et al. 2013;

Page 3 of 12 38

Andersson et al. 2014), cultural economics (Carvalho et al. 2010; Wicker et al. 2012), transportation safety and economics (Hess et al. 2012; Kristiansen 2013) and a wide range of environmental services (Ojeda et al. 2008; Vo et al. 2012). Furthermore, studies regarding WTP are conducted in developed countries and mainly examine consumers’ WTP for renewable concentrating on environmental (Bain et al. 2012; Park et al. 2013) and health and social effects (Curtis 2012; de Bekker‐Grob et al. 2012). As the same time, environmental health and secure energy supplies are also addressed by many researches (Damigos et al. 2009; Bollen et al. 2010; Sovacool 2011). However, most of these studies used CVM and relied on choice experiments. Despite diversity of study design, all of them unanimously affirm that consumers generally have a positive attitude in terms of WTP. The goal is to elicit what farmers in West Selangor, Malaysia, are willing to pay for a planned adaptation programme to climate change. Survey questionnaires were used to ascertain the ‘WTP’ or ‘willingness to accept’, respectively, which means that the accuracy of CVMs depends on the quality of the survey instrument and how well people respond to the required assessments. The information component of the survey instrument, the explanation of the method of provision, payment vehicle, the decision rule and the time frame of payment (Boyle 2003) were all information presented to the farmers through a payment card. In this study, we developed a payment card according to Rowe et al. (1996). A payment card lists a series of values from which respondents choose an amount that best represents their maximum WTP. Payment card was designed with an exponential response scale that contains 24 cells. An exponential response scale is consistent with this hypothesis of measurement error increasing with WTP values. The values of the cells 2 through to 22 are computed (Table 1) by the following equation: Bn ¼ B1 ð1 þ K Þn

−1

where Bn =bid value, B1 =1 and K=range selected for the payment card so that the largest value on the payment card is (1.2860)21=200. The payment vehicle was the electricity bill of the respondents. The fund will be managed by the climate change adaptation fund for agriculture, a nongovernment organization (NGO) established specifically for that purpose.

38 Page 4 of 12


Table 1 Payment card Range RM 0–RM 200

Centre RM 16

RM 0

RM 2

RM 5

RM 10

RM 20

RM 45

RM 95

RM 200

RM 0.5

RM 3

RM 6

RM 12

RM 25

RM 55

RM 120

≥RM 200

a

RM 4

RM 8

RM 16

RM 35

RM 75

RM 155

Do not know

RM 1

a

Function 1.2860 (n−1)

USD1=MYR 3.43

Methodology Site selection and data collection The questionnaire used in this study was based on a survey among farmers of the Integrated Agricultural Development Area (IADA), West Selangor, Malaysia, which covers approximately 100,000 ha, of which 20,000 ha is for paddy, 55,000 ha for palm oil, 20,000 ha for coconut and 5000 ha for fruits and vegetable. There are approximately 10,300 paddy farming families who reside within the IADA who are involved in rice production. Survey design and sampling methods The study was conducted through direct face-toface interviews in order to obtain reliable responses from the respondents. The study area is located in the IADA North West Selangor which consists of eight areas including Sawah Sempadan, Sg. Burong, Sekinchan, Sg. Leman, Pasir Panjang, Sg. Nipah, Panchag Bedena and Bagan Terap. Of the eight areas and using a random sampling method, 50 farmers were selected with a total sample size of 400 (50×8). The survey was conducted in September 2013. The survey was confined to within the IADA, as it constitutes a prominent agricultural zone in Malaysia. The data was collected through interviews with heads of households who worked as rice farmers.

of the respondents). Section B enquired as to farmers’ perceptions and attitudes towards climate change. Section C consisted of CVM questions to estimate farmers’ WTP for a planned adaptation programme to climate change.

Specification of model A regression model was developed to explore the factors that might affect the WTP of farmers for a planned adaptation programme to climate change in North West Selangor, Malaysia. Respondents were offered a yes/no option to select their WTP for a planned adaptation programme to climate change. When the dependent variable is in the 0– 1 style, researchers can choose between logistic regression and probit regression (Wang and Elhag 2007). For this reason, in this study, logistic regression was selected as the evaluation method. It was assumed that the factors listed in Table 6 might affect WTP and these factors were included in the model as independent variables. The probability model of the WTP is P (Yi =1), which was represented as follows: The basic model of the logit estimation is as follows: " # h π i PðY ¼ 1ÞjX 1 ; …X p ¼ Loge Loge ð1Þ 1−π 1−PðY ¼ 1jX 1 ; …X p

¼ α þ β1 X 1 þ … þ βp X p ¼ α þ Design of the questionnaire The questionnaire consisted of sections A, B and C. Section A collected information on the farmers’ socioeconomic characteristics (i.e. household size, gender, ethnic group, religion, education level, income, type of farmers, farm size and family size

Xp j¼1

β jX j

ð2Þ

π is a conditional probability of the form P(Y= 1|X1,…, Xp). That is, it is assumed that success is more or less likely depending on a combination of values of the predictor variables. The log odd, as defined above, is also known as the logit transformation of π and the analytical approach described here is sometimes known


Page 5 of 12 38

as logit analysis. The logistic function takes the form of the following: Xp

PðY ¼ 1jX 1 ; …X p ¼

e

αþ

1þe

αþ

j¼1

β jX j

ð3Þ

Xp j¼1

β jX j

This can also be transformed into

PðY ¼ 1jX 1 ; …X p ¼

1 Xp

1þe

−α−

j¼1

ð4Þ β jX j

The nonresponse probability is P Y ¼ 0jX 1 …X p ¼ 1−p Y ¼ 1│X 1 …X p

¼ 1þe

αþ

1 Xp j¼1

ð5Þ β jX j

‘yes’ (=1) if the farmers state a positive WTP and ‘no’ (=0) when they are not WTP any amount. The independent variables employed to predict the probability of WTP are knowledge, attitudes, age, education, household income, farm size and types of farmers. Using the set of predictors, the LR equation for the log odds in favour of WTP is estimated to be P log ð6Þ ¼ bo þ bi þ x j 1−P using the partial coefficients, bi, informing the change to log odds of agreeing to pay for a planned adaptation programme to climate change in North West Selangor, Malaysia. The independent and dependent variables used in the logit analysis and their basic statistics are given in Table 2. A statistical analysis was employed to analyze the socioeconomics and motivation factors which influence farmer’s WTP for climate change adaptation.

Results and discussions The socioeconomic characteristics of the respondents Table 3 shows the age distribution of farmers in the study area. The ages of the respondents ranged between 18 and over 60 years. The greatest number of farmers (58.96 %) was from the age group between 46 and 60 years. Most of the farmers are middle aged. The

second largest group of farmers (50.38 %) was between 31 and 45 years. It could be concluded from this result that the population of rice farmers in the study area is middle-aged. This age distribution could have a positive impact on the adoption of new techniques of production. The research also found that the educational status of the farmers ranged with the majority having only secondary school education. The results show that 44.97 % of farmers had secondary education, while 21.30 % had tertiary education, 18.96 % had primary, 11.95 % had no formal education, and 2.85 % had university education. The distribution of education could have a positive impact on the adoption of new techniques of production. In regard to income brackets, this study found that only 7.27 % of the farmers had an income range of RM 2000 and below. The highest percentage of the farmers (33.77 %) had a monthly income ranging between RM 2000 to RM 4000 per month. Of the farmers, 29.87 and 20.78 % had monthly incomes ranging between RM 4000 and RM 6000, and RM 6000 and RM 8000, respectively. Only 5.71 % of the farmers had monthly incomes ranging between RM 8000 and RM 10,000, and 2.60 % of the farmers had monthly incomes greater than RM 10,000. According to the Malaysian Department of Statistics, more than half of Malaysian households earn a monthly income of less than RM 3000. The remaining households earn between RM 3001 and RM 4000 (12.9 %), RM 4001 and RM 5000 (8.6 %), RM 5 001 and RM 10,000 (15.8 %) and above RM 10,000 (4.9 %) (DOSM 2009). Motivating factors for WTP This study found that the majority of the respondents (74 %) were willing to pay for a planned adaptation programme to climate change in North West Selangor, Malaysia, as shown Fig. 1. The most important motivation factors for their WTP were that 85 % of the total respondents mentioned climate change impacts on agricultural production, 83 % mentioned avoiding future natural disasters, and 78 % of the respondents are concerned for the risks posed by climate change, which are presented in Fig. 2. In this study, farmers were asked to rank the motivation factors which they feel very important to motivate them for WTP for climate change adaptation. The most important item was to be ranked 1, then second most important 2 and so on. All items had to be ranked, and no rank could be used more than once. Having collected

38 Page 6 of 12


Table 2 Variables included in the logit models Variables

Description of the Variables

Category

Dependent variable WTP

WTP for climate change adaptation 1=Willing to pay, 0=not willing to pay

Independent variables Education

Educational status

Household size

1=No formal education, 2=primary, 3=secondary, 4= tertiary, 5=university Number of household members

Farm size

Area of farm

Income

Monthly family income

1=RM 2000 and less, 2=RM 2001–4000, 3=RM 4001– 6000, 4=RM 6001–8000, 5=RM 8001–10,000 and above 10,000

The environment has the right to be protected irrespective of the costs

Motivation factors for WTP

1=Strongly disagree, 2=disagree, 3=neutral, 4=agree and 5=strongly agree

I care about the environment in general


I feel responsible for my contribution to climate change


Impacts on agricultural production


Concern for the risk posed by climate change


Table 3 Socioeconomic characteristics of the respondents (N= 385) Demographics

Frequency

Percentage

27

7.02

Age (years) 18–30 31–45

194

50.38

46–60

150

38.96

60 and above

14

3.64

No formal education

46

11.95

Primary

73

18.96

Secondary

173

44.94

Tertiary

82

21.30

University

11

2.85

28

7.27

Education level

the data from 283 farmers from IADA, North West Selangor, Malaysia, we ranked all the motivating factors based on mean values. The lowest mean value was assigned the rank 1 which indicates that the most important factors to authors discussed the ranking of motivation factors as shown Fig. 3. It indicates that Malaysian farmers assign rank 1 to ‘impacts on agricultural production, while ranks 2, 3, 4, 5, 6, 7, 8 and 9 for ‘to avoid future natural disasters’, ‘concern for the risk posed by climate change’, ‘I care about the environment in general’, ‘I feel responsible for my contribution to climate change’, ‘to protect flora and fauna on this earth’, ‘to reduce future economic damage costs’ and ‘to protect future generations for WTP to adapt climate change’, respectively.

Income