Amount of information and the willingness of consumers to pay for food traceability in China

Accepted Manuscript Amount of information and the willingness of consumers to pay for food traceability in China

Shaosheng Jin, Yan Zhang, Yining Xu PII:

S0956-7135(17)30058-0

DOI:

10.1016/j.foodcont.2017.02.012

Reference:

JFCO 5447

To appear in:

Food Control

Received Date:

12 November 2016

Revised Date:

08 February 2017

Accepted Date:

09 February 2017

Please cite this article as: Shaosheng Jin, Yan Zhang, Yining Xu, Amount of information and the willingness of consumers to pay for food traceability in China, Food Control (2017), doi: 10.1016/j. foodcont.2017.02.012

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT Highlights • Chinese consumers would prefer a food traceability system with detailed information rather than abbreviated information. • The willingness to pay (WTP) for traceability is about 10% higher with detailed information than abbreviated information. • Educational level, self-reported health, risk attitude, and other factors affected WTP by consumers. • The main concerns of Chinese consumers are quality certificates and chemical fertilizer/pesticide details.

ACCEPTED MANUSCRIPT 1

Amount of information and the willingness of consumers to pay for food

2

traceability in China

3

Shaosheng Jin a, *, Yan Zhangb, Yining Xub

4

a

5

University, 866 Yuhangtang Rd, Hangzhou 310058, P.R. China

6

b

7

University, 866 Yuhangtang Rd, Hangzhou 310058, P.R. China

China Academy for Rural Development (CARD), School of Public Affairs, Zhejiang

Department of Agricultural Economics and Management, School of Management, Zhejiang

8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 *

Corresponding author. Tel.: +86 571 88981490; fax: +86 571 88981522 E-mail address: [email protected] (S.S. Jin)

1

ACCEPTED MANUSCRIPT 24

Amount of information and the willingness of consumers to pay for food

25

traceability in China

26 27

Abstract

28

There is no consensus about whether the food traceability system planned for construction in

29

China or other countries should record detailed information like the beef traceability system

30

in Japan, or simple abbreviated information similar to that provided in the USA. Using apple

31

as a research subject, we adopted random nth price experimental auction to investigate the

32

willingness to pay (WTP) for traceability based on abbreviated and detailed information

33

among consumers in China. Totally 88 participants attended the experimental auction. The

34

results showed that consumers had a positive WTP for both types of food traceability

35

system, but the average premium that consumers were prepared to pay for traceability with

36

detailed information was 10% higher than that with abbreviated information. Males, married

37

subjects, and those with a relatively low educational level placed a higher premium on

38

traceability with detailed information, but consumers with good self-reported health did not

39

want to pay a high premium for traceability with detailed information. The results also

40

showed that consumers were most interested in a food traceability system that provides

41

quality certificates and details of the chemical fertilizers/pesticides used in food production.

42

We discuss the implications of these results for the implementation of a food traceability

43

system.

44 45

Keywords: China, consumer, information, food traceability system, willingness to pay

46 47 2

ACCEPTED MANUSCRIPT 48 49

1. Introduction

50

Information asymmetry often leads to increased anxiety, uncertainty, and rapidly declining

51

confidence among consumers (Hobbs, 2004; Houghton et al., 2008). To restore consumer

52

confidence, it is essential and effective to provide them with more food-related information

53

(Golan et al., 2004; van Rijswijk & Frewer, 2012), which can be achieved via traditional food

54

labels (e.g., Kehagia et al., 2007) and food traceability systems using modern technology (e.g.,

55

Golan et al., 2004; Hobbs et al., 2005; Liao et al., 2011). Labeling is a conventional method for

56

food information provision and it still plays an important role in communicating with consumers

57

(Kehagia et al., 2007). However, the space limitations of simple paper labels restrict the amount

58

of information that can be conveyed (Verbeke & Ward, 2006; Jin & Zhou, 2014). Due to

59

continuous improvements in technology and devices, barcodes, radio frequency identification,

60

wireless sensor networks, an electronic nose coupled with mass spectrometry, and optical

61

systems are now used widely in food traceability systems (Peres et al., 2007; Chrysochou et al.,

62

2009; Aung & Chang, 2014). Thus, the capacity to provide food safety and quality information

63

via food traceability systems is much greater (Jin & Zhou, 2014).

64

In terms of the amount of information conveyed, there are two types of food traceability

65

system, which provide abbreviated information or detailed information. For example, a beef

66

traceability system may provide abbreviated information, such as the beef traceability system

67

employed in the USA1, which is simply a record-keeping system for controlling the supply chain,

68

facilitating food safety control, differentiating the attributes of foods, and monitoring animal

69

diseases (Golan et al., 2004; Schulz & Tonsor, 2010). This is a voluntary traceability system,

70

which is motivated mainly by economic incentives (Souza-Monteiro & Caswell, 2004). The 1

The National Animal Identification System is a voluntary program and it is the most comprehensive system in the USA for implementing food traceability (Schroeder et al., 2009).

3

ACCEPTED MANUSCRIPT 71

second type of system provides detailed information, e.g., the Japanese beef traceability system.

72

According to Jin and Zhou (2014), the Japanese Beef Traceability Law requires much more

73

detailed information2, and thus the mandatory Japanese beef traceability system has more depth

74

and breadth than the EU3 traceability system (Souza-Monteiro & Caswell, 2004). Therefore, the

75

beef traceability systems used in the USA lag far behind those in Japan in terms of the amount of

76

information provided (Smith et al., 2005).

77

In practice, there are many barriers to the implementation of a food traceability system with

78

detailed information, including liability among the participating producers (Breiner, 2007;

79

Schulz & Tonsor, 2010), the reliability of technology (Schroeder et al., 2009; Schulz & Tonsor,

80

2010), standard limitations (Bosona & Gebresenbet, 2013), and the willingness to provide

81

information (Golan et al., 2004). Another major concern is the expense of providing information

82

(Golan et al., 2003). Food traceability systems are expensive and complex, which could lead to

83

financial problems (Bosona & Gebresenbet, 2013) because greater amounts of information and a

84

more detailed traceability system will incur higher costs (Souza-Monteiro & Caswell, 2004). For

85

food producers, the critical issue is who will pay the cost (Souza-Monteiro & Caswell, 2004;

86

Breiner, 2007). Thus, producers do not want to provide detailed information if they have to bear

87

the additional cost.

2

The full name of the Japanese Beef Traceability Law is “Law for Special Measures Concerning the Management and Relay of Information for Individual Identification of Cattle,” which was implemented to allow full traceability from farm to fork in 2004 by the National Livestock Breeding Center with the support of the Ministry of Agriculture, Forestry, and Fisheries. The following information is required: individual identification number, date of birth or country of origin, sex, individual identification number of the maternal parent, location (prefecture name) of the raising facilities, start and end of breeding in the breeding facilities, date of slaughter, breed of cattle, name of the exporting country (for imported cattle), title and location of the abattoir where the cattle were slaughtered, and the country of origin (for imported cattle) (Clemens, 2003; Jin and Zhou, 2014). Excluding the information required by law, beef retailers can provide additional information voluntarily to facilitate better assurance of food safety and quality, e.g., Jusco Supermarkets (Aeon Company, Ltd) provide consumers with the story of how the meat was produced, photographs and the name of the producer on the packaging, BSE testing details, an official stamp from Aeon, etc. (Clemens, 2003). 3 The EU is a major driver in establishing global standards that are leading to the introduction of a traceability system worldwide (Souza-Monteiro & Caswell, 2004).

4

ACCEPTED MANUSCRIPT 88

Similar to many other developing countries, China is in the preliminary stages of

89

implementing a food traceability system, but there is no consensus regarding the amount of

90

information that should be recorded in the food traceability system. Information comes at a cost,

91

so it is very important to identify the attitudes of consumers and their preferences regarding food

92

traceability systems containing different amounts of traceability information. However, previous

93

studies of food traceability have focused mainly on the willingness to pay (WTP) for traceability

94

per se among consumers (e.g. Dickinson & Bailey, 2002; Hobbs et al., 2005; Loureiro &

95

Umberger, 2007; Ubilava & Foster, 2009; Lee et al., 2011; Ortega et al., 2011; Zhang et al.,

96

2012; Wu et al., 2012; Lu et al., 2016) and the results of these studies suggest that consumers

97

from different countries or regions are willing to pay a premium for food with traceability

98

attribute (Jin & Zhou, 2014).

99

Meanwhile, attention has also been paid to food traceability systems. From the perspective

100

of food industry, some studies analyzed the economic incentives/motives/benefit (e.g., Hobbs et

101

al., 2005; Bosona & Gebresenbet, 2013; Aung & Chang, 2014; Menozzi et al., 2015) and barriers

102

to establish food traceability systems (Bosona & Gebresenbet, 2013). Also there are some other

103

studies focusing on how to develop food traceability systems (e.g., Feng et al., 2013; Hu et al.,

104

2013). From the perspective of consumers, van Rijswijk et al. (2008) investigated consumers’

105

perception of food traceability systems. As food traceability systems represent a good means of

106

information provision, several recent studies investigated the types of traceability information

107

that consumers were interested in. For example, Wu et al., (2016) investigated consumers’ WTP

108

and preference rankings for different kinds of traceability information, including specific

109

information related to farming, slaughter and processing, distribution and marketing, and

110

government certification. Based on a national representative sample of 6243 Japanese

111

consumers, Jin and Zhou (2014) reported that harvest date, production method, and production

5

ACCEPTED MANUSCRIPT 112

method certification are the items of most interest to Japanese consumers. Generally, the existing

113

literature shows that easy-to-understand, quick-to-process information (van Rijswijk et al., 2008)

114

and information of quality assurances (Hobbs et al., 2005) are more preferred than technical

115

information of traceability (Gellynck & Verbeke, 2001).

116

Despite above valuable contributions, prior researches have not assessed the premiums that

117

might be paid for traceability with different amounts of information recorded by a food

118

traceability system. This paper seeks to fill this gap with the following goals:

119 120 121 122 123 124

1) To compare the WTP among Chinese consumers for traceability with abbreviated and detailed information. 2) To investigate the factors that affect the WTP premiums among consumers for traceability with abbreviated and detailed information, and 3) To identify the specific types of food safety and quality information that interest Chinese consumers.

125

The remainder of this paper is organized as follows. In the next section, we describe the

126

background and details of the food traceability system in China. Section 3 explains the methods

127

employed and the data. The results and discussion are presented in Section 4. In Section 5, we

128

give our conclusions and discuss the implications of this study.

129 130

2. Background regarding the food traceability system in China

131

China began to explore the implementation of a food traceability system in the early 2000s,

132

when the Management Regulations for Animal Vaccination Identification Tag were released in

133

2002, which stipulate that livestock must wear immunity ear tags and that an immunity archives

134

management system should be established. However, progress in the construction of a

135

traceability system has been driven mainly by food safety issues. In particular, the EU imposed 6

ACCEPTED MANUSCRIPT 136

mandatory traceability on imported beef, aquatic products, and vegetables in 2004 due to BSE,

137

which prompted the Chinese government to enact tracing and tracking guidelines for exit aquatic

138

products, beef, vegetables, and fruits in order to promote the export and exchange of agricultural

139

products. In addition, two important laws, i.e., the Agricultural Product Quality Safety Law and

140

Food Safety Law, both require that food enterprises establish records regarding procurement,

141

production, processing, packaging, and circulation for the food supply chain. However, due to

142

high costs and technical constraints, only a limited number of food categories were covered and

143

the development of the food traceability system was slow before 2006 (Wu et al., 2012; Bai et

144

al., 2013).

145

The development of a food traceability system in China has progressed rapidly since 2007.

146

The production of a Certificate and Invoice Asking System and Purchase and Sale Ledger System

147

were encouraged by the State Administration for Industry and Commerce to improve the

148

management of food circulation, where nine categories for 69 types of major products (45 types

149

are food products) had to be implemented for mandatory electronic supervision with a code

150

attached to the package. In order to further E-enable the Certificate and Invoice Asking System

151

and Purchase and Sale Ledger System, and to improve the circulation and packaging of meat and

152

vegetables, which are the largest components of the typical Chinese “shopping basket,” the

153

Ministry of Commerce and the Ministry of Finance began to fund 10 capable cities, i.e.,

154

Shanghai, Chongqing, Dalian, Qingdao, Ningbo, Nanjing, Hangzhou, Chengdu, Kunming, and

155

Wuxi, as pilots to establish a food circulation traceability system in 20104. However, it is still

4

From 2011 to 2014, another four batches of 48 pilot cities were included in this program. The second batch of pilot cities (2011): Tianjin, Shijiazhuang, Harbin, Hefei, Nanchang, Jinan ( in Shandong province ), Haikou, Lanzhou, Yinchuan and Urumchi; The third batch of pilot cities (2012): Beijing, Taiyuan, Hohhot, Changchun, Zhengzhou, Changsha, Nanning, Guiyang, Xi’an, Xining, Suzhou, Wuhu, Weifang, Yichang and Mianyang; The fourth batch of pilot cities (2013): Qinhuangdao, Baotou, Shenyang, Jilin, Mudanjiang, Xuzhou, Fuzhou, Yantai, Zibo, Luohe, Xiangyang, Xiangtan, Zhongshan, Zunyi and Tianshui; The fifth batch of pilot cities (2014): Lhasa, Jinzhong, Haidong, Tongren, Shihezi, Wuzhong, Weihai and Linyi.

7

ACCEPTED MANUSCRIPT 156

difficult to identify the origins of meat and vegetables if food safety problems occur, so there is

157

an urgent need to “informationize” the traceability system nationwide. This will require that the

158

pilot cities implement a unified acquisition index, coding rules, transmission format, interface

159

specification, and traceability procedures to ensure the smooth communication of information

160

between different regions, as well as among different traceability technology modes. At the

161

initial stage, large wholesaling markets, large and medium-sized supermarket chains, and

162

designated mechanized slaughterhouses are the main targets. The circulation traceability system

163

for meat and vegetables may serve as a successful case to drive the implementation of

164

traceability systems for other agricultural products. Subsequently, China will establish

165

traceability systems for tea, milk powder, aquatic products, etc., in different regions.

166

The Chinese food traceability system can provide much more information after the

167

implementation of the E-enabled Certificate and Invoice Asking System and Purchase and Sale

168

Ledger System. However, a related controversial topic is whether the food traceability system

169

under construction in China should record detailed information similar to the beef traceability

170

system in Japan or simply convey abbreviated information like that in the USA to improve

171

supply-side management. This is a crucial issue for China because it is still in the preliminary

172

stage of implementing a food traceability system, which covers a limited number of food

173

categories.

174 175

3. Methodology and data

176

3.1 Methodology

8

ACCEPTED MANUSCRIPT 177

To elicit the WTP among consumers regarding traceability using abbreviated and detailed

178

information, we employed apples5 as the research subject and we conducted an experimental

179

auction. Experimental auction is a kind of revealed preference method, which has advantages

180

over stated preference methods such as contingent valuation method (CVM) and choice

181

experiment (CE) (Alfnes & Rickertsen, 2003; Lusk et al., 2007; Bougherara & Combris, 2009).

182

Involving real money and real goods, the experimental auction can create a non-hypothetical

183

setting where participants have the greatest incentive to reveal their true values (Lusk, 2003), and

184

sincere bidding is the weakly dominant strategy for participants (Melton et al., 1996; Shogren et

185

al., 2001; Lusk & Shogren, 2007; Chern & Chang, 2012). Thus, experimental auction has been

186

extensively used in estimating consumers’ WTP and preferences for new products and product

187

extensions in recent years (e.g., Alfnes & Rickertsen, 2003; Lusk et al., 2007; Chern & Chang,

188

2012; Corrigan et al., 2012; Wu et al., 2016).

189

Specifically, this study adopted random nth price sealed-bid auction method (e.g., Shogren

190

et al., 2001; Gracia et al., 2011; Lee et al., 2011; Wu et al., 2016). It shares the characteristics of

191

both Becker-deGroot-Marschak (BDM) (Becker et al., 1964) and the Vickrey second price

192

auction mechanism (Vickrey, 1961) that all participants have a reasonable chance of winning

193

regardless of their true values (Shogren et al., 2001; Lusk et al., 2007; Lusk & Shogren, 2007). It

194

works as follows: In a quiet experimental environment, each participant submitted a sealed bid

195

and all bids were then ranked in descending order by their values before randomly drawing a

196

number n ∈ {2, 3,...,𝑘} (where k is the number of bidders) from a uniformly distribution, where

197

the n–1 highest bidders purchased a unit of the auctioned good at the randomly drawn nth price.

5

Apples are a good choice for valuation experiments because of their worldwide popularity, consumer familiarity, and level of consumption, and there is little heterogeneity among experiment respondents in their understanding of this product (Costanigro et al., 2014).

9

ACCEPTED MANUSCRIPT 198

The experiment was conducted during January and February in 2013. A pre-test was

199

conducted at Zhejiang University in Hangzhou to assess the understandability, content and

200

length of the auction mechanism and questionnaire. A total of 15 respondents participated in the

201

pre-test. Following the feedback received, we adjusted the procedure of auction and ambiguous

202

expressions. During the formal auction process, the subjects were identified through random

203

sampling, consumers aged 18 years and above were eligible. All of the subjects signed up

204

voluntarily, and they were recruited by leaflet and on-site recruitment in Hangzhou city,

205

Zhejiang Province, which is among the first batch of pilot cities implementing a food traceability

206

system in China. 99 consumers signed up for the experimental auction and 88 attended finally.

207

The auction design had nine sessions and each comprised 6–12 participants.

208

Two treatments were designed to evaluate the preferences of consumers for the two types of

209

food traceability system, i.e., in the first scenario, apples were provided with abbreviated

210

information comprising “brand, producer, place of origin, size, harvest date, shelf life, storage

211

instructions, and E-business website”; and in the second scenario, apples were provided with

212

detailed information comprising the abbreviated information mentioned above as well as

213

“contact details, pesticide residuals, logistics information, nutritional content (calories, fat,

214

dietary fiber, protein, carbohydrate, carotene, vitamin A, vitamins B1, B2, and B3, vitamin C,

215

vitamin E; cholesterol, retinol equivalent, K, Ca, Fe, Zn, P, Na, Mg, Mn, Cu, and Se).”

216

The experiment was divided into two stages. First, a brief presentation was given to explain

217

the principles and operation of the auction, before a real money practice auction was conducted

218

using induced values, which is standard in evaluations of goods during experimental auctions

219

(Lusk & Shogren, 2007), with a pencil to strengthen the participants’ understanding of the

220

auction mechanism. Similar to the design of the actual apple auctions, each participant was

221

endowed with a pencil, they were then asked to submit their sealed WTP to upgrade the endowed 10

ACCEPTED MANUSCRIPT 222

pencil into higher quality pencil in three consecutive rounds. The binding round was then

223

determined randomly and the auction winners paid for the upgraded pencil.

224

Second, a series of formal food auctions were conducted as follows. Step one: Initially, the

225

experimental instructions were explained in both written and oral form. Next, a unique ID

226

number, 500 g of apples with no traceability information, and 10 CNY6 were given to each

227

participant. Step two: Another 500 g of apples with a barcode were provided but the other

228

characteristics were the same. The barcode linked to a website with abbreviated information.

229

Each subject submitted their sealed truthful WTP to upgrade the given apples to apples with

230

abbreviated information. All the bids were then ranked in descending order and a random

231

number n was drawn. The highest bid, lowest bid, and random nth bid (market price) were

232

posted. Three additional rounds were repeated. Step three: Another 500 g of apples with a

233

barcode were provided (the other characteristics were remain the same) but the barcode linked to

234

detailed information, and four rounds of auction were then conducted in the same manner as step

235

two. Step four: One round was randomly selected from all eight trials. The auction winner paid

236

for the upgraded apples and all the participants received their participation fee. A questionnaire

237

was completed after the auction to obtain information from the subjects, which was used for

238

modeling their WTP.

239

3.2 Data Summary statistics for the selected sample characteristics are reported in Table 1.

240 241 242

[Insert Table 1 here]

243

6

At the time of the auction, 1$= 6.30 CNY, and the market price for apples with no traceability information was 6 CNY/500 g.

11

ACCEPTED MANUSCRIPT 244

Approximately 63% of the participants were married and 60% were females, which is

245

higher than that in China in 20137, it is reasonable considering the food buyers in China are

246

mainly females. Almost 85% of the participants had a monthly household income below 11000

247

CNY and 61% had a college or above degree. The monthly household income and education

248

level of the sample are relatively higher than common Chinese. The majority of the participants

249

self-reported their good health and about 89% of subjects were fond of apples, while 46% of the

250

subjects were concerned about news regarding food safety. The risk response of consumers can

251

be decoupled into risk perception and risk attitude according to Pennings et al. (2002). In terms

252

of risk perception, 28.4% of the participants perceived a risk that the apples they consume may

253

contain chemical residues (score < 4), 58% of the subjects did not perceive risks regarding apple

254

quality and safety (score > 4), whereas the remaining 13.6% of the participants were risk

255

perception neutral (score = 4). In terms of risk attitude, the average score was 3.49, which shows

256

that the subjects were generally risk averse. On average, most subjects (62.5%) were risk averse

257

(score < 4), about 26% of the subjects had a high willingness to accept the risk when consuming

258

apples (score > 4), whereas the remaining 11.5% of the participants were risk attitude neutral

259

(score = 4).

260 261

4. Results and discussion In this section, we present our analysis of the bids under the two treatments, the factors that

262 263

influenced WTP among consumers, and the types of information that consumers preferred.

264

4.1 Comparison of WTP among consumers

265

The average bids by subjects in each round are reported in Table 2. Among consumers, the

266

WTP for apples with abbreviated traceability information ranged from ca 1.85 CNY to 2.22 7

According to China Statistical Yearbook 2014, the ratio of females in China was 48.76% in 2013.

12

ACCEPTED MANUSCRIPT 267

CNY, whereas the average WTP for apples with detailed traceability information was ca 2.7

268

CNY. The stabilization indices for the last two rounds under the abbreviated and detailed

269

traceability information treatments were 1.28 and 1.35, and 1.47 and 1.42, respectively, which

270

shows that the bids tend to stabilize8.

271 272

[Insert Table 2 here]

273 274

The mean bids under the two treatments are reported in Table 3, where the results indicate

275

that compared with ordinary apples, consumers would be prepared to pay 34.3% and 44.5%

276

premiums for apples with abbreviated and detailed traceability information, respectively. This

277

result agrees with that reported by Zhang et al. (2012) who found that Chinese consumers place a

278

positive WTP premium on a food traceability system. Different amounts of traceability

279

information influence the behavior of Chinese consumers, where the significant result of a

280

simple mean equality t-test indicates that the Chinese consumers in the present study were

281

sensitive to the information treatments (p < 0.01). Detailed traceability information resulted in

282

higher WTP values and, on average, they were prepared to pay a 10% higher premium (about

283

0.61 CNY/500 g) for apples with detailed traceability information rather than similar apples with

284

abbreviated traceability information.

285 286

[Insert Table 3 here]

287 288

4.2 Factors that influenced the WTP premium of consumers

8

The stabilization index was calculated by dividing the mean bid prices by the standard deviation in each round (Lee et al., 2011).

13

ACCEPTED MANUSCRIPT 289

A regression model was used to investigate how various factors affected the preferences of

290

consumers for the provision of different amounts of information by a food traceability system.

291

According to Costanigro et al. (2014), the differences in the WTP can be represented by the first-

292

difference

293

, where the change in WTP is a = WTPdetailed ‒ WTPabbreviated WTP(detailed ‒ abbreviated) i i i

294

stochastic function of the demographic and individual characteristics of subjects xi, and an error

295

term μi. The random variable ∆bid(detailed ‒ abbreviated) is the observed counterpart of ∆ i

296

, where ∆bid(detailed ‒ abbreviated) , and thus = biddetailed ‒ bidabbreviated WTP(detailed ‒ abbreviated) i i i i

297

we adopted the following regression model:

298

between

the

detailed

and

abbreviated

∆bidi = β0 + β'xi + μi,

information

scenarios:

∆

μi~N(0,σ2),

299

where ∆bidi is the difference between consumer i’s bids for detailed information and

300

abbreviated information scenarios; xi is a vector of independent variables, including the

301

demographic and individual characteristics of subjects; μi is an error term, which is assumed to

302

have a normal distribution; and β' measures the partial effects of xi on E(∆bidi|xi).

303 304

The estimated parameters for the WTP model are shown in Table 4. R-squared = 0.427 and Prob>F = 0.0000. Overall, the results appear to be reasonable.

305 306

[Insert Table 4 here]

307 308

According to the regression results, educational level had a negative influence on the ∆bidi

309

of the participants. Compared with consumer who had the lowest average educational level,

310

highly educated consumers were not prepared to pay a higher premium for apples with detailed

311

traceability information compared with those with abbreviated traceability information. As in the 14

ACCEPTED MANUSCRIPT 312

second scenario four additional types of information (i.e. nutritional content, pesticide residuals,

313

contact details, logistics information) were included in the traceability system, a possible reason

314

for the above result is that the highly-educated consumers already have more nutrition related

315

knowledge of food products (Drichoutis, 2005; De Vriendt et al., 2009) and thus are more likely

316

to show less interest in added information.

317

The negative sign for self-reported health showed that the consumers with bad self-reported

318

health have a higher WTP premium for apples with detailed traceability information. For the

319

consumers with bad self-reported health, they pay more attention to the nutritional content and

320

nutritional value of food (Drichoutis et al., 2009). As is reported in Drichoutis (2005), the

321

consumers with bad self-reported health use label of nutrient content more often. Since such

322

information they interested/concerned is available within detailed traceability information, the

323

consumers with bad self-reported health would like to pay a higher premium for apples with

324

detailed information. By comparison, the consumers with good self-reported health gave a lower

325

WTP premium to apples with detailed traceability information. It is possible that the consumers

326

with good self-reported health consider that food is safe provided that a food traceability system

327

exists, regardless of the amount of information that it conveys.

328

The response of consumers to the risk related to food is likely to shape their likelihood of

329

purchasing food and their WTP for food (Pennings et al., 2002). In the present study, we also

330

found that both the risk perception and risk attitude significantly affected the WTP for food

331

among consumers, which is similar to the results obtained by de Jonge et al. (2008) and Lim et

332

al. (2014). The negative sign for risk perception indicates that consumers who perceived a risk

333

would not like to pay a higher premium for apples with detailed traceability information

334

compared with the consumers who did not perceive a risk. For consumers with higher level of

335

risk perception, they were more likely to have less confidence in the fruit quality, and would not

15

ACCEPTED MANUSCRIPT 336

like to pay a higher WTP premium to apples with detailed traceability information. In addition,

337

the positive sign for risk attitude means that consumers with a higher willingness to accept the

338

risks of eating apples will have a higher WTP for apples with detailed traceability information.

339

As risk attitude captures consumers’ willingness to consume fruits across different risky

340

situations (Schroeder et al., 2007), risk tolerance consumers will not reduce the consumption of

341

fruits even if there is a potential risk. Our result shows that risk tolerance consumers would like

342

to pay more for apples with detailed traceability information than risk averse consumers,

343

probably because the detailed traceability information helps in purchase decision and thus reduce

344

uncertainty.

345

4.3 Demand for specific food safety and quality information

346

The value of a food traceability system depends critically on the information that it conveys

347

(Aung & Chang, 2014), so the type of food safety and quality information recorded are also

348

important for the development of a food traceability system. In our study, the respondents were

349

also asked to rank their preferences for eight different types of food safety and quality

350

information (quality certificate, chemical fertilizers and pesticides used in production processes,

351

harvest date, production standards, nutrition, place of origin, producer, and circulation process),

352

and the statistical result is shown in Figure 1.

353 354

[insert Figure 1 here]

355 356

Among the eight types of information, “quality certificate” was most important and only

357

5.68% consumers were not interested in this information. A quality certificate is a readily

358

interpretable indicator of quality and it is easy to process (Hobbs et al., 2005; van Rijswijk et al.,

359

2008). The second type of information preferred by respondents (91%) was “chemical fertilizers 16

ACCEPTED MANUSCRIPT 360

and pesticides used in production processes.” This is probably because China ranks among the

361

highest users of fertilizer (Calvin et al., 2006) and pesticides (Fang & Zhu, 2014), thereby

362

resulting in excessive residues in the harvested product, which is one of the main reasons for

363

enhancing food safety in China (Fang & Zhu, 2014). The “harvest date” and “production

364

standards” both ranked third (86.36%) in the list of preferences. The four main types of

365

information differ slightly from those reported by Jin and Zhou (2014) based on a study in Japan,

366

where they found that the harvest date, production method, certification of production method,

367

and pesticides (drug) used in production were the four main types of information preferred by

368

Japanese consumers.

369

As shown in Figure 1, “nutrition” (73%), “place of origin” (64%), “producer” (52%), and

370

“circulation process” (51%) were the four main types of information that consumers did not

371

prefer. Nutrition knowledge is not as popular as easily understandable information such as

372

quality certificate, harvest date, etc. probably because most Chinese consumers have low levels

373

of nutrition knowledge. Liu et al. (2015) stated that up to date there is no educational campaign

374

to promote nutrition knowledge neither nationally nor regionally in China, most consumers

375

cannot understand scientific nutrition information well.

376 377

5. Conclusion and implications

378

Food traceability systems were originally designed to facilitate food supply chain

379

management, but they can also provide an effective medium for information provision. At

380

present, there is no consensus regarding whether a food traceability system should record

381

detailed information or simply convey abbreviated information. Cost is one of the main barriers

382

to information provision but it is very important to understand the attitudes and preference of

383

consumers regarding food traceability systems that provide different amounts of food traceability 17

ACCEPTED MANUSCRIPT 384

information. In this study, we determined the premiums that consumers are prepared to pay for

385

food traceability systems with different amounts of traceability information in China. Our results

386

have many implications for China and other countries regarding the implementation of a

387

traceability system.

388

Chinese consumers had a positive WTP for traceability with both abbreviated and detailed

389

information, but the WTP for traceability with detailed information (average premium of 44.5%)

390

was higher than that for traceability with abbreviated information (average premium of 34.3%),

391

where the difference in the premiums was about 10% (0.61 CNY/500 g). The size of the

392

premium and the difference between the premiums can provide a reference for pricing produce

393

with different amounts of traceability information.

394

In terms of the factors that affected WTP among consumers, consumers with good self-

395

reported health and highly educated consumers were not prepared to pay a higher premium for

396

traceability with detailed information. The behavior of consumers is also driven by risk attitude

397

and risk perception, and we found that consumers who are risk tolerance were more likely to pay

398

a higher premium, whereas those who perceived risk were less likely to pay for traceability with

399

detailed information. These results suggest that consumers with different social demographic

400

characteristics differed in their preference for the amount of information provided. Thus, social

401

demographic characteristics and market segmentation should be considered when deciding the

402

amount of information recorded in a food traceability system. If the products introduced into a

403

market are mainly targeted at those with poor self-reported health, a low educational level, and

404

who are not averse to risk, then a food traceability system with detailed information is preferable

405

according to our results.

406

In terms of specific information, our results showed that the most popular information

407

among Chinese consumers was a “quality certificate,” followed by details of the “chemical

18

ACCEPTED MANUSCRIPT 408

fertilizers and pesticides” used in the food production process, as well as the “harvest date.” By

409

contrast, information about the food producer and food circulation was the least preferred. The

410

information recorded by a food traceability system should meet consumer demands (Verbeke,

411

2005; Karipidis et al., 2009) because it can significantly affect perceptions of food that

412

consumers eat (Dickinson & Bailey, 2002). Our results may provide an appropriate reference for

413

policy makers and food industry stakeholders when deciding the types of information that should

414

be recorded by a food traceability system. Under the constraint of a limited cost budget, a quality

415

certificate is the most important information, followed by details of fertilizers/pesticides.

416

This study is subject to some limitations. First, we focus on citizens of Hangzhou, which

417

puts limits to generalize the findings to Chinese population. A more representative sample for

418

China is expected in future study. Second, our study differentiates food traceability system by

419

amount of information, and the results suggest that most Chinese consumers are willing to pay

420

for traceability with detailed information recorded by a food traceability system, but we do not

421

mean that the more information provided in the food traceability systems the better. Considering

422

the time constraints and information processing capacity of consumers, excess information may

423

prevent consumers from making optimal choices (Teisl & Roe, 1998; Salaün & Flores, 2001),

424

the philosophy that “more information is better” runs the risk of “information overload” (Jacoby

425

et al., 1974). Therefore, it would be useful to determine the appropriate amount of information

426

for recording in a food traceability system in future research.

427 428 429 430

Acknowledgements

19

ACCEPTED MANUSCRIPT 431

The authors would like to thank research assistant Qiyan Zeng and Yiyun Zhang for their

432

helpful support. The authors gratefully acknowledge support from the Fundamental Research

433

Funds for the Central Universities (SSEYI201102), the National Natural Science Foundation of

434

China (NNSFC-71273233, 71333011) and the Major Program of the Key Research Institute of

435

Chinese Ministry of Education (No. 15JJD790032).

436 437 438

References

439

Alfnes, F., & Rickertsen, K. (2003). European consumers' willingness to pay for US beef in

440

experimental auction markets. American Journal of Agricultural Economics, 85(2), 396-405.

441

Aung, M. M., & Chang, Y. S. (2014). Traceability in a food supply chain: Safety and quality

442 443 444 445 446 447 448 449 450

perspectives. Food control, 39: 172-184. Bai, J., Zhang, C., & Jiang, J. (2013). The role of certificate issuer on consumers’ willingness-topay for milk traceability in China. Agricultural Economics, 44(4-5), 537-544. Becker, G. M., DeGroot, M. H., & Marschak J. (1964). Measuring utility by a single-response sequential method. Behavioral science, 9(3), 226-232. Bosona, T., & Gebresenbet, G. (2013). Food traceability as an integral part of logistics management in food and agricultural supply chain. Food Control, 33(1): 32-48. Bougherara, D., & Combris, P. (2009). Eco-labelled food products: what are consumers paying for? European Review of Agricultural Economics, 36(3), 321-341.

451

Breiner, S. J. (2007). Perceptions and attitudes of cow-calf producers toward emerging

452

technologies and policy issues in the beef cattle industry. Doctoral Dissertation, Kansas State

453

University, Manhattan, Kansas. Calvin, L., Gale, F., Hu, D., & Lohmar, B. (2006). Food

454

safety improvements underway in China. Amber Waves, 4(5), 16-21. 20

ACCEPTED MANUSCRIPT 455 456

Chern, W. S., & Chang, C. (2012). Benefit evaluation of the country of origin labeling in Taiwan: Results from an auction experiment. Food Policy, 37(5), 511-519.

457

Chrysochou, P., Chryssochoidis, G., & Kehagia, O. (2009). Traceability information carriers.

458

The technology backgrounds and consumers’ perceptions of the technological solutions.

459

Appetite, 53(3), 322-331.

460

Clemens, R. (2003). Meat traceability and consumer assurance in Japan (MATRIC Briefing

461

Papers. Paper 9). Iowa: Midwest Agribusiness Trade and Information Center, Iowa State

462

University. http://lib.dr.iastate.edu/matric_briefingpapers/9/ Accessed 15.10.22.

463

Corrigan, J. R., Drichoutis, A. C., Lusk, J. L., Nayga, R. M., & Rousu,

M. C. (2012). Repeated

464

rounds with price feedback in experimental auction valuation: an adversarial collaboration.

465

American Journal of Agricultural Economics, 94(1), 97-115.

466

Costanigro, M., Kroll, S., Thilmany, D., & Bunning, M. (2014). Is it love for local/organic or

467

hate for conventional? Asymmetric effects of information and taste on label preferences in an

468

experimental auction. Food Quality and Preference, 31, 94-105.

469

de Jonge, J., van Trijp, H., Goddard, E., & Frewer, L. (2008). Consumer confidence in the safety

470

of food in Canada and the Netherlands: The validation of a generic framework. Food Quality

471

and Preference, 19(5), 439-451.

472

De Vriendt, T., Matthys, C., Verbeke, W., Pynaert, I., & De Henauw, S. (2009). Determinants of

473

nutrition knowledge in young and middle-aged Belgian women and the association with their

474

dietary behaviour. Appetite, 52(3), 788-792.

475 476 477 478

Dickinson, D. L. & Bailey, D. (2002). Meat traceability: Are U.S. consumers willing to pay for it? Journal of Agricultural and Resource Economics, 27(2), 348-364. Drichoutis, A. C. (2005). Nutrition knowledge and consumer use of nutritional food labels. European Review of Agriculture Economics, 32(1), 93-118.

21

ACCEPTED MANUSCRIPT 479 480 481 482

Drichoutis, A. C., Lazaridis, P., & Nayga, R. M. (2009). Would consumers value food-awayfrom-home products with nutritional labels? Agribusiness, 25(4), 550-575. Fang, B., & Zhu, X. (2014). High content of five heavy metals in four fruits: Evidence from a case study of Pujiang County, Zhejiang Province, China. Food Control, 39, 62-67.

483

Feng, J., Fu, Z., Wang, Z., Xu, M, & Zhang, X. (2013). Development and evaluation on a RFID-

484

based traceability system for cattle/beef quality safety in China. Food Control, 31(2), 314-325.

485

Food Standards Agency (2002). Traceability in the food chain: A preliminary study. Food

486 487 488

Standards Agency: Food Chain Strategy Division. Accessed 14.02.14. Gellynck, X., & Verbeke, W. (2011). Consumer perception of traceability in the meat chain. Agrarwirtschaft, 50(6), 368-374.

489

Golan, E., Krissoff, B., Kuchler, F., Calvin, L., Nelson, K., & Price, G. (2004). Traceability in

490

the U.S. food supply: economic theory and industry studies. Agricultural Economic Report

491

Number

492

https://tgmsystem.com/wp-content/uploads/2014/11/USDA-Traceability-Report.pdf/ Accessed

493

14.04.20.

494 495 496 497 498 499 500 501

830,

US

Department

of

Agriculture,

Economic

Research

Service.

Golan, E., Krissoff, B., Kuchler, F., Nelson, K., Price, G., & Calvin, L. (2003). Traceability in the U.S. food supply: dead end or superhighway? Choices, 18(2), 17-20. Gracia, A., Loureiro, M. L., & Nayga, Jr. R. M. (2011). Valuing an EU animal welfare label using experimental auctions. Agricultural Economics, 42(6), 669-677. Grunert, K. G., Bech-Larsen, T., & Bredahl, L. (2000). Three issues in consumer quality perception and acceptance of dairy products. International Dairy Journal, 10(8), 575-584. Hobbs, J. E. (2004). Information asymmetry and the role of traceability systems. Agribusiness, 20(4), 397-415.

22

ACCEPTED MANUSCRIPT 502

Hobbs, J. E., Bailey, D., Dickinson, D. L., & Haghiri, M. (2005). Traceability in the Canadian

503

red meat sector: Do consumers care? Canadian Journal of Agricultural Economics, 53(1), 47-

504

65.

505

Houghton, J. R., Rowe, G., Frewer, L. J., Van Kleef, E., Chryssochoidis, G., & Kehagia, O.,

506

Korzen-Bohr, S., Lassen, J., Pfenning U., & Strada, A. (2008). The quality of food risk

507

management in Europe: Perspectives and priorities. Food Policy, 33(1), 13-26.

508 509 510 511 512 513

Hu, J., Zhang, X., Moga, L. M., & Neculita, M. (2013). Modeling and implementation of the vegetable supply chain traceability system. Food Control, 30(1), 341-353. Jacoby, J., Speller, D. E., & Kohn, C. A. (1974). Brand choice behavior as a function of information load. Journal of Marketing Research, 11(1), 63-69. Jin, S., & Zhou, L. (2014). Consumer interest in information provided by food traceability systems in Japan. Food Quality and Preference, 36, 144-152.

514

Karipidis, P., Athanassiadis, K., Aggelopoulos, S., & Giompliakis, E. (2009). Factors affecting

515

the adoption of quality assurance systems in small food enterprises. Food Control, 20(2): 93-

516

98.

517

Kehagia, O., Chrysochou, P., Chryssochoidis, G., Krystallis, A., & Linardakis, M. (2007).

518

European consumers' perceptions, definitions and expectations of traceability and the

519

importance of labels, and the differences in these perceptions by product type. European

520

Society for Rural Sociology, 47(4), 400-416.

521

Lee, J. Y., Han, D. B., Nayga, R. M., & Lim, S. S. (2011). Valuing traceability of imported beef

522

in Korea: an experimental auction approach. Australian Journal of Agricultural and Resource

523

Economics, 55(3), 360-373.

23

ACCEPTED MANUSCRIPT 524

Liao, P., Chang, H., & Chang, C. (2011). Why is the food traceability system unsuccessful in

525

Taiwan? Empirical evidence from a national survey of fruit and vegetable farmers. Food

526

Policy, 36(5), 686-693.

527

Lim, K. H., Hu, W., Maynard, L. J., & Goddard, E. (2014). A taste for safer beef? How much

528

does consumers’ perceived risk influence willingness to pay for country-of-origin labeled

529

beef. Agribusiness, 30(1), 17-30.

530 531 532 533

List, J. A. & Shogren, J. F. (1999). Price information and bidding behavior in repeated secondprice auctions. American Journal of Agricultural Economics, 81(4), 942-949. Liu, R., Hoefkens, C., & Verbeke, W. (2015). Chinese consumers’ understanding and use of a food nutrition label and their determinants. Food Quality and Preference, 41, 103-111.

534

Loureiro, M. L., & Umberger, W. J. (2007). A choice experiment model for beef: What U.S.

535

consumer responses tell us about relative preferences for food safety, country-of-origin

536

labeling and traceability. Food Policy, 32(4), 496-514.

537 538 539 540

Lu, J., Wu, L., Wang, S., Xu, L., & Xu, L. (2016). Consumer preference and demand for traceable food attributes. British Food Journal, 118(9), 2140-2156. Lusk, J. L. (2003). Using experimental auctions for marketing applications. Journal of Agricultural and Applied Economics, 35(2): 349–360.

541

Lusk, J. L., Alexander, C., & Rousu, M. C. (2007). Designing experimental auctions for

542

marketing research: the effect of values, distributions, and mechanisms on incentives for

543

truthful bidding. Review of Marketing Science, 5.

544 545

Lusk, J. L., & Shogren, J. (2007). Experimental Auctions: Methods and Applications in Economic and Marketing Research. New York: Cambridge University Press.

24

ACCEPTED MANUSCRIPT 546

Melton, B. E., Huffman, W. E., Shogren, J. F., & Fox, J. A. (1996). Consumer preferences for

547

fresh food items with multiple quality attributes: evidence from an experimental auction of

548

pork chops. American Journal of Agricultural Economics, 78(4), 916-923.

549

Menozzi, D., Halawany-Darson, R., Mora, C., & Giraud, G. (2015). Motives towards traceable

550

food choice: A comparison between French and Italian consumers. Food Control, 49, 40-48.

551

Ortega, D. L., Wang, H. H., Wu, L., & Olynk, N. J. (2011). Modeling heterogeneity in consumer

552

preferences for select food safety attributes in China. Food Policy, 36(2), 318-324.

553

Pennings, J. M. E., Wansink, B., & Meulenberg, M. T. G. (2002). A note on modeling consumer

554

reactions to a crisis: The case of the mad cow disease. International Journal of Research in

555

Marketing, 19(1), 91-100.

556

Peres, B., Barlet, N., Loiseau, G., & Montet, D. (2007). Review of the current methods of

557

analytical traceability allowing determination of the origin of foodstuffs. Food Control, 18(3):

558

228-235.

559

Pieniak, Z., Verbeke, W., Vermeir, I., Bruns, K., & Olsen, S. O. (2007). Consumer interest in

560

fish information and labelling: Exploratory insights. Journal of International Food & Agri-

561

Business Marketing, 19(2-3), 117-141.

562 563 564

Salaün, Y., & Flores, K. (2001). Information quality: meeting the needs of the consumer. International Journal of Information Management, 21(1), 21-37. Schroeder, T. C., Blasi, D. A., Brester, G. W., Crosby, C., Dhuyvetter, K. C., Freeborn, J.,

565

Pendell, D. L., Smith, G. C., Stroade, J., & Tonsor, G. T. (2009). Benefit cost analysis of the

566

national animal identification system. Report for U.S. Department of Agriculture, Animal and

567

Plant Health Inspection.

568

http://animalid.aphis.usda.gov/nais/naislibrary/documents/plans_reports/Benefit_Cost_Analysi

569

s_NAIS.pdf

Accessed 14.08.07.

25

ACCEPTED MANUSCRIPT 570 571

Schulz, L. L., & Tonsor, G. T. (2010). Cow-calf producer perceptions regarding individual animal traceability. Journal of Agricultural and Applied Economics, 42(4), 659-677.

572

Schroeder, T. C., Tonsor, G. T., Pennings, J. M., & Mintert, J. (2007). Consumer food safety risk

573

perceptions and attitudes: impacts on beef consumption across countries. The BE Journal of

574

Economic Analysis & Policy, 7(1), Article 65.

575 576 577 578

Shogren, J. F., Margolis, M., Koo, C., & List, J. A. (2001). A random nth-price auction. Journal of Economic Behavior & Organization, 46(4), 409-421. Smith, G. C., Tatum, J. D., Belk, K. E., Scanga, J. A., Grandin, T., & Sofos, J. N. (2005). Traceability from a U.S. perspective. Meat Science, 71(1), 174-193.

579

Souza-Monteiro, D. M., & Caswell, J. A. (2004). The economics of implementing traceability in

580

beef supply chains: trends in major producing and trading countries. Working Paper No. 2004-

581

6,

582

http://papers.ssrn.com/sol3/papers.cfm?abstract_id=560067 Accessed 14.04.22.

583 584 585 586

Department

of

Resource

Economics,

University

of

Massachusetts

Amherst.

Teisl, M. F., & Roe, B. (1998). The economics of labeling: an overview of issues for health and environmental disclosure. Agricultural and Resource Economics Review, 27, 140-150. Ubilava, D., & Foster, K. (2009). Quality certification vs. product traceability: Consumer preferences for informational attributes of pork in Georgia. Food Policy, 34(3), 305-310.

587

van Rijswijk, W., Frewer, L. J., Menozzi, D., & Faioli, G. (2008). Consumer perceptions of

588

traceability: a cross-national comparison of the associated benefits. Food Quality and

589

Preference, 19(5), 452-464.

590

van Rijswijk, W., & Frewer, L. J. (2012). Consumer needs and requirements for food and

591

ingredient traceability information. International Journal of Consumer Studies, 36(3), 282-

592

290.

26

ACCEPTED MANUSCRIPT 593 594

Verbeke, W. (2005) Agriculture and the food industry in the information age. European Review of Agricultural Economics, 32(3), 347-368.

595

Verbeke, W., & Ward, R. W. (2006). Consumer interest in information cues denoting quality,

596

traceability and origin: an application of ordered probit models to beef labels. Food Quality

597

and Preference, 17(6), 453-467.

598 599 600 601

Vickrey, W. (1961). Counterspeculation, auctions, and competitive sealed tenders. The Journal of Finance, 16(1): 8-37. Wu, L., Wang, H., Zhu, D., Hu, W., & Wang, S. (2016). Chinese consumers' willingness to pay for pork traceability information-the case of Wuxi. Agricultural Economics, 47(1), 71-79.

602

Wu, L., Xu, L., Zhu, D., & Wang, X. (2012). Factors affecting consumer willingness to pay for

603

certified traceable food in Jiangsu Province of China. Canadian Journal of Agricultural

604

Economics, 60(3), 317-333.

605 606

Zhang, C., Bai, J., & Wahl, T. I. (2012). Consumers' willingness to pay for traceable pork, milk, and cooking oil in Nanjing, China. Food Control, 27(1), 21-28

27

ACCEPTED MANUSCRIPT

1

Figure Quality certificate

94.32%

Chemical fertilizers and pesticides used in production processes

90.91%

Production standards

86.36%

Harvest date

86.36% 72.73%

Nutrition

63.64%

Place of origin Producer

52.27%

Circulation process

51.14% 0%

2

Do not want to know

20%

Do not care

40%

60%

80%

100%

Always want to know

3 4 5

Fig. 1. Demand for specific safety and quality information provided by food traceability systems

6 7

1

ACCEPTED MANUSCRIPT Tables

1

Table 1. Summary statistics based on the selected sample characteristics

2

Variables

Definition and coding

Percent

Mean

SD

Gender

0 = female; 1 = male

-

0.40

0.492

Age

Age of participant ( 18 years)

-

39.14

18.007

Marital status

0 = unmarried; 1 = married

-

0.63

0.487

Children

If the family have any children below 18 years of age: yes = 1, no = 0

-

0.38

0.487

-

-

-

-

Educational level Education

1 = Junior high school or lower

20.45%

2 = Senior high school or technical secondary school

18.18%

3 = Bachelor or college degree

47.73%

4 = Masters or above

13.64%

Household income per month (CNY) 1 = less than 5000 2 = 5000–6999

25.01% 23.86%

3 = 7000–8999

18.18%

4 = 9000–10999

18.18%

5 = 11000 and above.

14.77%

Self-reported health

1 = healthy, 0 = other

-

0.93

0.254

Preference for apples

Degree of preference for apples:1 = like, 0 = other

-

0.89

0.318

News

Concerned with news about the safety of agricultural products:

-

0.47

0.498

-

4.44

1.065

-

4.63

2.086

-

4.39

2.120

-

4.31

2.108

-

3.49

1.044

Income

Risk perception

Consumer 1 = care, 0 perception = other of food safety: 1 = Did not perceive a risk 7 = Perceived a risk At present, the fruit market is generally safe, although incidents such as excessive pesticide residual, and the illegal use of preservatives and industrial wax have occurred occasionally. At present, fruits containing chemical substances comprise a large proportion of the fruit market and they are very harmful. At present, fruits containing chemical substances do little harm to the health of consumers.

Risk attitude

Consumer attitude towards food safety: 1 = Risk aversion 7 = Risk tolerance

1

ACCEPTED MANUSCRIPT Although I often hear about bad news such as excessive pesticide residues, and the illegal use of preservatives and industrial wax, it

-

3.71

2.295

-

3.08

2.118

-

3.67

2.159

does not affect my fruit purchasing behavior. I never worry about pesticide residues, preservatives, and industrial wax when eating fruit. I cannot tolerate the unacceptable health risk when eating fruit containing chemical substances.

3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 2

ACCEPTED MANUSCRIPT 23

Table 2. Mean bids in each round and the stabilization index for bid price Treatment

24

Abbreviated information

Detailed information

Trial

Trial

1

2

3

4

1

2

3

4

Minimum

0

0

0

0

0

0

0

0

Maximum

6.8

6.5

6.5

6.0

9.0

8.0

8.0

8.0

Median

1.5

2.0

2.0

2.0

2.5

2.2

2.0

2.2

Mean

1.85

2.04

2.13

2.22

2.67

2.71

2.69

2.63

Standard deviation (SD)

1.43

1.40

1.66

1.64

1.72

1.80

1.84

1.85

Mean/SD

1.29

1.46

1.28

1.35

1.55

1.51

1.47

1.42

Unit: CNY, N = 88

25 26 27 28 29 30 31 32 33 34 35 36 37 38 3

ACCEPTED MANUSCRIPT 39

Table 3. Mean bids and t-test for equality of the WTP means between the information treatments 　

　

Mean

Median Standard Deviation Mean WTP Difference t-value

Abbreviated

2.06

2.0

1.537

Detailed

2.67

2.2

1.796

Information 40

0.61***

Note: *** Significant at 1% level; average price of ordinary apples sold on markets was 6 CNY/500 g.

41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 4

9.30

ACCEPTED MANUSCRIPT 61

Table 4. Regression results on bids ∆bid

Coefficient

Robust Std. Err.

Gender

0.244

0.269

Age

–0.009

0.013

Marital status

0.735

0.570

–0.095

0.323

-

-

–0.376

0.405

Undergraduate

–0.967**

0.445

Masters and above

–1.539***

0.523

-

-

5000–6999

–0.149

0.346

7000–8999

–0.132

0.416

9000–10999

–0.030

0.469

11000 and above

0.381

0.441

Self-reported health

–0.959*

0.510

Preference for apples

0.214

0.266

News

0.461

0.303

Risk perception

–0.219**

0.110

Risk attitude

0.248**

0.123

3.036

1.006

Children under 18 years of age Education Junior high or lower High school

Income less than 5000

Constant N = 88

F(16, 71) = 4.980

Prob > F

= 0.000

R-squared = 0.427 62

*, **, *** Significant at the 10%, 5%, and 1% levels, respectively. 5