Assessment of food fraud vulnerability in the spices chain: An explorative study

Accepted Manuscript Assessment of food fraud vulnerability in the spices chain: an explorative study

I.C.J. Silvis, S.M. van Ruth, H.J. van der Fels-Klerx, P.A. Luning PII:

S0956-7135(17)30262-1

DOI:

10.1016/j.foodcont.2017.05.019

Reference:

JFCO 5625

To appear in:

Food Control

Received Date:

30 March 2017

Revised Date:

12 May 2017

Accepted Date:

13 May 2017

Please cite this article as: I.C.J. Silvis, S.M. van Ruth, H.J. van der Fels-Klerx, P.A. Luning, Assessment of food fraud vulnerability in the spices chain: an explorative study, Food Control (2017), doi: 10.1016/j.foodcont.2017.05.019

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

Fraud assessment reveals vulnerabilities due to opportunities, motivations, and controls Interviewed spice chain actors assigned food fraud vulnerability overall as medium

Key risks are simple adulteration, detection difficulty, price and market competition

Hard fraud mitigation vary considerably among interviewed spice actors

ACCEPTED MANUSCRIPT 1

Assessment of food fraud vulnerability in the spices chain: an explorative study

2

I.C.J. Silvisab ([email protected]), S.M. van Ruthab* ([email protected]) (corresponding author), H.J. van der Fels-Klerxa ([email protected]), P.A. Luningb ([email protected])

3 4 5

a RIKILT

7

Wageningen University and Research, P.O. Box 230, 6700 AE Wageningen, The Netherlands

8

b Food

6

9

Quality and Design Group, Wageningen University and Research, P.O. Box 17 / bode 30 6700 AA Wageningen, The Netherlands

10

* Corresponding author

11

Keywords: control measures, fraud, fraud mitigation, fraud indicators, motivations, opportunities, spices, vulnerability

12 13

1

ACCEPTED MANUSCRIPT 14

Abstract: Recent scandals have increased the need to strengthen companies’ ability to combat fraud within their own organizations and across their supply chain. Vulnerability assessments are a first step towards the inventory of fraud vulnerability and fraud mitigation plans. Spices are reported frequently in the international food fraud databases. In the current study the fraud vulnerabilities of various actors in the spices supply chain were examined. The SSAFE food fraud vulnerability assessment tool, which comprises of 50 indicators categorized in opportunities, motivations, and control measures was applied for getting insight into these fraud vulnerabilities. Eight companies participated in the study: a trader, two importers, two business to business companies, and three business-to- business/ business-to-consumer enterprises. The ease to adulterate spices combined with the complexity of fraud detection create considerable opportunities to commit fraud (high vulnerability), whereas opportunities associated with supply chain transparency and fraudulent incidences in the past were judged as medium vulnerable. The high competition level in the sector together with the high added value of spices are perceived as important economic drivers to commit fraud (high vulnerability). Cultural/behavioural factors such as ethical business culture were considered to contribute to the actual fraud vulnerability to a lesser extent. The implementation of both the hard and soft control measures varied widely among the actors. Hard fraud specific measures are merely lacking or are at a very basic level. For soft control measures of the own company, the scores were higher. From the results of the full assessments can be concluded that the various actors perceived the level of food fraud vulnerability in the spices chain as medium vulnerable.

2

ACCEPTED MANUSCRIPT 15

1. Introduction

16 17

Food fraud scandals and issues in the last few years have reinforced the need to

18

understand the vulnerability to fraud in food chains. The food industry is generally

19

vulnerable to crime and the spice industry is mentioned as one of the most vulnerable

20

ones, in addition to meat, fish, and olive oil industries (Morling & McNaughton, 2016).

21

For example, in 2014, ground peanut shells were discovered in powdered cumin.

22

This caused a major recall because of the allergenic properties of the peanut material,

23

which is a severe risk to those that suffer from a peanut allergy (Sayers et al., 2016).

24

Investigations revealed that fraudulent activity and not accidental contamination was

25

behind the incident. The main motivation of the company was the economic benefit

26

from the addition of cheaper bulk material to the premium quality cumin.

27

Food fraud involves the deliberate substitution, addition, tampering or

28

misrepresentation of food, food ingredients or food packaging, or false or misleading

29

statements made about a product for economic gain (Spink & Moyer, 2011a). This

30

definition has been widely adopted by various authors (e.g. Pustjens, Weesepoel, &

31

van Ruth, 2016; Avery, 2014; GFSI, 2014), and by internationally acknowledged

32

bodies such as the Global Food Safety Initiative (GFSI). The addition of a cheaper

33

ingredient is the most common type of economically motivated adulteration (EMA)

34

(Capuano & van Ruth, 2012), which can result in thousands of euros from illegal

35

profits (Moyer, DeVries, & Spink, 2016). Food fraud can be committed by any

36

individual person or group involved in the whole supply chain, including suppliers,

37

food manufacturers, retailers and importers (Johnson, 2014). Adulteration is the

38

preparation of foods for sale by replacing valuable with less valuable ingredients or

39

constituents.

3

ACCEPTED MANUSCRIPT 40

In general, herbs and spices represent an attractive category for potential offenders,

41

because the products have a high value by weight and consumers have a limited

42

capacity to detect adulteration (Schaarschmidt, 2016; Moore, Spink, & Lipp, 2012).

43

Common authenticity issues associated with spices are the addition of lower value

44

product foreign and product own material (Peter, 2011), which may include addition

45

of unapproved ‘enhancements’, such as dyes (Haughey, Galvin-King, Ho, Bell, &

46

Elliott, 2015) to cover up the extension. Ground spices are particularly prone to

47

adulteration, because the milling or grinding step changes the shape of both the

48

spice and adulterant to a powder, which makes it difficult to detect adulterants in the

49

final product.

50

Although it is the governments’ responsibility to set clear legal requirements it is the

51

responsibility of the industry to mitigate food fraud risks (Spink & Moyer, 2011b).

52

However, such measures are not yet widely adopted in current food safety

53

management systems. In the past few years, several initiatives to analyse, measure

54

and/or mitigate food fraud risks have been developed because of the raised

55

awareness. For example, the U.S. Pharmacopeia Convention (USP) developed the

56

USP tool to assist food industries and regulators in developing and applying

57

preventive management systems to identify the most vulnerable ingredients within

58

their supply chains and to choose valid situation-specific mitigation measures (USP,

59

2014). Grocery Manufacturers Association (GMA) established a tool for the purpose

60

of assuring the integrity of brand and safety of food products (Kerney, 2010).

61

Moreover, the British Retail Consortium (BRC) version 7, a private food safety

62

standard, added a module on food fraud and provides food companies guidance on

63

how to do a vulnerability assessment (BRC, 2015). Furthermore, SSAFE (2016) has

64

published a science-based food fraud vulnerability self-assessment tool (SSAFE

4

ACCEPTED MANUSCRIPT 65

FFVA tool), which is based on the routine activities theory (Cohen & Felson, 2016). It

66

consists of 50 questions which consider the three theory’s key elements:

67

opportunities (suitable target), motivations (motivated offender), and control

68

measures, the scientific background has been reported by van Ruth, Huisman, and

69

Luning (2017).

70

In principle, the SSAFE FFVA tool is developed as a basis for companies to self-

71

assess their business, but it can also be used to compare companies (multiple

72

respondents) and to analyse a specific chain.

73

The aim of the current study is to get insight in potential fraud vulnerabilities of

74

various actors in the spices supply chain by applying this new tool.

75 76

2. Materials and methods

77 78

2.1. The Food Fraud Vulnerability Assessment (FFVA) approach

79 80

2.1.1.Theoretical aspects of the FFVA

81

The principal structure of the FFVA is based on the routine activities theory (Cohen &

82

Felson, 2016) and the “design rules” as used in the development of diagnostic tools

83

for Food Safety Management System (FSMS) assessment (Kirezieva, Jacxsens,

84

Uyttendaele, Van Boekel, & Luning, 2013; Luning et al., 2009; Luning, Bango,

85

Kussaga, Rovira, & Marcelis, 2008). The routine activities theory defines the three

86

key elements leading to crime: a suitable target, a motivated offender, and the

87

absence of guardianship. These key elements were modified to suit food fraud and

88

are the centre of the FFVA: i.e. opportunities, motivations and control measures. The

89

“design rules” include focus on key factors/activities, identify indicators to analyse

5

ACCEPTED MANUSCRIPT 90

crucial aspects of these factors/activities, formulate questions linked to the indicators,

91

and develop grids to enable a differentiated assessment. Grids depict typical

92

descriptions that reflect for example, a high, medium, or low risk situation for the

93

particular factor/activity. The situations are linked to a score system to enable the

94

development of spider web diagrams to visualise the profiles (Luning et al., 2011;

95

Sampers et al., 2010). The overall principle of the FFVA tool is reflected in the

96

formula: opportunities x motivations x control measures = actual fraud vulnerability.

97

So, more opportunities and motivations will increase fraud vulnerability, whereas

98

control measures can counteract these vulnerabilities. The terms “risk” and

99

“vulnerability” are used interchangeably and are therefore defined explicitly. The

100

following definition of vulnerability applies and originate from USA food regulations

101

(DHS, 2015): “A physical feature or operational attribute that renders an entity open

102

to exploitation or susceptible to a given hazard.”

103

The tool was tested, discussed and adapted based on multiple workshops in The

104

Netherlands (Zaandam), USA (Washington), and Singapore (Singapore) with

105

representatives of global food industry actors.

106 107

2.1.2. Practical aspects of the FFVA

108

The FFVA consists of 50 indicators (Table 1) each with a related question and

109

corresponding assessment grid to enable companies to judge their actual situation

110

with respect to the key risk factors related to opportunities, motivations, and control

111

measures, which provide an overall profile of their fraud vulnerability. Potential

112

opportunities, motivations, and control measures for food fraud are assessed related

113

to both the internal organization and the external environment of the company. The

114

environment consists of multiple levels: i.e. the company, the direct suppliers and

6

ACCEPTED MANUSCRIPT 115

customers, the industry segment, and the national and/or international environment.

116

The various environmental levels are all considered in the FFVA.

117

Opportunities related fraud factors of raw material and final product include

118

indicators, such as the complexity of adulterating spices and whether the technology

119

to adulterate is common knowledge or complex. In addition to these technical

120

indicators, there are indicators to analyse opportunities in time and space, such as

121

the accessibility to materials in production and the transparency of the network. The

122

questions and answers have the following template. The question linked to the

123

indicator “complexity of adulteration” reads: “Is it simple or complex to adulterate the

124

raw material”? The assumption is that easy alteration of the composition of raw

125

materials provides opportunities for potential offenders to commit fraud. Three

126

answer options are provided, one of which need to be selected. Low vulnerability

127

answer option 1 is: “Composition of the materials cannot be modified and products

128

can only be replaced, i.e. it concerns large objects such as fruit”. Medium

129

vulnerability answer option 2 is: “Composition of the raw materials can be modified by

130

mixing with low-quality product-own material or foreign material, i.e. as is feasible

131

with grinded products (e.g. powders, grinded beef, etc.)” and high vulnerability

132

answer option 3 is: “Composition of the raw materials can be modified by mixing with

133

low-quality or foreign material (e.g. powders, ground meat, etc.) and by altering

134

valuable food components (e.g. protein content)”.

135

Motivations related fraud factors concern economic aspects as well as cultural and

136

behavioural facets. For instance, prices, supply and demand, and value-adding

137

attributes of the materials are important economic factors, as well as the level of

138

competition in the sector and the economic health of the business. Behaviour and

139

culture related aspects include for instance business strategy, ethical business

7

ACCEPTED MANUSCRIPT 140

culture, and corruption level of the country in which the company and/or supplier is

141

based. These factors can enhance fraudsters’ motivations to commit fraud.

142

The control measures are divided in soft and hard control ones. Hard control

143

measures are well observable and can be tested (Drechsler, Halff, Huisman, & Post,

144

2012). They affect the ‘hard’ aspects of an organization such as planning, control,

145

tasks and responsibilities. Soft controls are non-tangible behaviour influential factors

146

in an organization and focuses on e.g. personality of employees and behaviour. The

147

soft control measures can influence motives, loyalty, integrity, inspiration, norms and

148

values of employees and are on a more personal level. A subdivision is provided for

149

the environmental layers, i.e. the internal hard and soft controls, and the external

150

controls at the level of the direct suppliers/customers and the wider (inter) national

151

environment including law enforcement.

152 153

2.2. Case study design: the spices chain network

154 155

2.2.1. Selection and characteristics of respondents

156

Eight companies, members of the European Spices Association (ESA), participated

157

in the study. Seven were based in the Netherlands and one in Germany. Six

158

interviews were conducted face-to-face and two assessments were carried out by

159

online survey. The characteristics of the companies and the participants are listed in

160

Table 2. Altogether, the eight companies represent trader, importer, business-to-

161

business enterprises (b2b), and business to business/business-to-consumers

162

enterprises (b2b/b2c) in the spices chain. Some of their processing activities overlap,

163

but the spices they trade or produce may differ.

164

8

ACCEPTED MANUSCRIPT 165

2.2.2. The vulnerability assessments

166

The original questions in English were translated into Dutch, and some questions

167

were adjusted to spices. For example, the word ‘raw material’ in the survey was

168

substituted with ‘spice’. A questionnaire was sent by e-mail one week prior to the

169

interview to prepare respondents for the face-to-face interview. They had time to

170

consult additional documents and ask experts in their organization about certain

171

questions. This was necessary for the interviews to ease the conversation with the

172

representatives of the spice companies. The interviewer asked the 50 questions one

173

by one. The interviewer interpreted the answers and allocated the answer to one of

174

the options in the grid, which was discussed with the respondent. The duration of the

175

interview was between 1.5 and 2 hours. The first author performed the face-to-face

176

interviews and all were voice recorded with permission of the respondents. The

177

conversations were replayed for data analysis.

178 179

2.2.3. Data analysis

180

After the interview was completed, the options were transformed to the score system,

181

to enable a frequency analysis. A high vulnerability situation for the opportunities and

182

motivations corresponds to a score of 3 (e.g. the knowledge required for adulteration

183

is generally available). A medium vulnerability situation obtains a score of 2 (e.g.

184

advanced technologies, methods, facilities and knowledge are required to adulterate

185

the raw materials) and a low vulnerability situation corresponds to a score 1 (i.e.

186

technologies and/or methods to adulterate the raw materials are neither available,

187

known, or reported). For the control measures, a score of 1 is assigned to high

188

vulnerability situation (i.e. no specific fraud focus in control), a score of 2 to a medium

189

vulnerability situation (e.g. some basic/simple fraud related measures in place), and 3

9

ACCEPTED MANUSCRIPT 190

to answers linked to a low vulnerability situation (e.g. fraud dedicated measures in

191

place). For all 50 questions, the most frequently given answer (i.e. the mode) to a

192

certain situation (and corresponding score) was determined.

193 194

3. Results and discussion

195

3.1. Overall food fraud vulnerability profiles

196

The results of the FFVA are presented in the spider web diagrams in Fig. 1 showing

197

the scores for the most frequently given (mode) answers for each indicator. The

198

indicators are grouped per category, “opportunities”, “motivations” and “control

199

measures”. The dashed line shows the second score with the highest frequency, in

200

case of identical frequencies for two answers (ties). A larger surface corresponds

201

with a higher fraud vulnerability. For the control measures, the larger the surface area,

202

the more measures are in place and the more fraud specific/dedicated they are.

203

Overall, the respondents scored between 2-3 (medium to high vulnerability) for the

204

opportunities, and scored between 1-2 (low to medium) for the motivations. For the

205

control measures mean scores distributed widely from 1 to 3. Eleven indicators out of

206

19 scored 1 (high vulnerability, low level of control) indicating that fraud specific

207

measures are merely lacking or are at a very basic level. For soft control measures of

208

the own company, the scores were higher (low vulnerability, high level of control).

209 210

3.2. Food fraud profiles of individual actors

211

The individual scores for all indicators and actors are presented in Table 3. The

212

scores for the low, medium, and high vulnerability situations for the opportunities,

213

motivations, and control measures related indicators are coloured green, orange, and

214

red respectively. The actors in this case study operated independently from each

10

ACCEPTED MANUSCRIPT 215

other and the importers were not the suppliers of the other actors presented in the

216

table.

217 218

3.2.1. Opportunities- related fraud factors

219

Technical opportunities

220

All the respondents rated the technical opportunities (indicator 1-5) as medium to

221

high vulnerability,demonstrated with red (high vulnerable) and orange (medium

222

vulnerable) per actor in the table. This is in line with the fact that spices are

223

commonly easy to adulterate when they are milled (Everstine, Spink, & Kennedy,

224

2013). The technology required for adulteration is generally available and is not

225

complex. When the spices are milled with a potential adulterant, the shape is similar

226

and detection of adulterant material requires advanced analytical techniques

227

(Sasikumar, Swetha, Parvathy, & Sheeja, 2016). One of the reasons why companies

228

buy spices from their suppliers in their whole form (not milled) is because they wish to

229

ensure that the material is free from adulterants. To the “complexity of counterfeiting”

230

indicators, (indicators 6-7) different scores were assigned. Only b2b/b2c (h)

231

answered with a score 1 (low vulnerability). However, counterfeiting is a full imitation

232

of the genuine product passed off deceptively as genuine. This kind of full

233

replacement of spices with other (non) plant-based material is not an issue, only

234

partial replacement and mixing.

235 236

Opportunities in time and space

237

“Production lines/processing activities” (indicator 8) scored 1 (low vulnerability) for all

238

companies, except from trader (a) (medium vulnerable) but all companies have

239

specific requirements in terms of food defence. The accessibility should be strictly

11

ACCEPTED MANUSCRIPT 240

authorised to particular persons to prevent intentionally contamination by people who

241

want to do harm (Guide to developing a food defence plan for Food Processing

242

Plants, 2008). All participants mentioned that their facilities allow little interference

243

and modification between batches and accessibility for unauthorised personnel

244

during day and night is not possible. “Transparency chain network” (indicator 9),

245

scored 2 (medium vulnerability) by most respondents, because they consider their

246

chain to some extent transparent. However, trader(a) and importer(b) – and (c) at the

247

beginning of the chain assigned score 3 (high vulnerability), because they perceive

248

their supply chain as not being transparent. Transparency in the chain is important,

249

because without visibility into the supply chains and due to the dispersed nature of

250

today’s supply chains, there are multiple opportunities to commit fraud (Manning &

251

Soon, 2014). Supply chain transparency cannot be easily achieved, because it

252

requires a solid foundation and continuous improvement over time (Linich, 2014).

253

According to Linich (2014), a four-step plan could support in mitigating risks on fraud

254

due to lack of chain transparency. This plan includes identifying and prioritizing risks;

255

visualizing risks; using transparency levers to close information gaps and managing

256

and monitoring resulting information. With respect to “Fraudulent incidences in past”

257

(indicators 10-11), only one company b2b (d) assigned a score 1 (low vulnerability).

258

All other respondents assigned a score 2 (medium vulnerability), because they

259

acknowledged fraudulent incidences that occurred in the past, such as the addition of

260

the harmful adulterations with the forbidden colorant Sudan red in capsicum products

261

and the peanut-shell in cumin case. Other examples of less harmful adulterations are

262

low quality pepper and various kinds of foreign matter in whole or ground pepper,

263

and the removal of essential oil in nutmeg (Peter, 2011).

264

3.2.2. Motivations-related fraud factors

12

ACCEPTED MANUSCRIPT 265

Economic drivers

266

“Supply and pricing raw materials” (indicator 12) scored 3. It is considered as highly

267

vulnerable, because the prices of spices by weight are high compared to other food

268

materials and they can vary considerably. It is a seasonal product and the quality is

269

affected by multiple factors such as climate, harvesting and large variation is

270

common (Hübert, Tiebe, & Banach, 2016). Spices also contain valuable contents

271

such as volatile oils, which are important value determinants. Measuring the volatile

272

oil (V/O) helps to identify whether the spice has been adulterated, such as addition of

273

foreign materials; addition of low quality materials or addition de-oiled or defatted

274

material (spent). Other more spice specific compounds are piperine in black pepper

275

and safranol in saffron (Peter, 2011).The indicator 13 “Valuable components or

276

attributes raw materials” scored 3 (high vulnerability) for all actors, because the value

277

of the spices is largely determined by the purity. Absence of impurities is a measure

278

of the amount of foreign and extraneous matter, for example insect contamination,

279

but also product foreign adulterants. Furthermore, geographical origin determines the

280

value of spices as well as the type of production system. Organic spices will be

281

dearer than spices from the conventional production. For indicator 31”price

282

assymetries”, different scores were given because this is dependent on the type of

283

spice the company trades.

284

Overall, the level of competition in the spices industry is high (indicator 30), which is

285

reflected by the fact that the respondents assigned a score 2 or 3 (medium and high

286

vulnerability). The high competition originates from multiple reasons. There is an

287

increase in scarcity of raw materials and suppliers have to comply with strict buyer

288

requirements such as, quality, food safety, and traceability (Manning & Soon, 2014).

289

Moreover, the market entry requirements are becoming stricter as a result of

13

ACCEPTED MANUSCRIPT 290

technological advances and food safety scandals. These trends and the rising prices

291

are changing the market place (CBI 2015). Even though the competition is high, the

292

respondents rated their “economic conditions own company” (indicator 14) with 1 or 2

293

(low and medium vulnerability). The economic condition of their suppliers is judged

294

as medium vulnerability, except for trader (a), who mentioned that his supplier are

295

primitive farmers and therefore highly vulnerable. The spice sector as a whole it is

296

rated as medium vulnerable as well (indicators 20, 26).

297 298

Culture and behaviour

299

“Organizational strategy own company” (indicator 15) and “ethical business culture

300

own company” (indicator 16) were assigned a score 1 (low vulnerability). This is due

301

to the fact that the companies interviewed strive for long-term financial goals and

302

sustainable relationships with their suppliers. However, for the actors in their

303

environmental layer, they assigned a score 1 (low vulnerability) and 2 (medium

304

vulnerability), i.e. for the ethical business culture of the suppliers and sector

305

(indicators 21, 22, 28). None of the companies was involved in criminal offences

306

(indicator 17) with a score 1 (low vulnerability), but interestingly, for the criminal

307

offences of their suppliers and customers (indicators 23, 24, 27) they all assigned a

308

score 2 (medium vulnerability) as they lack concise information. Spice companies

309

commonly operate in and with countries in which the corruption level is high (indicator

310

18 and 25)but whether this is a potential high risk, is dependent on the spice and

311

origin of the spicethe company trades with. More than half of European imports

312

come from developing countries (97% of total imported volume) (CBI: Trends: Spices

313

and Herbs in Europe, 2016). These countries are rated as highly corrupt, based on

314

the Corruption Perception Index, from “Transparency International” (Corruption

14

ACCEPTED MANUSCRIPT 315

Perceptions Index 2015, 2016). The spice companies give different scores to

316

indicator 29. Criminal offences occurred in the chain according to 5 respondents, but

317

importer (c), b2b (d) and b2b/b2c (h) gave answer 1 ( low vulnerability).

318 319

3.2.3. Control measures related risk factors

320

Hard control measures

321

Measures dedicated to fraud control are not common in the spice industry. “Fraud

322

monitoring system raw materials” (indicator 32) scores range from 1 to 3. The

323

medium and large size companies have raw material controls with fraud monitoring,

324

such as b2b(e),b2b/b2c(f) –and (h) score 3 (low vulnerability), whereas the small

325

companies such as trader(a) and importer(b) score 1 (high vulnerability), because

326

they do not have the money and resources to build a fraud mitigation plan. This is

327

also another explanation of why they show more often high vulnerability in their

328

scores. . According to Professor Elliot from Queens University Belfast - who reviewed

329

the horse meat incident and made recommendations tackling fraud in the Elliot

330

Review for the UK Food Standards Agency - said in an interview with (Levitt, 2016)

331

that smaller businesses do not have the resources to map out dangers of food fraud

332

in their supply chain and says bigger companies should help smaller ones to

333

safeguard consumers. When it comes to adulterations, the difficulty is that offenders

334

are seeking for adulterants that have not been reported so far. Therefore, companies

335

are not conscious of what to find in their product and how to determine the presence

336

of undeclared substances. “Fraud monitoring system final products” and “Systematics

337

and autonomy of verification of fraud monitoring system” (indicator 34, 33, 35),

338

scored 1 (high vulnerability) for most respondents. However, B2b(d) is the exception

339

and has a more dedicated final product fraud monitoring system (low vulnerability).

15

ACCEPTED MANUSCRIPT 340

Most other companies believe that a final product monitoring system is irrelevant,

341

because it suggests that you doubt the integrity of your product if such a system is in

342

place. This is in agreement with the alien conspiracy theory, which describes that

343

crime is often not perceived as part of the society/environment and shaped by the

344

society itself, but rather a problem of “outsiders” that threaten society (Kleemans,

345

2013). The medium- and large size companies have a comprehensive information

346

system including information on mass balance flows. The same was mentioned for

347

their tracking and tracing system (low vulnerability) (indicators 36, 37). Mass balance

348

traceability is a pre-requisite within the food supply chain for ensuring extrinsic quality

349

(Manning & Soon, 2014). The elementary conditions in which suppliers usually

350

operate, explain the dominant score 1 (high vulnerability) for the indicators “Fraud

351

control system supplier”, ”Mass balance control supplier” and “Tracking and tracing

352

system supplier” (indicators 42, 43, 44). Especially the larger b2b/b2c companies

353

have a fraud contingency plan in place when they suspect fraudulent products of

354

suppliers (indicator 50). The plan involves breach of contract with the supplier.

355 356

Soft controls

357

“Ethical code of conduct own company” (indicator 39) was assigned score 3 (low

358

vulnerability) except for trader(a) (small company), which assigned score 1 (high

359

vulnerability). When present, code-of-conduct rules are advertised in the organisation

360

via brochures and trainings. Scores 1 and 3 (high and low vulnerability) were

361

assigned to “Integrity screening own employees” (indicator 38). A few companies

362

questioned if integrity screening is allowed in terms of privacy regulations. “Whistle

363

blowing own company” (indicator 40), scores varied from 1 to 3 (high, medium and

364

low vulnerability). When employees suspect unethical behaviour and malpractices,

16

ACCEPTED MANUSCRIPT 365

they should be able to report this to a confidential mediator. A whistle blowing system

366

protects those that accuse. For the companies who have a system, it is usually a

367

dependent person (medium vulnerability). B2b(e) has an independent person (low

368

vulnerability).

369

Food fraud is an issue widely emphasized by governmental and non-governmental

370

institutions that results in strict requirements to which the buyer has to conform (CBI:

371

Trends: Spices and Herbs in Europe, 2016). The indicator (number 41) “Contractual

372

requirements supplier”, which is related to this topic, got a score 3 (low vulnerability)

373

by all the companies with exception of trader(a). For “Social control chain network”

374

(indicator 45) score 2 was assigned (medium vulnerability), which means the

375

enterprises perceive medium social control in the spice chain. Companies indicated

376

that during the European Spice Association (ESA) meetings, companies are warned

377

about fraud and an “adulteration awareness document” has been distributed (ESA,

378

2015).

379

The control measures “Fraud control industry”, “Specificity national food policy” and

380

“Law enforcement chain network” (indicators 46-49) scored 1 (high vulnerability).

381

Even though the ESA provides an adulteration awareness document, there is control

382

by auditing parties on a limited scale. The reason for this is that the interest in

383

tackling food fraud at that level has only recently developed. The interviewees

384

perceived that laws and policies that address particular fraud issues are not actively

385

enforced in their country.

386 387 388 389

3.3. Methodology consideration The SSAFE tool is based on the routine activities theory and is operationalised in grids with qualitative descriptions and assigned scores to enable a differentiated

17

ACCEPTED MANUSCRIPT 390

assessment of the vulnerability inherent to opportunities, motivations, and control

391

measures to mitigate fraud. The outcome of the assessment should be used

392

qualitatively because data uncertainty is common in early stage assessment of

393

vulnerabilities, as emphasized by John Spink who takes COSO concepts into

394

consideration (Spink, Moyer, & Speier-Pero, 2016). The Committee of Sponsoring

395

Organizations of the Treadway Commission (COSO) is an initiative of five private

396

sector organizations (i.e. the institute of internal auditors) and develops frameworks

397

and guidance on enterprise risk management, internal control and fraud deterrence.

398

Furthermore, the tool is initially designed as a self-assessment tool for companies.

399

However, in this research the companies filled in the survey or discussed it in the

400

interview. It is commonly known that in face-to-face interviews, the chance on social

401

desirable answering is high (Bradburn, Sudman, & Wansink, 2004). Especially for

402

sensitive topics as fraud, respondents might have the tendency to answer according

403

to societal norms and answers that will be best valued by the interviewer (Goffman &

404

Edinburgh Social, 1958). However, the decisions—and the data—must be justifiable.

405 406

4. Conclusions and outlook

407 408

The current assessment of the spices chain reveals that the vulnerability to fraud in

409

the chain is overall perceived as medium vulnerable by the various respondents.

410

Technical opportunities and economic drivers scored high vulnerability across the

411

board, opportunities in time and place as well as culture and behaviour related

412

motivations scored medium vulnerable. The control measures varied widely and

413

especially the smaller sized companies in this study lacked control measures. This

414

kind of fraud vulnerability studies allow comparison of the vulnerability of different

18

ACCEPTED MANUSCRIPT 415

supply chains. Furthermore, the fraud vulnerability assessments of individual actors

416

are a solid base for further development of the companies’ fraud mitigation plans.

417 418 419 420

Acknowledgements

421

This research was executed in the framework of the EU-project SPICED (Grant

422

Agreement: 312631) with the financial support from the 7th Framework Programme

423

of the European Union, European Commission - Directorate-General Enterprise &

424

Industry. This publication reflects the views only of the authors, and the European

425

Commission cannot be held responsible for any use which may be made of the

426

information contained therein. Furthermore, authors acknowledge co-financing of the

427

project through the ‘Kennisbasis’ funding programme by the Ministry of Economic

428

Affairs of the Netherlands and financial support of the PhD project by Intertaste, the

429

Netherlands. Authors are grateful to the eight companies for their participation in the

430

assessment

431

19

ACCEPTED MANUSCRIPT Table 1. Indicators for the three key elements opportunities, motivations, and control measures and their numbering used in the food fraud vulnerability assessment Opportunities Motivations Control measures 1. Complexity of 12. Supply and pricing raw 31. Price asymmetries adulteration raw materials materials 2. Availability technology 13. Valuable components or 32. Fraud monitoring and knowledge to attributes raw materials system raw materials adulterate raw materials 3. Detectability 14. Economic conditions 33. Verification of fraud adulteration raw materials own company mon. system raw materials 4. Availability technology 15. Organizational strategy 34. Fraud monitoring and knowledge to own company system final products adulterate final products 5. Detectability 16. Ethical business culture 35. Verification of fraud adulteration final products own company monitoring system final products 6. Complexity of 17. Criminal offences own 36. Information system counterfeiting company own company 7. Detectability of 18. Corruption level country 37. Tracking and tracing counterfeiting own company system own company 8. Production lines/ 19. Financial strains 38. Integrity screening processing activities supplier own employees 9. Transparency chain 20. Economic conditions 39. Ethical code of network supplier conduct own company 10. Historical evidence 21. Organizational strategy 40. Whistle blowing own fraud raw materials supplier company 11. Historical evidence 22. Ethical business culture 41. Contractual fraud final products supplier requirements supplier 23. Criminal offences 42. Fraud control system supplier supplier 24. Victimization of supplier 43. Mass balance control. supplier 25. Corruption level country 44. Tracking and tracing supplier system supplier 26. Economic conditions 45. Social control chain sector network 27. Criminal offences 46. Fraud control industry customer 28. Ethical business culture 47. National food policy sector 29. Historical evidence 48. Law enforcement branch of industry local chain 30. Level of competition in 49. Law enforcement sector chain network 50. Fraud contingency plan 432

20

ACCEPTED MANUSCRIPT 433

Table 2.

434

Characteristics of the interviewed enterprises and people Company/

Company Interviewees

Business

size

Trader(a)

small

Head

Importer(b)

small

Head

Importer(c)

large

Vice-president

B2B(d)

medium

Quality manager

B2B(e)

medium

Spices flavourist & MVO coordinator

B2B/B2C(f)

large

Strategic buyer and head of customer quality

B2B/B2C(g) large

Quality manager

B2B/B2C(h) large

Quality manager and director of sustainability

435 436

21

ACCEPTED MANUSCRIPT 437

Table 3

438

Fraud factors inherent to opportunities, motivations, and control measures. The

439

scores for the indicators and actors are presented in Table 2, categorized per actor.

440

“b2b” stands for business-to-business and “b2b/b2c” for a combination of a business-

441

to-business and business-to- consumer enterprise. The mode scores for the low,

442

medium, and high vulnerability situations for the opportunities, motivations, and

443

control measures related indicators are coloured green, orange, and red respectively. Opportunities-related fraud factors

Indicator

(a) trader

(b) importer

(c) importer

(d) b2b

(e) b2b

(f) b2b/b2c

(g) b2b/b2c

Technical Complexity of adulteration Availability technology and knowledge to adulterate Complexity counterfeiting

1,3 2,4,5 6,7

In time and space Production lines /processing activities

8

Transparency chain network

9

Fraudulent incidences in past

10,11

Motivations-related fraud factors Economic drivers Supply & pricing materials Valuable components or attributes raw materials

12

Price asymmetries

31

Level of competition in sector

30

Economic conditions own company

14

Economic condition supplier

20

Economic conditions sector

26

13

Culture and behaviour Organizational strategy own company

15

Organizational strategy supplier

21

Ethical business culture own company

16

Ethical business culture supplier

22

Ethical business culture sector

28

Criminal offences own company

17

Criminal offences supplier

23

Criminal offences customer

27

Corruption level country own company

18

Corruption level country supplier

25

22

(h) b2b/b2c

ACCEPTED MANUSCRIPT Historical evidence branch of history

29

Victimization of supplier

24

Control measures related factors Hard control measures Fraud monitoring system raw materials

32

Fraud monitoring system final products

34

Fraud control system supplier Systematics and autonomy of verification of fraud monitoring system

42 33,35

Information system own company

36

Mass balance control supplier Tracking and tracing system own company

43

Tracking and tracing system supplier

44

Fraud contingency plan

50

37

Soft control measures Ethical code of conduct own company

39

Integrity screening own employees

38

Whistle blowing own company

40

Contractual requirements supplier

41

Social control network

45

Fraud control industry

46

Specificity national food policy Law enforcement chain network

47, 48 49

444 445

23

ACCEPTED MANUSCRIPT 446

447 448

Fig. 1

449

Spider web diagrams for opportunities (indicator 1-11), motivations (indicator 12-30),

450

and control measures (indicator 31-50) for scores with highest frequencies

451

(continuous line). In case of identical frequencies for two answers (ties), the second

452

score with highest frequency is presented as well (dashed line). Numbers with

453

corresponding indicators are listed in Table 1.

454

24

ACCEPTED MANUSCRIPT 455

References

456 457 458

Avery, J. (2014). Fighting food fraud. European Parliamentary Research Service, 1–7. Retrieved from http://www.europarl.europa.eu/RegData/bibliotheque/briefing/2014/130679/LDM_BRI(2014)130679_REV 1_EN.pdf

459 460

Bradburn, N., Sudman, S., & Wansink, B. (2004). Asking questions: The definitive guide to questionnaire design for market research, political polls, and social and health questionnaires.

461 462 463

BRC Global Standard for Food Safety Issue 7. (2015). Retrieved from http://www.brcglobalstandards.com/Portals/0/library/files/newsletters/BRC Global Standards newsletter February 2015.html

464 465

Capuano, E., & van Ruth, S. M. (2012). QA: Fraud Control for Foods and Other Biomaterials by Product Fingerprinting. In Latest Research into Quality Control. InTech. https://doi.org/10.5772/51109

466 467 468

CBI: Trends: Spices and Herbs in Europe. (2016). The Hague . Retrieved from https://www.cbi.eu/sites/default/files/market_information/researches/trends-europe-spices-herbs2016_0.pdf

469 470 471

CBI Field of Competition: Spices and Herbs. (2015). The Hague . Retrieved from https://www.cbi.eu/sites/default/files/market_information/researches/competition-europe-spices-herbs2015_0.pdf

472 473 474

Cohen, L. E., & Felson, M. (2016). SOCIAL CHANGE AND CRIME RATE TRENDS : A ROUTINE ACTIVITY APPROACH *. Source: American Sociological Review American Sociological Review, 44(4), 588–608. Retrieved from http://www.jstor.org/stable/2094589

475

Corruption Perceptions Index 2015. (2016). Retrieved from http://www.transparency.org/cpi2015

476

DHS. (2015). National Critical Infrastructure Security and Resilience Research and Development Plan.

477 478 479

Drechsler, H., Halff, M., Huisman, P., & Post, F. (2012). Toepassen soft controls: balanceren tussen wens en werkelijkheid. Retrieved October 6, 2016, from http://www.softcontrols.nu/docs/toepassen-soft-controlsbalanceren-tussen-wens-en-werkelijkheid.pdf

480

ESA. (2015). European Spice Association Quality Minima Document, rev 5(December).

481 482 483

Everstine, K., Spink, J., & Kennedy, S. (2013). Economically motivated adulteration (EMA) of food: common characteristics of EMA incidents. Journal of Food Protection, 76(4), 723–35. https://doi.org/10.4315/0362-028X.JFP-12-399

484 485

GFSI. (2014). GFSI position on mitigating the public health risk of food fraud. Retrieved from file:///D:/Thesis/5.Keten studie/References/Food_Fraud_Position_Paper.pdf

486 487

Goffman, E., & Edinburgh Social. (1958). The presentation of self in everyday life. Edinburgh University of Research Centre (Vol. 55). The Overlook Press The Overlook Press.

488 489

Guide to developing a food defence plan for Food Processing Plants. (2008). Retrieved from https://meathaccp.wisc.edu/additional_info/assets/Guide Food Processing.pdf

490 491 492

Haughey, S. A., Galvin-King, P., Ho, Y. C., Bell, S. E. J., & Elliott, C. T. (2015). The feasibility of using near infrared and Raman spectroscopic techniques to detect fraudulent adulteration of chili powders with Sudan dye. Food Control, 48, 75–83. https://doi.org/10.1016/j.foodcont.2014.03.047

493 494

Hübert, T., Tiebe, C., & Banach, U. (2016). Electronic Noses and Tongues in Food Science. Electronic Noses and Tongues in Food Science. https://doi.org/10.1016/B978-0-12-800243-8.00012-3

495 496

Johnson, R. (2014). Food Fraud and “ Economically Motivated Adulteration ” of Food and Food Ingredients. Retrieved from https://www.fas.org/sgp/crs/misc/R43358.pdf

497

Kerney, A. T. (2010). Consumer product fraud: Deterrence and detection.

498 499 500

Kirezieva, K., Jacxsens, L., Uyttendaele, M., Van Boekel, M. A. J. S., & Luning, P. A. (2013). Assessment of Food Safety Management Systems in the global fresh produce chain. Food Research International, 52(1), 230–242. https://doi.org/10.1016/j.foodres.2013.03.023

501 502

Kleemans, E. R. (2013). Theoretical Perspectives on Organized Crime. Oxford University Press. https://doi.org/10.1093/oxfordhb/9780199730445.013.005

503

Levitt, T. (2016). Three years on from the horsemeat scandal: 3 lessons we have learned.

25

ACCEPTED MANUSCRIPT 504

https://doi.org/10.1017/CBO9781107415324.004

505 506 507 508

Linich, D. (2014). The path to supply chain transparency: A practical guide to defining, understanding, and building supply chain transparency in a global economy. Retrieved from https://dupress.deloitte.com/content/dam/dup-us-en/articles/supply-chaintransparency/DUP785_ThePathtoSupplyChainTransparency.pdf

509 510 511

Luning, P. A., Bango, L., Kussaga, J., Rovira, J., & Marcelis, W. J. (2008). Comprehensive analysis and differentiated assessment of food safety control systems: a diagnostic instrument. Trends in Food Science and Technology, 19(10), 522–534. https://doi.org/10.1016/j.tifs.2008.03.005

512 513 514 515

Luning, P. A., Jacxsens, L., Rovira, J., Osés, S. M., Uyttendaele, M., & Marcelis, W. J. (2011). A concurrent diagnosis of microbiological food safety output and food safety management system performance: Cases from meat processing industries. Food Control, 22(3–4), 555–565. https://doi.org/10.1016/j.foodcont.2010.10.003

516 517 518

Luning, P. A., Marcelis, W. J., Rovira, J., Van der Spiegel, M., Uyttendaele, M., & Jacxsens, L. (2009). Systematic assessment of core assurance activities in a company specific food safety management system. Trends in Food Science & Technology. https://doi.org/10.1016/j.tifs.2009.03.003

519 520

Manning, L., & Soon, J. M. (2014). Developing systems to control food adulteration. Food Policy, 49, 23–32. https://doi.org/10.1016/j.foodpol.2014.06.005

521 522 523

Moore, J. C., Spink, J., & Lipp, M. (2012, April). Development and Application of a Database of Food Ingredient Fraud and Economically Motivated Adulteration from 1980 to 2010. Journal of Food Science. https://doi.org/10.1111/j.1750-3841.2012.02657.x

524 525

Morling, A., & McNaughton, R. (2016). A 2016 Baseline FOOD CRIME ANNUAL STRATEGIC ASSESSMENT. Retrieved from https://www.food.gov.uk/sites/default/files/fsa-food-crime-assessment-2016.pdf

526 527 528

Moyer, D. C., DeVries, J. W., & Spink, J. (2016). The economics of a food fraud incident – Case studies and examples including Melamine in Wheat Gluten. Food Control. https://doi.org/10.1016/j.foodcont.2016.07.015

529 530

Peter, K. V. (2011). Handbook of herbs and spices (1st ed., p. 319). Cambridge: Woodhead Publishing Limited. Retrieved from http://vanveenorganics.com/ebooks/Herbs. Handbook of Herbs and Spices Vol 1.pdf

531 532 533

Pustjens, A. M., Weesepoel, Y., & van Ruth, S. M. (2016). Innovation and Future Trends in Food Manufacturing and Supply Chain Technologies. Innovation and Future Trends in Food Manufacturing and Supply Chain Technologies. https://doi.org/10.1016/B978-1-78242-447-5.00001-0

534 535 536

Sampers, I., Jacxsens, L., Luning, P. A., Marcelis, W. J., Dumoulin, A., & Uyttendaele, M. (2010). Performance of food safety management systems in poultry meat preparation processing plants in relation to Campylobacter spp. contamination. Journal of Food Protection, 73(8), 1447–1457.

537 538 539

Sasikumar, B., Swetha, V. P., Parvathy, V. A., & Sheeja, T. E. (2016). 22 – Advances in Adulteration and Authenticity Testing of Herbs and Spices. In Advances in Food Authenticity Testing (pp. 585–624). https://doi.org/10.1016/B978-0-08-100220-9.00022-9

540 541 542

Sayers, R. L., Gethings, L., Wallace, A., Semic-Jusufgic, A., Simpson, A., Barran, P., … Mills, E. N. C. (2016). How Much of a Problem Is Peanut in Ground Cumin for Individuals with Peanut Allergy? Journal of Allergy and Clinical Immunology, 137(2). https://doi.org/10.1016/j.jaci.2015.12.597

543 544 545

Schaarschmidt, S. (2016). Public and private standards for dried culinary herbs and spices—Part I: Standards defining the physical and chemical product quality and safety. Food Control, 70, 339–349. https://doi.org/10.1016/j.foodcont.2016.06.004

546 547

Spink, J., & Moyer, D. C. (2011a). Backgrounder: Defining the Public Health Threat of Food Fraud. Retrieved from http://foodfraud.msu.edu/wp-content/uploads/2014/07/food-fraud-ffg-backgrounder-v11-Final.pdf

548 549

Spink, J., & Moyer, D. C. (2011b). Defining the Public Health Threat of Food Fraud. Journal of Food Science, 76(9), R157–R163. https://doi.org/10.1111/j.1750-3841.2011.02417.x

550 551

Spink, J., Moyer, D. C., & Speier-Pero, C. (2016). Introducing the Food Fraud Initial Screening model (FFIS). Food Control, 69, 306–314. https://doi.org/10.1016/j.foodcont.2016.03.016

552 553

van Ruth, S.M., Huisman, W., & Luning, P.A. (2017). Food fraud vulnerability and its key factors. Trends in Food Science and Technology, submitted.

554

26