The role of cooperatives in food safety management of fresh produce chains: Case studies in four strawberry cooperatives

Food Control 62 (2016) 299–308

Contents lists available at ScienceDirect

Food Control journal homepage: www.elsevier.com/locate/foodcont

The role of cooperatives in food safety management of fresh produce chains: Case studies in four strawberry cooperatives Klementina Kirezieva a, Jos Bijman b, Liesbeth Jacxsens c, Pieternel A. Luning a,∗ a

Food Quality and Design Group, Wageningen University, Netherlands Management Studies Group, Wageningen University, Netherlands c Department of Food Safety and Food Quality, Ghent University, Belgium b

a r t i c l e

i n f o

Article history: Received 12 July 2015 Received in revised form 21 October 2015 Accepted 27 October 2015 Available online 10 November 2015 Keywords: Food safety management systems Cooperatives Supply chain

a b s t r a c t Recent outbreaks with fresh produce have raised questions regarding management of quality and safety in the complex supply chains, where cooperatives play a central role. The overall objective of this article was to investigate the role of cooperatives in food quality and safety management in the fresh produce chain, focussing on the food safety management systems implemented on the farms. More specifically, we raise the question how a more market-like or a more hierarchy-like governance of transactions via cooperatives affect the quality and safety management system of the members. The research employed case studies in four cooperatives in Belgium and the Netherlands, each with different size and per cent of contractual sales. Data was collected with a diagnostic tool for assessment of food safety management systems (FSMS) on the farms, and semi-structured interviews with the quality assurance managers of the cooperative firms. Twenty-eight strawberry farms were assessed with the diagnostic tool, seven per each cooperative. Cooperatives play a double role in managing quality and safety in the food supply chain. They are responsible for the supply chain management, including tactical decisions about coordination of quality and safety requirements between customers, cooperative firms and their farmers. At the same time, they are selling the products of their members and make strategic decisions about the governance of transactions in the supply chain, which ultimately may have an impact on the supply chain management and the FSMS on the farms. Farmers in cooperatives with more hierarchical relationships showed better operation of control activities (score 3), and advanced assurance activities at score 4 (advanced level), more science-based, adapted and tested for their effectiveness. This was associated with more effort put in supply chain management by the cooperative to support collaboration and coordination in the chain. However, the largest cooperative had moderate scores for several key control functions, suggesting that large cooperatives with complex business functions may suffer from lower commitment of members, leading to lower FSMS performance at farms. © 2015 Elsevier Ltd. All rights reserved.

1. Introduction Each company in the food chain needs to implement a food safety management system, and it is a legal requirement in many parts of the world, for instance within the European Union (EC, 2002). Food safety is considered as one of the aspects of food quality, and a food safety management system is a part of the overall quality management system of the company (Luning & Marcelis, 2009a). Quality management systems encompass activities of com∗

Corresponding author. E-mail address: [email protected] (P.A. Luning).

http://dx.doi.org/10.1016/j.foodcont.2015.10.038 0956-7135/© 2015 Elsevier Ltd. All rights reserved.

panies aimed to direct, control and coordinate quality, including formulating policy, setting objectives, planning, controlling, assuring and improving (Luning & Marcelis, 2007; Luning & Marcelis, 2009b). Over the last years, however, the focus of quality management is moving from intra-company to inter-company within a supply chain (Kaynak & Hartley, 2008). This is especially important in fresh produce chains due to the perishability of the products, global sourcing and long supply chains (Blackburn & Scudder, 2009; Rong, Akkerman, & Grunow, 2011). Recently, studies have begun to focus on the alignment between supply chain governance structures and food quality assurance schemes (Trienekens & Wognum, 2013; Wever, Wognum, Trienekens, & Omta, 2010).

300

K. Kirezieva et al. / Food Control 62 (2016) 299–308

However, to our best knowledge insights are lacking about the influence of supply chain governance on the food safety management systems implemented at farm level. From a transaction cost economics perspective, supply chain governance structures differ in their level of coordination, ranging from market-based to hierarchy based structures (Raynaud, Sauvee, & Valceschini, 2005). Market-based refers to spot market, where goods are sold and delivered immediately (for instance via an auction where a clock sets the price) through contracts which can be verbal, formal or equity-based. In hierarchy-based structures, transactions are controlled administratively. Full coordination on product, transport and other requirements can be realized in hierarchybased structures by defining requirements in contracts, while spot market is commonly anonymous and coordination is limited. The auction is a form of a spot market during at which the auctioneer begins with a high asking price, which is lowered on a clock until some participant is willing to accept the auctioneer’s price, or the minimum acceptable price is reached (McAfee, Dinesh Satam, McMillan, & Dinesh, 1987). The farmer-owned marketing cooperative is a governance structure that coordinates the transactions in many European fresh produce supply chains (Bijman et al., 2012; Hendrikse & Bijman, 2002). Cooperatives, however, differ greatly in the extent to which they apply more market-based versus more contract-based governance of the transactional relationships with their members. According to Ménard (2004, 2007), a cooperative is a hybrid governance form, as it combines pooling of resources, coordination through contracts and combining competition and cooperation. Chaddad (2012) has further elaborated on this conceptualization of the cooperative as a hybrid governance structure, emphasizing that cooperatives use both market-like and hierarchy-like mechanisms of governance. In other words, the relationship between the farmer-member and the cooperative firm is partly market-based and partly contract-based. The impact of the cooperative governance structure on food quality and safety assurance is an on-going debate. Several studies showed that cooperatives provide more support activities to their members and therefore deliver better and more uniform quality than investor-owned companies (Cechin, Bijman, Pascucci, Zylbersztajn, & Omta, 2013; Hoffmann, 2005). Others, however, claim that members of cooperatives produce lower quality compared to farmers delivering to non-cooperative buyers (Pennerstorfer & Weiss, 2013). The latter studies have argued that the dispersed ownership and democratic decision-making structures of cooperatives lead to low incentives to increase quality. In this study we are not so much interested in the ownership structure, but in the governance of the transaction between members and their cooperative. More specifically, we raise the question how a more market-like or a more hierarchy-like governance of transactions affects the quality and safety management systems of the members. The latter can thereby affect the quality and safety of produce. The overall objective of this article is to investigate the role of cooperatives in food quality and safety management in the fresh produce chain, focussing on the food safety management system (FSMS) implemented on the farms. 2. Theoretical foundation Food quality and safety management are among the most challenging functions of marketing cooperatives, as these organisations need to work in the interests of their members as well as meet the stringent requirements of wholesalers and retailers. To achieve this, not only the cooperative firm but also the cooperative members (the farmers) need to implement quality management systems (Fig. 1, operational decisions level). Within such quality management system, a particular part that is aimed at controlling and as-

Fig. 1. Key concepts in the study and their interrelatedness in view of the quality management decisions taken in a company.

suring safety is called a food safety management system (FSMS) (Luning, Bango, Kussaga, Rovira, & Marcelis, 2008, 2009). Quality and safety management systems are supporting the operational decisions on the farms or in the companies by monitoring and controlling food quality and safety and by providing assurance (Luning et al., 2009). Furthermore, to answer the market requirements and to realise the demanded quantity and quality (including safety), cooperatives apply supply chain management practices (Fig. 1; tactical decisions level) to coordinate production, processing and distribution activities with different steps of the supply chain (Doukidis, Matopoulos, Vlachopoulou, Manthou, & Manos, 2007; Fish, 2011). This, for instance, may involve tactical decisions about the choice of quality assurance standards in the supply chain or type of logistics (Wever et al., 2010). Supply chain management is affected by the strategic decisions taken about the governance and types of transactions in the supply chain (Fig. 1; strategic decisions level; Richey, Roath, Whipple, & Fawcett, 2010). Fig. 1 represents the hypothesised interrelations among these three levels of decisionmaking, which are elaborated in the following sections. 2.1. Food quality and safety management systems Quality management systems represent all the activities used to direct, control and coordinate quality in a company, including the organisational structure, responsibilities, processes, procedures, and resources that facilitate the achievement of quality management. These systems are the result of translating requirements from the market and regulatory environment into the specific production activities and organisational structures of the company (Luning & Marcelis, 2009b). These requirements can be put by public standards or recommendations (like code of practice according to Codex Alimentarius). Furthermore, private stakeholders in the fresh supply chain, such as retailers, can also demand certification against private standards (e.g. GlobalGAP, Tesco’s Nurture) (Kirezieva et al., 2015). 2.2. Supply chain management Quality management systems can be influenced by the overall supply chain management of the company, which is aimed at coordinating activities with suppliers and customers and to achieve

K. Kirezieva et al. / Food Control 62 (2016) 299–308

certain levels of quality and safety throughout the supply chain (Foster, 2008). This is targeted through tactical decisions, such as the selection and control of suppliers (e.g. based on certification), coordination of deliveries according to harvesting time and required quality, and the organisation of logistics and inventory (Vorst, Kooten, & Luning, 2011). Still, the integration of quality and safety decisions into supply chain management faces difficulties as supply chain managers and quality managers have different perspectives on supply chain management activities; supply chain managers focus at reducing transaction costs, while quality managers focus on quality assurance (Fish, 2011; Pagell, 2004). Moreover, quality and supply chain management decisions can be contradictory. For instance, research has shown that the use of quality information in food supply chain logistics positively contributes to the final quality and safety of products, but at the expense of transportation costs (Rijpkema, Rossi, & Van Der Vorst, 2014). 2.3. Supply chain governance Supply chain management is affected by the strategic decisions about the overall governance of transactions in the supply chain. Previous research revealed that hierarchical governance structures are used with private reputational quality assurance certifications, whereas market-based governance structures are used with public standards (Raynaud et al., 2005; Wever et al., 2010). Still, acquiring certification is not a guarantee for quality (or safety) (Grabot et al., 2014; Havinga, 2006; Roth, Tsay, Pullman, & Gray, 2008). A question remains whether governance structures may influence food quality and safety management systems implemented in companies, and whether more hierarchical governance of transactions in the supply chains promote more advanced food quality/safety management systems. The latter is also the hypothesis behind this research. 2.4. Cooperatives in supply chain governance In the fresh produce chain, supply chain management decisions are commonly taken within the cooperatives as part of the overall strategy to supply large volumes of uniform quality to comply with requirements of customers. In practice, this means pursuing the reduction of product and production variation between different farms. To achieve these goals cooperatives have activities at the level of the cooperative firm, such as sorting, packaging, storage and transport. In addition, they facilitate the implementation of quality and safety management systems on the member-farms. These activities may differ between cooperatives, as they differently govern transactional relationships with their members. 3. Methodology 3.1. Field study design Cooperatives involved in the study were selected to represent different types of cooperatives in Belgium and the Netherlands – from small to big, from selling mostly via spot-market (auction) to having a higher number of contractual sales (Table 1). COOP1 is selling 70% of products via the auction, while this percentage is lower for the other three cooperatives. The rest of the production of the cooperatives is sold via direct contracts with retailers and wholesalers. To diagnose the performance of food safety management activities at primary production level, in each of the four cooperatives seven strawberry farmers have been visited. In total, 28 questionnaires were filled in during a face-to-face interview and an on-site visit. The sample selection occurred by randomly choosing berry farmers from the member list of each cooperative. Strawberries

301

have been chosen as a case study due to their sensitivity in terms of quality and safety, and the significant trade value in Belgium and the Netherlands (Jacxsens, Van Boxstael, & Uyttendaele, 2011). Moreover, strawberries are a seasonal product and are therefore suitable to sell via an auction system. All farmers were certified against the same quality assurance standard – GlobalGAP, thus isolating the effect of certification on the safety management system. 3.2. Diagnostic tool to assess food safety management systems’ performance A diagnostic tool has been used to collect the information about the FSMS (Kirezieva, Jacxsens, Uyttendaele, Van Boekel, & Luning, 2013; Kirezieva, Nanyunja et al., 2013). The tool allows assessment of the following elements: 1) the status level of FSMS (control and assurance activities) and their system output, and 2) the riskiness of the context, including product, production, organisational and chain characteristics. Context factors are defined as structural elements of a situation that can affect decision-making in the FSMS (Luning et al., 2011). Risk refers to situational characteristics inherent to the product, process, organizational and chain environment context that may create ambiguity (due to lack of understanding), uncertainty (due to lack of information) and vulnerability (due to a high susceptibility for microbial/chemical contamination while decisions must be made) in decision-making in the FSMS, thereby affecting the food safety output of the system (Luning et al., 2011). The basic principle behind the diagnostic tools is that a more risky context requires more advanced control and assurance activities to mitigate the actual risk on chemical and or microbial contamination (system output) (Kirezieva, Nanyunja et al., 2013; Luning et al., 2011). More advanced control measures and assurance activities are typified by being based on scientific knowledge, using (company) specific information, procedural driven, and independent positions (important for assurance). Ambiguity, uncertainty, and vulnerability in decision making can be reduced by using scientific knowledge, specific information, structured analysis, and independent positions/measures (Luning and Marcelis, 2007; Luning et al., 2011). The assessment of the FSMS focuses on the core control and assurance activities. Core control activities are assessed on how preventive measures, intervention strategies, and monitoring activities are designed and actually operated. Assessment of assurance activities focuses on the set-up of the systems, validation, verification and documentation. Indicators are selected to represent the current situation of a context factor, or FSMS activity. They are measured by creating a grid with three situational descriptions following the same criteria for context factors and FSMS activities, respectively. The grids for context indicators represent low (score 1), moderate (score 2) and high risk (score 3) to decision making, by following the criteria of vulnerability to contamination, uncertainty due to lack of information and ambiguity due to lack of understanding. The grids for FSMS activities represent low (score 1), basic (score 2), average (score 3), and advanced level (score 4). Similarly, the grids for the system output represent no information (score 1), poor (score 2), moderate (score 3), and good output (score 4). In total, the tool encompasses sixty-nine (69) indicators with corresponding situational grid descriptions (see some examples in Fig. 2). For the context factors, a low risk level (score 1) is given when the chance of microbial or chemical contamination is low, and when the organisation characteristics support appropriate decision-making and when chain characteristics levels less dependency. A moderate risk level (score 2) is given when there is a potential risk of contamination, constrained organisational support, and restricted dependency on the other chain actors. A highrisk level (score 3) is given when the risk of contamination is high,

302

K. Kirezieva et al. / Food Control 62 (2016) 299–308

Table 1 Characteristics of the cooperatives. Characteristic a

Size (number of members) Turnover (million euro)a Voting ruleb Auction clock sales, % Own brand Selling also non-members products Foreign partnerships Location a b c

COOP1

COOP2

COOP3

COOP4

300 165 1 member 1 vote 70% Yes No No Belgium

186 47 Proportionalc 30% Yes No No The Netherlands

354 353 Proportional 25–30% Yes Yes Collaboration with one cooperative The Netherlands

720 1400 Proportional 20–30% Yes Yes Several subsidiaries The Netherlands

In the end of 2012, based on the annual reports of the cooperatives. Turnover is the annual sales volume net of all discounts and sales taxes. Based on Kalogeras, Pennings, Benos, & Doumpos, 2013. Members are granted votes according to the number of their patronage equity shares.

organisational support is low, and chain dependency is high. For the FSMS activities, a low level is given when an activity is not implemented or not applicable. A basic level is given when activities are executed according to historical knowledge or own experience, using common materials or methods. An average level is given when standardised equipment and methods are used, sector standards or guidelines are followed. An advanced level is given when activities are performed systematically, modified and tested for the concrete farm situation, by following scientific knowledge and methods. For the system output, a poor output level is defined as the situation when sampling is ad-hoc, minimal criteria are used, and various complaints and remarks are received. A moderate output level is given when sampling is done regularly, several criteria are used, complaints and remarks are restricted. A good output level is given when sampling is done systematically, criteria are specific and tailored, complaints and remarks are not received. Furthermore, the tool includes an introduction part with general questions about characteristics of the farm: number of workers, standards and guidelines implemented and certified, etc. 3.3. Semi-structured interviews to characterise activities of cooperatives Next to the diagnostic tool, semi-structured interviews have been conducted with the quality assurance managers of the cooperatives. These interviews aimed at collecting additional information about cooperatives’ decisions about supply management activities such as logistics, supplier relationships and support activities to the members. The interviews included questions about logistics, investments in infrastructure and equipment, information and training of farmers on quality assurance and food safety, chemical, physical and microbiological control of the products, complaint registration and assurance activities (i.e. validation and verification). The support that the cooperative firm claimed to provide to the farmers was confirmed during the interviews with the member farmers. 4. Results 4.1. Status of FSMS in cooperative members Table 4 presents the results for the indicators of the context factors of product and production characteristics. Score 3 (high risk) was given for the indicators of microbial, pesticide residues and mycotoxin risk of final product, as all farmers were dealing with sensitive products in terms of food safety – strawberries. The risk for initial materials (i.e. growing seedlings, pesticides) was assessed as moderate for both risk of microorganisms and pesticide residues, as berries are sensitive to microbial contamination and

fungi and pesticides are sprayed. Microbial risk of initial materials scored even higher in COOP1 and COOP2 (3, high risk), compared to COOP3 and COOP4 (2, moderate risk), due to the use of treated seedlings. Farmers in all cooperatives scored as moderate risk (2) for the indicators of production characteristics, due to protected cultivation, moderate climate conditions and controlled source of water. Most indicators for organisation and chain characteristics scored as moderate risk (2) (Table 4). COOP1 and COOP2 demonstrated higher risks in their organisation and chain context due to low worker competences (high risk), and logistic facilities that were not specific for fresh produce (COOP1 = 2) or not controlled at all (COOP2 = 3). Table 5 presents the results for the indicators for control activities to analyse the design of the preventive measures, intervention strategies, and monitoring system. Most activities were designed according to guidelines and standards (e.g. GlobalGAP) and by using best available equipment and methods (score 3, average level). Two activities scored lower for COOP1, 2 and 4. These were hygienic design of equipment, premises and tools, which was not considered (COOP1 = score 1, low level) or at basic level (COOP2&COOP4 = score2); and sanitation program and which was based on own knowledge and experience (score 2, basic level). Furthermore, incoming materials control and corrective actions were at average level (score 3) for COOP1 and COOP3, whereas at basic level (score 2) at COOP2 and COOP4. None of the cooperatives, except COOP3 (score 3, average level), were taking microbiological samples (score 1, low level). Only COOP1 was using organic fertilizers, applied according to guidelines (score 3, average level). Table 5 shows the results for the indicators of the actual operation of control activities. COOP3 and COOP1 scored 4 (advanced level), corresponding to easily available and understandable procedures that were regularly updated. However, the other two cooperatives scored at average level (3). Workers in COOP3 and COOP4 were aware about the content of procedures and were strictly following them (score 4, advanced level), while workers in COOP1 and COOP2 were mostly working according to habits and controlled by supervisors (score 3, average level). All analytical analyses were done by external accredited labs (score 4, advanced level). Lower scores were given to actual performance of equipment and tools which was not known in the farms from COOP1 (score 1, low level). Actual storage capacity was not checked by farmers in COOP1 and COOP2 (score 2, basic level), and well-known and tested for COOP4. Farmers in COOP3 did not have any in-house storage facilities (score 1). Assurance activities scored mostly 2 (basic level), which means that it was done by own people and only upon problems (Table 5). An exception was the farmers in COOP3, which were validating and verifying their activities by an external expert (score 3, average level), and pro-actively translating stakeholder requirements (score

K. Kirezieva et al. / Food Control 62 (2016) 299–308

303

Fig. 2. Examples of indicators for context, FSMS activity and output from the diagnostic tool to assess food safety management systems’ performance.

4, advanced level) and using feedback information to update their activities (score 3, average level). All farmers had documentation and record keeping at average level (score 3), structured, up-todate and centrally accessible. Table 5 reveals the results for the indicators of the system output. All farmers were audited by two or more third parties (score 3 or 4), did not have major remarks (score 4) nor complaints about pesticide residues (score 4). COOP3, however, showed better results for several indicators, as farmers had few visual complaints (score 4), the cooperative has established a microbiological sampling plan (score 3) with strict criteria (score 3), and was recording the nonconformities of received products (score 4). All were sampling for

pesticide residues on a cooperative level (score 3). COOP1, 2 and 4 showed similar basic results for the indicators of microbiological sampling and non-conformities, as these were not systematically detected.

5. Discussion 5.1. Characteristics of cooperative members Table 2 shows the characteristics of the farmers interviewed in each cooperative. Farms in the four cooperatives were all

304

K. Kirezieva et al. / Food Control 62 (2016) 299–308

Table 2 Characteristics of the farmers in the cooperatives. Farm code

Size

Workforce

Sales, %

Sales, tons

Standards

Hectares

Tons

Permanent staff

Temporary workers

Clock

Contracts

Clock

Contracts

COOP1 A01 A02 A03 A04 A05 A06 A07

10.0 15.0 8.0 7.0 3.0 23.0 15.0

130 180 200 300 140 800 900 2650

2 3 2 2 1 4 4

30 50 30 40 20 60 70

100 100 100 100 100 100 100 100

– – – – – – –

130 180 200 300 140 800 900 2650

– – – – – – – –

GlobalGAP; GlobalGAP; GlobalGAP; GlobalGAP; GlobalGAP; GlobalGAP; GlobalGAP;

COOP2 B08 B09 B10 B11 B12 B13 B14

6.5 6.0 1.5 6.0 1.3 5.0 3.0

280 250 70 240 150 180 300 1470

2 3 2 1 4 2 3

40 50 50 10 9 15 20

50 – 20 100 90 90 80 61

50 100 80 – 10 10 20 39

140 – 14 240 135 162 240 931

140 250 56 – 15 18 60 539

GlobalGAP GlobalGAP GlobalGAP; AH Protocol GlobalGAP GlobalGAP GlobalGAP GlobalGAP

COOP3 C15 C16 C17 C18 C19 C20 C21

2.0 40.0 15.0 20.0 24.0 2.0 1.0

220 400 200 570 630 200 200 2420

2 4 2 2 3 1 2

50 70 60 40 50 9 6

40 – 20 30 – 50 30 24

60 100 80 70 100 50 70 61

88

132 400 160 399 630 100 140 1961

GlobalGAP GlobalGAP GlobalGAP GlobalGAP; QS GlobalGAP; QS; M&S; Tesco GlobalGAP GlobalGAP

COOP4 D22 D23 D24 D25 D26 D27 D28

2.2 15.0 3.0 20.0 15.0 2.0 2.1

170 350 450 350 350 150 160 1980

2 1 4 3 3 1 3

10 20 60 50 30 10 9

40 75 50 – 40 100 – 44

60 25 50 100 60 – 100 56

102 88 225 350 210

GlobalGAP GlobalGAP GlobalGAP; Tesco GlobalGAP GlobalGAP GlobalGAP GlobalGAP

40 171 100 60 459 68 262 225 140 150 846

160 1134

Flandria Flandria Flandria Flandria Flandria Flandria Flandria, ICQC

Table 3 Activities of cooperatives (results from the interviews). Activities

COOP1

COOP2

COOP3

COOP4

Logistics

Various logistic companies; mostly uncontrolled conditions

Various logistic companies; mostly uncontrolled conditions

Own logistic company; controlled conditions

Specifications

Packaging material, payment conditions, basic product quality and process requirements Information evenings when new methods or equipment are available

Packaging material, payment conditions, basic product quality and process requirements Training is organised when becoming a member and every year afterwards

Own logistic company; controlled conditions; provision and sanitation of crates Packaging, payment conditions, norms on product and process quality; via contract- also price and quantity Training is organised when becoming member, every year, and upon changes of standards, new methods and equipment

Control when products are received at the cooperative warehouse, corrective actions implemented by the cooperative Control when products are received at the cooperative warehouse Not applied yet

Control when products are received at the cooperative warehouse, corrective actions implemented by the cooperative Control when products are received at the cooperative warehouse Not applied yet

Complaints received and analysed by the cooperative None

Complaints received and analysed by the farmers Technical advisors upon request

Information/training

Chemical control

Physical control

Microbiological control

Complaint registration Support for assurance activities

Control when products are received at the cooperative warehouse, corrective actions implemented by the farmer Control when products are received at the cooperative warehouse Systematic control when products are received at the cooperative warehouse Complaints received and analysed by the cooperative Instructions about translation of new requirements from legislation or standards and validation of new methods e.g. for pest control (pesticides)

Packaging, payment conditions, norms on product and process quality; via contract- also price and quantity Training/education when becoming a member, and sporadic afterwards. Information, data and advice from the cooperative upon request. Control when products are received at the cooperative warehouse, corrective actions implemented by the farmer Control when products are received at the cooperative warehouse Systematic control when products are received at the cooperative warehouse Complaints received and analysed by the cooperative Instructions about validation of new methods e.g. for pest control (pesticides)

K. Kirezieva et al. / Food Control 62 (2016) 299–308

305

Table 4 Number of companies with same scores and median for indicators of context factors of the companies in the four cooperatives. Indicators

Frequency (n = 28) Low riska

Product characteristics Microbiological risk of initial materials Risk of initial materials due to pesticides Risk of initial materials due to mycotoxins Microbiological risk of final product Risk of final product due to pesticides Production characteristics Susceptibility of production system Risk of climate conditions of production environment Susceptibility of water supply Organisational characteristics Presence of technical staff Variability of workforce composition Sufficiency of operators’ competences Extent of management commitment Degree of employee involvement Level of formalization Sufficiency of supporting information system Chain characteristics Severity of stakeholder requirements Extent of power in supplier relationships Degree of information exchange in supply chain Sophistication of logistic facilities Supportiveness of food safety authority Degree of globalization of supply Specificity of external support Specificity of food safety legal framework

3 2

9 2 3 7 2 5 4 2 1 2 5 13 15 21 28 28

Medium risk

High risk

8 26b 6

17 22 28 28

11 28 22

8

14 21 10 21 15 20 24

11

24 24 20 7 10 7

4

16 7 8 4 2 3 2 3 8 3

COOP1 (n = 7)

COOP2 (n = 7)

COOP3 (n = 7)

COOP4 (n = 7)

3 2 3 3 3

3 2 3 3 3

2 2 3 3 3

2 2 2 3 3

2 2 2

2 2 2

2 2 2

2 2 2

2 2 3 2 2 2 2

2 2 3 2 2 2 2

2 2 2 2 2 2 2

3 2 2 2 2 2 2

2 2 2 2ac 1 2a 1 1

2 2 2 3a 2 1b 2 1

2 2 2 1b 1 1b 1 1

2 2 2 1b 1 1b 1 1

a Low (1), medium (2) and high (3) risk situations for product and process characteristics correspond to low, potential and high chance of microbiological or chemical contamination. For organizational and chain characteristics they correspond to supportive, constrained and lacking administrative chain conditions or low, restricted and high dependence on other chain actors. b With bold – the mode. c If the letters in the same row differ (a, b, c) then significant differences between groups (P < 0.05) based on Mann–Whitney U non-parametric test.

family-owned with one to four permanent workers and temporary workers during the high season ranging from four to seventy. All farmers were selling all of their products through the cooperatives. However, farmers in COOP1 were selling all their products through the auction clock, while the other three were selling both via the clock and contracts. The farmers from COOP2 had more sales via the auction clock compared to farmers from COOP3 and COOP4 (Table 2). All farmers were certified against GlobalGAP, and several also had certification against other private standards. Table 3 presents the services that the cooperatives provided to their members. While COOP1 and COOP2 were using various external logistic companies, COOP3 and COOP4 invested in their own with refrigerated conditions. COOP3 was even providing and sanitizing the crates. The cooperatives were providing information and training on quality and safety management to a different degree, with most efforts put by COOP3 which was organising trainings every year, and upon changes of standards, new methods and equipment. In the last year COOP3 and COOP4 began performing microbiological control. They also provided support to assurance activities of farmers, such as translation of new requirements into the food safety management system and validation of pest control methods.

other hand, cooperatives supported farmers in translating the continuously evolving requirements of the quality assurance standards and legislation into their farm-specific FSMS, so farmers are able to supply products with the required quality and safety levels. These functions were inter-related. Cooperatives provided information about pesticide residue levels, as required by the retailers, and they performed sampling and analysis when receiving the products from the farms (Table 5). To support farmers in low pesticide use, cooperatives organised expert meetings and trainings about the quality assurance standards required by the customers, provide instructions or assist in the validation of new methods (Table 3). It has been discussed before that cooperatives may be less able to provide higher quality levels. Quality standards are generally not required as a criteria for membership, members differ in their ability to implement these standards, and cooperative firms have to deal with diverse quality levels (Mérel, Saitone, & Sexton, 2009). However, results from our study suggest that the evolution away from the traditional auction sales led also to changes of strategies of cooperative firms in coordinating quality and thereby to higher quality levels.

5.2. The role of cooperatives in food safety management

Farmers in the cooperatives with higher numbers of contract sales (COOP3 and COOP4) demonstrated more advanced levels of FSMS activities and less risky organisation and chain, associated with the presence of a technical person with expertise in food safety, higher competences of workers and more sophisticated logistic facilities. These differences can be partially attributed to differences in the governance of transactions in the fresh produce sector in the Netherlands and Belgium. The fresh produce market in both countries includes a large number of small famers

The overall objective of this article was to investigate the role of cooperatives in food quality and safety management in the fresh produce chain, focussing on the food safety management systems implemented on the farms. The cooperatives in this study performed a key role in quality and safety coordination in the supply chain. On the one hand, cooperatives communicated quality and safety information between farmers and customers. On the

5.3. Supply chain governance

306

K. Kirezieva et al. / Food Control 62 (2016) 299–308

Table 5 Total number of companies with same scores (N = 28), and the median for the indicators of food safety management practices in the four cooperatives. Indicators

Preventive measures design Sophistication of hygienic design of equipment & facilities Specificity of maintenance program Adequacy of storage facilities Specificity of sanitation program Extent of personal hygiene requirements Sophistication of incoming materials control Sophistication of supplier control Specificity of fertilizer program Specificity of pesticide program Sophistication of water control Adequacy of irrigation method Intervention method design Adequacy of partial physical intervention Monitoring system design Appropriateness of risk analysis Adequacy of analytical methods for microbiological hazards Adequacy of analytical methods for pesticide residues Specificity of microbiological sampling plan Specificity of pesticides’ sampling plan Adequacy of measuring equipment Extent of corrective actions Actual operation of control activities Availability of procedures Compliance to procedures Actual hygienic performance of equipment & facilities Actual cooling and storage capacity Actual capability of partial intervention Actual measuring equipment performance Actual analytical equipment performance Assurance activities Sophistication translating external requirements Extent of systematic use of feedback information Sophistication validating preventive measures Sophistication validating intervention strategies Extent verifying people related performance Extent verifying equipment & methods performance Appropriateness documentation Appropriateness record-keeping system System output Comprehensiveness external evaluation Seriousness of remarks Type of microbiological complaints Type of chemical food safety complaints Type of visual quality complaints Advancedness of microbiological sampling Comprehensiveness of judgement criteria for microbial FS Advancedness of pesticides sampling Comprehensiveness of judgement criteria for chemical FS Type of non-conformities

Frequency (N = 28) 1a , b

2

3

4

11 2 3 7

8 4 3 8 6 10 5 1 2 9 8

9 19 16 11 18 9 19 4 12 8 19

3 6 2 4 2 1 4 14 8 1

7 3 19 3

COOP1 (n = 7)

COOP2 (n = 7)

COOP3 (n = 7)

COOP4 (n = 7)

1a 3 3 2 4a 3b 3 3 4 3a 3ab

2ab 3 3 2 3b 1a 3 1 3 2b 3a

3b 3 3 3 3b 3b 3 1 4 4a 3ab

2ab 3 3 2 2b 2ab 2 1 3 2b 2b

1

1

1

1

2 1a 4 1a 3a 3 3a

3 1a 4 1a 3a 3 2b

3 4b 4 3b 3a 3 3a

3 1a 4 3b 1b 3 2b

4 3a 1a 2 1 1 4

3 3a 2a 2 1 3 4

4 4b 3b 1 1 4 4

3 4b 1a 4 1 4 4

24

4

5 24 1 13 7 4 5

3

19

7

11 21 20 13

3 2 4 3

19 17 6 4

5 9 2 10

10

13 24

14 15 1 5

14 7 15 2 8 9 24 22

5 3 2 1 3 2 1

2a 2a 3 1 2a 2a 3 2a

2a 2a 2 1 2a 2a 3 3b

4b 3b 3 1 3b 3b 3 3b

2a 2a 3 1 2a 2a 3 3b

1

17 3

10 23 1 26 8

4 4 1 4 3 1a 1a 3 2 1a

3 4 1 4 3 1a 1a 3 2 1a

4 4 4 4 4 3b 3b 4 3 4b

3 4 4 4 3 1a 1a 3 3 1a

1 16 11 28 5 4 2 4 3 25 3 2 2 1 1 1 27 1 20 20 2 2 17

7 14 8

3 2 2 1 11

1 17 6 5 20 9 6

1 6 27 4 4 3

1 5 6 5

The values that appear most often for the indicators are represented in bold. a If online letters differ (a, b, c) then significant differences between groups (P < 0.05) based on Mann–Whitney U non-parametric test. b Situations 1, 2, 3, and 4 correspond to:– low situation (1) → absent, not applicable, unknown;– basic situation (2) → lack of scientific evidence, use of company experience/history, variable, unknown, unpredictable, based on common materials/equipment;– average situation (3) → best practice knowledge/equipment, sometimes variable, not always predictable, based on generic information/guidelines for the product sector;– advanced situation (4) → scientifically underpinned (accurate, complete), stable, predictable, and tailored for the specific food production situation.

organised in marketing cooperatives (Bijman, 2002; Gijselinckx & Bussels, 2012). The latter sell respectively about 95% and 85% of all fruits and vegetables (Bijman & Gijselinckx, 2012). The sales via cooperatives, however, are organised differently in Belgium and the Netherlands. While in Belgium products are still largely marketed through cooperative auctions which represent a spotmarket (Mondelaers & Van Huylenbroeck, 2008), in the Netherlands contractual relationships with retailers or wholesalers represent a much higher share of the market (Bijman & Gijselinckx, 2012). In general, the two cooperatives (COOP1 and COOP2) with predominant spot-market sales, one located in Belgium and one in

the Netherlands, showed less advanced FSMS than the cooperatives with more contractual (i.e., hierarchy-based) transactions. It was discussed before that the relative anonymity of the spot market may lead to lower information exchange between buyer and seller, a loss of quality due to the extra logistic step and the lack of incentive for famers to improve quality and safety (Bijman & Hendrikse, 2003). A collective reputation that does not reveal the identity of the individual producers can lead to lower quality levels (Winfree & Mccluskey, 2005). The cooperative (COOP3) with fewer products sold through the auction clock and more through contracts, demonstrated more advanced levels of control and assurance activities. The use of

K. Kirezieva et al. / Food Control 62 (2016) 299–308

contracts strengthens the coordination in the supply chain (Chaddad, 2012). In general, contracts are supposed to contribute positively to food safety and quality management, as members expect to get proper remuneration for their higher effort in producing quality (Cechin, Bijman, Pascucci, & Omta, 2013). Moreover, contracts commonly include specifications on quality and safety parameters, which require further coordination between members and the cooperative firm, and between the cooperative firm and the customers. COOP3 also invested in logistics (Table 3), which is associated with lower riskiness of context characteristics. Compared to the auction clock, where products from different farms are combined and sold in one lot, sales through contracts allow for specifying quality and safety requirements. The latter provide more incentives for farmers to enhance their FSMS. Furthermore, additional training and advise by the cooperative correlated to more advanced levels of FSMS activities, especially regarding the knowledge-intensive assurance activities. These efforts were triggered by demands from retailers and other buyers as the cooperative was selling a large percentage of the berries via contracts.

5.4. Commitment of cooperative members Another dimension of cooperatives that can possibly affect quality and safety management activities is their size and diversity of business functions. Developing complex business functions may lead to dissatisfaction, lack of trust, involvement and commitment of cooperative members (Nilsson, Kihlen, & Norell, 2009). Commitment is essential to implement a quality assurance strategy and to invest in more advanced FSMS activities (Cechin, Bijman, Pascucci, & Omta, 2013). In this study, COOP4 was much bigger than the other three cooperatives, and COOP3 and COOP4 were considerably more complex in terms of business functions. This was also reflected in their turnover (Table 1). Previous studies discussed that many Dutch cooperatives have introduced more complex business functions, acquired foreign subsidiaries and increased their product diversification to supply the needs of various customers, whereas Belgian cooperatives kept the traditional spotmarket sales and did not diversify (Bijman & Gijselinckx, 2012; Bijman, Hendrikse, & Van Oijen, 2013). Our study is in line with these findings, because COOP1, which was the smallest and having the highest percentage of auction clock sales, was located in Belgium. The other three cooperatives were located in the Netherlands and had higher percentage of contract sales (Table 1). The members of the largest and most complex cooperative (COOP4) in this study showed average and even basic levels for control and assurance activities, meaning that they were not putting effort into adapting their activities and checking their effectiveness in the particular farm. These results might be explained with lower commitment by the members to the strategies of the cooperative firm. This type of cooperative typically shows large heterogeneity among members also in terms of different market objectives (Bijman & Hendrikse, 2003). A previous study discussed the difficulties experienced by cooperatives with losing commitment of members due to heterogeneity and power struggles (Hendrikse, 2011). Moreover, (Harvey & Sykuta, 2006) provided evidence that trust and commitment of farmers to investorowned firms are lower compared to cooperatives. Evidence from another study suggest that cooperatives which provide information on quality improvements have more committed members (Trechter, King, & Walsh, 2002). In our study, there was some variation in the activities of each cooperative. Training was organised systematically only by COOP3, while others provided assistance only sporadically. Indeed, this was the cooperative that showed the most advanced FSMS (Table 3).

307

6. Conclusions and suggestions for future research This study contributes to the existing literature on agricultural cooperatives and the effect of governance of transactions on quality and safety management by linking it to the actual activities in FSMS implemented in the farms. Moreover, it expands the topic by defining the activities through which cooperatives play a role in food quality and safety management in the fresh produce chain. In this study we identified a two-step role of cooperatives in managing quality and safety in the supply chain. On the one hand, they are responsible for supply chain management, including tactical decisions about coordination of quality and safety between farmers, cooperative firm and customers. On the other hand, cooperative firms sell the products of their members and make strategic decisions about the governance of transactions in the supply chain, which ultimately may have an impact on the supply chain management and the FSMS on the farms. The case studies revealed that hierarchy-like governance of transactions has a positive association with quality and safety management systems on the fresh produce farms. This was linked to more effort put in supply chain (quality) management and activities of the cooperative firms to support collaboration and coordination in the chain, such as hazard monitoring, training and advise to the members to implement quality assurance standards, providing logistics, sorting and packaging of the products. On the contrary, cooperatives with predominantly market-based governance of transactions demonstrated less advanced FSMS, mostly following standards and missing some core activities. However, the low commitment and social capital in a cooperative with many members and complex business function may ultimately lead to a less advanced FSMS at the farm. This study was a first exploration of the factors affecting set-up and operation of FSMS in fresh produce supply chains as linked to supply chain quality management and the role of cooperatives. The case studies approach allowed us to dive into the differences between different governance mechanisms and their effect on FSMS implemented at farm level, but did not permit us to include different product groups. Moreover, the research was conducted only in cooperatives located in the Netherlands and Belgium, which limits the generalizability of conclusions towards other parts of the world. Further research should address a larger sample of cooperatives and farms of different production sizes and product groups, as these can influence behaviour and commitment of cooperative members. Moreover, comparison with farmers that are not cooperative members would contribute to better understanding the role of cooperatives to safety management. Longitudinal studies about changes in the supply chain governance structures and the relationship to quality management could be used to further inform food quality and safety management decisions. Acknowledgements This research has received funding from the European Community’s Seventh Framework Program (FP7) under grant agreement № 244994 (project Veg-i-Trade ‘Impact of Climate Change and Globalization on Safety of Fresh Produce – Governing a Supply Chain of Uncompromised Food Sovereignty’; http://www.veg-i-trade.org). We would like to thank to the farmers and cooperatives for their cooperation, and to Arjan Bruls for his support during data collection. References Bijman, J. (2002). Essays on agricultural co-operatives : Governance structure in fruit and vegetable chains. Coöperaties en beheersstructuur in groente- en fruitketens (PhD thesis). Erasmus University Rotterdam.

308

K. Kirezieva et al. / Food Control 62 (2016) 299–308

Bijman, J., & Gijselinckx, C. (2012). Support for farmers’ cooperatives: Fruit and vegetables cooperatives in the Netherlands and Belgium. Support for farmers’ cooperatives. European Commission. Bijman, W. J. J., & Hendrikse, G. W. J. (January 2003). Co-operatives in chains: institutional restructuring in the Dutch fruit and vegetables industry. Journal on Chain and Network Science, 95–107. Bijman, J., Hendrikse, G., & Van Oijen, A. (2013). Accommodating two worlds in one organisation: changing board models in agricultural cooperatives. Managerial and Decision Economics, 34, 204–217. Bijman, J., Iliopoulos, C., Poppe, K. J., Gijselinckx, C., Hagedorn, K., Hanisch, M., et al. (2012). Support for farmers’ cooperatives. European Commission. Blackburn, J., & Scudder, G. (2009). Supply chain strategies for perishable products: the case of fresh produce. Production and Operations Management, 18, 129–137. Cechin, A., Bijman, J., Pascucci, S., & Omta, O. (2013). Decomposing the member relationship in agricultural cooperatives: implications for commitment. Agribusiness, 29, 39–61. Cechin, A., Bijman, J., Pascucci, S., Zylbersztajn, D., & Omta, O. (2013). Quality in cooperatives versus investor-owned firms: evidence from broiler production in Paraná, Brazil. Managerial and Decision Economics, 34, 230–243. Chaddad, F. (2012). Advancing the theory of the cooperative organization: the cooperative as a true hybrid. Annals of Public and Cooperative Economics, 83, 445– 461. Doukidis, G. I., Matopoulos, A., Vlachopoulou, M., Manthou, V., & Manos, B. (2007). A conceptual framework for supply chain collaboration: empirical evidence from the agri-food industry. Supply Chain Management: An International Journal, 12, 177–186. Ec (2002). Regulation (EC) No 178/2002 of the European Parliament and of the Council of 28 January 2002 laying down the general principles and requirements of food law, establishing the European Food Safety Authority and laying down procedures in matters of food safety. Official Journal of the European Union, L 31, 1–24. Fish, L. A. (2011). Supply chain quality management. In D. Onkal (Ed.), Supply chain management – Pathways for research and practice. InTech. Foster, S. T. J. (2008). Towards an understanding of supply chain quality management. Journal of Operations Management, 26, 461–467. Gijselinckx, C., & Bussels, M. (2012). Support for farmers’ cooperatives: Country report Belgium. Support for farmers’ cooperatives. Wageningen, the Netherlands: European Commission. Grabot, B., Vallespir, B., Gomes, S., Bouras, A., Kiritsis, D., Mendes Dos Reis, J., et al. (2014). Supply chain quality management in agribusiness: An approach of quality management systems in food supply chains. Advances in production management systems. Innovative and knowledge-based production management in a global-local world. Berlin Heidelberg: Springer. Harvey, S. J. J., & Sykuta, M. E. (2006). Farmer trust in producer- and investor-owned firms: evidence from Missouri corn and soybean producers. Agribusiness, 22, 135–153. Havinga, T. (2006). Private regulation of food safety by supermarkets. Law & Policy, 28, 515–533. Hendrikse, G. W. J. (2011). Pooling, access, and countervailing power in channel governance. Management Science, 57, 1692–1702. Hendrikse, G., & Bijman, J. (2002). Ownership structure in agrifood chains: the marketing cooperative. American Journal of Agricultural Economics, 84, 104–119. Hoffmann, R. (2005). Ownership structure and endogenous quality choice: cooperatives versus investor-owned firms. Journal of Agricultural & Food Industrial Organization, 3, 1–26. Jacxsens, L., Van Boxstael, S., & Uyttendaele, M. (2011). Selection of case studies based on the vulnerability of fresh produce to food safety hazards and climate change in a globalized world. European Symposium of International Association of Food Protection. Ede, the Netherlands. Kalogeras, N., Pennings, J., Benos, T., & Doumpos, M. (2013). Which cooperative ownership model performs better? A financial-decision aid approach. Agribusiness, 29(1), 80–95. Kaynak, H., & Hartley, J. L. (2008). A replication and extension of quality management into the supply chain. Journal of Operations Management, 26, 468–489. 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, 230–242. Kirezieva, K., Luning, P. A., Jacxsens, L., Allende, A., Johannessen, G., Tondo, E. C., et al. (2015). Factors affecting the status of food safety management systems in the global fresh produce chain. Food Control, 52, 85–97.

Kirezieva, K., Nanyunja, J., Jacxsens, L., Van Der Vorst, J. G. A. J., Uyttendaele, M., & Luning, P. A. (2013). Context factors affecting design and operation of food safety management systems in the fresh produce chain. Trends in Food Science & Technology, 32, 108–127. 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 & Technology, 19(10), 522–534. Luning, P. A., & Marcelis, W. J. (2007). A conceptual model of food quality management functions based on a techno-managerial approach. Trends in Food Science & Technology, 18, 159–166. Luning, P. A., & Marcelis, W. J. (2009). A food quality management research methodology integrating technological and managerial theories. Trends in Food Science and Technology, 20, 35–44. Luning, P. A., & Marcelis, W. J. (2009). Food quality management: Technological and managerial principles and practice. Enfield Pub & Distribution Company. Luning, P. A., Marcelis, W. J., Rovira, J., Van Boekel, M. A. J. S., Uyttendaele, M., & Jacxsens, L. (2011). A tool to diagnose context riskiness in view of food safety activities and microbiological safety output. Trends in Food Science &Technology, 22(Suppl. 1), S67–S79. 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, 20, 300–312. McAfee, Dinesh Satam, McMillan, & Dinesh (1987). Auctions and bidding. Journal of Economic Literature, 25(2), 699–738. Ménard, C. (2004). The economics of hybrid organizations. Journal of Institutuional and Theoretical Economics, 160, 345–376. Ménard, C. (2007). Cooperatives: Hierarchies or hybrids? Vertical markets and cooperative hierarchies. Netherlands: Springer. Mérel, P. R., Saitone, T. L., & Sexton, R. J. (2009). Cooperatives and qualitydifferentiated markets: strengths, weaknesses, and modeling approaches. Journal of Rural Cooperation, 37, 201. Mondelaers, K., & Van Huylenbroeck, G. (2008). Dynamics of the retail driven higher end spot market in fresh food. British Food Journal, 110, 474–492. Nilsson, J., Kihlen, A., & Norell, L. (2009). Are traditional cooperatives an endangered species? About shrinking satisfaction, involvement and trust. International Food and Agribusiness Management Review, 12. Pagell, M. (2004). Understanding the factors that enable and inhibit the integration of operations, purchasing and logistics. Journal of Operations Management, 22, 459–487. Pennerstorfer, D., & Weiss, C. R. (2013). Product quality in the agri-food chain: Do cooperatives offer high-quality wine?. European Review of Agricultural Economics. Raynaud, E., Sauvee, L., & Valceschini, E. (2005). Alignment between quality enforcement devices and governance structures in the agro-food vertical chains. Journal of Management & Governance, 9, 47–77. Richey, R. G., Roath, A. S., Whipple, J. M., & Fawcett, S. E. (2010). Exploring a governance theory of supply chain management: barriers and facilitators to integration. Journal of Business Logistics, 31, 237–256. Rijpkema, W. A., Rossi, R., & Van Der Vorst, J. G. A. J. (2014). Effective sourcing strategies for perishable product supply chains. International Journal of Physical Distribution & Logistics Management, 44, 494–510. Rong, A., Akkerman, R., & Grunow, M. (2011). An optimization approach for managing fresh food quality throughout the supply chain. International Journal of Production Economics, 131, 421–429. Roth, A. V., Tsay, A. A., Pullman, M. E., & Gray, J. V. (2008). Unraveling the food supply chain: strategic insights from China and the 2007 recalls. Journal of Supply Chain Management, 44, 22–39. Trechter, D. D., King, R. P., & Walsh, L. (2002). Using communications to influence member commitment in cooperatives. Journal of Cooperatives, 17, 14–32. Trienekens, J., & Wognum, N. (2013). Requirements of supply chain management in differentiating European pork chains. Meat Science, 95, 719–726. Vorst, J. G. A. J. V. D., Kooten, O. V., & Luning, P. A. (2011). Towards a diagnostic instrument to identify improvement opportunities for quality controlled logistics in agrifood supply chain networks. International Journal on Food System Dynamics, 2, 94–105. Wever, M., Wognum, N., Trienekens, J., & Omta, O. (2010). Alignment between chain quality management and chain governance in EU pork supply chains: a transaction-cost-economics perspective. Meat Science, 84, 228–237. Winfree, J. A., & Mccluskey, J. J. (2005). Collective reputation and quality. American Journal of Agricultural Economics, 87, 206–213.