Since consumers cannot easily observe credence attributes, they may rely on a supplier and/or brand to deliver the expected quality. Thus, consumer confidence in credence attributes was conceptualised as the degree to which consumers perceive the food system (i.e. government, food manufacturers, farmers and retailers) and food brands (i.e. fresh chicken meat and packaged green salad) as trustworthy. Stated differently, both trust in the food actors and brand trust are expected to contribute to consumer confidence in food quality attributes, and ultimately, influence repurchase intentions and brand loyalty. These relationships are portrayed in Figures 6.9 and 6.10 and are analysed individually below. Overall, the model fit is satisfactory (RMSE=.050, CFI=.984 for chicken; RMSE=.042, CFI=.988 for salad), thereby confirming that the proposed network of relationships in which consumer confidence is embedded fits the data.
134
Figure 6.9: SEM of the drivers and consequences of confidence: chicken
135
6.2.3.1 Trust in the Food System and Consumer Confidence
Is it possible that consumers are confident in the food they eat, yet they mistrust some institutions responsible for supplying high quality and safe food? (Ekici, 2004). According to the current findings, the answer to the question is “No”. In fact, results provide strong support for the hypothesized relationship (H1) between confidence in credence attributes (Confid) and trust in the food system (Strust) for chicken (βstand=.47, p<.05) and salad (βstand=.61, p<.05). That is,
when Strust goes up by 1 standard deviation, Confid goes up by 0.47 standard deviations for chicken, suggesting that a higher degree of trust in market actors tends to make consumers feel more confident about fresh chicken. For salad, one standard deviation in Strust is associated with a .61 standard deviation increase in Confid, indicating that the more consumers confer trust in actors of the food chain, the more confident they are about the quality of packaged salad.
Not only is the result consistent with the hypothesized assertion, but it also lends support to extant findings reported from different studies in the food context, showing that a higher level of institutional trust is associated with higher levels of confidence in food (e.g., Berg, 2004; De Jonge et al, 2004; De Jonge et al., 2006; De Jonge et al., 2008b). According to De Jonge et al. (2004: 840), “trust in regulators and actors in the food chain is a minimum requirement for confidence in the safety of food, assuming such trust is one of the mechanisms by which confidence is created and sustained.” On the other hand, the current result suggests that when trust in the food system is damaged, it may erode consumer confidence in credence qualities. Many examples of the consequences of past and recent food safety events can be highlighted. For instance, BSE or so-called “Mad Cow” disease during the 1990s was a pivotal point in the loss of confidence in public and private bodies among many European consumers. It has been suggested that the lack of confidence in beef was connected to mistrust in the British government (Kjærnes, Harvey and Warde, 2007).
6.2.3.2 Brand Trust and Consumer Confidence
While the hypothesized relationship between brand trust (Btrust) and confidence (Confid) is confirmed for chicken (βstand=.22, p=.002), it is rejected for salad (βstand=0, p=.987). The
finding reveals that a higher degree of trust in chicken brands tends to increase consumer confidence in the quality and safety attributes of chicken. This result aligns with those of Bredhal (2003) who shows, in an analysis of the use of quality cues with regard to branded beef,
136
that brand is the predominant quality signal for consumers to form expectations about the healthiness and the eating quality of unprocessed meat brands. However, the present results suggest that consumers’ trust in salad brands is not a significant predictor for confidence in food attributes. This means that degree to which trust in brands influences consumer confidence depends upon the category of the food product (meat versus produce) and perhaps on the range of the brands available on the market. Potential explanations of the variability in results are further explored in Chapter 7.
A comparison between the effect of Strust (βstand=.47) and Btrust (βstand=.22) on Confid in
chicken indicates that the impact of trust in the food actors as a group is twice the effect of brand trust. As such, consumer trust in the food system appears to be more influential in leading to confidence in credence attributes than trust in individual food products. While brands as signalling mechanisms are useful, perhaps trusting them is not sufficient for consumers to make confident expectations about credence qualities and may not work for every food category. According to Singh and Sirdeshmukh (2000), as the relationship between a consumer and an agent evolves (through repetitive purchase), consumers rely more on trust expectations than on the provided signals (e.g., brands) and premiums to judge product quality. Furthermore, the authors speculate that “ for ongoing exchanges, the trust mechanism for affecting performance expectations and price perceptions will be increasingly more prominent relative to the influence of signaling investments and price premiums established by market agents” (Singh and Sirdeshmukh, 2000: 164).
6.2.3.3 Trust in the Food System and Brand Trust
Results from both products indicate that trust in the food system (Strust) has a positive and significant impact on brand trust (Btrust), which supports H3. Furthermore, the magnitude of the influence is about the same for either product: βstand=.69 with p<.05 for chicken, and
βstand=.71 with p<.05 for salad. This infers that increased trust in the actors within the food
supply chain fosters consumer trust in food brands, regardless of the type of product. In fact, previous work suggests that most Canadians assume that production standards and practices adopted in the Canadian system adhere to strict guidelines that are well enforced, allowing them to some extent to have confidence in food (AAFC, 2007).
137
6.2.3.4 Repurchase Intentions and the Trusting Constructs (Confid, Strust)
As theorised, the path coefficient for hypothesis H16 is significant for both chicken
(βstand=.31, p<.05) and salad products (βstand=.39, p=.002), suggesting that trust in the food
system (Strust) positively influences repurchase intentions (Purchas). This yields support to a number of previous studies reporting that trust in market actors such as retailers (e.g., Macintosh and Lockshin, 1997) or farmers (e.g., Yee, 2002) enhances repurchase intentions. For instance, in an EU project on “Food Risk Communication and Consumers’ trust in the Food Supply Chain”, Cavicchi et al. (2005) found that safety information provided by food chain actors (farmers, processors, retailers) has a very large positive impact on purchase intentions for chicken meat in Italy and France. Likewise, Yee (2002) shows a significant and positive causal relationship between consumer trust in livestock farmers and their likelihood of purchasing meat.
Unexpectedly, the influence of confidence (Confid) on repurchase intentions is not significant for either chicken (βstand=-.09, p=.421) or salad (βstand=-.06, p=.673), which does not
validate H15a. This reveals that positive repurchase intentions are better explained by trust in the food system rather than by confidence in credence attributes. Perhaps for consumers who are uncertain about the overall quality of the food products, trusting the food system appears to be more important than trusting food products.
6.2.3.5 Brand Loyalty and the Trusting Constructs
The direct path coefficient between brand trust (Btrust) and brand loyalty (Loyalty) reveals significant support for H17 for chicken (βstand=.27, p<.05) and salad (βstand=.30, p<.05).
The result implies that trusting a brand tends to be an influential factor in establishing brand loyalty. The finding aligns with many previous marketing studies. For instance, Chaudhuri and Holbrook (2001) found a strong positive relationship between brand trust and brand loyalty (purchase loyalty and attitudinal loyalty) based on a data set of 107 brands of 41 different product categories (ice cream, cheese, cereal, bacon, canned fruit, perfume, computers, gasoline, etc.).
The effect of consumer confidence (Confid) on brand loyalty (Loyalty) was not significant for either chicken (βstand=.13, p=.091) or salad (βstand=.13, p=.085). This hints that Confid is a not an important predictor for brand loyalty for the produce and meat categories,
138
thereby rejecting the postulated relationship of H15b. The finding suggests that confidence in food quality and food safety may be not sufficient to secure brand loyalty, a result supported also by Tan, Hishamuddin and Devinaga (2011) in the context of fast food brands. As such, it appears that loyalty does not flow automatically from confidence in individual brands.
Despite the difference in proportions of brand shoppers in both samples (30% buyers of branded chicken against 70% buyers of branded packaged salad), the non-significant result may suggest that there is no difference in confidence in generic or branded food. Perhaps loyalty to a particular food product does not necessarily imply that the product should be branded. With stores such as Superstore that sell the same product (say chicken) under branded and generic versions, it may be that many consumers do not perceive significant differences between store brands or generic versions (at least for fresh food), so choosing any version (i.e. branded or generic raw chicken) could be the same for them.
6.2.3.6 Repurchase Intentions and Brand Loyalty
The relationship between repurchase intentions (Purchas) and brand loyalty (Loyalty) is corroborated for chicken (βstand=.42, p=.004), but not for salad (βstand=.21, p=.162). This result
infers that repurchase intentions enhance consumers’ commitment (loyalty) to a particular brand of chicken but not to salad brands. The finding relative to the chicken product is in line with a number of studies within both food and non-food contexts. For instance, Gogoi (2013) found that purchase intentions have a significant impact on the development of brand loyalty for a private label in the Apparel industry. The difference in results suggests that purchase and loyalty in the food context vary among product categories. While not directly explored in the survey, if a consumer does not intend to repurchase the same brand of salad, that does not necessary mean that brand may not be of higher quality. Perhaps, the consumer may want to try new alternatives.
Synthesis
The significance of the structural relationships of sub-model 3 is summarized in Table 6.7 that shows all relationships are supported for chicken except for H15, and four paths are not supported for salad. For both products, trust in the food system enhances overall consumer confidence in food, contributes to brand trust, and leads to positive repurchase intentions. Moreover, consumer confidence is not a predictor either for repurchase intentions or for brand
139
loyalty for the chicken and the salad products. The effect of brand trust on brand loyalty is supported for both products. Nevertheless, trust in chicken brands contributes to building consumer confidence, which is not the case for packaged green salad. Similarly, repurchase intentions leads to brand loyalty for chicken but not for salad. The discrepancy between results across both products suggests that the characteristics of a food product category may influence brand-level effects such as brand trust and brand loyalty.
Table 6.7: Summary of the significance of the structural relationships
Hypothesized paths Chicken Salad
H1: System trust → Consumer confidence Supported Supported H3: Brandtrust→Consumerconfidence Supported Not supported H3: System trust → brand trust Supported Supported H15a: Consumer confidence → Repurchase intentions Not supported Not supported H15b: Consumer confidence → Brand loyalty Not supported Not supported H16: System trust → Repurchase intentions Supported Supported H17: Brand trust → Brand loyalty Supported Supported H18: Repurchaseintentions→Brandloyalty Supported Not supported