A behavioural reasoning perspective on the consumption of local food. A study on REKO, a social media-based local food distribution system

Consumption of local food is increasing due to its convenience and multiple perceived benefits. However, the factors shaping consumer attitude towards local food and brand love for local food have received limited academic attention. The study thus investigates the under-explored concept of self-identity, attitude towards local food, and brand love, along with contextual variables with respect to local food consumption. Behavioural Reasoning Theory (BRT) is adopted as the theoretical frame in which self-identity is utilised as value, supporting the environment and positive product perception about the food sold by REKO are the reasons for choosing local food, and negative product perception about the food sold by REKO and price are reasons against selecting local food from REKO. Data from 2045 Finnish consumers of local food were collected and analysed, with the findings indicating that self-identity is associated with attitude towards local food, and reasons for as well as against the consumption of local food distributed by REKO. Moreover, reasons for, as measured through supporting the environment and positive product perception, and reasons against, as measured by negative product perception, are associated with attitude towards local food. However, along with reasons for, only price, representing reasons against, is associated with brand love, which, in turn, is associated with attitude towards local food. Reasons and attitude towards local food also have a mediation effect on the hypothesised associations. The findings contribute to improved theoretical insights on BRT, brand love, and local food distribution systems and offer useful managerial recommendations to help promote local food consumption and their underlying distribution systems.


Introduction
Local food is considered highly important by the local communities. It offers multiple benefits to consumers, such as transparency of production (Skallerud & Wien, 2019), a natural product content (Lim & Hu, 2016), awareness about the value chain and origins of the products (Holcomb et al., 2018), and other benefits (Costanigro, Kroll, Thilmany, & Bunning, 2014;Hempel & Hamm, 2016). Local food consumption is becoming popular, especially in developed regions, including the United States (Hedberg & Zimmerer, 2020) and Europe (Skallerud & Wien, 2019). Various national governments also promote and support local food products (Jensen et al., 2019), which indicates that the production of these foods can be expected to grow in the years to come. To capitalise on these growing trends, several restaurants have incorporated local food into their menu (Shafieizadeh & Tao, 2020). Indeed, the market for local food was expected to generate a revenue of more than 20 billion USD in 2020 (Zhang et al., 2020).
Due to this increasing popularity and rise in the consumption of local food, scholars have focused on acquiring a deeper knowledge of the varying perspectives on local food consumption (Farmer & Betz, 2016). Although the related literature has expanded recently, much of the scholarship has been centred on the local food system in the context of rural areas and their development (e.g., Mundler & Laughrea, 2016), the distribution of local food (Hedberg & Zimmerer, 2020), and consumer behaviour (Birch & Memery, 2020;Choe & Kim, 2019). At the same time, despite the debate centred on local foods, the sales of such products have remained low, accounting for only 8% of total in-store grocery sales (Paloviita, 2014) in the first half of the last decade in Finland. In more recent years, i.e., from 2015 to 2019, there has only been a moderate increase in the interest of Finnish consumers in local food (Niinimäki, 2020).
Scholars have noted that the absence of a large-scale distribution channel to deliver local food is the biggest obstacle to the area's growth (Korpela, 2019). A more serious challenge with local food, however, is that it is usually produced in small quantities by small firms, which restricts the availability and distribution of such food to the mainstream audience (Korhonen et al., 2017). Since many consumers classify local food as food that has been grown/processed within the 100-mile vicinity of the outlet from where it is sold (Statista, 2018), it is quite plausible to say that distribution systems are inextricably tied with the perception towards local foods. This argument can be further supported based on the fact that if local foods were sold through the regular grocery stores, their unhindered supply might not be assured as these stores have multiple products. Such stores might ask for bulk deals that individual, small, and less organised local food producers may not be able to provide. However, if the consumers are assured that there are dedicated channels of distribution for such products, they might be ready to change their food choices to include more local options in their diets. In the present study, we thus examine consumer behaviour towards local foods in tandem with their distribution systems.
Using this perspective, we seek to address the following three research gaps: (a) scholars have investigated the different drivers of local food consumption but have been unable to explain the poor reception of local food by consumers (Zhang et al., 2020); there are limited studies on behavioural reasoning processes to explain the consumption of local food, with no past study having investigated local food consumption in connection with their associated delivery systems, despite the fact that the two may be closely related; (c) no study has explored brand love for local foods and their associated distribution systems that could explain the loyalty of consumers towards certain local foods in connection with their distribution systems.
We posit that examining the consumer behaviour towards local foods in the context of their distribution systems can thus help the stakeholders concerned, such as farmers, producers, and policymakers, to create sustainable structures and strategies that can drive local food consumption so that it becomes a way of life. Given that consumers, producers, group moderators, and policymakers can understand the dynamics of local food consumption and strategise their resources, we propose to examine the consumer behaviour of Finnish local food consumers since it offers an opportunity to examine not only the perception toward local foods but also their associated distribution system. Specifically, we examine REKO (Rejäl konsumtion or fair consumption), which is a social media platform-based local food delivery system developed in Finland. Indeed, REKO has become a synonym for local food. Recent studies have noted this fact and investigated consumer behaviour in this context (Kumar et al., 2021b).
REKO groups have established themselves as one of the most popular food delivery channels (Salminen, 2019). These groups operate through Facebook, where consumers and producers work together to decide the order and deliveries. It is widely present in Finland, Sweden, and Norway, with members of these groups varying from a few hundred to multiple thousands (Salminen, 2019). REKO further offers numerous benefits to both consumers and producers. For example, consumers can receive local food directly from the farm, which is fresh, tasty, and nutritious (Salminen, 2019), and can obtain a sense of satisfaction by contributing to the local economy through such purchases (Aitojamakuja, 2020). On the other hand, producers receive greater value for their products by eliminating middlemen and experience lower transportation costs, fewer storage wastages, and have almost no need to use preservatives (Salminen, 2019). REKO thus promotes environmentally-friendly behaviour by reducing the emission of harmful gases as a result of the shorter supply chain as well as optimising the utilisation of resources (Aitojamakuja, 2020).
Taking the discussion further, we propose to examine the behavioural response of the existing consumers of local food in the context of its associated distribution system through Behavioral Reasoning Theory (BRT; Westaby, 2005) since the integration of multiple perspectives related to behavioural reasoning can offer greater insights into the determinants of consumers' decisions (Tandon et al., 2020). The current study utilises BRT as a theoretical framework for two reasons. First, BRT theorises that individuals' values and beliefs offer the approach (reasons for) and avoidance (reasons against) for their global motive (e.g., attitude), which further influences behavioural intentions (Sahu et al., 2020). Second, the theory utilises multiple contextual variables, which are useful in explaining specific food consumption behaviour (Tandon et al., 2020). BRT has also been used in several areas, such as sustainable logistics (Peterson & Simkins, 2019) and environment-friendly products (Tandon et al., 2020). Through a review of the literature, the current study has found that self-identity (Carfora et al., 2017) is a suitable choice for the value/belief component of BRT, as it is likely to influence the reason for and reason against the consumption of local food. Selfidentity is defined as an overall assessment between the total received and total expected benefits, which is based on the consumers' environmental desire and pro-environmental behaviour (Confente, Scarpi, & Russo, 2020). Furthermore, self-identity is an important factor that promotes pro-environmental behaviour, including local food consumption.
We have identified as well that supporting the environment and positive product perception about local food distributed by REKO are reasons for consuming local food (e.g., Memery et al., 2015), whereas the negative product perception about such food and price act as reasons against it (e.g., Aschemann-Witzel & Zielke, 2017). The 'reasons for' component of BRT promotes consumption, whereas the "reasons against" component inhibits the consumption (Tandon et al., 2020). Lastly, to estimate the outcome of the behavioural reasoning process, the study has utilised attitude and brand love as the outcome variables for local food consumption. To offer robust findings, we have also investigated the mediation effect of reasons and attitude towards local food on the proposed associations.
In specific terms, the study attempts to address three research questions (RQs). RQ1: How is self-identity associated with reasons for (supporting the environment, positive product perception), reasons against (negative product perception, price), and attitude towards consumption of local food distributed by REKO? RQ2: How are reasons associated with attitude and brand love for local food distributed by REKO, and how is attitude towards local food associated with brand love? RQ3: Do reasons and attitude towards local food mediate the proposed associations? We analysed the data collected from 2045 Finnish consumers of local food distributed by REKO towards this end.
The present study offers three unique contributions. To begin with, is is the first empirical work to use and empirically validate BRT to explain behavioural reasoning perspectives related to consumer behaviour toward local food and its associated distribution system. Second, it offers new findings by investigating a relatively new and less-studied sales and distribution system, i.e., REKO, which is based in Finland. Lastly, the study goes beyond behavioural intentions and investigates a more global and under-examined construct, i.e., brand love, by extending the previous findings (e.g., Birch & Memery, 2020) to explain the attitude towards local food and brand love for local food and its distribution system. Thus, the study offers a deeper understanding of brand love for local food and its associated distribution system.

Local food
Although there is no formal definition of local food (Feldmann & Hamm, 2015), it can be considered to be food distributed through a short supply chain or local food system in which production and consumption are in proximity to one another (Skallerud & Wien, 2019). Some studies have classified products as local food when the consumption is within a 100-mile radius from the production (e.g., Korhonen et al., 2017), while others have used administrative boundaries, i.e., country or state borders (Darby et al., 2008), to label products as being local. However, a few studies have used additional factors, such as product type (e.g., Lim & Hu, 2016), production technique (traditional is local) (e.g., Skallerud & Wien, 2019), size of the firm (small firms produce local) (e.g., Autio et al. 2013), and recipes (specific to the local area) (e.g., Witzling & Shaw, 2019) to classify products as such. The understanding of local food also varies across geographies. For example, the North American view of local food is dependent on environmental sustainability and the principle of social justice (e.g., Holloway et al., 2007). In comparison, the European view is based on aligning small farms and their associated agricultural activities with development at the national level (e.g., Skallerud & Wien, 2019). Thus, it can be inferred that the understanding of local food varies among scholars (e.g., Ricketts Hein et al., 2006).
Past studies have suggested that the purchase of local food is driven by multiple factors, such as supporting local economic activities (Jensen et al., 2019) or health benefits received from the food's intrinsic qualities of freshness and a natural taste (Feldmann & Hamm, 2015). Scholars have also observed that out of multiple factors which motivate consumers to buy local food, supporting the local environment is a strong driver of local food consumption (Memery et al., 2015). In fact, supporting the local environment has been found to be more important than other variables related to food choices (e.g., local, environmentally friendly, and animal-friendly) (Weatherell et al., 2003). Personality traits have also been found to impact consumers' preference for local products (Bazzani et al., 2017). Furthermore, local food studies are also country-specific. For example, Arsil et al. (2014) identified that saving money and health benefits are important factors for the Javanese ethnic group in Indonesia, while Ditlevsen et al. (2020) concluded that different types of Danish consumers prefer local food.
Moreover, most consumers have a positive disposition towards local food, which is then reflected in their purchase intentions (Bianchi & Mortimer, 2015). However, Penney and Prior (2014) showed that positive intent is not always predictive of buying behaviour, suggesting that in addition to attitude towards local food, other factors need to be considered while investigating local food buying behaviour. For instance, Birch and Memery (2020) argued the effect of the environment on local food purchase while other scholars have suggested the role of individual values and beliefs along with intrinsic product qualities (Memery et al., 2015;Jensen et al., 2019). Therefore, the current study utilises additional variables to attitude towards local food, such as value/belief and reasoning (both for and against), to explore the brand love for local food through the theoretical lens of BRT.

Behavioural reasoning theory (BRT)
BRT interlinks variables, such as beliefs, values, global motives, intentions, and behaviour (Westaby, 2005), to posit that individuals with beliefs and values have reasons for and against performing the desired behaviour (Sahu et al., 2020). The theory accepts that such reasons impact motives and intentions by acting as vehicles to shield and legitimise a person's behaviour, thereby assisting and protecting their self-esteem (Sahu et al., 2020). In terms of values, BRT clarifies the change in individuals' expectations and why and how these reasons share an important association among beliefs and values, motives, intentions, and behaviour (Westaby 2005). In BRT, values and beliefs are theorised to affect the reasons that individuals use to clarify their behaviour (Sahu et al., 2020;Westaby, 2005). Furthermore, BRT classifies the reasons to perform a behaviour into two categories: the reason for and the reason against. Accordingly, these reasons directly influence attitude towards local food. BRT also assumes that an individual evaluates an alternative by identifying reasons for and against. Furthermore, BRT suggests that these reasons also influence intentions, as individuals with reasons behind their behaviour are more comfortable in performing and justifying the action (Ryan & Casidy, 2018;Westaby, 2005).
This study utilises BRT as its theoretical frameowrk to explore the behavioural reasoning perspectives associated with brand love towards local food. In this direction, the present study examines how the consumers' values and beliefs shape reasons for and reason against along with attitude towards local food, which, in turn, influence brand love for local food.

Brand love
Brand love is the passionate and emotional association of satisfied customers with a given brand or trade (Carroll & Ahuvia, 2006). The idea is derived from theories of love, brand attachment, constructive evaluation, favourable emotions, and representation of love (Ahuvia, 2005;Mody & Hanks, 2020), along with satisfaction, loyalty, and recommendation (Carroll & Ahuvia, 2006;Sarkar, 2014). Brand love is considered an overarching construct that encompasses satisfaction and represents a more general behavioural construct (Bagozzi, Batra, & Ahuvia, 2017;Islam & Rahman, 2016). Scholars have examined brand love in multiple contexts as well, including impulse buying (Sarkar, 2014) and authenticity (Mody & Hanks, 2020). Kumar et al. (2021a) found that health consciousness and environmental concern can indirectly predict brand love for natural products. In addition, Bagozzi, Batra, and Ahuvia (2017) developed a scale to measure brand love, while Batra et al. (2012) explored this concept in several products, identifying three key factors of brand love: positive emotional attachment, self-integration with the brand, and passion-guided behaviour. Studies have also suggested that brand love develops over a long period and exists beyond a transactional relationship as it is close to one's concept of self (Bagozzi et al., 2017;Batra et al., 2012;Carroll & Ahuvia, 2006). Despite the importance and criticality of the brand love concept in the consumer behaviour literature, almost no study has explored brand love for local food, which is a significant research gap. This study thus fills this gap by conceptualising and empirically testing a research model that includes brand love as an outcome variable.

Research model
The conceptual model of the study is based on the BRT framework, as presented in Fig. 1. The study focuses on self-identity as a value for two reasons. First, the extant literature has explored the influence of selfidentity in multiple contexts, including pro-environmental consumption (Confente et al., 2020). As such, it can be applied in the context of local foods and related distribution systems as well. The use of selfidentity is also apt since consumers' consumption orientation is guided by their beliefs, characteristics, and lifestyle, which is captured by this construct (see Johe & Bhullar, 2016). Second, self-identity has been shown to both, directly and indirectly, influence behavioural intentions (Carfora et al., 2017), such as adoption (Barbarossa et al., 2017) or purchase intentions (Confente et al., 2020), which can be extended to investigate attitude towards local food and brand love, as previous studies have confirmed the association between intentions, attitude towards local food, and brand love (e.g., Hegner et al., 2017). Similarly, we identified reasons for and against based on our qualitative study and previous literature. Our review revealed that a feeling of obligation to support the environment and the benefits received from the quality of the product are reasons that can cause consumers to consume local food.
However, some product attributes and expensive pricing can act as potential reasons against this type of consumption. Thus, our conceptual model comprises supporting the environment and positive product perception as reasons for and negative product perception and price as reasons against. Furthermore, the study utilises attitude towards local food and brand love as outcome variables, of which brand love has remained under-investigated in the local food context. Lastly, no prior study has investigated the concept of brand love for local food distribution systems, even though it is a more holistic representation of consumer's behaviour (Ahuvia, 2005), as it is broader than satisfaction and intentions (Sarkar, 2014). These study measures are described in Table 1.

Self-identity and reasons
Self-identity is a cognition-based construct that describes 'who am I?' (Hogg, 2001). Scholars have extended this understanding to the environmental context, suggesting that self-identify is the consumers' overall evaluation and identification with pro-environmental behaviour, which reflects their environmental desire and sustainable expectations (Barbarossa et al., 2017;Khare & Pandey, 2017). Furthermore, previous studies have found an association between self-identity and proenvironmental behaviour. Barbarossa et al. (2017) found that selfidentity is related to intentions to adopt electric cars, while Johe and Bhullar (2016) concluded that self-identity increased the intentions to consume environmentally-friendly products by influencing attitude towards local food and group norms. Confente et al. (2020) further indicated that the green self-identity determines the perceived value, which helps in deciding whether to consume such products. Scholars have also suggested that self-identity supports proenvironmental actions (Gkargkavouzi, Halkos, & Matsiori, 2019), which include supporting the environment (e.g., Memery et al., 2015) and intrinsic quality of food (e.g., Memery et al., 2015;Birch et al., 2018). Pro-environmental consumers are likely to take the necessary steps to support the environment, such as supporting the local economy and local farmers (Memery et al., 2015). Meanwhile, they ensure that they may benefit from these pro-environmental actions, such as the consumption of local food, which is perceived to be healthy, pure, fresh, and free from harmful chemicals (e.g., Jensen et al., 2019;Lim & Hu, 2016). Based on the above arguments, we posit that self-identity may be associated with positive drivers of choosing local food, which include supporting the environment and the positive product perception about  Brand love is considered to be more overarching than customer satisfaction.
Carroll & Ahuvia, 2006 local food distributed by REKO. Thus, it can be hypothesised as follows, which partially addresses research question 1.

H1.
Self-identity is positively associated with supporting the environment. H2. Self-identity is positively associated with the positive product perception of local food distributed by REKO.
Furthermore, previous studies have indicated that self-identity helps consumers assess the fit between their beliefs about the product and their existing beliefs about the self (Confente et al., 2020;Gkargkavouzi et al., 2019). A poor fit indicates that products will not be liked by consumers and, therefore, will act as a reason against buying them (Khare & Pandey, 2017). In the context of local food, a poor fit could explain why consumers hold a negative evaluation of it. The reasons against choosing the food may also be associated with physical appearances, such as its shape, colour, or taste (Zhang, Chen, & Hu, 2019), or with the price of this food, which is usually considered higher than that of its non-local counterpart (e.g., Printezis & Grebitus, 2018;Profeta & Hamm, 2019). Based on the above argument, we can hypothesise as follows, which partially addresses research question 1.

H3.
Self-identity is negatively associated with the negative product perception of local food. H4. Self-identity is negatively associated with the price of local food.

Self-identity and attitude towards local food
Attitude represents consumers' favourable or unfavourable predilection towards a particular behaviour (Tandon et al., 2020;. Prior studies have found an association between self-identity and attitude (Khare & Pandey, 2017). Björk and Kauppinen-Räisänen (2016), for example, indicated that individuals differ in their attitudes towards local food, which shapes their food-related behaviour, in turn. Choe and Kim (2018), meanwhile, concluded that the high quality of food influences the attitude towards local food as well as destinations offering it. Gkargkavouzi et al. (2019) and Carfora et al. (2017) also suggested positive association of self-identity with attitude along with personal norms, subjective norms, and perceived behavioural control. We argue that consumers' self-identity will allow consumers to assess the extent of fit between their existing pro-environmental values and beliefs about local food. Consumers may find a fit for local food in their lifestyle, values, image, or personal characteristics, thereby suggesting a favourable attitude. Thus, it can be hypothesised as follows, which partially addresses research question one. H5. Self-identity is positively associated with the attitude towards local food.

Supporting the environment, attitude and brand love
Supporting the environment refers to the contribution that consumers offer to environmental sustainability (Kumar & Smith, 2018). Scholars have found that consumers consume local products not only because of their perceived health benefits but also because they contribute to the welfare of the environment (Low et al., 2015). Critical of the increasing quantity of food imports, consumers consider local food to be superior alternative to non-local food and more climate and environmentally friendly (Feldmann & Hamm, 2015). Kumar and Smith (2018) further found that consumers who were concerned for the environment had a stronger attitude towards local food, which influenced their purchase intention as a result. In addition, Witzling and Shaw (2019) concluded that liberal and adventurous consumers had a strong preference for local food due to its environmental benefits. The study argued that consumers desiring to support the local environment are likely to develop a favourable orientation towards local food, which may cause them to form a passionate bond with it. Thus, it can be hypothesised as follows, which partially addresses research question two.
H6. Supporting the local environment is positively associated with the attitude towards local food. H7. Supporting the local environment is positively associated with brand love for local food.

Positive product perception, attitude towards local food and brand love
Positive product perception or intrinsic quality refers to the holistic evaluation of local food, which is estimated by characteristics, such as its freshness, purity, or safety (Memery et al., 2015). Scholars have confirmed that food quality, such as taste or freshness (Birch et al., 2018), is a key determinant of local food consumption. These foods offer greater transparency over how the food is grown and what kind of chemicals are used (Shafieizadeh & Tao, 2020). Scholars have shown that the intrinsic quality of food is the key driver of intentions and satisfaction (Zhang, Chen, & Hu, 2019). In the case of local food, consumers are motivated by characteristics that offer multiple healthrelated benefits along with the satisfaction of acting in a proenvironmental way (Choe & Kim, 2019). These benefits, which are offered by the food's freshness or purity, may shape consumers' overall evaluation, thereby helping in the formation of a favourable attitude towards local food. Furthermore, pro-environmental consumers intend to accept the local food due to its perceived environmental, local, and health related benefits. Given these benefits, consumers may develop a meaningful and passionate connection with local food, which evokes a sense of love and emotional attachment with local food. Thus, it can be hypothesised as follows, which partially addresses research question 2.

H8.
Positive product perception about local food distributed by REKO is positively associated with the attitude towards local food. H9. Positive product perception about local food distributed by REKO is positively associated with brand love for local food.

Negative product perception, attitude towards local food and brand love
Product attributes or negative product perception refer to the elements of a product through which its benefits are communicated and delivered (Cahyani & Iriani, 2019). These include design, features, and style, which in the context of local food, translates to shape, colour, and taste. Scholars have found that product attributes have an association with attitude and choice variables (e.g., Wen Wan, Peng Chen, & Jin, 2017). Wang and Yu (2016) found that product attributes, namely, functional, sensory, and branding, influence perceived values such as hedoninc and utilitarian, which determine intentions to repurchase ready-to-drink coffee beverages. Cahyani and Iriani (2019) revealed that the product attribute of being halal-specific was the most dominant influencer of consumers' attitude towards local food served in a fastfood restaurant. Similarly, Wang and Somogyi (2018) confirmed that the attributes of familiarity, consumption accompany, and sensory attributes are positively associated with the attitude towards local food towards shellfish.
These studies indicate that product attributes act as stimuli and influence the effect, cognition, and behaviour, which, in turn, shapes the overall evaluation of the product (Cahyani & Iriani, 2019). The product attributes associated with food items are appearance, shape, colour, and taste. In the case of local foods, these attributes may be perceived negatively by consumers since local foods are allowed to grow 'as nature intended', causing them to have less-than perfect shapes, be off-colour or have less-pronounced flavour. In other words, the local farm produce may not resemble the food obtained from non-local sources that are transformed in certain ways to achieve visibly and commerciallyappealing size, shape, taste, and appearance. Due to this, we argue that the negative product perception contributes to the unfavourable overall evaluation of local food distributed by REKO. Furthermore, consumers are emotionally connected to specific product attributes (Wen Wan, Peng Chen, & Jin, 2017), which may apply to local food as well. This suggests that consumers may develop a strong but negative emotional connection with the local food, which reflects (albeit adversely) the brand love for local food. Thus, it can be hypothesised as follows, which partially addresses research question 2.

H10.
Negative product perception about local food distributed by REKO is negatively associated with the attitude towards local food. H11. Negative product perception about local food distributed by REKO is negatively associated with brand love for local food.

Price, attitude towards local food and brand love
Price is an important criterion for food purchase. However, the extant findings offer a mixed result regarding the role of price in local food purchase. For example, Zepeda and Nie (2012) found that consumers placed a higher priority on price, which was negatively associated with the consumption of local food. Kumar and Smith (2017) also indicated that consumers may not purchase local food due to price. On the other hand, Feldmann and Hamm (2015) observed that consumers did not perceive this food as expensive, possibly negating the purchase barrier. Sirieix, Kledal, and Sulitang (2011), meanwhile, indicated that price is the key benefit of local food. These studies thus provide mixed findings for local food with respect to price. Consumers may expect a lower price, while, at the same time, they may interpret low price as indicative of poor quality (Aschemann-Witzel & Zielke, 2017). Therefore, we argue that higher price inhibits the favourable evaluations of local food. Consumers may perceive local food as expensive and, thus, may prefer non-local products, which they can obtain at a lower price. In turn, they may form a negative attitude towards local food. Moreover, this higher price is not welcomed by consumers and may generate negative emotions associated with the local food. Thus, it can be hypothesised as follows, which partially addresses research question 2.

H12.
Price is negatively associated with the attitude towards local food H13. Price is negatively associated with brand love for local food

Attitude towards local food and brand love
Attitude is the holistic and relatively enduring evaluation of a person, entity, or product that can be either positively or negatively valenced (Hegner et al., 2017). Previous studies noted a positive association of attitude with brand love, such as the research by Han et al. (2019), who observed that product and service quality determined attitude and thereby positively influenced brand love. Navaneethakrishnan and Sathish (2020), meanwhile, found that attitude determines brand love for online purchases. Similarly, Hegner et al. (2017) noted that attitude influences brand love, which further influences uniqueness, pleasure, intimacy, and memories for fashion products. Although brand love is less researched in the context of food, we posit that REKO-distributed local is perceived by consumers as providing multiple benefits, such as health-related or environmental ones, which are then favourably evaluated by consumers (Barska & Wojciechowska-Solis, 2020). Consumers receiving greater such benefits are thus likely to be more emotionally connected with local food offered by REKO. Moreover, Batra et al. (2012) indicated that attitude is favourable for brand love, which may plausibly be extended to REKO and the local food context. Thus, it can be hypothesised as follows, which partially addresses research question 2.
H14. Attitude towards local food is positively associated with brand love for local food distributed by REKO.

Mediation effect
Along with the hypothesized direct effects, the study also explores the mediation effect of reasons and attitude on all of the relationships. Past research that has used the BRT framework suggested that these three variables mediate the relationship between predictor and outcome variables (Sahu et al., 2020). For instance, Tandon et al. (2021) confirmed the mediating role of reasons for and reasons against in the association between values and attitude. They also found that attitude mediates the association of both reasons for and against with purchase intentions. Furthermore, Ryan and Casidy (2018) revealed the mediating role of reasons in the association of values with attitude. We thus posit that consumers' self-identity is likely to influence the reasons and attitude towards local food, which, in turn, determines brand love. In addition, we aniticipate that both types of reasons mediate the relationship between self-identity and attitude. Thus, it can be hypothesised as follows, which addresses research question 3.

H15a-d.
The reasons for (supporting the environment and positive product perception) and reasons against (negative product perception and price) mediate the association between self-identity and attitude towards local food distributed by REKO. H16a-d. Attitude towards local food mediates the association between the reasons for (supporting the environment and positive product perception), reasons against (negative product perception and price), and brand love for local food distributed by REKO.

Control variables
Various factors can confound the results. Due to this, the present study utilised certain sociodemographic variables as control variables. This approach was guided by previous studies, which have suggested that demographic characteristics may influence individuals' consumption patterns (e.g., Kumar et al., 2021a;Witzling & Shaw, 2019). For example, Witzling and Shaw (2019) indicated that demographic variables, namely, income, age, education and gender, influenced the consumption of local food. Sirieix et al. (2011) also indicated the influence of demographic variables on local products. Based on the above studies, we posit that sociodemographic characteristics may influence the consumption of local food. We thus utilised age, gender, household size, and economic background as control variables, thereby controlling their effect on the dependent variables, i.e., attitude towards local food and brand love.

Research context
The context of this study was a social media-based local food distribution system. REKO (Rejäl komsumtion) is a Facebook-based online group that was founded in 2013 in Nordic countries (Sweden, Norway, and Finland). As of today, there exist more than 500 such REKO groups or other online sales and distribution channels for local food (Aitojamakuja, 2020), with 195 in Finland alone. The member size of these groups ranges from a few hundred to multiple thousands (Einiö, 2020). For example, REKO Helsinki (8,142 members) and REKO Espoo (6,620 members) are some examples of REKO groups based in the Helsinki capital region. These groups are continually growing as they provide direct contact between local food suppliers and consumers without any intermediaries. REKO operates on four principles, which include no retailing, a focus on local products, ethical production and transparency of processes (Aitojamakuja, 2020). Anyone following these rules can also start a REKO group, although it should contain the word REKO in its name.

Data collection: Qualitative study
To better understand the perception of consumers about local food distributed by REKO, we conducted a qualitative study through semistructured interviews, wherein we interviewed both consumers (n = 9) and non-consumers (n = 10) of local food. The respondents were selected on the basis of their familiarity with REKO and local foods. We used semi-structured interviews for the qualitative study since these can clarify the reasoning behind respondents' perceptions, which is of considerable importance from our study's perspective. Of the 10 nonconsumers interviewed, all were working professionals, between 26 and 70 years in age. In addition, seven were female. Of the nine local food consumers interviewed, however, the gender ratio was much more even, with five men and four women. These respondents were randomly chosen from two REKO gatherings. The working status of the interviewees was not inquired. However, their ages were estimated to range from 25 to 60 years old. Our analysis of the responses helped us identify the consumers' perceptions related to local food distributed by REKO.

Data collection: quantitative study
Data for the hypothesis testing were collected through a crosssectional survey of 2045 REKO consumers. For the data collection, we contacted the moderators of 167 Finnish REKO groups through Facebook. Of these, 132 permitted us to collect data from the groups' members, with two more declining to participate and no response given from the remaining 33. The moderators posted the survey questionnaire on their respective Facebook pages. In addition, two sets of reminders were also given before the deadline of the survey. The collected survey data was part of a larger project aiming to understand consumer behaviour towards the local food distributed through REKO .
The items for the study constructs were either adapted versions of pre-validated scales or developed based on the qualitative study. Selfidentity was adapted from Huh (2011) and Koh and Noh (2009), the positive product perception was adapted from Memery et al. (2015), attitude from Honkanen et al. (2006), and brand love from Carroll and Ahuvia (2006). Items for three constructs, namely, supporting the environment, negative product perception, and price were developed based on inputs from the qualitative study. This preliminary questionnaire was first presented to the expert panel (two professors and two practitioners from the area) for their feedback. We then pilot tested it with 10 Finnish consumers of local food. The feedback from the study was used to revise the preliminary questionnaire and prepare the final survey. This step was taken to ensure the questionnaire's face and content validity. The response for all items was collected on a five-point Likert scale where 1 represented "strongly disagree", and 5 represented "strongly agree". The final list of study measures and measurement items are presented in Table 3.

Data analysis method
The present study analysed the data using the popular covariancebased structural equation modelling (CB-SEM) technique, frequently used by studies published in the area of consumer behaviour (e.g., Talwar et al., 2019;. Thus, we used the two-step approach, wherein we first generated the measurement model through confirmatory factor analysis to evaluate the validity and reliability of the study measures. Thereafter, we tested the proposed hypotheses through structural path analysis using AMOS 22. Prior to applying CB-SEM, we examined the multivariate characteristics of the data to ascertain its suitability for CB-SEM, as suggested (e.g., Kaur et al., 2020).

Preliminary analysis
After discarding missing and invalid responses, 2045 responses were analysed. The demographic profile of respondents is presented in Table 2. Before performing the analysis, we examined the data for its distribution. The values of skewness and kurtosis, being within − 3 to + 3, confirmed the normalcy of distribution. Multicollinearity, meanwhile, was investigated through VIF values, which ranged from 1 to 3, thus revealing the absence of multicollinearity (Hair et al., 2010). Common method bias was also examined by performing Harman's single factor test. The results highlighted that a single factor could account for only 23.7% variance, which is much less than the cut-off value of 50% (Podsakoff et al., 2012).

Measurement model
To begin, we conducted a confirmatory factor analysis (CFA) to examine the reliability and validity of the constructs. Internal consistency was estimated through Cronbach's alpha values, which were found to be above the cut-off value of 0.7 for each construct (Hair et al., 2010). The composite reliability values for all constructs were also above the cut-off value of 0.7 (Fornell & Larcker, 1981), as were the factor loadings for all items (Table 3), thereby exceeding the recommended cut-off of 0.5 (Hair et al., 2010). Furthermore, the average variance explained (AVE) values (Table 4) surpassed the threshold value of 0.5 and confirmed the convergent validity. We also established discriminant validity as the inter-construct correlation values were smaller than the square root of respective AVE values (Fornell & Larcker, 1981). Lastly, all fit indices values were well-within the indicated limit (i.e. χ 2 /df = 2.05, CFI = 0.99, TLI = 0.99, RMSEA = 0.02), confirming a good model fit (Hair et al., 2010;Tabachnick & Fiddel, 2007).

Control variables
The results of the analysis revealed that age, gender, economic background, and household size did not have a confounding effect on attitude towards local food. Similarly, age, gender, and household size had no confounding effect on brand love, while economic background did (-0.05*).

Mediation analysis
In addition to the direct effect, the study also investigated the mediating role of reasons and attitude towards local food in the association between self-identify, attitude towards local food, and brand love. Results are presented in Tables 5 and 6. The outcome of the analysis indicated that the two reasons for (supporting the environment, positive product perception) and reasons against (negative product perception, price) partially mediated the association between selfidentify and attitude towards local food (Table 5) (H15a-d). Moreover, attitude towards local food was found to partially mediate the association of the two reasons for (supporting the environment, positive product perception) and reasons against (price) with brand love (H16ac) (Table 6). Additionally, attitude towards local food fully mediated the association of negative product perception with brand love (H16d).

Discussion
The study investigated three research questions that were addressed through the associated hypotheses. In response to RQ1, which queried about the nature of the potential association of self-identity with reasons for (supporting the environment, positive product perception), reasons against (negative product perception, price), and attitude towards consumption of local food distributed by REKO, we tested five hypotheses, focused on the direct effect of self-identity on reasons for, reasons against, and attitude towards local food distributed by REKO (H1 to H5). In line with our expectations, the findings showed that self-identity is positively associated with the reasons for and negatively associated with reason against choosing local food distributed by REKO. Specifically, self-identity was found to be positively associated with supporting the environment (H1) and positive product perception (H2). These findings are in concordance with previous studies (e.g., Aitken et al., 2020;Gkargkavouzi et al., 2019;Memery et al., 2015), implying that the consumers whose food consumption practices are aligned with their lifestyle, values, image, and characteristics will have a positive disposition in the form of reasons for buying local food from REKO. These reasons for include the perception that buying local food from REKO is the right thing to do for the environment since it helps reduce emissions and is ecologically-friendly. Furthermore, we found that consumers feel that the local food distributed by REKO is healthy, safe, and pure and hence should be bought. In essence, consumers perceive that local food distributed by REKO is aligned with their self-identity beliefs, which make them aware of their responsibility to support the environment and  Note: Composite reliability = CR, Average variance extracted = AVE, Maximum shared variance = MSV, Average shared variance = ASV, Supporting the environment = STE, Positive product perception = PPP, Price = PR, Negative product perception = NPP, Attitude = ATT, Brand love = BL, Self-identity = SI. motivate them to consume food that has a certain positive product perception. Similarly, as proposed, self-identity was found to be negatively associated with negative product perception (H3) and price (H4), in consonance with previous studies in other contexts (e.g., Confente et al., 2020;Printezis & Grebitus, 2018) and not in consonance with Profeta and Hamm (2019). This indicates that the consumers whose food consumption practices are aligned with their lifestyle, values, image, and characteristics will have a negative disposition in the form of reasons against buying local food from REKO. These reasons against include the perception of consumers that the product attributes, such as the shape, colour, and taste of the food sold at REKO, are deterrents that inhibit them from buying it. In addition, we found that consumers also perceive that the food sold by REKO costs more than food sold at normal grocery stores and thus does not offer sufficient value for money.
As a part of our response to RQ1, we also proposed and examined the association of self-identity with the attitude towards local food (H5). The results supported the hypothesis, which was developed based on prior studies in different contexts (e.g., Johe & Bhullar, 2016;Khare & Pandey, 2017). This result confirms that the self-identity beliefs of consumers lead them to feel that buying local food sold by REKO is good and pleasant. In sum, all hypotheses formulated to address RQ1. were supported by the results of statistical analysis.
To respond to RQ2, which inquired about the association of reasons for and reason against with attitude towards local food and brand love for local food sold by REKO and attitude with brand love, we proposed H6 to H14. Our findings indicated that consumers feel that buying food  Note: Supporting the environment = STE, Positive product perception = PPP, Price = PR, Negative product perception = NPP, Attitude = ATT, Brand love = BL, Self-identity = SI. Note: Supporting the environment = STE, Positive product perception = PPP, Price = PR, Negative product perception = NPP, Attitude = ATT, Brand love = BL, Self-identity = SI.
distributed by REKO is ecologically-friendly, which causes them to have the attitude that buying local food sold by REKO is good and pleasant, thereby supporting H6. Similarly, H7 was supported, indicating that the perceived pro-environmental aspect of buying local food from REKO causes the consumers to develop passionate brand love and attachment towards food distributed by REKO. These findings are in line with previous studies in other contexts (e.g., Birch et al., 2018;Memery et al., 2015). In a similar vein, the results revealed that positive product perception is positively associated with attitude towards local food (H8) and brand love (H9), affirming the findings of studies in different areas (e.g., Birch et al., 2018;Zhang et al., 2019). The results imply that the perception that the local food distributed by REKO is healthy, safe, and pure causes consumers to find buying food from REKO to be pleasant, which leads them to develop a passionate attachment to it. Next, the findings indicated that negative product perceptions related to the shape, colour, and taste of the food sold at REKO were negatively associated with attitude towards local food, thus supporting (H10), in consonance with previous studies (e.g., Cahyani & Iriani, 2019;Wang & Somogyi, 2018). At the same time, it is important to state that we had proposed taste to be a part of negative product perception for REKO-distributed local food on the basis of our qualitative study. The empirical results also support the conceptualisation. However, this finding is in contradiction with many prior studies that have argued taste to be a positive driving factor in the case of local food choices (e.g., Ditlevsen et al., 2020;Costanigro, Kroll, Thilmany, & Bunning, 2014;Denver & Jensen, 2014;Choe & Kim, 2018). A potential reason behind our qualitative and quantitative finding, in opposition to the prior literature, could be that the negative perception about taste is based on the fact that these foods are not transformed or enhanced in any way to make them sweeter or have a more pronounced flavour. Rather they have subtle or under-stated taste, which is compared by consumers with non-local foods sold through commercial supply chains, where additives may be used when possible to boost the flavour or give them a stronger taste. Thus, the negative view about the taste of local food distributed by REKO may actually be an expression of how the consumers perceive its taste as compared to non-local food, which may be transformed from a commercial perspective. Nonetheless, very few studies have been conducted to examine local foods in the context of their distribution system (e.g. Kumar et al., 2021b), and it is suggested that before any conclusion is drawn, more studies, by collecting data from a larger sample and different regions should be undertaken. Another possible reason could be that the appearance or colour of the local food creates a dissonance among consumers, causing them to develop a negative disposition towards it. In contrast to the support for H10, H11, proposing a negative association of negative product perception with brand love, is not supported. Rather, we found a surprising statistically significant positive association. Such an association, however, is not theoretically plausible. Therefore, we contend that more studies need to be undertaken to clarify the nature of the association between the two variables.
Price, the second reason against, was negatively associated with brand love, as proposed, but not with attitude towards local food, thereby indicating support for H13, but not for H12. The support for H13 is in alignment with previous findings in varying contexts (e.g., Aschemann-Witzel & Zielke, 2017). This outcome implies that consumers perceive that the local food sold by REKO is more expensive than the food sold by other outlets, a perception that creates a negative feeling towards it. In turn, this negative valuation adversely impacts the consumers' attitude towards this food and reduces their love and attachment to it. In the case of the association of price with attitude (H12), we found a statistically significant positive association that is confounding and logically less probable. Although consumers may perceive a higher price as signalling a certain quality/value, which may thereby increase their positive attitude, further research is required before drawing any conclusions.
With regard to the association of attitude with brand love, the results of the statistical analysis confirmed a positive relationship between the two. Thus, H14 is supported, in agreement with previous studies (e.g., Barska & Wojciechowska-Solis, 2020;Hegner et al., 2017). This outcome implies that the general feeling of pleasantness towards food sold by REKO increases love and attachment towards this food.
Finally, we addressed RQ3, which proposed the possibility of indirect effects along with the direct paths discussed above, by examining the mediation effect of reasons for, reasons against, and attitude. The results indicate support for all hypotheses (H15a-d, H16a-d). Our confirmation of the mediation effect of supporting the environment and positive product perception on the association between self-identity and attitude towards local food distributed by REKO indicates that the mechanism of forming a positive attitude towards such food is more complex than a simple alignment with self-identity beliefs. Rather, it has a much broader perspective. Similarly, the support for the mediation effect of negative product perception and price on the association between self-identity and attitude towards local food distributed by REKO indicates that no matter how aligned consumption of local food distributed by REKO is with consumers' self-identity beliefs, perceptual and practical considerations related to the food's appearance and value for money erodes the consumers' positive attitude towards it. Furthermore, the support for the mediation effect of attitude on the association of the two reasons for (supporting the environment and positive product perception) and the two reasons against (negative product perception and price) with brand love for local food distributed by REKO confirms the complex dynamics of consumer behaviour.
The results of our analysis revealed that age, gender, economic background, and household size have no confounding effect on attitude. Similarly, age, gender, and household size have no confounding effect on brand love. However, the economic background has a confounding effect on brand love (− 0.05*).
These findings suggest that irrespective of their age, gender, or the number of members in their household, consumers have a favourable disposition and emotional attachment with local food distributed by REKO. However, their economic background does have an effect on the extent of their brand love for local food distributed by REKO.

Theoretical implications
The three key theoretical contributions are: First, it utilised BRT as a theoretical foundation to argue the association between self-identity, reasons, attitude towards local food, and brand love to conceptualise the complex dynamics of consumer behaviour towards local food distributed by REKO. BRT theorises what guides consumers' actions and rationalisations (Westaby, 2005). By extending it to capture the behavioural response of local food consumers, we enrich the literature in two ways: (a) by clarifying the mechanism of interaction among the considerations impinging upon local food consumers' decision-making process in the case of a well-organised distribution system (REKO), and (b) by supporting the theoretical validity and versatility of BRT in explicating multi-faceted consumer behaviour in non-technology contexts as well. Thus, the study offers enhanced understanding of BRT in varied contexts (e.g., Ryan & Casidy, 2018;Sahu et al., 2020;Tandon et al., 2021).
Second, although scholars have suggested that self-identity beliefs may lead to a behavioural response in consumers (Barbarossa et al., 2017;Confente et al., 2020;Khare & Pandey, 2017), the extant literature has offered a limited understanding of how self-identity shapes attitude towards local food, which, in turn, gets reflected in the behavioural response of brand love. Moreover, brand love as a behavioural outcome related to local food and its associated distribution systems has remained under-explored in prior studies in the area. Our research thus offers a clearer conceptualisation of these associations, thereby augmenting the theoretical understanding of these important expressions of consumers' psychosocial profile and their responses to products and services. Furthermore, by investigating both the direct and the indirect interplay of self-identity, attitude towards local food, and brand love, our study provides a broader coverage of consumers' decision-making process.
Lastly, the study offers insights from Finland regarding the beliefs that drive Finnish consumers to purchase local food distributed by REKO. No prior studies have examined the behavioural aspect of Finnish consumers related to local food, especially against the backdrop of a specific distribution channel. As such, we offer novel findings to enrich the local food choice behaviour research (e.g., Korhonen, Kotavaara, Muilu, & Rusanen, 2017). Specifically, the study enhances the literature on local food distributed through social media channels (e.g., O'Hara & Lin, 2020).

Managerial implications
The three key managerial implications are: First, by revealing the significant association of supporting the environment and positive product perception with attitude towards local food and brand love towards local food and REKO, the study reveals the key aspects that can be leveraged by to improve marketing plans and draw new members. For example, these networks can design their promotional communication to underscore the environmental and health benefits of the food they distribute. Similarly, since price emerged as a significant reason against that erodes brand love in this context, the producers and managers can work around the price or offer discounts on certain days to improve consumer perception about buying food from REKO thereby increasing their customers' attachment to it. Also, managers can try to utilise appropriate communication tools to reinforce the fact that food products distributed by REKO offer benefits and positive attributes superior to non-local food products, thus offering a greater value for money. This is likely to offset the perception that consumers have about the REKO distributed food being high on price. Furthermore, since our findings reveal that the consumers have a negative perception of REKOdistributed local food in terms of shape, colour, and taste, the marketers should try to build their communication narrative around better explicating the reasons behind these three aspects. In other words, they should emphasise that the REKO-distributed local food is 'as nature intended' and it has not been tempered with during the growing or packaging process to modify the shape, enhance colour, or make the taste more prominent. In addition, the marketers can get testimonials from dieticians that not interfering with the naturalness of the food makes it a healthier and safer consumption alternative. These efforts are likely to improve the perception of consumers, who would otherwise expect price discounts by reaching erroneous conclusions about product quality driven by the appearance (in terms of shape and colour) of food products, as argued by prior studies (Hartmann et al., 2021).
Second, there are ongoing alternative food movements across the globe (Kondoh, 2014), such as teikei in Japan and Community Supported Agriculture (CSA) in the United States (Ostrom 2007), which are based on a certain partnership between consumers and producers. Such groups can benefit from the findings of our study regarding how to promote consumers' positive attitude towards local food and brand love towards food distributed by them.
Lastly, our findings are useful for policymakers and governments who wish to promote the consumption of local food given their perceived environmental and health benefits. By revealing that price erodes the brand love of consumers towards food distributed by REKO, our study attracts the attention of policymakers towards re-evaluating their existing policies or creating new ones that can reduce any levies or taxes imposed on local food distribution. This can help these distribution systems reduce the prices they charge consumers. Furthermore, since consumers value positive product perception and proenvironmental aspects of local food distributed by REKO, governments can institute a certification program wherein a designated logo/QR code is allowed to be used by local food distributors who meet the specified criteria. Such a move can help strengthen brand love for local food distributed by systems/channels that are able to meet these criteria and thereby can use the logo/code.

Limitations
Despite offering some interesting findings, the study has three main limitations. First, the data was collected from members of a particular social media channel for local food distribution in Finland, which limits the generalisation of these findings to other geographies and distribution channels, such as Helsinki food hub or Lähiapaja. Future studies can address this limitation by conducting replication studies on REKO group members from other countries or using our conceptual model to examine consumer behaviour towards other food distribution channels for buying and selling local food. Second, the study collected data during the ongoing COVID-19 pandemic, which has significantly disrupted supply chains, including social media marketing channels. Therefore, the findings may not be applicable in the case of a normal situation with no external stressors. Researchers can test our conceptual model in the post-COVID period and compare the findings with the ones from this study. In addition, future studies may incorporate some other contextspecific factors while investigating brand love for local food. Finally, the present study collected data through a cross-sectional survey at only one point in time. Future studies can thus use a longitudinal research design that can offer more insights about changing consumer preferences and behaviour over time.