An Empirical Test of Brand Love and Brand Loyalty for Restaurants during the COVID-19 Era: A Moderated Moderation Approach

: Although brand love–loyalty relationships can deepen, the literature does not include systematic and empirical investigations demonstrating when perceived value and relationship duration are valuable in enhancing the brand love–loyalty relationship. This study investigates the effects of relationship duration, perceived value, and restaurant type on the relationships between brand love and brand loyalty during the coronavirus disease of 2019 (COVID-19) pandemic. In particular, hedonic value rather than utilitarian value is hypothesized to negatively enhance the relationship between brand love and brand loyalty when consumers continue to have long-lasting relationships with a restaurant brand. Using data collected from an online research ﬁrm in Korea, the ﬁndings revealed that brand love negatively inﬂuences brand loyalty. However, the impact of brand love on brand loyalty increases when customers seek hedonic value. Our ﬁndings also demonstrate that consumers who sought hedonic value strengthened the brand love–loyalty link compared to consumers who sought utilitarian value, particularly one with a short-lasting relationship. Consumers who sought utilitarian value through a long-lasting relationship strengthened the same relationship, although the increased correlation was not statistically signiﬁcant. Furthermore, brand loyalty gradually decreases at ﬁne-dining restaurants, whereas it sharply increases at takeaway restaurants.


Introduction
"One may naturally argue that, when someone loves someone else or something, he will be loyal to what he loves. Or the longer she has fun with him, the more loyal she will be to what she likes. When a person views a restaurant according to its function, he does not need or should go to the restaurant he likes. They can choose any restaurant whenever they want. In fact, he may even visit one restaurant as long as he gets what he needs. So, do researchers need to test the proposition?" [1] The answer is "yes," because, as this quote highlights, there are two important issues that have not yet been addressed in previous studies: perceived value and relationship duration before the improvement of a brand love-brand loyalty link. People are particularly likely to visit a restaurant if they perceive a good feeling at the restaurant. In exploring how restaurants can manage existing customers, researchers and managers have emphasized the importance of customer relationships [2][3][4][5]. Customers often depend on restaurant visit length (short-term vs. long-term) and perceived value (utilitarian vs. hedonic) when they select a restaurant. Regarding relationship duration, for example, the concept of relationship marketing that focuses on identifying and retaining long-term customers can be directly linked to restaurant selection. In addition, as not all consumption experiences in the restaurant sector evoke the same emotional states, customer behavior for selecting a restaurant can rely on both utilitarian and hedonic values [6].

•
Does the brand love-brand loyalty link matter amidst the COVID-19 pandemic? If so, what happened? • Human love weakens over time [14]. Does this imply that the longer the relationship with a particular restaurant brand, the weaker the relationship becomes, which ultimately weakens brand loyalty? • Does the hedonic value strengthen the relationship between brand love and brand loyalty? • Consumer behaviors have changed after the impact of COVID-19. What is the difference in the brand love-brand loyalty link between fine-dining and takeaway restaurants?
To answer these questions, we propose a conceptual model to address the brand love-loyalty link that comprises direct, indirect, and moderated moderation effects. The proposed model provides ample evidence of the brand love-loyalty relationship using existing theories such as brand equity, customer value, and customer relationship marketing.
To add to the existing literature on brand equity in service marketing, this study contributes to extending the existing literature not only by providing evidence to determine types of consumer value, but also by identifying the different influences of relationship duration. This leads us to broaden the knowledge of restaurant (or brand) experience in this research stream by introducing the two indicators of the brand love-loyalty mechanism.
The remainder of the study is organized as follows. We develop research hypotheses regarding two factors that could influence the brand love-loyalty link. Next, we describe the data collection procedure and test our hypotheses. We then discuss the findings of this study and their implications and outline further directions.

Literature Review
Although "love" has been a popular topic in the psychology literature, brand love has also been an interesting research subject for consumer researchers. Having been adapted from Sternberg's [15] triangular theory of interpersonal love in the domain of psychology, brand love, which concretizes the concept of love that occurs in human relationships, is a concept that is imbued deeply in brands and includes an emphasis on consumer self-congruence [5,16]. This concept focuses on the interdependent relationship between consumers and brands. In particular, the stronger the interdependence, the stronger the attachment relationship [17]. In particular, when creating restaurant brand experiences, managers usually try to establish consumers' emotional linkage using three love components (i.e., intimacy, passion, and commitment) [18]. Table 1 shows an overview of the selected literature to compare our study with existing work.
The first step in achieving a complete understanding of brand love is to understand how researchers define the construct when consumers refer to their love for a particular restaurant brand. Batra et al. [7] demonstrated that general conceptualizations of brand love are difficult to apply because of several compelling reasons. Alternatively, they have emphasized a deep understanding of how consumers experience a brand and build valid connections. In particular, interpersonal interaction with customer experiences is key to improving brand love in the hospitality industry [5]. This approach is consistent with other studies addressing emotional attachment [8,19], self-brand connections [20], and passion after repeat experiences [16]. In line with these observations, we defined brand love as the degree of consumers' positive emotional connections after they experience a brand. This definition highlights the connection and emotional attachment between a brand and consumers. Ultimately, it elicits the memory associated with the experience and increases repeat purchases (or revisits) of the brand (or restaurant) [21].
However, brand love is a somewhat different concept from satisfaction. The principal difference between brand love and satisfaction is that satisfaction is based on the expectancy-disconfirmation paradigm through transactions with a particular restaurant, whereas brand love can traverse the use of a restaurant. For example, although most Korean consumers have no experience of visiting Michelin star restaurants, they have a love for a Michelin star restaurant (e.g., Gordon Ramsay restaurants, Pierre Gagnaire, Guy Savoy, etc.). Although clearly distinguishing brand love from satisfaction is somewhat difficult, it might lead to theoretical clarity if systematic scale measures are developed. However, scholars commonly accept passionate feelings as one of the principal characteristics of brand love [2,8,22]. We demonstrate that brand love is formed through a passionate feeling about a self-brand connection. That is, the dynamic nature of love within a consumerbrand relationship is linked to the requirement to maintain this passionate feeling toward a brand [11]. This understanding is in line with the definition of brand love.
In particular, researchers have recommended that establishing an emotional bond (or focus on the concept of brand love) beyond satisfaction is critical to improving brand loyalty [8,23,24]. Robert [7] demonstrated that if customers were emotionally attached to a particular brand, they would believe that the brand was superior to other brands. Conceptually, when a customer is satisfied with a particular restaurant, they form a special attachment to the restaurant [25]. Carroll and Ahuvia [8] highlighted that satisfaction was the basis for creating brand love. These observations are consistent with the palette theory, which establishes the evolution of brand colors [26]. According to this theory, the starting point is the satisfaction experience. Subsequently, a change in brand color can create brand Sustainability 2021, 13, 9968 4 of 17 love beyond satisfaction [27]. The cycle of satisfaction [28] and the phase-wise sequential order of brand loyalty [29] also support this process.

Topic Key Findings Selected Papers Relevant for This Study
The concept of brand love Brand love, which concretizes the concept of love that occurs in consumer object relations, is a key concept that is deeply imbued in brands.
This focus is on the interdependency between consumers and the brand. [15][16][17] This study adds to the work on brand love-loyalty relationships to emphasize the importance of the shopping value and relationship duration in influencing the level of brand loyalty evaluations.
Th role of brand love The congruence literature in the marketing and psychology fields shows that consumers build emotional connections with a brand that is congruent with their attachment. [8,9,16,20] Applying the basic congruence, this study examines how the relationship between shopping value and relationship duration influences the brand love-brand loyalty link in the restaurant sector.
Interaction among satisfaction, brand love, and loyalty Consumers in particular establish an emotional bond with a brand beyond satisfaction to improve brand loyalty. Brand love can be formed based on satisfaction or emotional connection, subsequently evolving into loyalty. [21,[23][24][25][26][27] Our research excludes satisfaction; however, we measure brand love in terms of emotional attachment beyond satisfaction, thus expanding the bases of brand love through shopping motivation (or value) and relationship duration.

The Brand Love-Loyalty Linkage
Most hospitality studies have included a focus on brand love as a key driver of brand loyalty [3][4][5]. In this study, we defined brand loyalty as a deep commitment to rebuy (or revisit) a preferred brand (or restaurant) consistently in the future [29]. A loyal consumer is a committed consumer, suggesting that a consumer with a positive emotional connection to a brand will continue to display a strong relationship with the brand (being committed) [30]. Similarly, Dick and Basu [31] demonstrated that brand loyalty should be greater under conditions of positive emotional mood or affect. This approach is consistent with the causal relationship between brand love and brand loyalty. The consumer-brand relationship model of Fournier [2], the causal approach of Carroll and Ahuvia [8], and the system approach of Batra et al. [9] all suggest this causal relationship.
We argue that brand love and brand loyalty are closely related concepts at both attitudinal and behavioral levels because brand love is mostly expressed in a self-oriented manner and reflects a passionate desire [8,9], which enhances the emotional connection with a particular brand [22,32]. In this perspective, brand love creates emotional bonds with favorite brands (e.g., the congruence of self-expression or the congruence between the consumer's identity and the brand's identity), whereas brand loyalty plays a critical role in brand choice [33,34]. Therefore, we focused on the reasons other studies expend resources on attitudinal and behavioral standpoints.
Promoting restaurant brand loyalty is a critical objective for restaurant owners not only with respect to retaining existing customers, but also to prevent them from switching to competitors during the COVID-19 pandemic. More specifically, converting existing customers into loyal customers and generating excellent performance is more desirable because acquiring new customers is difficult given the coronavirus situation. For example, consider a restaurant choice situation that patronizes only one brand. One possible explanation could involve the absence of experience of other restaurant brands and, thus, indifference to other restaurant brands. Another explanation is that consumers heard of other restaurant brands (or search for restaurants using mobile devices), visited restaurants, or ordered from restaurant menus through mobile apps, and found that restaurants could differ in quality, package, staff service, and so forth, identified a particular restaurant brand they could connect to emotionally, and now prefer one particular brand to other brands with which they have a weaker emotional connection. This scenario indicates that the customer is likely to form a strong emotional connection with the restaurant brand. If the customer feels happier, this emotional connection can lead to a willingness to revisit (or repurchase) the restaurant brand in the future.
Among numerous determinants of brand loyalty, brand love can strengthen emotional ties between consumers and brands [2,34]. In particular, when consumers feel psychologically dependent on a particular brand, they tend to be loyal through emotional connections, indicating that consumer-brand attachment increases [35]. Logically, the anthropomorphism of brand love (e.g., some consumers have the tendency to attribute human qualities to brands) highlights that the consumer-brand attachment is important. This demonstrates that if consumers are devoted to the objects they own, they are more likely to use them [36]. This is the reason why brand love leads to brand loyalty. The higher-order brand-love prototype model also establishes that brand loyalty is a function of brand love [9]. Previous studies have also revealed a causal relationship between brand love and brand loyalty [8,30,34,37].
Meanwhile, the COVID-19 pandemic negatively affects restaurant performance. For example, because there are many restrictions on restaurant reservation and operation, it is inconvenient for consumers to attend restaurants. The more restrictions on the consumers' use, the more they may experience negative feelings about a particular restaurant. That is, negative emotions are negatively related to positive brand loyalty [38,39]. Thus, we hypothesize the following: Hypothesis 1 (H1). Brand love has a negative effect on brand loyalty during the COVID-19 pandemic.

Hedonic Value and Utilitarian Value
The principal reason for using a particular restaurant brand is that consumers expect to gain benefits from their consumption experience [40]. For example, if consumers feel love for a brand through a consumption experience, they feel emotionally connected to it and perceive additional benefits after subsequent purchasing (or visiting). In this case, perceived values (hedonic and utilitarian) are a pivotal construct in the consumer behavior research. Perceived value is often identified by the hedonic and utilitarian values that consumption experiences or consumption motivation provide [41]. Hedonic value involves an emphasis on the consumers' intrinsic need for a brand, whereas the focus of utilitarian value is on external outcomes (or intended consequences) such as the performance of a brand itself [42]. Specifically, hedonic value reflects the consumers' potential pleasure/amusement and emotional worth [43], whereas utilitarian value means that a brand is selected efficiently from a functional or task-related standpoint [44].
Hedonic and utilitarian values are important in the hospitality industry [6,45,46]. However, restaurant consumption (or brand experience) also varies across consumers who may differ in the two types of value [47]: hedonic consumers are more emotionally connected to the restaurant (or brand) through, for instance, pleasure, fun, and excitement, than consumers who seek a utilitarian value [39]. That is, hedonic value perceived through the restaurant brand experience helps consumers establish a strong emotional attachment, resulting in brand love [2,48]. In particular, Kuikka and Laukkanen [49] found that the relationship between emotional construct and brand loyalty was strengthened when hedonic value was high. Because hedonic value is more closely linked to emotions, the brand love-loyalty link is strengthened by the moderating role of hedonic value rather than by the moderating role of utilitarian value. As such, hedonic value can strengthen the intrinsic attachment relationship between consumers and brands through the consumption experience, thereby increasing brand loyalty. These arguments lead us to hypothesize the following: Hypothesis 2 (H2). Hedonic value amplifies the positive effect of brand love on brand loyalty more than utilitarian value.

The Moderated Moderation of Relationship Duration
The customer relationship stage includes a focus on the development process of the mutual relationship between two parties through an emphasis on the pivotal role of time across which the relationship evolves [50]. As such, the maturing of mutual relationships over time can be applied to consumer-brand relationships in the restaurant sector. This indicates that the outcome of the relationship duration affects loyalty [51].
The brand love-loyalty link tends to become stronger over time if this relationship is not terminated. In particular, the longer the relationship duration, the more beneficial the relationship between the two constructs. Because brand love can depend on relationship duration [15], the longer the relationship, the stronger the brand love, which can enhance brand loyalty.
Relationship duration is also related to consumer value [9,52]. Improving brand love and loyalty through intrinsic needs, rather than through utilitarian value that can be externally evaluated, would require a long-lasting relationship. We expect that the relationship between brand love and brand loyalty changes over time and varies between hedonic and utilitarian consumers. In addition, because hedonic value tends to be shortterm and spontaneous [53], the COVID-19 pandemic can limit the consumers' hedonic preference, resulting in enhancing negative emotions. Thus, we hypothesize the following: Hypothesis 3 (H3). Hedonic value, rather than utilitarian value, negatively enhances the relationship between brand love and brand loyalty when consumers continue to have long-lasting relationships with a restaurant brand.

Comparison of Fine-Dining Restaurant and Takeaway Restaurant
Since the COVID-19 pandemic has disrupted consumer behaviors [13], restaurant visit frequency and loyalty may also be changed. This is because the lockdown and social distancing often prevent consumers from accessing restaurants. Traditionally, consumers tend to love fine-dining restaurants more than fast-food or takeaway ones [54]. However, it is inevitable that they now prefer to visit takeaway restaurants or use home delivery services. If their consumption experiences accumulate, their brand love improves, which, in turn, affects brand loyalty. We expect that the brand love-brand loyalty link is higher in a fine-dining restaurant than in a takeaway restaurant (see Figure 1).

Data Collection
We collected data from a professional online research company (PMI, one of the major online research companies) in Korea. We selected fine-dining restaurants (i.e., Western,

Hypothesis 4 (H4).
In fine-dining restaurants, brand love strengthens brand loyalty compared to in takeaway restaurants.

Data Collection
We collected data from a professional online research company (PMI, one of the major online research companies) in Korea. We selected fine-dining restaurants (i.e., Western, Japanese, Chinese, and Korean styles) located in Seoul, the capital city of South Korea, because they provided adequate employee training, interior design, and food and beverage quality. In addition, more fine-dining restaurants are located in Seoul than in other areas, indicating the relative advantage of respondent selection.
Sample selection criteria were limited to customers who had used a particular restaurant twice or more during the COVID-19 pandemic. Although we mainly focused on fine-dining restaurants, we also checked the respondents' takeaway restaurant consumption from the same respondents. They frequently used these takeaway restaurants due to the lockdown and social distancing. Thus, we excluded subjects who were not qualified.
The online survey was conducted late May to early June 2020. Respondents participated in the survey via email. The survey was similar to a Google Survey format. For example, only one question is presented on the screen at a time, and respondents cannot move on the next question until they answer the question on the screen. We selected 203 respondents who matched our sample criteria, after excluding those who had no experience of fine-dining restaurants (n = 58) during the COVID-19 pandemic or gave unreliable responses (n = 19). The period of data collection was very stringent, and many restaurants were limited in business. Therefore, at least in Korea, restaurants tended to modify their strategies in the direction of strengthening existing customers.
The effective response rate was 72.5%. We also checked for the non-response bias from a random sample of 40 participants with nonparticipating respondents whose sample selection criteria were similar. No significant differences existed between the two samples (the original research sample vs. the random sample).
The respondents' ages ranged from 20 to 53 years, with a mean age of 34.6 years (SD = 3.5). Of the respondents, 53.7 percent (n = 109) were female (46.3% were male). The majority (72%) of the respondents had a college degree, with an average restaurant-use period of 1.4 years. More than half of the respondents (68.5%) earned a monthly income of $4000-$8000, followed by less than $4000 (19.3%) and $8000 and above (12.2%). The mean consumer expenditure at fine-dining restaurants was approximately USD 72.

Measurements
In the surveys, we adapted all the scale items from previous studies [6,8,[55][56][57]. Because we focused on Korean consumers, two bilingual researchers followed the translationback translation procedure [55] to create a Korean version. We measured brand love and brand loyalty by using 10 items and four items, respectively, adapted from Carroll and Ahuvia [8]. Perceived value (hedonic and/or utilitarian) was measured by using six items adapted from Carroll and Ahuvia [8], Jones et al. [56], and Ryu et al. [6]. Several items were modified from original items because of the nature of this study.
As Table 2 shows, we employed multi-item scales using a five-point Likert scale (1= "strongly disagree" and 5 = "strongly agree," or 1 = "never" and 5 = "as often as possible"). However, we measured the relationship duration using a single item by using both short-term (during and post-COVID-19) and long-term customers (pre-COVID-19). The relationship duration was divided into short-term (n = 109) and long-term duration (n = 94) based on the COVID-19 pandemic. The criterion of the short-term duration was for respondents to have initially visited a restaurant after COVID-19 (e.g., after the early of January 2020). In contrast, the long-term duration criterion was for respondents to have frequently visited a restaurant before the COVID-19 pandemic. We identified two types of restaurant: fine-dining restaurants (0) and takeaway restaurants (1). In so doing, we asked respondents to check which type of restaurant they preferred to visit.
Finally, Petrick [57] highlighted that emotional construct measures may be easier to interpret and more informative in the proposed model if the value measurement is corroborated with consumption experience. In line with this observation, perceived value constitutes information in the brand love-loyalty linkage model.

Measurement Reliability and Validity
We assessed the reliability and validity of scales using SPSS 23 and AMOS 23. We first conducted a confirmatory factor analysis (CFA) using three latent constructs, resulting in an acceptable fit: χ 2 = 473.836 (df = 167), CFI = 0.951, TLI = 0.913, and RMSEA = 0.067. Specifically, the justification of values for CFI and TLI should be above 0.9 [58]. The RMSEA should be less than 0.07 [59]. As Table 2 shows, all Cronbach's alpha values ranged from 0.77 to 0.87. Furthermore, composite reliability (CR) exceeded the recommended level of 0.7 (brand love = 0.89, brand loyalty = 0.92, and perceived value = 0.95), indicating that the internal consistency of each construct was significant [59].
Most factor loadings exceeded 0.70, although two items had loadings of 0.67 and 0.68, respectively. In structural equation modeling, factor loadings between 0.60 and 0.70 are not valid cut-off values [60], indicating that our loadings were acceptable. Average variance extracted (AVE) values were above the thresholds (0.50), supporting adequate convergent validity.
Finally, we assessed the discriminant validity suggested by Fornell and Larcher [61]. We checked whether the AVE exceeded the squared correlation. We further tested Chisquare differences between constrained and unconstrained models for all possible pairs of constructs [62]. As Table 3 shows, all of the results were significant. Thus, discriminant validity was supported.

Moderated Moderation Results
The moderated moderation regression produced coefficients including a two-way and a three-way moderator. The equation of the moderated moderation effect is as follows: = i1 + (b1 + b 4 M + b5W + b7WM)X + b2M + b3W + b6MW + eY, where M is the perceived value (two-way moderator) and W is the relationship duration (three-way moderator). This equation reveals that X (brand love) has two moderators on Y (brand loyalty). One is b1 + b5W. The other one is b4 + b7W, which depends on M. Thus, it is acceptable that the effect of X on Y depends on W moderated by M.
To test the research hypotheses, we used Process Macro, as suggested by Hayes [63]. As Table 4 shows, the effect of brand love on brand loyalty (β = −2.185, p < 0.05) was significantly negative. This result supports H1. More specifically, customer restrictions on a particular restaurant during the COVID-19 pandemic stimulated negative feelings about the restaurant. Thus, brand love had a negative effect on brand loyalty during that period. The direct effect of perceived value on brand loyalty was also significant (β = −6.173, p < 0.01). The relationship between perceived value and brand love on brand loyalty (H2) was positively supported (β = 1.848, p < 0.01). These results demonstrated that differences between hedonic and utilitarian consumers appeared because of the moderating effect of the perceived value rather than the direct effect of perceived value on brand loyalty. As presented in Figure 2, the brand love-loyalty link was higher for hedonic value consumers than for utilitarian consumers.  Although we did not establish a particular hypothesis, the direct effect of relationship duration on brand loyalty was significant (β = −5.811, p < 0.01). In the traditional wisdom of a customer-brand relationship, long-term customers tend to be more loyal to the brand. Interestingly, this result indicates that short-term customers who initially visited restaurants during and post-COVID-19 were more loyal. In other words, they were willing to visit new restaurants despite the COVID-19 pandemic.
H3 posits that hedonic value rather than utilitarian value can enhance the relationship between brand love and brand loyalty when the relationship lasts longer. This was a negative and significant moderated moderation effect of relationship duration (β = −1.213, Although we did not establish a particular hypothesis, the direct effect of relationship duration on brand loyalty was significant (β = −5.811, p < 0.01). In the traditional wisdom of a customer-brand relationship, long-term customers tend to be more loyal to the brand. Interestingly, this result indicates that short-term customers who initially visited restaurants during and post-COVID-19 were more loyal. In other words, they were willing to visit new restaurants despite the COVID-19 pandemic.
H3 posits that hedonic value rather than utilitarian value can enhance the relationship between brand love and brand loyalty when the relationship lasts longer. This was a negative and significant moderated moderation effect of relationship duration (β = −1.213, p < 0.01). In particular, we focused on the negative effect (−) of coefficient and found an interesting slope of relationship duration, which is presented in Figure 3. The marketing literature has emphasized the importance of long-term relationship when a restaurant looks for building loyal customers, whereas the slope of this study was reversed. The latter may be acceptable because customers who visited a restaurant after COVID-19 (short-term duration in this study) love the restaurant with high levels of brand loyalty. Most customers were reluctant to visit the restaurant because of COVID-19, but these customers increased their frequencies of visit during that period. This argument is directly linked to the direct effect of relationship duration (particularly in the short-term duration) on brand loyalty. looks for building loyal customers, whereas the slope of this study was reversed. The latter may be acceptable because customers who visited a restaurant after COVID-19 (shortterm duration in this study) love the restaurant with high levels of brand loyalty. Most customers were reluctant to visit the restaurant because of COVID-19, but these customers increased their frequencies of visit during that period. This argument is directly linked to the direct effect of relationship duration (particularly in the short-term duration) on brand loyalty. Our final hypothesis (H4) was to investigate the moderating role of restaurant type on the brand love-brand loyalty link. Using Process macro (M = 1), brand love negatively affected brand loyalty (β = −0.682, p < 0.01). In particular, as is shown in Table 5, the moderating effect of restaurant type had a positive effect on the brand love-brand loyalty link. In particular, brand loyalty gradually decreased among fine-dining restaurants, whereas brand loyalty sharply increased at takeaway restaurants (see Figure 4). Thus, H4 was supported.  Our final hypothesis (H4) was to investigate the moderating role of restaurant type on the brand love-brand loyalty link. Using Process macro (M = 1), brand love negatively affected brand loyalty (β = −0.682, p < 0.01). In particular, as is shown in Table 5, the moderating effect of restaurant type had a positive effect on the brand love-brand loyalty link. In particular, brand loyalty gradually decreased among fine-dining restaurants, whereas brand loyalty sharply increased at takeaway restaurants (see Figure 4). Thus, H4 was supported. Meanwhile, it is important to note that the standardized coefficient cannot mathematically exceed the bound of (−1 and 1). As this study deals with one dependent variable, there is no multicollinearity issue. However, the standardized coefficient can exceed the bound of (−1 and 1) if the relationship between the independent variable and moderated variable is weakly correlated [64]. As shown in Table 3, this correlation is relatively low. Thus, our standardized coefficients are acceptable.  Meanwhile, it is important to note that the standardized coefficient cannot mathematically exceed the bound of (−1 and 1). As this study deals with one dependent variable, there is no multicollinearity issue. However, the standardized coefficient can exceed the bound of (−1 and 1) if the relationship between the independent variable and moderated variable is weakly correlated [64]. As shown in Table 3, this correlation is relatively low. Thus, our standardized coefficients are acceptable.

Discussion
Drawing upon the consumer-brand relationship and consumption value theories, we examined the effects of brand love directly and indirectly (through perceived value and relationship duration). More specifically, we explored how both perceived value and relationship duration moderated the brand love-loyalty link. Using a survey design method, our analysis provided four sets of results in the restaurant sector: • Brand love had a direct and negative effect on brand loyalty during the COVID-19 pandemic. In particular, the effect of the brand love-loyalty link depended on the roles of the moderators.

•
The brand love-loyalty link among hedonic consumers became stronger than among utilitarian consumers when the relationship duration was short.

•
The effect of utilitarian value on the brand love-loyalty link increased when the relationship duration was shorter.

•
While brand loyalty gradually decreased at fine-dining restaurants, it sharply increased at takeaway restaurants.

Theoretical Implications
Over the last two decades, the brand literature has principally focused on the direct relationship between brand love and brand loyalty [2,8,9,65]. Since the COVID-19 pandemic has had a critical impact on the restaurant industry, the traditional wisdom of brand love has dramatically changed. While the brand literature has highlighted the significance of brand love to improve brand loyalty, our findings show that the relationship

Discussion
Drawing upon the consumer-brand relationship and consumption value theories, we examined the effects of brand love directly and indirectly (through perceived value and relationship duration). More specifically, we explored how both perceived value and relationship duration moderated the brand love-loyalty link. Using a survey design method, our analysis provided four sets of results in the restaurant sector:

•
Brand love had a direct and negative effect on brand loyalty during the COVID-19 pandemic. In particular, the effect of the brand love-loyalty link depended on the roles of the moderators.

•
The brand love-loyalty link among hedonic consumers became stronger than among utilitarian consumers when the relationship duration was short.

•
The effect of utilitarian value on the brand love-loyalty link increased when the relationship duration was shorter. • While brand loyalty gradually decreased at fine-dining restaurants, it sharply increased at takeaway restaurants.

Theoretical Implications
Over the last two decades, the brand literature has principally focused on the direct relationship between brand love and brand loyalty [2,8,9,65]. Since the COVID-19 pandemic has had a critical impact on the restaurant industry, the traditional wisdom of brand love has dramatically changed. While the brand literature has highlighted the significance of brand love to improve brand loyalty, our findings show that the relationship between the two constructs is negative-at least in the Korean restaurant sector. While this is probably temporary, brand love is affected by negative circumstances. These circumstances trigger the "love-becomes-hate" effect in the marketing literature [66]. This indicates that brand love might hold a cynical interrelationship without positive feelings if people face an uncertain and negative circumstance.
Several studies have included examinations of moderating effects of a specific construct (e.g., experience, price, and perceived risk) on the brand love-loyalty link [67,68]. Drawing upon the consumer-brand relationship and consumption value theories, we provide an improved understanding of the moderating effects of perceived value and relationship duration on the brand love-loyalty link. Specifically, relationship marketing theory demonstrates that consumers who seek hedonic value tend to pursue their pleasure, regardless of the duration of their relationship with a particular brand [69]. In contrast, the moderating effect of hedonic value on brand love-loyalty is considerably higher than that of utilitarian value when the relationship duration is short. This finding means that understanding the direct effect of brand love on brand loyalty can enhance the understanding of the passionate feeling of the consumer-brand relationship, and that the effect of the brand love-loyalty link varies when perceived value and relationship duration are involved. As such, we found that hedonic consumers increased the effect of the proposed link in short-lasting relationships.
This study contributes to the discussion about whether the consumer-brand relationship depends on relationship duration and results in advantageous restaurant performance [70,71]. Our results partially support the view that the consumer-brand relationship depends on relationship duration. However, it is important to note that a long-term relationship with a particular restaurant does not always increase the brand love-loyalty link. That is, the type of relationship duration has different effects on the brand love-loyalty link depending on the perceived value. These findings contribute to determining how the consumer-brand relationship should be controlled and managed.

Managerial Implications
An insight of this study that managers could benefit from is that relationship duration can systematically influence the level of brand love or brand loyalty. Long-term relationships with customers do not always improve brand loyalty. This study includes the suggestion that not all customers respond positively when the relationship with a particular brand is long. Given the finding that hedonic-value consumers are likely to repurchase in short-lasting relationships, restaurants may not benefit from maintaining a long-term relationship with such customers. They may need to change the view of what brands hedonic consumers are expected to consume after their first-time restaurant visit.
Regarding utilitarian-value consumers, who shop to achieve efficient outcomes [56], short-term management of customers should be avoided. Depending on whether longterm management is involved, strategies for successful customer management may differ. Given the naturally long-term customer management, for example, restaurants should try to avoid promoting sensory experience by displaying fascinating photos. In contrast, restaurants should try to invite customers who seek the long-term, efficient consumption of a particular restaurant experience and, subsequently, offer them special benefits on customer services such as free cooking classes by the chef. Alternatively, restaurants could provide valuable information to improve the customers' understanding of brands such as introducing online food delivery platforms during the COVID-19 pandemic.
Since both lockdown and social distancing have affected restaurant selection, consumer behaviors have changed with respect to visiting restaurants. In particular, takeaway restaurants (or home delivery services) have increased in popularity, and this trend is consistent with our findings. A possible way to increase brand loyalty is by favorably enhancing consumer feelings. Assuming that customers will return to their normal behavior once the COVID-19 pandemic subsides, both takeaway and fine-dining restaurants could implement a showcase of fast, clean, and hygienic processes (e.g., not just food storage, but a delivery process with hygienic packaging). This demonstration should be mandatory among all the restaurant types and could offer a "WOW" factor, resulting in the translation of brand love into brand loyalty. Here, win-back offer worth (WOW) is the perceived overall value of the offer extended to defected customers in an effort to attract these customers back to their previous restaurant [72]. The concept of WOW enables managers to develop win-back offers that may address the customers' specific needs to facilitate their frequent restaurant visits in the future.

Limitations and Future Research
Although this study provides insights into why perceived value and relationship duration are critical in a consumer-brand relationship, it has limitations that should be addressed in future research. For example, this study included a focus on fine-dining restaurants in Korea. Because these restaurants are highly competitive, testing the brand love-loyalty link should include alternative restaurant brands (e.g., fast-food and finedining restaurants) to improve the generalization of our findings.
Finally, researchers might also consider the relative differences in the restaurant visiting itself and consider how these differences influence the consumers' subsequent consumption (or revisits). In addition, future research could focus on which relationship duration produces different responses between male and female customers. For example, male customers who seek stability and efficiency in a service context are more likely to stay with a particular brand than female customers are, although their brand loyalty is likely to decrease over time [73]. A comparative investigation of gender difference might provide significant insights into how to effectively manage brand love and its relevant influencers [74].

Conclusions
The effects of brand love on brand loyalty can vary substantially, whereas the literature does not include systematic and empirical investigations demonstrating when perceived value and relationship duration are valuable in enhancing the brand love-loyalty link. In particular, the brand love-loyalty link should be an anomaly, contrary to what should have happened during the COVID-19 pandemic. Drawing upon the consumer-brand relationship and consumption value theories, we examined the effects of brand love directly and indirectly (through perceived value and relationship duration). More specifically, we explored how both perceived value and relationship duration moderated the brand loveloyalty link.
Our findings reveal that brand love negatively influences brand loyalty. However, the impact of brand love on brand loyalty increases when customers seek hedonic value. Our findings also show that consumers who sought hedonic value strengthened the brand love-loyalty link compared to consumers who sought utilitarian value, particularly one with a short-lasting relationship. While most customers are reluctant to visit the restaurant because of COVID-19, customers who initially visited restaurants during and post-COVID-19 increased their frequencies of visit during that period. Finally, brand loyalty gradually decreased at fine-dining restaurants, whereas it sharply increased at takeaway restaurants. Thus, our study adds to this research stream by identifying the type of restaurants required for the importance of practical management.