It’s all about the brand: place brand credibility, place attachment, and consumer loyalty

Despite the growing attention brand credibility attracts in recent marketing literature, little is known about the credibility of place brands and their relationship with consumers’ loyalty. The present study builds on brand signalling and attachment theory to investigate whether credible place brands affect consumers’ attachment formation, their subsequent word of mouth (WOM) and intention to return. Survey data from an Alpine winter sport destination reveal that place brand credibility is a key antecedent of consumers’ attachment. Moreover, consistent with brand signalling theory, credible place brands exert a particularly strong direct effect on consumers’ WOM. A positive effect on their intention to return, however, will result only if they feel attached with the place brand beforehand. Implications for marketing scholars and brand managers are offered.


Introduction
Clear, credible, and consistent brand signals help consumers to make inferences about a brand and its promise (Erdem and Swait 1998;Rao et al. 1999). Fulfilling a brands' promise is critical, as it impacts consumers' behaviour, which in turn influences overall firm performance, competitiveness, and profitability (Karanges et al. 2018;Henkel et al. 2007). Since it is difficult for consumers to evaluate the true quality of a brand, credible brand signals serve as vital means to reduce uncertainty about the quality of tangible and intangible brand attributes (Bettman and Park 1980). Accompanied by increasing global competition, rising digitization aspects and international affairs that heighten uncertainty, the notion of brand credibility has become more important than ever, particularly in the field of place branding (Alexander et al. 2020;Zenker et al. 2017). This is not surprising, because an attractive place brand represents a destinations' major asset to overcome substitutability and lacking differentiation (Pike 2005). Similarly, today's consumers are more flexible in their decision-making and favour place brands that keep their promises and provide unique, memorable travel experiences (e.g. Volgger and Pechlaner 2014). To compete successfully, brand managers thus need to understand what contributes to consumers' place attachment and how to gain their loyalty. But how does place brand credibility affect consumers' attachment and later behavioural outcomes?
Although brand credibility has received considerable attention in previous research in marketing (e.g. Erdem and Swait 1998), retailing (Guido et al. 2011), services management (Bougoure et al. 2016), consumer behaviour (Dholakia 1987), human-computer interaction (Shan 2016), and philosophy (Umeogu 2012), the credibility of place brands remains largely unexplored. Among the few contributions, Veasna et al. (2013) analysed the impact of destination source credibility on tourists' satisfaction at heritage sites. Relatedly, Zenker et al. (2017) examined the influence of place brand complexity on both residents' and tourists' behaviour. Yet it is striking that emotional, cognitive, and behavioural consequences of credible place brands received little attention in previous research. Particularly in the tourism industry, the primary field of application for place branding, consumers frantically look for brand experiences that convey credibility and trust. In the current research, we thus investigate the underlying mechanism of how informational cues (i.e. credible brand signals) influence consumers' place attachment and their subsequent response behaviour (Bianchi and Pike 2011). We assume that the perception of informational cues first involves a stimulus, followed by emotional and cognitive processing (i.e. attachment formation), and a favourable response towards the branding efforts of a place (Zeithaml 1988;Bolton and Drew 1991).
The present study makes three core contributions. First, in line with brand signalling theory (Erdem and Swait 1998), we postulate that the trustworthiness of information about a place exerts a persuasive effect on consumers' opinion (Bougoure et al. 2016). From this perspective, brand credibility can be achieved by acquiring informational cues from place brands. Consumers should thus only develop a favourable opinion towards a place when these cues are viewed as credible and reliable. Second, consistent with attachment theory (Mikulincer and Shaver 2007), we argue that place attachment represents an important emotional, cognitive, and functional mediating mechanism among informational cues (e.g. credible brand signals) and consumers' later responses. That is, when consumers perceive a place brands' marketing efforts to be trustworthy (Grisaffe and Nguyen 2011;Veloutsou and Moutinho 2009), the place brand becomes imbued with so much meaning that consumers use it to create a desired selfconcept and are likely to enter a relationship with the brand (Escalas 2004). Particularly when consumers have formed strong attachment with the place beforehand, place brand credibility should lead to increased behavioural outcomes. Third, we argue that consumers' sense of connectedness with a place can have a direct effect on their loyalty. Specifically, a high degree of place attachment should result in an increased desire to return to the setting and to spread positive word of mouth (WOM).
The purpose of this paper is to examine whether credible place brands facilitate brand-self connection and lead to a favourable approach behaviour. We first conceptualize the mechanism on how credible marketing efforts impact consumers' attachment formation and their loyalty by drawing on brand signalling (Erdem and Swait 1998), attachment theory (Mikulincer and Shaver 2007), and the self-expansion process (Aron et al. 2013). In a next step, consistent with our positivist perspective on brand signalling, we empirically test our research model using a survey-based approach (SEM) in an Alpine destination. The results provide novel insights for brand managers, showing how place attachment and consumer loyalty can be managed and improved by a deliberate placement of signals.

Brand credibility
Credibility can be defined as the degree to which an object is considered as a reliable and truthful source of information (Tirole 1988). The credibility of brands plays a significant role when consumer uncertainty arises due to imperfect and asymmetric information characterizing most product and service offerings (Erdem and Swait 2004;Erdem et al. 2006). In this context, brands can serve as credible signals to influence consumers' interpretations and actions through prior marketing communication strategies (Duncan and Moriarty 1998).
The concept of brand credibility is based on Hovland's et al. (1953) early work on the influence of source credibility on communication effectiveness. The authors conceptualize source credibility using a two-dimensional model composed of source expertise, the extent to which a communicator (i.e. a brand) is perceived as a valid source, and trustworthiness, the degree of confidence in the communicator's intent (Hovland et al. 1953). Branding research has further advanced this concept and identified three core dimensions determining the overall quality of brand signals (Erdem and Swait 1998;2004): Brand signal clarity describes the receivers' perceptions of unambiguity within the information conveyed by the signaller (Nguyen 2009). Clear brand signals should be free from bias, facilitate timely understanding and reduce delays or potential misunderstandings. Brand signal credibility depicts whether the information sent out by the signaller is perceived as truthful and reliable by its receivers (i.e. consumers). Lastly, brand signal consistency refers both to the degree of convergence of exposure with the brand across all touchpoints and channels, but also to the extent to which the signal remains consistent over time (Erdem et al. 2006).
Prior research considered brand credibility as an antecedent or consequence in a series of consumer responses. In marketing and consumer behaviour, brand credibility was found to be an important predictor for purchase intentions (e.g. Tae et al. 2010) and attitudes towards the product, ad, or service provider (e.g. Lutz et al. 1983). Credible sources have been asserted greater attitude change power than sources having less of these dimensions, unless they are perceived as meaningful and trustworthy (Hovland and Weiss 1951;Johnson et al. 1968). Conversely, low levels or a complete lack of brand credibility were found to reduce consumers' quality perceptions and intentions to purchase, whereas strong and positive source credibility enhanced brand image and brand value, resulting in higher purchase intention (e.g. Goldsmith et al. 2000).
More recently, brand credibility has been shown to predict brand preference and purchase intentions in the case of luxury restaurants (e.g. Jin et al. 2015) and to serve as a moderator among online brand experience and behavioural intentions in a destination context (e.g. Jiménez- Barreto et al. 2020).

The credibility of place brands
Place branding and the development of memorable destination brands have become key elements for destinations to compete in international terms (García et al. 2012;Zenker et al. 2017). Place branding (e.g. Kavaratzis and Hatch 2013) deals with the application of branding principles to places and their adjustment to the specific conditions of a setting. Place brands thereby not only identify and differentiate destinations from each other, they also convey a memorable experience uniquely associated with the setting and reinforce the collection of pleasurable memories (Tung and Ritchie 2011). Effective place branding thus reassures quality experiences and offers a fruitful means for brand managers to establish a unique selling proposition (Blain et al. 2005).
In the persuasion and decision-making process, the credibility of a place brand plays a key role as it reduces search costs for information (Erdem et al. 2006) and lowers perceived uncertainty (Erdem and Swait 2004). Brand managers, however, typically know more about their brand's offering than potential customers. In the light of this information asymmetry, it becomes difficult for consumers to evaluate the true quality of a brand. Credible brand signals (e.g. Spence 1974) thus serve as informational cues to enhance the believability of tangible and intangible brand attributes and help consumers in their decision-making. As signallers, destination management organizations hold privileged information about their brand and use a wide range of marketing mix elements and communication channels (e.g. written, personal, digital, non-verbal, or symbolic communication) to reduce consumers' uncertainty (Karanges et al. 2018).
Since consumers typically lack complete information about the brand, they will carefully observe, interpret, and respond to brand signals depending on their own experience. The quality of the transmitted signals then depends on consumers' communication competence, access, familiarity, and flexibility (Sitkin 1992).

Place brand credibility and place attachment
For a place brand, it is crucial that its claims are truthful and believable to fully exploit its potential of lowering information-processing costs and perceived risk (Erdem and Swait 1998). When consumers trust a place brand and believe that it does not intend to break its promise, it is likely that a persuasive effect on their attachment follows (i.e. giving them a "sense of place"). Credible place brands can thus serve as vital means for self-expansion (Aron et al. 2013), as they positively influence the formation of place attachment.
The notion of brand attachment is rooted in psychology and was initially developed to understand infant's attachment to a primary caregiver (Bowlby 1969). Here, early meaningful relationships lead to the formation of experience-based internal working models of the self and others, which form the basis for perception, feeling, and behaviour in all later meaningful relationships (Bowlby 1969). Marketing literature adapted this concept and suggests that consumers can also develop attachment to market entities, including product brands (Schouten and McAlexander 1995), celebrities (Thomson 2006), or places (Williams and Stewart 1998). Park et al. (2010) thus define brand attachment as the strength of the bond that connects the consumer with the self, exemplified by rich and accessible cognitive representations that involve thoughts and feelings about the brand and the brand's relationship with the self. By categorizing the brand as part of the self, consumers then develop a sense of oneness with the brand and establish emotional and cognitive links that connect the brand with the self (Park et al. 2010).
Adapted to a destination context, attachment refers to a cognitive, emotional, and functional bond that consumers establish with a place (Halpenny 2006;Yuksel et al. 2010). Drawing on attachment theory (Mikulincer and Shaver 2007), consumers possess an intrinsic motivation for self-expansion, a desire to incorporate others (i.e. place brands) into their conception of the self (Aron et al. 2013). The greater the inclusion of the entity (i.e. the place brand) in the self-concept of the consumer, the stronger the bond becomes in consumers' minds (Park et al. 2010). Over time, this relationship between the self and the entity evolves, and so does one's attachment: consumers develop a feeling of "oneness" with the place and view its resources as their own (Aron et al. 1992). Importantly, attached consumers are not merely passive recipients of the place brand and its resources, they also invest social, financial, and time resources to maintain their relationship and their process of self-expansion (Park et al. 2010). Consequently, they engage in all kinds of behaviours to ensure a continuation of their brand relationship over time, particularly when their attachment is strong (e.g. Mikulincer 1998). Place brands with credible, clear, and consistent signals (Erdem and Swait 1998) should thus ease the process of self-expansion and foster the feeling of perceived oneness (Aron et al. 2013;Park et al. 2010). Consistent with prior studies from relationship marketing (e.g. Bolton et al. 2004;Morgan and Hunt 1994), positing that trustworthy brand signals positively influence brandself connection, we hypothesize: H1 Place brand credibility positively affects consumers' place attachment.

Consumers' response behaviour
Consumers tend to engage in all kinds of behaviours when attachment is strong (Mikulincer 1998). Likewise, attachment itself requires all kinds of behaviours to maintain the relationship across time (Park et al. 2010). Consistent with attachment theory (Mikulincer and Shaver 2007), positing that attachment manifests in rich cognitive representations which become reactivated at later stages and lead to behavioural outcomes (e.g. Sweeney and Swait 2008;Hosany et al. 2017), we argue that place attachment impacts consumers' loyalty, measured in terms of intention to return and WOM. Based on this rationale, the subsequent hypotheses are formulated: H2 Place attachment positively impacts consumers' intention to return to a place.
H3 Place attachment positively impacts consumers' WOM activity.
The effect of place brand credibility on consumers' cognitive and behavioural responses constitutes a particularly important gap that has remained largely unaddressed so far. Even though previous studies showed that source credibility impacts purchase intention in the context of online brand advertisements (e.g. Visentin et al. 2019), price sensitivity (e.g. Erdem et al. 2002), and corporate brand equity (e.g. Hur et al. 2014), it remains largely unclear whether these findings also persist in the context of place brands. In other words, do credible place brands directly influence consumers' intention to return to a place and their engagement in positive WOM? More formally, we therefore hypothesize: H4 Place brand credibility leads to increased intention to return among consumers.
H5 Place brand credibility leads to positive WOM among consumers.
Whereas previous research focused on either exploring the antecedents (e.g. Bougoure et al. 2016;Reitsamer et al. 2016) or consequences of place attachment in isolation (Prayag and Ryan 2011;Kyle et al. 2004), extant studies did not investigate the role of place attachment as mediator in greater depth (Hosany et al. 2017). This is surprising, because attachment per se represents an emotional, cognitive, and/or functional mediating mechanism among environmental inputs (e.g. a credible place brand) and consumers' later behavioural responses (Mikulincer and Shaver 2007). We thus assume that place brand credibility will lead to increased WOM activity and revisit intention (Thomson et al. 2005), particularly if consumers have formed a strong attachment with the destination beforehand. More formally: H6a Place attachment mediates the path from place brand credibility to intention to return.
H6b Place attachment mediates the path from place brand credibility to WOM.

Sample
To test our hypothesized relationships, a self-administered field survey was conducted in the Alpine destination Lech-Zürs during winter season 2016/17. Located in the Arlberg Fig. 1 Research model region in Austria, Lech-Zürs has grown to become one of the world's leading winter sport destinations with 305 km of adventurous slopes. The offered activities revolve around skiing, snowboarding, freeriding, cross-country skiing, tobogganing, and winter hiking. Besides its scenic and natural resources, the destination offers the world's highest density of gourmet dining restaurants, sport events such as 'The White Ring Speed Race', live music and après ski entertainment throughout the season and hosts business meetings on a regular basis, such as the 'Rotary Club winter meeting'. Data collection was administered by four scheduled interviewers and resulted in 302 returned questionnaires. The final, useable sample consists out of 243 questionnaires (41.6% female, on average 41-50 years old, 84% repeat time visitors; see Table 1). Only consumers' who had spent at least one full day in the destination were asked to participate in the study to ensure visitation of the place and its major spots.

Measurement scales
All study constructs were measured using reliable multiitem scales drawn from the literature. Place brand credibility (α = 0.95) was operationalized with six statements from the brand credibility scale developed by Veasna et al. (2013) (e.g. "Information claims from X are believable"). For place attachment, eight items (α = 0.94) from Prayag and Ryan (2011) were used (e.g. "I identify strongly with this destination"). Intention to return (α = 0.94) was operationalized by three statements suggested by Lam and Hsu (2006) (e.g. "The likelihood to visit X in the next 2 years is high"). Finally, we used three items from Zeithaml et al. (1996) to measure word of mouth (α = 0.85) (e.g. "I would say positive things about this destination to other people"). All items were measured on a seven-point Likert-type scale anchored by "1-strongly disagree" to "7-strongly agree". All scales have been used in a destination context and performed well in terms of consistency, reliability, and validity in previous research (e.g. Jiménez-Barreto et al. 2020; Wang et al. 2020). Table 2 provides the full inventory of items used, their factor loadings and corresponding reliability.

Measurement model
In a first step, internal consistency was assessed using EFA, item-to-total correlation, and Cronbach's alpha as firstgeneration criteria (e.g. Churchill 1979). Model parameters were further estimated with the help of confirmatory factor analysis (CFA). For an assessment of the psychometric properties of our scales, a four-factor measurement model was estimated using full information maximum likelihood estimation in Mplus 7.0 (Muthén andMuthén 1998-2012). Four cases were eliminated before model estimation due to missing values. Because there is no single conclusive test of significance of the model fit (Hu and Bentler 1999), we evaluated overall model fit based on a combination of non-inferential fit indices. As the chi-square statistic is sensitive to sample size and is not relied solely as a basis for model fit assessment, we used CFI, TLI, SRMR, and RMSEA for evaluating the fit of our measurement and structural models (Kline 2015;Hair et al. 2014). The measurement model thus showed an acceptable fit (χ 2 = 338.83, df = 164, CFI = 0.95, TLI = 0.94, RMSEA = 0.067, SRMR = 0.047). All composite reliability estimates exceeded 0.86, indicating internal consistency for our latent variables (Bagozzi and Yi 1988). Convergent validity was established since factor loadings of all constructs exceeded 0.50 and were significant (t > 1.96, p < 0.05) (Anderson and Gerbing 1988). Results also suggest strong discriminant validity, as the intercorrelations between pairs of constructs in Table 3 were less than the square root of the average variance extracted (AVE) estimates of the two constructs (Fornell and Larcker 1981;Hair et al. 2014).
Since we used a self-administered survey design, it is important to rule out common-method variance (CMV) as a source of bias in our results (Podsakoff et al. 2012). To control for CMV, procedural and statistical (ex-ante) approaches were used. Among procedural remedies, we inspected all scales to reduce ambiguity, separated the measurement of predictor and outcome variables in our questionnaire and ensured anonymity and privacy to our respondents (Malhotra et al. 2006). In a second step, we controlled for statistical remedies by introducing a common latent factor (CLF) to capture the common variance among all observed variables in our model. Following the approach suggested by Homburg et al. (2010), we compared an unconstrained CLFmodel with a fully zero-constrained CLF-model. The results demonstrate that the chi-square difference between both models is insignificant (χ 2 = 353.36, df = 149 vs. χ 2 = 373.78, df = 168; p > 0.05), thus confirming that common-method variance is not a major concern in our study.

Structural model and hypothesis testing
Hypotheses were tested using structural equation modelling (Mplus 7.0) and a bootstrapping mediation analysis. The structural model revealed an acceptable fit by exceeding the cut-off criteria suggested by Hu and Bentler (1999) (χ 2 = 338.83, df = 164, CFI = 0.94, TLI = 0.93, RMSEA = 0.067, SRMR = 0.047). The results confirm that brand signalling theory represents a valid and useful framework for research on place brand credibility (Table 4), as indicated by the significant, positive impact of destination brand credibility on place attachment (β = 0.676, p < 0.001; full support for H1). Subsequently, place attachment was found to exert a highly significant direct effect on consumers' intention to return (β = 0.489, p < 0.001) and WOM (β = 0.399, p < 0.05), thus supporting H2 and H3. A bootstrap estimation with 1000 resamples in Mplus 7.0 (χ 2 = 437.61, df = 164, CFI = 0.94, TLI = 0.93, RMSEA = 0.084, SRMR = 0.047) further demonstrates that place attachment mediates the effect of place brand credibility on consumers' WOM activity (β = 0.269, p < 0.001; support for H6b). The direct effect, however, turned out to be much stronger (β = 0.519, p < 0.001; support for H5), suggesting complimentary mediation (Zhao et al. 2010). Given that brand signals are credible and trustful, consumers will directly spread positive WOM, with attachment not necessarily relevant to cause WOM activity. Lastly, comparing the non-significant direct effect of place brand credibility on intention to return (β = 0.162, p > 0.05; rejection of H4) with the highly significant mediation effect via place attachment (β = 0.330, p < 0.001; support for H6a), an indirectonly mediation of place attachment results. In other words, only if consumers' feel attached with the place brand, they will remain loyal and return to the destination. Comparing the magnitude and size of the obtained effects, consumers' intention to spread positive WOM reveals to be much more sensitive to credible signals and strong attachment than intention to return. This becomes evident when juxtaposing the size of the direct effects that place brand credibility exerts on WOM (β = 0.519, p < 0.001) and intention to return (β = 0.162, p > 0.05), but particularly when inspecting their total effects (β = 0.789, p < 0.001 vs. β = 0.493, p < 0.001). Table 4 summarizes the obtained findings.

Alternative models
In addition to the integrative model illustrated in Fig. 1, we tested for alternative, theoretically plausible models (e.g. Hennig-Thurau et al. 2002). Given that our current model argues for complimentary mediation of place attachment, we decided to choose a model without direct effects of our antecedent variable (place brand credibility) on our central dependent variables WOM and intention to return (Model 1) as conceptual alternative. In addition, we inspected a model where the roles of our antecedent place brand credibility and our mediator (place attachment) are reversed (Model 2). Moreover, to control for the frequency of previous visitation, we added first versus repeat time visitation as covariate to our integrative model (Model 4).
We tested the aforementioned models using ML-SEM with Mplus 7.0. Comparisons based on non-inferential fit indices and probabilistic model selection criteria (AIC, BIC, and Log-Likelihood) confirm that the hypothesized model (Model 3) including direct effects of place brand credibility on our dependent variables is superior to all alternative models (see Table 5). Swapping place brand credibility with place attachment, by contrast, did not result in better model fit (Model 2), which confirms our theoretical arguments provided earlier. Model 4 shows slightly better (i.e. lower) alternative model fit criteria without changing the significance of the hypothesized relationships from Model 3. Given the lower non-inferential fit indices and the comparably high proportion of repeat time visitors (i.e. 84.0%), which represents a limitation in our sample, Model 3 is considered as superior.

Discussion
Credible place brands are a key antecedent of consumers' attachment formation and their response behaviour. Confronted with an array of destination choices (Qu et al. 2011), memorable place brands that are clear, precise, and truthful in their information claims help consumers in making judgments, forming bonds, and developing loyalty towards a place. While place attachment directly affects consumers' loyalty in terms of WOM, it is an even more powerful mediating factor for their intention to return. Particularly if consumers' feel attached to a place, credible brand attributes will cause them to come back. By contrast, if they do not obtain a feeling of perceived oneness with the place during their stay, also the most credible and Even though attachment shows a significant mediation effect between brand credibility and WOM, the direct effect reveals to be much stronger. This circumstance further underlines the predictive power of credible brands (Erdem et al. 2002), as they have the potential to directly affect consumers' WOM behaviour, with place attachment not necessarily being given. Importantly, place brand credibility should be considered as antecedent to attachment formation in managerial decision-making, as our hypothesized model performed superior to alternative models inspecting its role as a mediating variable. Managers should thus focus on credible signalling in the first place, as this will positively affect consumers' attachment. Particularly when consumers are uncertain in their decision-making, informational cues can signal quality, such as, for example, consistency in product quality over time, charging price levels that fit the destination and its offerings or providing warranties such as free cancellation policies. Although the success of such marketing efforts will depend on market characteristics, consumer and competitive behaviour, it is important for managers to show long-term commitment to their signalling approach and assure consumers that their brand promises are kept (Erdem and Swait 1998). In other words, destination managers should invest steadily in credible marketing communication, as it constitutes a fundamental antecedent when building attachment with consumers. Managers could, for instance, launch (digital) communication campaigns to reinforce their brand's ability to reduce risk, generate group identification and reinforce its trustworthiness (e.g. by means of storytelling, innovative VR content, or well-known testimonials). Similarly, brand extensions in the sense of linkages among smaller destinations (such as strategic mergers or seasonal cooperation among destinations) could serve as categories where risk or collective identity matter.

Limitations and future research directions
Among limitations, it needs to be acknowledged that our study tested the notion of place brand credibility in a communitytype destination for the first time. Although our assumptions are strongly rooted in theory (i.e. brand signalling, attachment theory, self-expansion process) and our research model shows low complexity and a modest number of parameters, alternative modelling techniques (e.g. a variance-based approach such as PLS-SEM) or alternative measurement approaches for place attachment (e.g. focusing on its functional or emotional facets) might lead to different results. Results from different destination types (e.g. corporate destinations) or other cultural areas might furthermore vary considerably from our findings. Further research could provide deeper insights through testing and replicating our model in different settings. Particularly the scales for place brand credibility and place attachment could greatly benefit from further replication and a deeper exploration of alternative emotional and cognitive processing routes. Second, as the study only considers consumers' perceptions and their attachment at a certain point during their stay, future studies could analyse how place attachment and consumers' responses develop over time (e.g. Mittal et al. 1999). Longitudinal designs have previously been used to study variations in emotional responses over the context of a vacation (Nawijn et al. 2012), a similar approach could thus make sense for place attachment, particularly throughout later post-consumption stages. Third, future research should take sample-specific characteristics into account, as our study was made up of a sportive, international sample with a comparatively high net   income (65.6% > 3000 EUR/month) and a rather high degree of repeat time visitors (84%). It would be interesting to study different consumer segments (e.g. first timers, purely recreational holidaymakers) or other forms of tourism and leisure (e.g. rural tourism, adventure tourism). Finally, this study pursued a consumer-centric view on place brand credibility in the tourism industry. Since tourists view place brands differently, develop attachment under varying circumstances and for different purposes than other stakeholder groups, it is of great importance to expand the scope of application towards residents, local entrepreneurs, or public policymakers in the future.