Consumer Reshoring Sentiment and Animosity: Expanding Our Understanding of Market Responses to Reshoring

Although an increasing body of research has focused attention on reshoring, namely, a company’s decision to relocate activities back to the home-country, how and when reshoring impacts market responses remains largely understudied and requires specific theoretical and empirical consideration. Our study complements existing research on reshoring by adopting a demand-side perspective. Through multiple, experimental and survey-based studies, conducted in two countries (Italy and the US), we verified that the interplay between Consumer Reshoring Sentiment (CRS) and Consumer Animosity (CA) leads to specific emotional reactions (gratitude and relief) which, in their turn, affect relevant market responses (positive word of mouth, willingness to buy, advocacy behaviors). Our work provides interesting insights for practitioners and international managers evaluating reshoring; they can capitalize on the positive market responses to reshoring by considering both reshoring sentiments characterizing the home-country market and consumer animosity associated with tensions between the countries involved in the reshoring decision.


Introduction
Reshoring is defined as a voluntary company decision to relocate its activities back to the home-country, after having implemented an offshoring decision in the past, regardless of the ownership of the activities reshored (Ellram 2013;Gray et al. 2013). The aim of our study is to complement existing research on reshoring in international business (IB) by (a) adopting a demand-side perspective (Priem et al. 2012) and (b) considering the effects on market responses (e.g., willingness to buy reshored products) taking into account not only the home-country characteristics (e.g., Consumer Reshoring Sentiment, CRS; Grappi et al. 2018), but also the specific country from which the company has reshored its activities. We propose that the host-country from which a company reshores plays a role in shaping home-country consumer responses. No research to date has taken into consideration this effect, and this gap calls for specific research in the reshoring context centered on consumer demand characteristics able to affect company decisionmaking processes (e.g., Siqueira et al. 2015). In order to address this gap, we use a moderated mediation approach to answer the questions "when" (i.e., under what conditions) and "how" do CRS reactions influence market responses (word of mouth, willingness to buy, advocacy behaviors). We introduce the construct of Consumer Animosity (CA) (Klein et al. 1998) into the conceptual network as the moderator variable ("when" question), and we examine the mediating emotional processes of gratitude and relief transforming reactions to reshoring into actions supporting the company ("how" question).
Literature on reshoring has primarily focused on the development of a shared definition (e.g., Ellram 2013;Gray et al. 2013) and on identifying its geographical boundaries (e.g., Dachs and Kinkel 2013) or its instigators from a firm-side perspective (e.g., Canham and Hamilton 2013;Ellram et al. 2013;Fratocchi et al. 2016;Wu and Zhang 2014). Recent research (e.g., Grappi et al. 2015Grappi et al. , 2018) has adopted a demand-side perspective in analyzing reshoring, in response to the need for specific consideration of consumer characteristics in strategic decision-making. Nevertheless, much remains to be discovered about the role of demand factors in company decision processes. By expanding demand-side issues in reshoring decisions, we aim to contribute to IB research by interjecting new elements into the existing knowledge base about determinants of the decision to reshore. To this end, we take into specific consideration actual and potential end users of products of reshoring companies (i.e., home-country consumers). In this way, reshoring is considered a "tool" able to enhance utility for companies through emphasizing value-creation for consumers. This utility via reshoring arises as a consequence of the interplay between home-country characteristics of CRS and host-country characteristics of CA.
A recent stream of research highlights that CA drives consumer negative responses toward companies associated with an offending nation involved in disputes of different natures such as, for example, economic or political quarrels (Riefler and Diamantopoulos 2007). CA, defined as "remnants of antipathy related to previous or ongoing military, political, or economic events" (Klein et al.

3
Consumer Reshoring Sentiment and Animosity: Expanding Our Understanding… 1998, p. 90), has been found to negatively affect consumers' willingness to buy foreign products (e.g., Klein 2002;Klein et al. 1998). Recent studies (Antonetti et al. 2019;Harmeling et al. 2015) also show the important role of negative emotions (both threat and extreme) in explaining CA effects on consumer responses.
In our research, we advance these suggestions by proposing an extended explanatory model of market responses to reshoring, taking into consideration (a) CRS effects, (b) the role played by CA in better qualifying these effects, (c) consumer emotional reactions, and (d) relevant market responses to reshoring (i.e., positive word of mouth, willingness to buy, and advocacy behaviors). The aim is to better assess how reshoring can increase home-country consumers' perceived value of company offerings, depending on the specific host-country from which the company reshored. In other words, CA provides boundary conditions for the effect of CRS on market responses and introduces conditions governing the impact of CRS that have not been proposed before. Moreover, the introduction of emotional mediators constitutes a contribution of our research, beyond the interaction of CRS and CA. Relief and gratitude were introduced to answer the question, how does CRS influence market responses. It does this by inducing feelings of relief and gratitude in consumers enroute to impacting market responses. Adopting this demand-side perspective, we extend current understanding of the reshoring decision-making process by providing new empirical evidence on the role that demand-side issues, together with the firmside criteria identified in previous research, can play.
Our work provides guidance for international managers operating in different countries and evaluating reshoring decisions. Our findings suggest the importance of considering not only positive sentiments that home-country consumers share towards reshoring, but also the effects that international disputes and tensions may have on the same consumers in terms of animosity toward the host-country, as well as accompanying emotional reactions. Although many companies might not consider CA that home-country consumers feel toward the host country when making location decisions, there may be cases where animosity is so strong amongst consumers (Antonetti et al. 2019) that it could affect their consumption decisions and company image and reputation that it should be taken into account. For example, CA has been found to contaminate perceptions of product quality (Antonetti et al. 2019). This might allow managers to more accurately detect adverse opinions toward the host-country from which they are reshoring, and thus help in determining if and when to leverage this information (e.g., by communicating the identity of the host-country or not) so as to capitalize on the positive responses of the homecountry market.
The remainder of the paper is structured as follows: We initially review the reshoring literature, examine past research on CA, and then build a foundation for the central role played by emotions in this context. The empirical assessment of the proposed conceptual framework is presented next. We conclude by discussing theoretical and practical implications of our work.

3 2 Conceptual Background and Hypotheses
The relevance of reshoring has exploded in the past few years. Although data on reshoring are fragmented and it is still difficult to map the phenomenon (Gray et al. 2013), some empirical research has added to our understanding of the nature of reshoring (e.g., Dachs and Kinkel 2013;Fratocchi et al. 2016;Kinkel 2012). Recently a data base (https ://resho ring.eurof ound.europ a.eu/) has emerged with the aim of monitoring reshoring in Europe. Alongside these developments, which are primarily descriptive, other research has focused on the decision-making processes of reshoring companies, adopting a firm-side perspective. The key factors of cost advantages (e.g., labor cost gap reduction) and benefits connected to specific resources and competences (e.g., higher level of production quality) came to light as pivotal input in decision-making to reshore (for a detailed review, see Barbieri et al. 2017;Grappi et al. 2018).
The numerous findings provided by existing research, on firm-side aspects of reshoring overlook demand-side worth. Recent studies have begun to consider the value of demands-side considerations. One study revealed that consumers share positive beliefs about reshoring (i.e., reshoring increases consumer perceived value of the company's offers), that lead to positive market responses (Grappi et al. 2015). More recently, Grappi et al. (2018) introduced the concept and value of CRS, defined as an appraisal structure of consumers formed by relevant, generalized, long-term positive beliefs connected to reshoring. They also demonstrated that CRS is related to criteria pertinent to strategic management decisions by showing that CRS affects consumer willingness to reward the reshoring company as, for example, by influencing decision to buy its products (i.e., reshoring increases utility created within the company's value system).
Despite these contributions, there is still a need to deepen the knowledge base about how demand factors complement the existing literature on reshoring. This call provides the main motivation for our research. Although recent studies have advanced our understanding of how reshoring impacts market responses, they suffer from some limitations. First, no research to date has examined possible effects of the host-country from which a company reshored. Previous research on offshoring proposed that specific characteristics of the countries involved (e.g., cultural distance) affect offshoring (Musteen et al. 2017;Pisani and Ricart 2018) and suggested taking into consideration the country where the company delocalizes activities (i.e., the host-country) in order to better understand market responses (Durvasula and Lysonski 2009;Thelen and Shapiro 2012;Thelen et al. 2011). Building on this intuition, we propose that the country from which a company reshores is instrumental in shaping the home-country consumers' evaluations. In order to address this gap, we introduce the construct of CA (Klein et al. 1998) into the conceptual framework when considering reshoring from a demand-side perspective, with the aim of examining the specific influence the host-country has in shaping market responses to reshoring. Second, in order to better capture the underlying mechanisms of such responses, we introduced the role played by individual emotional reactions. Recent research reveals that emotions are at the root of individual responses to both reshoring (e.g., Grappi et al. 2015) and CA (e.g., Antonetti et al. 2019;Harmeling et al. 2015;Nes et al. 2012). We build on literature in psychology (Lazarus 1991;Lazarus and Folkman 1984;Roseman 1996Roseman , 1984 to develop an extended model that captures the underlying emotional mechanisms of consumer responses to reshoring.

Boundary Conditions Induced by Consumer Animosity
A specific stream of research focused on CA shows that animosity drives consumer negative responses toward companies associated with an offending nation involved in a dispute, such as political or economic disagreements (for a review, see Riefler and Diamantopoulos 2007). Disputes and tensions among countries are increasingly present throughout the world, and the examination of their impact on consumer evaluations and behaviors towards products of companies from the offending nations has gained attention (Klein 2002;Klein et al. 1998;Nijssen and Douglas 2004).
From past research (Bilkey and Nes 1982;Sharma 2011), we know that numerous elements (e.g., brand name, symbols) are used by consumers to infer country of origin information and to guide decision making. Research on CA shows that international disputes can affect consumption of foreign products from a specific country. In the seminal study by Klein et al. (1998), CA was defined as consumers' strong feelings of dislike toward a specific country due to its political, military, or economic behavior. Subsequent studies recognized different types of animosity (Nes et al. 2012, for example, considered animosity as a four-dimensional construct) and examined their effects across different target countries, product categories, and times. CA adversely influences product evaluations (Shoham and Gavish 2016) and negatively affects willingness to buy foreign products independently of product quality judgements (Funk et al. 2010;Klein 2002).
Considering the specific research area of company location decisions, recent research examining offshoring (Durvasula and Lysonski 2009;Thelen and Shapiro 2012;Thelen et al. 2011) suggests that the host-country should be taken into consideration in order to better understand the market. One of the central results points to animosity as one of the forces that heightens the likelihood of consumers opposing offshoring. Inspired by this research, we propose that in examining the reshoring decision, the country from which the company reshored plays an important role in shaping home-country consumer evaluations; CA felt by home-country consumers toward the host-country shapes market responses. Specifically, we hypothesize that CA toward the host-country induces an opposite effect, increasing positive influence of CRS on market responses to reshoring. Thus, in line with previous research (Grappi et al. 2018), which shows that consumer positive beliefs about reshoring (i.e., CRS) foster favorable market responses, we hypothesize that this main mechanism is better explained by considering the positive boundary condition of CA.
Hypothesis 1: The positive influence of CRS on market responses towards the reshoring company is moderated by the level of CA felt by home-country consumers toward the host-country from which the company reshored.

Emotional Reactions Toward Reshoring
Well established research shows that consumer evaluations of a product are not based solely on classical cues such as brand name or perceived quality, but also on consumer emotions associated with product (e.g., Holbrook and Hirschman 1982). In line with emotional attachment theory (Bowlby 1979), emotions toward a specific element (connected with a product) can influence the way a consumer interacts with that product. The element connected with such emotions can be also a specific place or country (Obermiller and Spangenberg 1989;Thomson et al. 2005). The role of emotions has been supported by several factors which demonstrate that preferences for domestic (vs foreign) products are also motivated by feelings toward a particular country (Riefler and Diamantopoulos 2007;Verlegh 2007). Previous research recognized that countries can have brand-like properties (e.g., Dinnie 2008) that are important for explaining consumer behavior; the more they are in line with the consumer's relevant beliefs and thoughts, the more positive the associated emotions (Bagozzi et al. 1999).
Extant research demonstrated the role that emotions play in affecting the reshoring decision, both on firm-side (Benstead et al. 2017;Fratocchi et al. 2016) and demand-side (Grappi et al. 2015) perspectives. On the firm-side, companies' strong emotional attachment to their home-countries favors decisions to reshore (Benstead et al. 2017). Emotional factors can impact the attraction and the perceived advantages of reshoring, such as those connected with the support of the community and the development of its economy (Canham and Hamilton 2013).
On the demand-side, cognitive-affective theory suggests that individuals' cognitive appraisals of an event guide which types of emotions are felt by individuals in response to that event (Lazarus and Folkman 1984;Roseman 1996). Exposed to a company decision to reshore, an individual cognitive appraisal can be based on the one hand, on the consumer's beliefs about the reshoring decision (i.e., CRS), but on the other hand, on consumer's aversion beliefs associated to the host-country itself (i.e., CA). In accordance with Grappi et al. (2015), the company's reshoring decision has a positive impact on consumer responses through the mediating influence of gratitude. Reshoring evokes ethical or moral reactions in the public (Gray et al. 2013;Smith 2014) (e.g., stopping the exploitation of labor in underdeveloped countries), and such considerations result in a favorable appraisal of reshoring and subsequent positive consumer reactions through the mediating effects of gratitude (Grappi et al. 2015). Furthermore, following a similar line of reasoning, the additional positive emotional experience of relief occurs when an unpleasant experience ends (Roseman 1996(Roseman , 2017Sauter 2017). In this specific context, the reshoring decision can be evaluated by individuals as the end of a threatening and unpleasant event (e.g., alleviation of job loss in the home-country due to company offshoring decisions), and consequently the elicitation of feelings of relief result.
We propose that the relationship between CRS and the positive emotions of gratitude and relief will be moderated by CA toward the host country felt by home-country consumers. These predictions are in line also with recent research (e.g., Antonetti et al. 2019;Harmeling et al. 2015;Nes et al. 2012) that shows that CA has a role in shaping individuals' emotional reactions and subsequent 1 3 Consumer Reshoring Sentiment and Animosity: Expanding Our Understanding… behaviors. In line with this research, we propose an important role for CA in shaping the intensity of the emotions of gratitude and relief felt by home-country consumers in response to reshoring. We expect that CRS influences the two emotions through the moderating role of CA (see Fig. 1).

Hypothesis 2: The positive influence of CRS on gratitude is moderated by the level of CA felt by home-country consumers toward the host-country from which the company reshored; then gratitude, in turn, positively affects market responses in favor of the reshoring company.
Hypothesis 3: The positive influence of CRS on relief is moderated by the level of CA felt by home-country consumers toward the host-country from which the company reshored; then relief, in turn, positively affects market responses in favor of the reshoring company.

Methods
Three field studies, two experiments and one survey-based, were conducted to test our hypotheses. We developed these studies, on the one hand to answer the call for experiments in IB research, until now rare and underused, to validly isolate causal variables, and on the other hand, to acknowledge the importance of conducting research in realistic settings, thereby offering more convincing findings in terms of ecological validity (Zellmer-Bruhn et al. 2016). The first two studies were run in Italy, the third in the US so as to cross-validate results and increase their reliability and generalizability. Before proceeding to test the proposed model, we conducted specific pretests on Italian and US consumers to discover the specific countries towards which consumers harbor feelings of animosity.

Qualitative Pretest
Following Riefler and Diamantopoulos (2007), we first conducted a qualitative pretest to identify the countries toward which Italian consumers feel high/low levels of animosity. Data were collected during the third week of January, 2018. Eighty-four consumers, ranging from 19 to 55 years of age (52% female), indicated the target country of their animosity. Respondents spontaneously named the foreign country for which they felt their strongest dislike and freely listed reason(s) for their animosity. The countries most often mentioned were Germany and China (23% and 13% of the sample, respectively). Subsequently, we asked the same consumers, following the same procedure, to identify countries toward which they felt low levels of (or no) animosity. The countries most often spontaneously mentioned were Spain and Norway (39% and 14% of the sample, respectively).

Quantitative Pretest
Data were collected during the first week of February, 2018. Based on Klein et al. (1998), we asked 44 respondents (59% were female; average age = 31 years) to rate their level of animosity toward the four countries previously identified (Germany, China, Spain, and Norway). The order of the countries was randomized among respondents and we retained only fully completed questionnaires (N = 136). Analysis of variance ( In order to improve the ecological validity of the study, we selected China as the high-CA country and Spain as the low-CA country. In fact, both are manufacturing countries from which a company might decide to reshore its activities. In this way respondents were asked to think realistically, thereby increasing the reliability of the results.

Study 1
The goal of Study 1 was to better understand consumer responses to reshoring by investigating the effects of (a) the level of CRS characterizing home-country consumers and (b) the level of CA connected to the country from which the company reshored. A onefactorial experimental design was chosen, in which CRS is a measured variable and CA the manipulated variable.

Participants
Data were collected during the last 2 weeks of March, 2018. Italian adults were randomly approached as they shopped in two different city-center areas in Italy by three interviewers. A convenience quota sample was selected; to improve the representativeness of the sample, the interviewers selected respondents based on the proportion of men and women and approximate age range in accordance with data on the Italian population (http://www.istat .it). The total number of people contacted was 186. Of these, 15 refused to participate; 24 did not correctly answer to the manipulation checks (i.e., they did not remember the host-country) and consequently were excluded. 1 The use of such a manipulation test followed common safe-guards used in studies to ensure that respondents read and paid attention to the questionnaire. Thus, Study 1 included 147 respondents; 48.3% men; 27.9% between 18 and 29 years of age, 32.7% between 30 and 49 years of age, 37.4% between 50 and 70 years of age, and the remaining 2% over 70 years of age. Undergraduate or higher educated respondents accounted for 11.6% of the sample, followed by respondents with a high school education (36.7%) or less (51.7%).

Procedure
Firstly, respondents answered items of the CRS scale (Grappi et al. 2018). Then a scenario describing the reshoring decision of a realistic, albeit fictitious, company was presented. Two different narratives were used to manipulate the country from which the company reshored, China was used as the high-CA country, and Spain the opposite. Each participant answered only one of the two versions of the questionnaire. The manipulations were carried out on the basis of the following scenarios: "CN5" is an important Italian company and is very well known in the market. Over the years, its products have been purchased by millions of consumers worldwide; "CN5" is, in fact, present in 130 countries and in about 61,000 stores. In 2011, "CN5" chose to move all its activities to China (Spain). Since that year all its activities have been carried out outside national borders, and more in detail in China (Spain). At the end of 2015, however, "CN5" decided to leave China (Spain) and reshore all its activities to Italy.
The name of the reshoring company was a fictitious one. At the beginning of the questionnaire it was clearly indicated to respondents that even if we used a fictitious company name, the description of the reshoring case was real. After participants had read the scenario, we asked them to answer subsequent questions on the basis of their own opinions and points of view. Consumer relevant behavioral responses to company reshoring (i.e., consumer willingness to pay for company products and participate in positive word of mouth) were measured, as well as their level of animosity, ethnocentrism, and some general demographic information. Consumer ethnocentrism (CE) and demographic information were used as control variables in the study.

Measures
Participants responded to a series of multi-item measures on 7-point scales to capture the constructs under study. An Italian marketing professor translated the English-language questionnaire into Italian, and two different marketing professors then independently translated the questionnaire back into English to verify its accuracy. Two bilingual English-Italian speakers then refined the survey. We used the items of the scale developed by Grappi et al. (2018) to measure CRS (reliability index α = 0.94). Willingness to buy (reliability index α = 0.82) and positive word of mouth (reliability index α = 0.96) were measured using three items each (adapted from Grappi et al. 2013;. CE (reliability index α = 0.90) was measured using six items selected from the CETscale (items 4, 6, 7, 8, 13, and 16 of the original CETscale; Shimp and Sharma 1987), consistent with recent studies that used subscales of the original one (e.g., Gineikiene and Diamantopoulos 2017;Harmeling et al. 2015;Sharma 2011). CA was measured at the end of the questionnaire to control for the level of animosity felt by respondents connected with the manipulation of the experimental design. Consistent with recent research (e.g., Gineikiene and Diamantopoulos 2017) that used two items based on Klein et al. (1998) to measure overall CA, we used the following items; "In general, I don't like China (Spain) very much" and "I dislike China (Spain)" (reliability index r = 0.68). Exploratory factor analyses showed that each item loaded highly on the corresponding factor and crossloadings were below 0.25 (factor loadings ranging between 0.74 and 0.98). 2

Results
CA was verified by a manipulation check in the final section of the questionnaire. Respondents rated their level of CA toward the host-country (Klein et al. 1998;Riefler and Diamantopoulos 2007). The two conditions showed significant mean differences for CA [M china = 2.57, M spain = 1.94, t(145) = 3.50, p < 0.01].
To determine whether or not CRS, CA, and their interaction significantly affect hypothesized outcomes, we conducted a moderation analysis using Model 1 of the 1 3 Consumer Reshoring Sentiment and Animosity: Expanding Our Understanding… PROCESS macro (Hayes 2013). Thus, for each dependent variable (positive word of mouth and willingness to buy), we tested whether the magnitude of CRS's effect on outcome variables depends on a third variable, that is, the moderator (CA). CE was controlled in the analysis. Results are detailed in Table 1. The top panel of Table 1 shows results for positive word of mouth, the bottom panel of Table 1 shows results for willingness to buy. As hypothesized, significant interaction effects between CRS and CA were found for both dependent variables, supporting hypothesis 1. The bootstrap confidence intervals show that each of these effects is statistically significant (see Table 1).

Discussion
Results support hypothesis 1; CRS and CA interact in shaping relevant market responses to reshoring. Results show that in both CA conditions (high vs. low), the stronger the CRS the higher the intentions to buy products of the reshoring company and to spread positive words in favor of the company. However, in the low CA condition, these effects are more accentuated compared to the high CA condition. As a consequence, the moderating effect of CA is particularly evident when consumers share less intense sentiments toward reshoring.

Study 2
Study 2 develops an experimental design examining the moderating effects of CA on gratitude and relief, which in turn are hypothesized to mediate the effect of CRS on relevant consumer responses, that is, consumer advocacy behaviors in favor of the reshoring company. To strengthen results, we include as a rival explanation for mediation, the emotion of happiness. Notice that happiness is the general category of positive emotions, whereas gratitude and relief are specific instances. Thus, we provide a tough test of mediational effects of gratitude and relief, while holding overall happiness constant.

Participants
Data were collected during the last 2 weeks of May and the first week of June, 2018. The same procedures described in Study 1 were used for recruiting respondents in Study 2. The total number of people contacted was 253. Of these, 16 refused to participate in the research; 22 did not correctly answer the manipulation check and were excluded. 3 Therefore, Study 2 included 215 respondents: 50.2% men; 24% between 18 and 29 years of age, 36.4% between 30 and 49 years of age, 36.8% between 50 and 70 years of age, and the remaining 2.8% over 70 years of age. Undergraduate or higher educated respondents accounted for 12.5% of the sample, followed by respondents with a high school education (35.3%) or less (52.2%).

Procedure
By use of an experimental design, we measured first the CRS of each respondent, and then we manipulated the company decision to reshore from high-vs low-CA countries. We used the same two scenarios illustrated in Study 1. The model analyzes the effect of the independent variable (CRS), X, on the outcome variable (advocacy behaviors), Y, where the effect passes through the mediators, M1 (relief), M2 (gratitude), and M3 (happiness as a rival explanation), and the effects of X on the Ms are conditioned by a moderator, W (CA-high vs. low). Each respondent was randomly assigned to one of the two experimental groups and answered only one of the two different versions of the questionnaire.

3
Consumer Reshoring Sentiment and Animosity: Expanding Our Understanding…

Measures
Participants responded to a series of multi-item measures (adapted from previous research and listed in Table 2) on 7-point scales. We followed the same doubleback translation procedure described in Study 1. The convergent and discriminant validity of the measures were assessed. Using structural equation modeling (LIS-REL), we ran a confirmatory factor analysis (CFA) on measures (see Table 2). The fit of the model was very good [χ 2 (df) = 441.95 (179); CFI = 0.98; NNFI = 0.97; RMSEA = 0.07; SRMR = 0.05]. All the average variances extracted (AVE) were above the recommended threshold of 0.50, and the analysis of the correlations between the variables did not include the value 1 by use of confidence intervals. With respect to the manipulation checks, we retained only respondents that correctly selected the country from which the company reshored on a question in the final section of the questionnaire. We also checked the level of CA toward the foreign country (Klein et al. 1998;Riefler and Diamantopoulos 2007), and the conditions showed significant mean differences [M china = 4.06, M spain = 1.80; t(213) = 11.28, p < 0.01]. Thus, respondents showed higher CA toward China compared to Spain, as expected.

Results
Results showed there is a difference between the two groups in terms of advocacy behaviors [t(213) = 8.92, p < 0.01]. The condition in which the company reshored from Spain (M = 3.82; SD = 1.28) had a significantly lower intention to adopt advocacy behaviors compared to the condition in which the company reshored from China (M = 5.18; SD = 0.94). Then we conducted a moderated mediation analysis using Model 7 of the PROCESS macro (Hayes 2013) for testing the hypothesized model. This model allows the indirect effects of CRS on advocacy behaviors through relief and gratitude (and with happiness considered as a rival explanation) to be moderated by the level of CA. The evidence of the moderation of the effect of CRS on the emotional mediators gives insight into the contingent nature of CRS's effect on the dependent variable, advocacy behaviors, through the mediators, depending on the moderator (CA). Such a process is called moderated mediation (Hayes, 2013). In the analyses, age, gender and CE, measured as in Study 1 (reliability index α = 0.90) (M = 3.57; SD = 1.54), were used as controls.
Our findings (see Table 3; Fig. 2) confirm that gratitude and relief constitute important mediators between CRS and the intention to adopt advocacy behaviors, and the effect of CRS on the two mediators is moderated by CA. As regards happiness, considered as a possible rival explanation of the mediation mechanisms, this emotion is affected by CRS but not moderated by CA, and also does not affect advocacy behavior. 4 4 To control for possible demand effects due to the CRS measures, we conducted a new study following the same procedures illustrated for Study 2 but changing the way in which CRS is measured; we moved the CRS scale to the end of the questionnaire after having assessed all the other variables. We collected data during the last 2 weeks of March 2019 on a sample of 79 adult Italian consumers: 46.8% men; 27.4% between 18 and 29 years of age, 32.9% between 30 and 49 years of age, 34.3% between 50

Discussion
Results support hypothesis 2 and hypothesis 3; gratitude and relief play an important mediating role and this role is shaped by the interplay between CRS and CA. Results show that in both CA conditions (high vs. low), the stronger the CRS, the stronger the emotions of gratitude and relief felt by consumers. In the low CA condition, these effects are more accentuated compared to the high CA condition. Taking the perspective of CRS, these results suggest that when consumers are characterized by a low level of CRS, CA positively affects both gratitude and relief. Conversely, when consumers share a high level of CRS, CA does not affect the two emotions (see Fig. 2). In other words, results show that when consumers share a low level of CRS, if the company reshores its activities from a high animosity country the emotions of gratitude and relief are positively affected in a stronger way compared to the case of reshoring from a low animosity country. Gratitude and relief then favor consumer advocacy behaviors. N = 215 ***if p < 0.001 Relief (

Study 3
Study 3 was a survey-based study conducted on US consumers. One of the aims was to strengthen our findings by verifying the proposed model in a different country. Moreover, in order to better assess the role played by CA within this context, this study examined in depth the effects that high levels of CA have on

3
Consumer Reshoring Sentiment and Animosity: Expanding Our Understanding… market responses to reshoring. Thus, we designed this study by developing a realistic reshoring case in which an American company decided to reshore to the US from China (country toward which US consumers feel high levels of animosity, as detailed in Sect. 3.4.1). We measured the CRS of US consumers asking them to consider China as the specific country from which the company reshored (an example of a contextualized item of the CRS scale is "The quality of the products that the company reshored in the US is higher compared to the quality of products previously made in China"). The contextualization of CRS provides a more realistic scenario that gives the opportunity to deeply analyze the interplay between CRS and CA and reinforce the external validity of our findings.

Pretests
Two preliminary studies, similar to those implemented in the Italian market, were conducted to identify the country toward which US consumers feel high levels of CA.

Participants
Data were collected during the first 2 weeks of September, 2018. Respondents were selected to reflect as much as possible the demographic characteristics of the American population, according to the United States Census Bureau (http://www.censu s.gov). A commercial firm was paid to collect data based on its expertise in conducting surveys in the US. A total of 345 consumers participated in the survey; about 13% of them were disqualified for incorrectly answering the manipulation check. 5 The final sample was comprised of 300 respondents, living in all the 50 States; 54.3% women; average age of 45 years (SD = 14.30). Undergraduate or higher educated respondents accounted for 43.7% of the sample, followed by respondents with a high school education or some college or associate's degree (54%) or less (2.3%); 17% of respondents had a total household income lower than $25,000, 27.7% between $25,000 and $50,000, 38.7% between $50,000 and $100,000, and 16.6% higher than $100,000.

Procedure
The questionnaire first illustrated the case of a US company reshoring from China using the scenario, adapted to the case in hand. All respondents then answered to items measuring all the relevant variables for the study. The reshoring decision was verified in the final section of the questionnaire. We retained only the questionnaires where participants remembered correctly the country from which the company reshored, as noted above.

Measures
Participants responded to the same measures used in Study 1 and Study 2. The convergent and discriminant validity of the measures were assessed using structural equation modeling (LISREL). A CFA was run and the fit of the model was excellent

Results
We ran the analyses using Model 7 of the PROCESS macro (Hayes 2013) for testing hypotheses. This model allows the effects of CRS on the mediators (relief, gratitude, and happiness considered as a rival explanation) to be moderated by the level of 1 3 Consumer Reshoring Sentiment and Animosity: Expanding Our Understanding… CA. At the same time, CE, age, gender, level of education, and average household income were used as controls in the analyses. Our findings confirm that gratitude and relief constitute important mediators between CRS and market responses. The effect of CRS on the two mediators is further qualified by the moderating role of CA. In line with results of Study 2, findings showed that the CA moderating effect is stronger in the presence of less intense sentiments toward reshoring. Subsequently, the two emotions of gratitude and relief, in turn, positively affect market responses (i.e., willingness to buy, word of mouth, and advocacy behaviors) (see Table 4). These results further support hypothesis 2 and hypothesis 3.
With respect to the role of the control variables considered, CE directly affects market responses, consistent with previous research (e.g., Herche 1992; Sharma 2011). 6 Finally, considering happiness, included in the model as a rival explanation of the mediation mechanisms, this emotion was directly affected by CRS but was not moderated by CA, and affected advocacy behavior, but not word of mouth and willingness to buy.

Discussion
The boundary conditions played by CA influencing felt gratitude and relief, as well as the positive direct effects of the two emotions on relevant market responses, were further confirmed. This Study, conducted in the US, cross-market verified the proposed model in a different country and strengthened the external validity of results by adopting a more naturalistic perspective via a survey study.

Discussion and Conclusions
Our study complements the existing research in IB on reshoring by proposing that the host-country from which a company reshored can play a role in shaping the home-country consumer responses. By adopting a demand-side perspective, we developed and tested an extended explanatory model of market responses to reshoring: we verified that the effects of CRS on market responses are regulated by the influence of CA; the interplay between the CRS and CA drives specific emotional 6 To strengthen the testing of our hypotheses, we also run analyses that explicitly took into consideration the role of CE as possible moderator in the model. CE is defined as the "beliefs held by consumers about the appropriateness, indeed morality, of purchasing foreign-made products" (Shimp and Sharma 1987: 280). It encompasses issues such as the morality of buying imported products and a personal prejudice against imports. Ethnocentric consumers consider buying national products as part of their duty to their country (Zeugner-Roth et al. 2015). Thus CE can be considered as a contiguous dimension from the theoretical point of view with the constructs under study, and the validity and reliability of the proposed model can be improved by taking its possible moderation role into consideration in the model in Fig. 1. Thus we control for this by inserting CE into the hypothesized model. Results showed that CE does not change the mechanisms underlining market responses to reshoring; CE did not moderate the effects of (a) CRS on emotions, (b) emotions on market responses, and (c) CRS on market responses (with the only exception of word of mouth). From a theoretical point of view, our study expands the current understanding of the reshoring decision-making process by providing new empirical evidence on the role that demand-side characteristics play, thus contributing to the emerging literature examining demand-side predictions in the reshoring context. We proposed and tested a wide perspective of market responses to the company reshoring decision, by demonstrating a role for relevant consumer input (CRS and CA). We recommend that these should be taken into consideration together with other relevant, firm-side elements identified in the current literature (e.g., Canham and Hamilton 2013;Casson 2013;Ellram et al. 2013;Fratocchi et al. 2016;Gray et al. 2013;Kinkel 2014;Wu and Zhang 2014). In this way, our work demonstrates that reshoring can be considered validly as a way to increase value within the company by emphasizing value-creation for consumers and by leveraging, when appropriate, the host-country involved in reshoring. We respond to the recent call of scholars to examine the role of consumer demand characteristics (i.e., CRS and CA) in affecting the company's decision-making process (Priem et al. 2012;Siqueira et al. 2015). On the one hand, we strengthen the existing knowledge base about reshoring in a demand-side perspective by verifying the pivotal role of CRS in favoring positive market responses to reshoring (as proposed by Grappi et al. 2018); on the other hand, we add to this by showing that the mechanism through which CRS operates can be conditioned by the degree of animosity that consumers feel toward the host-country from which the company reshored.
Disputes and tensions among countries are relevant in explaining consumer evaluations and behaviors towards products of companies from the offending nation (e.g., Klein 2002;Klein et al. 1998;Nijssen and Douglas 2004). Our work demonstrates the need to consider the country from where the company reshored its goods and the degree of animosity felt by consumers toward the offshored country to better understand market responses to reshoring. In this way we contribute to the body of research which acknowledges the effects of CA on consumer responses toward companies or products associated with a foreign country (e.g., Funk et al. 2010;Klein 2002). We demonstrated that CA moderates the positive effect of CRS on market responses; the relationship between CRS and market responses is positive in both CA conditions (high CA vs. low CA) meaning that, regardless of the level of CA, those with high levels of reshoring sentiment show more positive market responses to reshoring. However, this effect is stronger for consumers low (vs. high) in CA toward the host-country. As a consequence, the moderating effect of CA is particularly evident when consumers share less intense sentiments toward reshoring. Overall, findings support the important role of this additional demand-side characteristic (CA) in understanding market responses to reshoring (e.g., Harmeling et al. 2015;Riefler and Diamantopoulos 2007).
Other relevant theoretical contributions of our study are connected with the role of emotions in explaining market responses to reshoring. We proposed that consumers feel emotions toward specific elements connected with a product (as the country of origin) and that these emotions can influence the way consumers interact with that 1 3 product or with the company producing it. Recent research demonstrates that emotions can affect consumption decisions (e.g., Oberecker and Diamantopoulos 2011) and consumer responses (e.g., Antonetti et al. 2019;Grappi et al. 2015;Harmeling et al. 2015). Starting from Grappi et al. (2015), we built on this by considering the role of gratitude and adding another important emotion (relief) in our conceptual model explaining consumer responses to reshoring. We demonstrated that both gratitude and relief mediate the effects of CRS on market responses to reshoring; and the intensity of the two emotions depends on the level of animosity consumers feel toward the host-country. Thus, our research shows that consumers react with more than merely gratitude in response to reshoring; in this way we demonstrate the additional positive effect of relief in explaining market responses to reshoring. Gratitude and relief prompt a willingness to support reshoring companies.
Overall, our research contributes to the literature on reshoring by providing an integrated view of the elements that previous research adopting a demand-side perspective has identified as relevant (e.g., CRS, positive emotions). Our work also demonstrates the important role played by CA in shaping market responses to the major strategic move of reshoring.

Managerial Implications
Our research has practical relevance for international managers operating in different countries and dealing with company relocation decisions. Our work suggests CRS should be taken into consideration along with the possible tensions between the home and host countries involved in the reshoring decision. These tensions can favor strong levels of CA towards the host-country, and our study shows that CA toward the foreign country can enhance the positive effects of CRS on home-country market responses to reshoring.
Our research suggests that managers ought to recognize the possible presence of adverse thoughts and opinions toward the host-country and, if present, leverage this information. Acknowledging that the level of CA shared by the home-country consumers is only one of several pieces of information that a company can take into consideration in its decision process, its examination can be helpful to decide how to communicate the reshoring decision to the home-country market. In order to capitalize on the positive responses of the home-country market, if CA is high, managers can decide to disclose in their communications to the market the specific host-country from which the company reshored. In this way, they can magnify the emotional responses to reshoring and, consequently, the positive responses of the market to this decision. On the other hand, when CA is low, managers could decide to disclose the specific hostcountry only if this information induces a positive effect (i.e., together with high levels of CRS). Specifically, our work suggests that managers and practitioners should take into consideration both CRS and CA in home-country market segmentation and profiling. In view of the results of our research, "(re)made in the home-country" targeted campaigns can be very successful if the reshoring company is able to identify segments of home-country consumers characterized by strong CRS. This could be estimated with a random sample of consumers from the target market. For segments characterized by less intense CRS, but at the same time, strong CA toward the host-country from which the company reshored, a target campaign of "(re)made in the home-country from the specific, disliked host-country" could be very effective too. In general, our results suggest that reshoring companies could carry out an exhaustive analysis of the home-country market, segment it according to estimates of levels of CRS and CA and, then, target consumers strategically to achieve optimum responses.
Furthermore, our study demonstrates the important role played by the positive emotions of gratitude and relief in favoring positive market reactions. These findings suggest that a periodic monitoring of the evolution of these emotions should be carried out in order to identify which specific information connected to the reshoring decision maximizes gratitude and relief. By strategically managing the sources of information that consumers use to form their thoughts and emotions connected to reshoring, managers can get the most out of market responses to reshoring. The branding strategy of the reshoring company, for example, should build on these elements to suggest useful guidelines for influencing target consumers in the reshored market. The reshoring brand narratives of companies should leverage the emotions associated by consumers with reshoring and to the host-country from which the company reshored. For example, the renowned Kangol headwear brand publicized in November 2015 that it was bringing the manufacturing of its Kangol Patriot Tropic 504 back to the US from China, specifically to Pennsylvania in one of the oldest hat factories in the US. Kangol's president and CEO said that "This is historic time for a historic brand and a historic company. We have a chance to influence and be part of history" (D'Arienzo 2016), leveraging not only the home-country superior characteristics (compared to the host-country), but also other connected elements such as symbols of heritage and authenticity, which can arouse relevant consumer emotional reactions. In general, branding strategies able to embrace, among others, the positive beliefs forming reshoring sentiments of the home-country market (CRS) and the positive elements associated to the home-country location, can foster positive emotions, helping to make a reshoring decision successful, robust, and sustainable.

Limitations and Future Research
While our findings provide important implications regarding market responses to reshoring, some limitations should be noted. First, beyond our focus on three relevant market responses, future research could examine other relevant responses explaining these with the conceptual framework introduced herein. Moreover, our study relied on quantitative self-reported measures, which may restrict the conclusions drawn from the findings and this issue deserves investigation in future studies. Thus different types of consumer responses to reshoring (e.g., actual consumers buying behaviors of reshored products) should be considered in the future in order to broaden our understanding of how reshoring can increase the value created for companies. Further, the procedure followed to assess the effects of CRS could have led to possible demand characteristic effects. In Study 1 and Study 2 (differently from Study 3), CRS was measured before the relevant responses to reshoring. We conducted a specific study (see footnote 4) controlling for these possible effects, and we suggest that future studies further investigate this issue to demonstrate robustness of the findings.
Second, the focus in our study was limited to the general effect of CA without distinguishing between different types of conflict that could potentially drive specific consumer evaluations. Future research might consider if market responses differ based on specific sources of animosity (e.g., animosity based on economic reasons, animosity based on political issues, or animosity arising from cultural factors).
Third, we acknowledge that our framework might be expanded to include additional processes and variables. Although our findings demonstrate that CRS and CA play important roles in explaining market responses, they do not entail that these elements are the only demand-based characteristics affecting market reactions to reshoring. For example, consumer affinity (Oberecker and Diamantopoulos 2011) as an individual trait, or perceived authenticity (Beverland 2005) as a situational element might be useful to investigate as other moderators of responses to reshoring. Moreover, since our study is the first to examine the interplay between CRS and CA, its findings could be seen as suggestive rather than definitive and a call for further research is needed to strengthen them. Replications in other settings in which both CRS and CA co-exist are needed in order to extend the generalizability of our results as well.
Finally, our research is centered on the positive aspects of home-country consumer responses to reshoring. This focus was central and confirmed throughout the multiple, experimental and survey-based, cross-market studies presented herein. At the same time it must be acknowledged that home-country consumers might in certain cases react to associate reshoring with sympathetic sentiments toward the host country. For instance, concerns might be raised for the loss of jobs in the host country. The simultaneous presence of unpleasant feelings and positive emotions associated to reshoring (such as those identified in our research) can give rise to conflicting reactions. For example, if home-country consumers are grateful toward the reshoring company and thus inclined to support it by buying its products but at the same time feel guilty for possible negative effects that the reshoring decision can have on the host country, then this ambivalence could lead consumers to move away from this situation to reduce their discomfort. Such complex behavioral reactions deserve to be studied in future research. At the same time, it must be acknowledged that reshoring does not always mean that labor is re-located (as it can occur in the context of automation, for example). These specific cases may also have demandside effects that deserve to be examined in future studies.