1 Introduction

Country-Of-Origin (COO) refers to the impact information pertaining to a product’s place of origin has on consumer buying intention. COO research has invariably been based on the assumption that consumers view COO cues as diagnostic information, which they deliberately use in their evaluation of products (Katsumata and Song 2016). Consequently, a COOL cue may automatically stimulate country-related stereotypes that subsequently impact purchase intention (Herz and Diamantopoulos 2013).

Previous research focused on country of origin points to many consumers who frequently use label information as important criteria in formulating their purchase decisions and behavioral responses to a particular product (Wilcox 2015). In particular, consumers use countries’ manufacturing and technological abilities and reputation as main indications of product country of origin (Katsumata and Song 2016). In this regard, much of the extant literature conceptualizes the COO as encompassing perceptions of a sourced country's economic characteristics, political considerations, history, traditions and cultural attitudes, as well as the quality and reliability of a specific foreign product (Collins and Weiss 2015). Moreover, a review of many COO studies points to a vast majority focused on the impact of previous or ongoing events on consumer intention to buy foreign goods, where the same made-in labelling standards and regulations apply to all sourced countries and products. In recent years, however, particular labelling requirements were imposed on products originating from certain countries or geographic regions in order to ‘punish’ them by informing consumers that the labelled products have been imported from an undesirable source (Newman et al. 2014). Such “finger pointing” country of origin labelling aims to reduce the desirability of the purchase and consumption of the labelled products. For example, in 1918, in order to warn the European consumers of that time, the defeated countries, such as Germany, were obliged to put 'Made in Germany' labels (in English) on each product they exported (Cai et al. 2004; Morello 1984). An additional example is reflected in US-Canada trade relationships. Both countries have extensive requirements demanding that imported packaged products should bear some information about the product’s manufacturing location (Roff 2016). For instance, in 2008, a new US legislation required meat products’ labels to identify the country where the animal was born, raised, and slaughtered. Opponents of the US labeling law argued that it comprises a disguised trade barrier aimed at enabling consumers to easily identify US food products and to purchase them in lieu of imported food products (Kolodinsky 2012; Newman and Kopp 2009). Similarly, in 2009, the UK issued new guidelines concerning the labelling of products imported from the West Bank. The new guidelines require labelling to clarify whether West Bank products originate from settlements or from the Palestinian Authority. Denmark has similar legislation requiring that food products from Israeli settlements be accurately labelled.

Generally speaking, ‘root causes’ of most recent boycotts are of a political and economic nature, rather than due to traditional issues such as high prices (Al Shebil et al. 2011). In the context of the present research, the recent EU labelling decision (November 2015) reflects a policy which considers Israeli settlements illegal under international law. Nevertheless, Israeli settlements are at the center of the Arab/Muslim-Israeli animosity case, which is attracting the international attention of citizens of non-Arab and non-Muslim countries around the world. Moreover, many EU countries view Israeli settlements as seeking to gain economic power over the West Bank and Golan Heights regions, thereby causing real distress to the local population. Drawing on this, it could be argued that that while the stated purpose of EU regulations is to help consumers make informed purchasing decisions about the products they buy, opponents of the legislation (e.g., Israeli officials) contend that this decision is politically driven in that it enables consumers to more easily identify products made in ISt and accordingly, to avoid purchasing these products.

Yet, research on the influence of forced labelling regulations on consumer buying decisions is extremely limited. The sporadic studies focused on COO labelling actions indicate that generalizable knowledge about its impact on consumer behavior still remains scarce (Newman et al. 2014). For example, a study by Insch and Florek (2009) on the impact of labelling legislation on global trade flows concluded that new labelling requirements could lead to reductions in global trade. Another study by Kawashima and Sari (2010) examined the effect of product labelling on consumers who attach high importance to safety standards. The study demonstrated that COO had greatest impact on retail demand for domestic relative to imported food items (beef).

With this as the background, the present study fills this lacuna in the literature by developing and testing a conceptual framework that links the imposition of new labelling regulations with consumer buying behavior and testing the model empirically. In particular, the study investigates the extent to which EU decision to add a new label on products sourced from the Israeli settlements (ISt) would motivate consumers to consider boycotting these products. In addition, the study examines the “spillover” effect in the context of linked regions. For this purpose, we employ a research model that builds on information integration theory (IIT; Anderson 1971, 1981, 1982). IIT provides a theoretical lens that explains how behaviors are changed through the integration of new information cues with existing attitudes to make an overall judgment (Anderson 1971, 1981; Balduzzi and Tononi 2008; Kim et al. 2010). According to the theory, attitudes are formed and modified as individuals interpret new information and integrate it with their prior attitudes (Anderson 2008; Hastak and Hong 1991; Hofmans et al. 2012; Vidotto 2013).

The study of antecedents driving consumers to consider boycotting products made in disliked sources is being seen as increasingly important for exporting companies and retail executives, especially as the global competitive environment intensifies. Boycotts have critical effects on firms’ market share, profitability, viability, and future revenue stream. With increasing competition and rising costs, securing significant market share and profitability will necessitate exporters and retail executives handling boycott antecedents with more care and sensitivity to avoid animosity and consequent backlash in foreign markets, so as to stay ahead of their competitors.

The study begins with a background discussion of the antecedents of boycott participation. It then integrates relevant themes from this research field into a theoretical framework and suggests a series of hypotheses relating to the conditions in which boycott participation will occur in the context of a particular region (ISt), and expand to a linked country (Israel). The second part of the article presents a consumer behavior study designed to test these hypotheses, and presents its empirical findings. Finally, the paper discusses the implications of the analysis.

2 EU labelling regulations

COOL regulations refer to requirements by certain countries to label the origin of products imported from other particular countries or regions. Sometimes, the intent of such requirements is to render such products less desirable (Morello 1984).

In November 2015, the European Commission laid out guidelines for labelling products produced in Israeli settlements (ISt) in the West Bank and Golan Heights. According to the guidelines document, products originated from ISt, an indication limited to ‘product from the Golan Heights’ or ‘product from the West Bank’ would not suffice. Therefore, the label must also include the words ‘Israeli settlement’ in brackets. For example, expressions such as ‘product from the Golan Heights (Israeli settlement)’ or ‘product from the West Bank (Israeli settlement)’ should be used.

The present study investigates this link in the context of a third-country national (TCN) (the UK). TCN countries or nations are not directly involved in the animosity case (i.e. the Israeli-Palestinian conflict). However, the EU decision may evoke negative feelings driven by consumers’ concern for the local (non-Israeli) population. Such concern would be further strengthened by consumers’ awareness of and interaction with the animosity case, thereby activating negative emotions towards ISt. For example, we may experience negative emotions such as animosity if we consider how some individuals have been hurt or harmed by others (Batson 2006; Haidt 2003; Halimi et al. 2017; Prinz 2011).

3 Theoretical framework

3.1 Antecedents of boycott participation

A boycott is defined as a type of anti-consumption behavior, where consumers refrain from buying products made in a particular country or firm (Hoffmann 2011). Boycotts may be considered a rather extreme form of consumer behavior, when consumers are expected to refrain, rather than engage, in purchasing. Moreover, boycotts deliberate retaliatory actions usually driven by feelings of anger and animosity towards the target country or company. According to by Ben Mrad et al. (2014) feelings of anger can be translated into behavior when consumers believe their participation will make a difference and the boycotting action is effective. Furthermore, since boycotts negatively affect the targeted company market share and profitability they are potentially effective in forcing desirable changes in the actions of the boycotted firm or country (Chavis and Leslie 2009).

3.2 Information integration theory (IIT)

In assessing antecedents of boycott participation we rely on the information integration theory (Anderson 1971, 1981, 1982). The major concept of IIT is that individuals modify their attitudes or behaviors as they are introduced to new information cues (triggers or stimuli) (Anderson 1981; Bourdeau et al. 2007). However, to arrive at a final response, triggers must be cognitively internalized and analyzed in three stages (Anderson 1981; Carlson and White 2008; Hastak and Hong 1991). In the basic flow of IIT, triggers are first interpreted in the valuation stage, integrated during the second stage, and then a quantitative judgment (response) is developed and expressed during the final stage (Eggert et al. 2012).

Trigger interpretation refers to the process of internal valuating of the different stimuli components. In this process, individuals extract information from a new information cue and turn it into a derived value (Anderson 2008). Trigger integration occurs during the second stage. This process describes how individuals internalize and integrate information in a thought. For example, in the context of the current study, an individual might process new information derived from the EU labelling legislation as two different stimuli: utility (how EU regulations help consumers make better purchasing decisions about the products they buy), and emotion (the extent to which the new legislation evokes negative feelings which may motivate consumers to avoid buying products sourced by Israeli settlements).

Finally, a response is constructed during the final stage. This process, which is directly linked to the integration of multiple information cues, refers to the psychological process of imposing numerical values on the newly combined information. Thus, when information cues exist, after information is weighted and integrated, a consumer response or judgment is formulated and expressed as an observable variable based on the summed contribution of each cue (Anderson 1971; Kim et al. 2010).

Information integration theory has been applied frequently in various marketing research areas, such as country-of-origin effects (e.g., Hastak and Hong 1991; Roggeveen et al. 2007), brands and co-branding (Böger et al. 2018; Jeon and Baeck 2016; Kwon et al. 2008; McLelland et al. 2014: Saqib and Manchanda 2008), marketing communications (Kim et al. 2010; Navarro et al. 2009), operations management, marketing channels, and supply chain research (Pullman et al. 2001). For example, results from a study by Hastak and Hong (1991) pointed to significant country-of-origin (COO) effects on consumer judgements about perceived product quality and overall product evaluations. Similarly, based on IIT, Roggeveen, Bharadwaj and Hoyer (2007) substantiated the important role of country-of-origin (COO) in consumers’ service expectations from lesser-known firms.

Within the extensive literature on boycotts directed against entire countries, few studies have investigated antecedents that can be assigned to the category of triggers or new information cues. These studies (e.g. Aggarwal and Maamoun 2008; Ben Mrad et al. 2014; Heilmann 2016) have focused on international conflicts which motivated consumers to boycott products made in the targeted countries. No research has yet investigated the direct influence of new labelling actions on boycott participation within a particular context, nor on expanding boycotts to other linked regions/countries. However, some evidence exists in studies investigating the influence of new labelling legislation on global trade flows (Insch and Florek 2009), and on consumer demand for particular food items (Kawashima and Sari 2010).

The EU labelling regulations provide new COO information about products sourced in ISt. That being the case, in applying ITT in the context of this research, we assume that EU regulations comprise a trigger that can prompt an internal valuating process which in turn would drive consumers to consider new information cues, namely, consumer ethnocentrism, perceived economic threat, and product quality found in prior research to have critical effects on purchasing foreign-made products.

3.3 Economic threat

Economic threat refers to subjective perceptions and feelings about the effect of outgroup threats on other individuals’ economic security and well-being (Riek et al. 2006). This issue is highly relevant to the current study, which focusses on a third-country national (from the UK). As mentioned above, while TCNs are not directly involved in the animosity case, we assume that such populations would demonstrate negative sentiments towards Israeli settlements driven by their concern for the local (non-Israeli) population. For example, TCN consumers may experience moral emotions if they consider how the Palestinian economy has been harmed or hurt by Israeli settlements.

Empirical studies addressing the link between economic threat and negative emotions (e.g., animosity) provide evidence for the proposed relationship (Cheah et al. 2016; Cui et al. 2012; Huang et al. 2015). For example, a study by Huang et al. (2010) highlights the role of perceived economic threat as a factor that stimulates feelings of animosity. According to that study, perceived economic threat and economic hardship positively impact the phenomenon of consumer animosity, which in turn negatively affects consumers’ intention to purchase products originating from disliked countries. Based on this extensive research, it can be argued that the EU labelling regulations comprise information cues that will play a role in the subsequent information-integration process. Thus, based on IIT, one can claim that consumer ethnocentrism and perceived economic threat comprise trigger variables that promote negative emotions towards products sourced from ISt.

H1

There will be a positive relationship between perceived economic threat and animosity.

3.4 The moderating effect of product quality

Moderators are variables that affect the direction and/or strength of the relationship between predictor variables and outcome variables. In the context of product boycott participation, moderating factors such as product quality concern boycott effectiveness as they provide reasons to refrain from participation.

Prior COO studies examining the role product quality plays in consumers’ buying decisions regarding foreign products indicate a positive relationship between product quality and buying intention (Chrysochou et al. 2012; Ettenson and Klein 2005; Taewon and Kwon 2002; Zafar et al. 2013). Thus, the more favorable the product judgment, the greater the willingness to buy a foreign product.

Product quality has also been discussed in various studies that employed IIT (e.g. Hastak and Hong 1991; McLelland et al. 2014). A review of many such studies reveals that most studies investigated product quality that was first introduced to participants as a new information cue, then tested to provide evidence for the information integration approach. However, little attention has been devoted to investigating the influence of product quality as an information cue that can modify consumers’ negative thoughts and emotions associated with economic threat and ethnocentrism towards buying foreign-made products. Indeed, though highly ethnocentric consumers may view foreign products as an economic threat, one might expect that product quality can moderate the influence of ethnocentrism on perceived economic threat. Thus:

H2

Product quality judgment will moderate the influence of economic threat on perceived animosity towards products sourced in Israeli settlements.

3.5 Animosity

Research on consumer animosity provides a bridge between the literature on COO effects and the research stream on boycott participation. Animosity is related to COO in that it emphasizes negative emotions toward purchasing products from disliked sources (Gineikiene and Diamantopoulos 2017; Meshreki et al. 2018; Narang 2016; Roth and Diamantopoulos 2009). Animosity involves feelings of anger, open hostility, and negativity toward products originating in disliked sources (Amine et al. 2005; Klein et al. 1998). These antipathetic sentiments can easily motivate consumers to consider boycotting products from the country in question (Amine 2008; Cui et al. 2012; Fong et al. 2015; Maher and Mady 2010).

Indeed, most previous research has found animosity reduces consumer willingness to buy products from greatly disliked countries (Cui et al. 2012; Narang 2016). Moreover, many scholars have established a strong link between negative emotions and consumer intention to buy products produced by countries with conflicts (Cheah et al. 2016). According to Zafar et al. (2013), consumers who are highly antagonistic towards a particular country are prone to impulsive and quick response to products and services because of the past or on-going actions of the country. This is also evident in the context of the buying decisions of third-country nationals’ (TCN) consumers. A study by Halimi et al. (2017) highlights the role of negative emotions as a potential factor affecting third-country consumers’ decisions to boycott Israeli products. According to the study, consumers’ participation in the boycott against Israel was driven by their concerns about the Palestinians and Lebanese citizens. These concerns activated their negative emotions towards Israel, thereby motivating them to avoid the purchasing of Israeli products. Similar findings were reported in a study by Hoffmann (2013). The study results provide evidence of an animosity emotions-boycott intention link: The more strongly the individual perceives the hardship of other individuals, the stronger his or her feelings of animosity will be, and consequently, the more he or she will feel obligated to avoid purchasing products originated in disliked sources. (Hoffmann 2013). Thus:

H3

There is will be a positive relationship between perceived animosity and consumers’ intention to boycott the targeted country or region.

H4

There will be a positive relationship between animosity and perceived punishment.

3.6 Punishment as a promoter of boycott participation

Punishment refers to consumer’s belief that boycotting-participation comprises an effective action that will change the targeted country’s attitude (Klein et al. 2004). Many empirical studies which investigated consumer animosity point to a direct relationship with punishment. Strong hostile attitudes and negative emotions can easily prompt feelings of punishment among consumers, which in turn can lead them to consider boycotting as a form of punishment. Consequently, the impact of animosity is in line with views that hold punishment to primarily serve retaliatory purposes (Carlsmith et al. 2002; Darley and Pittman 2003; Bougie et al. 2003). A good example is the boycott of Danish products by Arab and Muslim consumers. In 2006, Danish tabloids published a series of cartoons depicting Islamic prophet Muhammad in an unfavorable manner. This promoted massive anger across the Middle East, and consumers of Danish products started a long boycott. Additional examples include the boycott of French products in the US over the Iraq War (2003), Turkey's boycott of Israel over the Gaza Strip conflict (2014), and the boycott of American products (2006) by Arab consumers in the Middle East as a reaction to the Israeli-Lebanese war (Ben Mrad et al. 2014; Heilmann 2016). Thus:

H5

There will be a positive relationship between perceived punishment and consumer intention to boycott ISt products.

3.7 Consumer ethnocentrism

Consumer ethnocentrism (CET) refers to beliefs held by consumers about the appropriateness and morality of purchasing domestic-made products while rejecting foreign-made products (Shimp and Sharma 1987). By definition, consumers with strong CET tend to have a more favorable perception of domestic products than they do of foreign ones (Kipnis et al. 2012; Wang and Chen 2004).

Consumer ethnocentrism has been studied extensively in the international marketing literature for its impact on product evaluations and buying intention of foreign products (Balabanis and Diamantopuolos 2004; Narang 2016; Yildiz et al. 2018). Previous research investigating the relationship between COO and CET point to a negative correlation between ethnocentrism and buying foreign-made products (Cilingir and Basfirinci 2014; Lewis and Grebitus 2016; Schnettler et al. 2011). For instance, a study by Kipnis and colleagues (2012) found ethnocentric consumers reluctant to purchase foreign products in general or at least less willing to buy foreign products from any country. Moreover, several scholars argue that CET reduces consumption of foreign products, since ethnocentric consumers view imported products as inferior to those produced in their own country. As such, ethnocentrism contributes to consumer intention to avoid buying foreign products in general and hence to the actual reduction of imports.

As noted above, ethnocentrism reflects consumers’ desire to protect local economy (Durvasula and Lysonski 2009). Moreover, according to Klein et al. (2006), ethnocentric consumers believe that a person buying foreign products is unpatriotic and amoral, and that such behaviour can hurt the economy and lower employment levels. Consequently, such consumers are more likely to refrain from buying foreign-made products. This is especially evident when ethnocentrism reflects a real concern for economic interests, such as losing jobs and high unemployment levels (Evanschitzky et al. 2008; Narang 2016; Sharma et al. 1995; Shoham and Gavish 2016).

H6

The higher the perceived ethnocentrism, the greater the tendency to avoid buying products manufactured in Israeli settlements.

3.8 Expanding boycotts to linked countries: the spillover effect

Boycott spillover occurs due to misperceptions about the national identity of a product, or when.

a product (or industry) from a particular country is targeted, but the negative consumer feelings expand to other products from that country (Al Shebil et al. 2011).

At the center of the current study is another type of boycott–spillover which occurs when products from a particular country/region are targeted, but the negative consumer emotions expand to products from linked countries/regions. Little attention has been devoted to studying antecedents of boycotts expanding to other regions or countries linked with the one targeted. However, some evidence may be found in studies that discuss boycott decisions directed towards ‘third-party’ countries that are indirectly associated with actions attributed to the offending country. In 2006, for instance, anti-American sentiment grew in the Middle East as a reaction to the Israeli-Lebanese war (Ben Mrad et al. 2014). In this war, the United States was perceived by Arab countries to have aligned itself with Israel. As a result, many Arab consumers boycotted American products as a reaction to their perception of USA foreign policies, even though the United States bore no responsibility for, and took no part in, the Lebanese-Israeli war.

Based on this, the current study suggests that consumer feelings of anger, defiance, and alienation towards products manufactured in a certain region (ISt), thereby considering boycotting products manufactured in this region, would be expanded to a related country (Israel), which in turn would drive consumers to consider boycotting products from the linked country as well. That is, when two regions are seamlessly linked, consumers can easily transfer their (negative) attitudes from one region (ISt) to the other (Israel).

H7

A boycott against products sourced from ISt will be spilled over to the linked country.

3.9 The moderating effect of country image on participation

Country image refers to consumer beliefs about a country, including its economic development, technological advancement, cultural, and political characteristics (Brijs et al. 2011; Kan et al. 2014). Consequently, country image is expected to affect product informational cues which impact consumer perception of quality (Collins and Weiss 2015; Gorostidi-Martinez et al. 2017; Meshreki et al. 2018). In effect, it is argued that consumers are able to make judgements about both country image and products (Jaffe and Nebenzahl 2006), and there is evidence that country image might play an even stronger role in determining product evaluations (Halkias et al. 2016). As such, country image variables negatively influence boycott effectiveness as they justify the avoidance of boycotts.

Prior research demonstrates a lack of agreement on the type of country image dimensions. However, many previous studies have assumed that country image is a multidimensional construct and the dimensions covered in prior research are varied (Lala et al. 2009). For example, Martin and Eroglu (1993) focused on political, economic, and technological dimensions. Lala et al. (2009) focused on economic condition, political structure, and so forth. Mahran and Nasser (2014) determined six dimensions, including quality and technology, whereas Laroche et al. (2005) focused on three dimensions, including country beliefs, people affect, and desired interaction. According to Lee and Ganesh (1999), country image is created not only by variables like economic and technological advancement, but also by product quality. Therefore, country image that can affect consumer behavior towards products from the country must be understood from beliefs generated by the production-oriented dimension of that country (Heslop et al. 2008). Thus, after a careful review of the literature, this study determined two dimensions of country image which appeared to be the most relevant: product quality and technological advancement. Both dimensions are directly related to consumer intention to purchase foreign-made products as these variables comprise inhibiting factors that can decrease consumer boycotting tendency. To the best of our knowledge, no study has yet evaluated the moderating effect of quality and technology dimension in the context of boycotting-participation behavior.

Several country image characteristics have been discussed in the international marketing literature, meaning some consumers may continue to buy the same products as in the past despite their animosity toward a particular country or firm (Gineikiene and Diamantopoulos 2017). Furthermore, given that boycott actions imply constraining previous consumption patterns, consumers are less likely to participate if the product is of perceived high quality and lacks adequate substitutes (Smith and Li 2010).

3.9.1 Perceived product quality

The product quality judgment–buying intention relationship has been studied in both domestic market research and international country-of-origin settings. In the latter context, Zafar et al. (2013), for example, claimed product quality judgments were directly related to consumer intention to purchase foreign-made products, suggesting that consumer product-judgments are positively correlated with their buying behavior. As such, the more favorable the product judgment, the greater the willingness to buy foreign products. Similarly, Chrysochou, Corsi, and Krystallis (2012) substantiated the important role of product quality judgment in influencing consumer purchase of foreign-made products. However, little attention has been devoted to the influence of perceived product judgment on boycott-participation intention. Studies investigating the product judgment—buying intention construct indicate a negative correlation to boycotting intention. Smith and Li (2010), for example, in a study on Chinese consumer willingness to boycott Japanese products claimed Japanese product quality directly affected judgments on boycott participation. That is, consumers who perceived Japanese products to be of higher quality than domestic goods were less likely to participate in a boycott. Drawing on this previous work, it can be argued that product quality judgment is an inhibiting factor as it decreases consumer boycott-participation tendency.

H8

Boycott spillover to a linked country or region will be moderated by quality judgment of products originating in the linked country.

3.9.2 Perceived technological development

Numerous studies have demonstrated that country image affects consumer preferences when shopping for foreign products. Consumers form images of countries, which in turn influence their intention to buy products from a specific country (Costa et al. 2016; Mahran and Nasser 2014). Specifically, the findings from these studies indicate consumers are most willing to buy products from highly economically developed countries (Gorostidi-Martinez et al. 2017). For instance, consumers recognize that manufacturing high-quality products requires high-level technological abilities. Consequently, they perceive such products to be superior when produced in countries with proved capabilities and efficiency (Magnusson et al. 2011).

H9

Boycott spillover will be moderated by perceived technological development of the linked country.

4 Application

The EU countries are an important trading partner for Israel with total trade approximately €48.6 billion in 2017. Exports of goods and services from Israel to the EU countries reached €20 billion in 2017, while imports of goods and services from the EU were €28.6 billion (European Commission 2019). Total exports of goods made in ISt reached €470 million. Of this, more than €200 million were directed to the EU countries (Schwartzman and Hoffman 2015).

Israeli settlements are civilian communities inhabited by more than 450,000 Israeli citizens, who have been steadily moving to the West Bank and the Golan Heights since they came under Israeli control in 1967. The settlements are subject to extensive parts of Israeli law, in addition to special legislation by the region’s military commander, which applies solely to Israeli residents (Yehuda et al. 2014). That being the case, Israeli settlements receive different legal treatment than cities and towns in Israel proper.

Hundreds of companies operate in the Israeli settlements, ranging from small businesses serving the local population to large factories which export their products worldwide. The most common settlement products sold in Europe include agricultural products such as dates, citrus fruits, and herbs, and manufactured products including cosmetics, carbonation devices, plastics, textile products, and toys.

We selected the United Kingdom as a representative country of the EU. The UK is a TCN country which has been an active member of the EU since 1973 and comprises Israel’s largest trading partner in Europe. Trade between the two reached £3.9 billion in 2017 (Secretary of State for International Trade 2019).

5 Research design and methodology

5.1 Questionnaire

The research instrument utilized in this study was a four-part structured questionnaire. The first part investigated respondents’ shopping patterns for food products. It explores the likelihood of individuals being exposed to, and influenced by, COO messages on product labels as required by the European Commission. To minimize possible effects of the research tool on results, we began with general questions about exposure to product labels per se. We then asked progressively more detailed questions about prior refrainment from buying foreign-made products.

The second part of the questionnaire concerned the first stage of ITT: the introduction and valuation of the new information cues. As such, a brief explanation of the EU COO labelling decision is provided. Then respondents were asked about their prior purchase of products manufactured in the Israeli settlements, their attitudes towards buying foreign products, and how the new labelling regulations impacted their feelings and attitudes regarding products originating in Israeli settlements. To investigate H1 and H2, the respondents were asked to evaluate two information cues: consumer the perceived quality of products manufactured in Israeli settlements and the extent to which these products comprise economic threat on Palestinian economy. Measures related to the importance of a cue and perceptions based on that cue were collected as well.

The third and the fourth parts includes multi-item scales designed to evaluate respondents’ intention to boycott products made in ISt, whereas the fourth part focuses on the “spillover” phenomenon. In this section, we explore the likelihood of boycott expansion to linked economies and factors that would moderate the spillover effect.

Six constructs and three moderating variables were developed which comprise the theoretical framework of this study (see Fig. 1). The constructs and moderating variables were presented to the subjects in question form, which they answered using a 7-point Likert scale. Table 1 depicts the items’ loading and Cronbach's alpha for the model components; all of the Cronbach’s alphas in the study are between 0.7 and 0.94 and, therewith, range from acceptable to excellent.

Fig. 1
figure 1

Model testing. *p < 0.05; **p < 0.01

Full size image
Table 1 CFA—Items’ factor loading and constructs’ reliability and validity measures
Full size table

The data for this study were collected during April–May 2016. A sample of 580 respondents was obtained from grocery shoppers across the UK. To attain a representative sample of shoppers, study participants were recruited via an online panel. This task was administrated by the international market research service provider Qualtrics Panels.

5.2 Descriptive Statistics

The socio-demographic profile of the respondents is presented with respect to their gender, age group, educational background, and annual household income. Female respondents constitute 52.2% of the total sample. In terms of age, 30% of the respondents are between 18 and 34 years old, 33.8% between 35 and 54, and respondents aged 55 or older accounted for 36.2%. Almost 28.3% of the participants have a senior high school education, 40% have a college or vocational school educational background, and an additional 29.1% have a university degree. Additionally, 61.2% have an annual household income of £30,000 or less, and 27.2% between £30,001 and £60,000. The rest of the respondents (11.6%) earn over £60,000 a year.

5.3 Validity and reliability

All variable items were subjected to an Exploratory Factor Analysis (EFA). The extraction method used was a principle component analysis with Varimax rotation. EFA yielded nine factors that explain 70.66 percent of the model’s cumulative variance. All items were satisfactorily loaded on the scales they were designed to measure (> 0.60). To ensure the items used to operationalize the constructs of interest were internally consistent and free of measurement errors, reliability analysis was carried out using Cronbach’s alpha. Internal consistency reliability coefficients for all research constructs were higher than the recommended levels of 0.7

5.4 Functional measurement

To test the IIT integration hypothesis, the current study employed the averaging model which assumes that individuals form a quantitative response by weighing the values of the new information variables, putting these weighted values together, and striking a balance. Thus, we propose the following averaging model of the influence of product quality and perceived economic threat on consumers’ feelings towards products made in Israeli settlements:

$$Rij = (We\,E - Ti + Wq \, Qj/\left( {We + Wq} \right).$$

where where Rij is the observed response to the ij double; We and Wq are the importance weights associated with economic threat (E-T) and product quality (Q). Note that the averaging process that governs the information integration is a linear model, which means that the effects of information about one attribute (e.g., economic threat) on response are independent of the level of the second attribute (e.g., product quality). This also implies that the meaning and importance of an information cue about the EU labelling regulations is assumed to be the same, regardless of whether it is combined with other favorable or unfavorable information. Additionally, linear integration suggests that the interaction between the two attribute factors should be statistically nonsignificant in an ANOVA model (Anderson 1981).

To determine if respondents are in fact integrating information as proposed by information integration theory, an ANOVA test was conducted. The results revealed no interaction for quality (F = 1.64; p = 0.1) and for economic threat ratings (F = 1.32; p = 0.18), thus providing support for the linear averaging-type rule.

To quantify the extent to which the perceptions of economic threat (E–T) and quality judgement (Q) influenced consumers’ animosity perception regarding products sourced from Israeli settlements, coefficients of E–T and Q were determined by regressing the relative values of the latter two variables on perceived animosity. As expected, the results (R2 = 0.51, t = 7.1, p < 0.01) indicate a negative influence of product quality on animosity perception (β = -0.13, p < 0.01), whereas economic threat and animosity are positively correlated (β = 0.72, p < 0.01). Taken together, we can conclude that the animosity perception arises from a weighted evaluation of products originating from the Israeli settlements.

5.5 Hypothesis testing

To test the hypothesized research structure, a structural model was built including all the antecedents and relevant relationships, enabling us to examine the hypothesized causal paths among the constructs by performing a simultaneous test. Structural Equation Modelling (SEM) employing SmartPLS 3.0 software was used for the analysis. PLS-SEM produces accurate estimates, even when assuming a common factor model (Ringle et al. 2013), and allows researchers to analyze the measurement model simultaneously with a large number of constructs and structural model relationships, as well as to adopt research models with both moderating and mediating relationships (Becker et al. 2012; Henseler and Chin 2010).

Bootstrapping samples of 5,000 were retained. The model with the best fit was retained as the final model.

Results indicate an acceptable fit of the model. The measures of discrepancy are below their corresponding HI95 value (Henseler 2017). SRMR (Standardized Root Mean Square Residual) is 0.054 (HI95 = 0.058), which is lower than the acceptable value of 0.08. The geodesic discrepancy (dG) is 2.015 (HI95 = 2.357), and the unweighted least squares discrepancy (dULS) is 0.666 (HI95 = 0.796). Moreover, NFI (Normed Fit Index) is 0.858, indicating a satisfactory level (Latan et al. 2019).

To assess the model fit with the data, it is recommended that the p values for the average path coefficient (APC = 0.34, p < 0.001) be lower than 0.05. In addition, it is recommended that the average variance inflation factor (AVIF = 1.15) be lower than 5 (ideally ≤ 3.3), (Kock 2017). Convergent validity was evaluated to validate the measurement model through investigation composite reliability (CR) and average variance extracted (AVE). Table 2 illustrates that all CR and AVE values meet the recommended threshold values. CR values are recommended to exceed 0.70 and AVE values should be greater than 0.50 (Fornell and Larcker 1981). Discriminant validity was further verified by using the Hetero Trait-Mono Trait (HTMT) criterion. Table 3 shows that the HTMT ratio is less than 0.90 (Hair et al. 2016), and all individual values fall within the confidence interval.

Table 2 Composite reliability, VIF, AVE, and correlation of constructs’ values
Full size table
Table 3 Heterotrait-Monotrait Ratio (HTMT)
Full size table

Variance Inflation Factor (VIF) was evaluated to check for the existence of collinearity. VIF values of less than 5 indicate no collinearity. All VIF values meet the recommended threshold values (VIF < 5.0) (Hair et al. 2015). Consequently, based on the current findings, this study demonstrates that the data show good fit with the proposed model which in turn, exhibits adequate reliability, construct validity, and lack of collinearity.

6 Results and discussion

Figure 1 displays the path coefficients and squared multiple correlations (R2) for the endogenous constructs. The proposed conceptual model allowed a moderate amount of variance in both dependent variables: boycott intention of products made in ISt and in Israel (40% and 52% respectively). The great majority of the hypothetical relationships were supported by the empirical model results, thus providing strong evidence for the hypotheses drawn from the theoretical framework. This means our conceptual model was able to capture both research goals: investigating the impact of forced labelling on consumer intention to boycott products sourced from ISt; and evaluating the influence of such a boycott on the linked country (Israel).

H1 predicts a positive relationship between perceived economic threat and animosity. The findings of the study support the hypothesized relationship (β = 0.75, p < 0.01) which support H1. Moreover, perceived economic threat is found to have the highest impact on influencing consumers’ animosity feelings. H2 suggests that product quality judgment moderates the influence of economic threat on animosity. The study results provide statistical support for the proposed moderating influence (β = −0.19, p < 0.01). This implies that consumers are less likely to view foreign products as an economic threat as they perceive the ISt-made products to be of high quality.

H3 and H4 suggest that consumer animosity affects both boycott-participation intention and perceived punishment. The results demonstrate statistical support for the proposed relationships (H3: β = 0.42, p < 0.01; H4: β = 0.47, p < 0.01). This means that animosity, reflected in strong negative emotions toward purchasing products sourced from ISt, comprises a major factor to directly prompt consumers to consider participating in a boycott against ISt. Similarly, as might be expected, consumer animosity has a direct and positive influence on perceived punishment which implies that high animosity levels comprise a factor in prompting feelings of punishment.

H5 predicts a positive relationship between perceived punishment and boycott-participation intention of the targeted region. The results indicate statistical support for the proposed relationship (H5: β = 0.15, p < 0.05).

H6 suggests a positive relationship between consumer ethnocentrism (CET) and boycott intention. The results reveal that the proposed relationship received statistical support (H6: β = 0.24, p < 0.01).

Concerning the second research topic (“spillover effect”), H7 suggests that a boycott directed toward products made in the targeted region will spill over to products sourced in the related country. The study results point to a significant and positive relationship, with a beta coefficient value of β = 0.66 (p < 0.01). This means consumers who clearly intend to refrain from buying products from the targeted region (ISt) tend also to consider boycotting products made the related country (Israel). That is, when products from ISt are targeted, the negative emotions (animosity and perceived punishment) associated with boycotting-participation intention can easily expand to another linked one (Israel).

Finally, H8 and H9 address the influence of two moderating variables: product quality judgement and perceived technology development of the linked country, on boycott expanding to the linked country, claiming that high levels of these factors decrease the spillover effects. The PLS results found product perceived quality has significant bearing on the intention to stop buying products labelled as imported from the linked country (Israel), as this variable moderated the intensity of the spillover effect (H8: β = -0.11, p < 0.05). This implies that even strongly ethnocentric consumers are less likely to consider boycott participation as a form of animosity and punishment, as they perceive the Israeli-made products to be of high quality. However, the study results didn’t provide statistical support for the proposed moderating influence of perceived technology advancement on boycott extension to the linked country (Israel) (H9: β = −0.04, p > 0.05).

7 Conclusion

This study examines the influence of new COO regulations on consumer buying decisions of foreign products. In particular, the study investigates the extent to which the EU decision to add a new label on products made in Israeli settlements (ISt) would drive third-country national consumers (the case of the U.K) to consider boycotting products originating in Israeli settlements and Israel. Hence, the motivation to include information about product origin as stemming from ISt on the related product labels. Within this framework, the objective of the reported study has been to estimate the potential of communicating ISt production information using product labels.

Drawing upon the information integration theory (IIT), we employed a research model that integrated new information cues (i.e., perceived economic threat and product quality judgement) and tested their influence on consumers’ feelings (perceived animosity) towards products sourced from ISt. As predicted, the findings show that EU labelling regulation, after information cues are weighted and integrated, comprise a factor that stimulates negative emotions (animosity) towards products manufactured in disliked sources (the Israeli settlements). Furthermore, it seems that these negative emotions are extremely intense, as they promote a strong desire to consider boycotting actions as an effective act of punishment. This conclusion is consistent with findings reported in past studies investigating the empathy feelings-boycotting intention link in the context of third-country nationals’ (TCN) consumers. Findings from these studies point to TCN’s consumers who may consider punitive actions against perceived immoral actions even if they are not directly involved in the animosity case (Fehr and Fishbacher 2004; Nelissen and Zeelenberg 2009). For example, according to Halimi, D’Souza and Sullivan-Mort (2017) concerns about Palestinian and Lebanese citizens were evoked by consumers’ awareness of the Israeli-Palestinian animosity case which, in turn, activated their animosity feelings towards Israel. Similar findings were reported in a study by Hoffmann (2013), which investigated perceived hardship-negative emotions-boycott intention links. Taken together, negative-emotions’ factors, namely, animosity and punishment were found to have significant impact on influencing boycott intention.

The study results shed light on boycott expansion to related countries. Specifically, the study discusses spillover effects wherein certain Israeli companies become unintentional victims of boycotts due to perceptions about their nationality. When countries or economies are linked, as are Israel and the ISt, a boycotting campaign targeted at one may expand to the other. Given the EU labelling requirements we tested H7 by measuring the readiness of respondents to boycott ISt as well as Israel. With β = 0.66 (p < 0.01), and R2 = 0.52, it seems that a respondent’s degree of readiness to boycott ISt is sufficient for estimating their readiness to boycott Israel. This conclusion provides evidence that negative consumer emotions towards products manufactured in disliked sources, thereby considering boycotting these products, would expand to a related country, which in turn would drive consumers to consider boycotting products from the linked country as well.

Finally, little attention has been devoted to the moderating influence of product judgment and perceived technology development of the linked country on boycott spillover. Drawing on the findings of the current study, it can be argued that only product quality judgment is inhibiting factor as it decreases the intensity of the spillover effects. This finding is consistent with Hoffmann (2011) who claimed that inhibitors provide reasons for avoiding boycotting.

The above indicates that forced labelling regulations prompt consumers to consider participating in a boycott directed against products made in disliked sources. This, in turn, can expand to other related economies. In addition, the study clearly revealed that incorporating IIT into the current study provides better understanding of the boycotting phenomenon in the context of international marketing studies.

8 Implications

8.1 Theoretical implications

This study contributes to the international marketing and retailing literature in several ways. First, the study serves as a pioneering effort to incorporate IIT into empirical marketing studies about forming and changing consumer attitudes, and the role these modified attitudes play in exploring consumer behavior. Second, the proposed theoretical framework identifies determinants of boycott participation and describes how each one affects consumer buying decisions. Third, the study provides new insights into the existing body of literature on consumer negative emotions and explains how these factors promote consumer intention to participate in the boycott. Finally, our study is the first to explicitly incorporate model key factors from four research approaches that differ with respect to their incentives, but also complement each other. The first is the COO labelling regulations found in some previous studies to better explain how consumers make decisions regarding purchasing foreign products. This is an important extension of the existing research, given that the literature on international marketing has paid little attention to the role played by new labelling initiatives on consumer buying decisions. The second is the boycott-participation literature which examines negative emotions that prompt consumers to consider participating in a boycott. The third is the “spillover effect” where a call for boycotting a targeted region or country expands to other associated countries. In this regard, the present study serves as a pioneering effort to incorporate a conceptual framework that deals with the new dimension.

8.2 Managerial implications

For managers, boycott participation has significant impact on exporting and retailing companies. Boycotts do carry the potential to significantly affect the targeted country’s international trade and reduce sales for individual firms (Chavis and Leslie 2009; Heilmann 2016). Consequently, boycotts may have critical effects on firms’ market shares and profitability. As such, marketers are required to handle any boycott promoter and trigger with more care and sensitivity. Marketing managers should, therefore, develop effective strategies to overcome the complex consequences of boycott antecedents in foreign markets. For example, managers can differentiate their offerings thereby emphasizing product quality due to advanced technology, and by using cues valued and trusted by consumers (e.g., manufacture in a country with a favorable product image). Furthermore, firms would consider manufacturing outside the ISt, for instance, in Israel. Firms also can consider moving part of their manufacturing operations to foreign markets. By doing so, they can conceal their ISt identity, thus reducing the boycott’s damage. However, such steps would increase business costs and may lead to a drop in demand from consumers looking for familiar products (Mehta and Bhanja 2018).

From yet another angle, the study findings pose major strategic challenges for retail executives. To start with, retailers should be attentive to customers who are particularly interested in boycotting products made in ISt. These customers require retail managers to stop sourcing products from the targeted countries, otherwise they may consider boycotting actions against the stores who continue to stock boycotted products. Moreover, retail managers interested in securing significant market share and profitability need to find alternative products to avoid boycotting customers actually switching to competitors.

8.3 Implications for public policy makers

Organizations that consider promoting the boycott of a certain target, as well as those that seek to protect their markets from such initiatives can benefit from the new knowledge presented in this study. The recent EU labelling regulations are political in nature. Such initiatives can drive consumers to show hostility towards products manufactured in specific places. Regardless of what companies do, they often cannot overcome the overwhelming negative impact of consumer feelings of animosity and punishment. Therefore, public policy makers must understand the motives behind boycotting specific targets and the reasons driving consumers in foreign markets to engage in boycotting behaviors, thereby monitoring consumer attitudes toward their company’s home country. Another important implication derives from the spillover effects of a particular boycott on associated countries. Though the results of the study highlight the meaningful moderating influence of both quality judgment and the linked country’s technological image on boycott participation, the knowledge that these are barriers to boycott expansion can serve public policymakers in designing successful marketing strategies that take country image characteristics into consideration.

9 Limitations and future research

While this study increases our understanding of the role of new labelling initiatives in buying foreign products, it is subject to some limitations related to the generalization of the findings; however, these limitations, in turn can indicate avenues for future research. First, it includes limited empirical testing of the conceptual model and hypothesized effects in a single setting (the UK), which might affect generalizability of the results. This limitation can be remedied by future studies with the inclusion of additional EU and other countries. Second, the proposed conceptual framework and empirical findings indicate the importance of selected inhibiting factors (i.e. quality judgment and technological image) on boycott participation. However, these factors would be ineffective in moderating the influence of animosity and punishment on consumer boycott intention, especially when intense feelings of hatred exist, and this must be explored further. Third, the current research addresses a new research topic, that of the ‘spillover effect’. Future studies should address this further. Finally, this research concentrates on boycott-participation intention in the context of shopping for grocery products. Our approach, however, can readily be extended to studying consumer buying decisions in other product categories and retail industries.