Purchase Unwillingness and Willingness of Indonesian Consumers Towards Israeli Products

This research aimed to measure factors influencing purchase willingness and purchase unwillingness on Israeli products. Four predictor variables including consumer animosity, product judgment, boycott participation, and boycott motivation were used. Data were collected by an online survey, and it attracted 337 participants. Three stages of data analysis were applied, those were exploratory analyses, confirmatory factor analyses, and Structural Equation Model (SEM). In total, there were seven hypotheses tested. This research finds a significant impact of animosity on product judgment, boycott participation, and boycott motivation. Furthermore, product judgment and boycott participation significantly affect purchase willingness. Meanwhile, boycott participation and boycott motivation significantly affect purchase unwillingness.


INTRODUCTION
In March 2016, the Organisation of Islamic Cooperation (OIC) held a summit in Indonesia. One of the outcomes of the summit was a declaration to boycott Israeli products. This movement was intended to exert pressure on Israel that had occupied Palestinian territory and as the support to Palestine as a member of the OIC. Consumer boycott occurs not only in developing countries, such as Indonesia but also in developed countries.
In developing countries, a boycott was mostly caused by religious and political triggers. Meanwhile, in developed countries, a boycott was mainly caused by economic triggers (Al Serhan & Boukrami, 2015). Furthermore, Seegebarth et al. (2011) distinguished consumers based on their boycott intention, namely the self-centred sceptics, the ambitious activists, the concerned waverer, and the mindless follower.
Many researchers have gleaned factors influencing consumer boycott participation and purchase willingness. However, there are some parts of this field of study that has lack of attention, particularly on the impact of boycott participation on purchase willingness and purchase unwillingness and impact of boycott motivation on purchase unwillingness. Therefore, this research aims to measure the impact of consumer animosity on product judgment, boycott motivation, and boycott participation, and its impacts on purchase willingness and unwillingness for Indonesian consumers towards Israeli products.
The theoretical framework ( Figure 1) partly is developed based on the research discussed. The framework has seven hypotheses and consists of four predictor variables such as consumer animosity, product judgment, boycott motivation, and boycott participation. In this case, customer animosity is linked to product judgment, boycott participation, and boycott motivation. In addition, product judgment is linked to purchase willingness. Meanwhile, boycott participation is linked to purchase willingness, and purchase unwillingness. In addition, boycott motivation is linked to purchase unwillingness.
Animosity is defined as an attitude of resentment and the expression of anger (Windom, 2012). Animosity has an effect on country image, ethnocentrism, product beliefs, and receptivity (De Nisco, Mainolfi, Marino, & Napolitano, 2013); product judgment, purchase intention, and ethnocentrism (Giang & Khoi, 2015); and boycott participation (Albayati, Mat, Musaibah, Aldhaafri, & Almatari, 2012;Smith & Li, 2010). In this research, customer animosity will be used as a predictor of product judgment and boycott participation. Giang and Khoi (2015) explained that animosity could influence product judgment negatively. Their research was to measure the intention of Vietnamese consumers to purchase Chinese household appliances. Although they demonstrated how customer animosity significantly influenced product judgment, some other researches showed the insignificance. For example, taking place in Saudi Arabia, Abosag and Farah (2014) examined the impact of animosity towards Danish products on the corporate brand image, customer loyalty, and product judgment. In this research, customer animosity was presented by religious motivation. They considered that religious animosity would stay longer in a person's heart comparing to other types of animosity. As a result, religious animosity had a significant impact on corporate brand image and customer loyalty. On the other hand, this animosity had an insignificant impact on product judgment. Moreover, Rose, Rose, and Shoham (2009) investigated the unwillingness of Arab Israelis to purchase British and Italian branded products. They reported the insignificant impact of animosity on product judgment of British and Italian branded products.
Furthermore, other researchers test the impact of customer animosity on boycott participation (Albayati et al., 2012;Smith & Li, 2010;Suhud, 2016). Smith and Li (2010) mentioned that customer animosity had a significant effect on boycott participation. Moreover, Albayati et al. (2012) measured the influence of boycott participation towards Danish products by Muslim consumers in Malaysia. They included selfefficacy, product, judgment, and customer animosity as the predictor variables. They also reported two significant findings, the effect of self-efficacy on product judgment and the effect of product judgment on boycott participation. They also found that customer animosity had an insignificant impact on boycott participation.
Another research was conducted by Suhud (2016) by examining the factors that influenced boycott participation against Israeli products in Indonesia. It was found that animosity influenced product judgment negatively, and affected boycott participation and boycott motivation positively. Besides that, boycott motivation influenced boycott participation positively.
H 4 = Product judgment affects purchase willingness positively and significantly Individuals, groups, or institutions are involved in various boycotts for some reasons. Companies wonder to understand what factors influence persons to commit in boycotting. Friedman (1985) defined boycott participation as an attempt by one or more parties to achieve certain objectives by urging individual consumers to refrain from making selected purchases in the marketplace.
Then, Klein, Smith, and John (2004) looked at four factors that predicted boycott participation. It included the desire to make a difference, the scope for self-enhancement, counterarguments that inhibited boycotting, and the cost to the boycotter of constrained consumption. Furthermore, Smith and Li (2010) measured boycott participation of Chinese consumers towards Japanese branded products by including animosity, efficacy and prior purchase. They found that these three factors could support consumers to be engaged in boycott activities.
In general, boycott participation can be influenced by customer animosity (Albayati et al., 2012;Smith & Li, 2010), corporate reputation and trust (Hoffmann & Müller, 2009), attitude, perceived control, and brand distrust (Chiu, 2016), involvement with a boycott cause, commitment to the brand to be boycotted, credibility of a call to participate in a boycott, and perceived success likelihood of a boycott (Albrecht, Campbell, & Heinrich, 2013), and brand credibility (Fazel, 2015). On the other hand, boycott participation can influence purchase willingness (Shah & Ibrahim, 2016). In this research, boycott participation is linked to purchase willingness and purchase unwillingness. However, there is a paucity of research that examines the impact of boycott participation on purchase unwillingness and purchase unwillingness.
Moreover, Shah and Ibrahim (2016) conceptualised the purchase willingness of Malaysian consumers towards foreign products by employing boycott participation, consumer animosity, consumer patriotism, and product judgment as predictor variables. To support the idea of the impact of boycott participation on purchase unwillingness, the researcher uses the concept of Shah and Ibrahim (2016). Hence, the other hypotheses are as follows.
Other researchers test the impact of boycott motivation on other variables, such as boycott participation and purchase intention (Suhud, 2016;Tian, 2010). In this research, boycott motivation is linked to purchase unwillingness. Meanwhile, Tian (2010) showed a negative impact of boycott motivation on purchase intention. There is a paucity of research that tests the impact of boycott motivation on purchase unwillingness. However, learning from the research conducted by Klein, Smith, and John (2004) and Smith and Li (2010), it shows a positive impact of boycott participation on purchase unwillingness, therefore, this is hypothesized as follows.
H 7 = Boycott motivation affects purchase unwillingness positively and significantly

METHODS
Respondents are approached by using a personal communication through an email and direct message of social media platforms. They are asked whether they have a willingness to participate in an online survey. Furthermore, to those who are willing, a link to the survey is distributed.
As a part of validation, indicators to measure each variable are selected from previous researchers. To measure consumer animosity, four indicators from Nakos and Hajidimitriou (2007) are considered and adapted. From the Nakos and Hajidimitriou (2007) and Nijssen, Douglas, and Bressers (1999), the researcher adapts indicators to measure product judgment. In addition, from Tian (2010), the researcher takes five indicators to measure boycott motivation. Furthermore, indicators from Klein et al. (2004) are taken and adapted to measure intention of boycott participation.

RESULTS AND DISCUSSIONS
There are 146 respondents who complete the instrument. This number is 43% of the total respondents who fill out the online survey. Some respondents give direct feedback to the researcher that they stop filling the instrument. They mention that the topic of this research is disturbing them. When respondents are asked whether they agree or disagree with the boycott movement towards Israel products, more than 100 respondents agree. However, 84 respondents do not support the idea that Israel should leave the Palestinian territory.
82 respondents claim that they do not follow the news related to Israeli product boycott and 97 respondents state that they can recognize Israeli products in the market. In terms of boycott experience, predominant respondents (123) state that they have an experience in boycotting previously. From 146 respondents, there are 61 males (41,8%) and 85 (58,2%) females. Regarding their occupation, 62 of respondents are still in schools or higher education institutions. 60 of them are employed. Additionally, 114 (78,1%) respondents are single, and 29 (19,9%) are married. The religion of respondents is Islam (120), Christian (15), Catholic (7), Buddhism (3), and other (1). Predominant respondents hold a higher school degree (55) and are followed by bachelor degree (53). Their ages are in the range of 17-20 (33), 21-24 (54), 25-30 (20) and the rests are older than 30.
For exploratory factor analysis, there are four indicators of customer animosity with a Cronbach's alpha score of 0,922. It is considered as reliable (Hair Jr., Black, Babin, Anderson, & Tatham, 2006). The factor ranges from 0,887 to 0,907. The result is in Table 1. Moreover, five indicators measure product judgement in exploratory factor analysis. This variable has a Cronbach's alpha score of 0,897. It is considered to be reliable (Hair Jr. et al., 2006). It can be seen in Table 2. Next, 11 indicators are picked to measure boycott participation. Based on the calculation of exploratory factor analysis, two dimensions are formed. The first dimension is 'promote change' consisting of eight indicators with a Cronbach score of 0,955. The second dimension is 'I will feel better about myself' with a Cronbach's alpha of 0,679. It can be seen in Table 3. In exploratory factor analysis, five indicators are used to measure boycott motivation. It develops two dimensions. The first dimension is product motivation consisting of three indicators with a Cronbach's alpha of 0,892. The second dimension is policy motivation with two indicators and a Cronbach's alpha score of 0,723. Table 4 shows the results. In exploratory factor analysis, purchase willingness develops two dimensions. The two dimensions tend to be opposite. As presented in the theoretical framework, these two dimensions are treated as two different variables. There are willingness to purchase with a Cronbach's alpha score of 0,661 and unwillingness to purchase with a Cronbach's alpha score of 0,808. It can be seen in Table 5. During calculation of structural model for hypotheses test, 'promote change' dimension, one of the two dimensions of boycott participation is dropped. Therefore, boycott participation is represented by 'feel better' dimension. Figure 2 achieves a fitness with a probability score of 0,645, CMIN/DF score of 0,912, CFI score of 1,000, and RMSEA score of 0,000. Table 6 presents the summary of hypotheses testing results. All paths tested have a greater C.R. score than 1,96 that show significance. Three of seven paths have a negative direction including the impact of customer animosity on product judgment and boycott participation (feel better), and the impact of boycott participation (feel better) on purchase unwillingness. The other four paths have a positive direction.
Considering that a dimension of boycott participation is left during hypotheses testing, the researcher builds another model with boycott participation 'promote change' as the dropped dimension. This fitted model has a probability score of 0,189, CMIN/DF score of 1,179, CFI score of 0,994, and RMSEA score of 0,035. Figure 3 shows the structural model.
As presented in Table 7, all indicators of promote change tend to be negative. On purchase willingness, promote change gives a negative effect whereas purchase unwillingness gives a positive impact. Furthermore, as a dependent variable, promote change is positively affected by animosity. In general, all paths have a greater C.R. score than 1,96.    All paths in the two models tested have a greater C.R. score than 1,96. It indicates a significance. The first hypothesis predicts the impact of customer animosity on product judgment. According to Giang and Khoi (2015), customer animosity has a negative impact on product judgment. Based on the calculation, this path has a C.R. score of -3,418. Therefore, H 1 is accepted. At the same time, this finding is against the finding by Abosag and Farah (2014) and Rose et al. (2009).
The second hypothesis predicts the impact of customer animosity on boycott participation. This path obtains a C.R. score of -4,520 in the first model. Therefore, H 2 is accepted. This finding supports the previous research conducted by Smith and Li (2010). On the other hand, in the second model, the path has a C.R. score of 14,528. These different results are rational. The different directions of impact between these two models are caused by the different contents of their indicators. The indicators of 'I will feel better about myself' are pro-boycott (the first model) and indicators of 'promote change' are anti-boycott (the second model).
The third hypothesis predicts the impact of customer animosity on boycott motivation. In the first model, this path has a C.R. score of 10,973 and 10,921 in the second model. Therefore, H 3 is accepted. This finding is significant with Suhud (2016).
Moreover, the fourth hypothesis predicts the effect of product judgment on purchase willingness. In the first model, the path has a C.R. score of 3,131, and in the second path, it has a C.R. score of 2,738. These findings are significant with the research by Čičić, Brkić, and Prašo-Krupalija (2003); Mrad, Mangleburg, and Mullen (2014); Shah and Ibrahim (2016).
The fifth hypothesis predicts the impact of boycott participation on purchase willingness. In the first model, the path has a C.R. score of -2,037, and in the second model, it has a C.R. score of -2,550. These findings are in line with the research by Shah and Ibrahim (2016). The higher the intention to be involved in boycott participation is, the less the respondents' intention to purchase Israeli products will be.
The sixth hypothesis predicts the influence of boycott participation on purchase unwillingness. In the first model, the C.R. score is 3,506, and in the second model, it is 2,570. Therefore, both hypotheses are accepted. The findings support the research by Shah and Ibrahim (2016).
The last hypothesis predicts the impact of boycott motivation on purchase unwillingness. In the first model, this path has a C.R. score of 10,874 whereas, in the second model, it has a C.R. score of 2,586. As discussed, there is a paucity of testing regarding the impact of boycott motivation on purchase unwillingness. However, the researcher refers to research conducted by Klein et al. (2004) and Smith and Li (2010) that show impact of boycott participation on purchase unwillingness.

CONCLUSIONS
This research aims to examine factors influencing purchase willingness of Indonesian consumers towards Israeli products. The testing includes consumer animosity, product judgment, boycott motivation, and boycott participation. Due to significance, in this testing, boycott participant is represented by 'I will feel better about myself' dimension. This research finds that customer animosity has a negative impact on product judgment (H 1 ) and boycott participation (H 2 ). Meanwhile, on boycott motivation, it has a positive impact (H 3 ). Furthermore, product judgment and boycott participation have a positive impact on purchase willingness (H 4 and H 6 respectively). In addition, boycott participation has a negative impact on purchase unwillingness (H 5 ), and boycott motivation has a positive impact on purchase unwillingness (H 7 ).
As the dimension of 'promote change', another model is tested with this dimension to represent boycott participation variable. As a result, customer animosity positively affects boycott participation (H 2 ) and boycott motivation (H 3 ). Whereas, on product judgment, it has a negative effect (H 1 ). Besides that, product judgment positively affects purchase willingness (H 4 ). Furthermore, boycott participation negatively affects purchase willingness (H 5 ) whereas it positively affects purchase unwillingness (H 6 ). Last, boycott motivation positively affects purchase unwillingness (H 7 ).
A boycott of Israeli products apparently is a sensitive issue. Although predominant respondents agree to the idea of boycotting Israel, about 60% respondents stop completing the instrument. This feedback suggests that the researcher and future research in this topic select the respondents. Muslims and those who connect to certain Islamic communities or organisations can be a good selection. Another recommendation for future research is to treat dimensions with contrast indicators as variables and tested separately.
Consumer animosity is a serious issue in business. It influences product judgment, boycott participation, and boycott motivation. In this case, Israel businesses can be a victim of foreign policy of Israeli Government. One thing that the businesses can do is to make their products unidentified by consumers as products made in Israel.