The Impact of Social Influence, Product Knowledge, and Fear of Missing Out (FOMO) towards Purchase Intention on Alcoholic Beverage in Bali

Based on the enactment of official government regulations in Governor Regulation No


INTRODUCTION
The enactment of official government regulations in Governor Regulation No. 1 of 2020 concerning the Governance of Balinese Fermented and Distilled Drinks stipulates the legality of arak Bali. It can be a prospective industry to improve the community's economic welfare by promoting it as a heritage-based product. Inadvertently, it can invite the capital owners to jump into the legalization of this traditional alcoholic beverage (Jessica, 2021). Arak Bali farmers work with the company whose purpose is to export it, such as farmers in Karangasem Regency with PT Niki Sake and farmers in Bondalem Village, Buleleng with PT Lovina (Jessica, 2021). In a research conducted by Wirastini (2015), in the past, the average income of arak Bali farmers only reached Rp1.259.615,00. However, today the production can reach 12 liters per day with the price of Rp35.000,00 per liter, and the amount of profit gained is Rp420.000,00. Thus, the arak Bali farmers can reap a profit of around Rp12.000.000,00 in 30 days (Jessica, 2021).
Arak Bali is a traditional Balinese alcohol made from the fermentation and distillation of Nira coconut. The distillation process is done traditionally, and arak Bali can contain more than 60% alcohol (Jessica, 2021). Nowadays, the Balinese people continue to expand this traditional alcoholic drink, as seen in the emergence of Micro, Small, and Medium Enterprises (MSMEs). Most enterprises offer arak Bali as the main product for their business. According to Nasution (2018), the culinary business tends to develop its menu to attract consumers back. Similarly, based on Febrian and Fadly (2021), a company must adapt to habits or conditions in each area to affect consumers' purchase intention.
Arak Bali is also used as the basic ingredient to be mixed with other flavors, such as processed fruits (oranges, salak, pineapple, mango, and others), vegetables (cucumber), flowers (champaka), and spices. Balinese people from all walks of life are accustomed to seeing or carrying out the habit of consuming arak Bali. For example, teenagers in Bali are also very familiar with this tradition in society, namely mearakan (Ardyanti & Tobing, 2017).
One factor influencing teenagers to consume arak is a group reference, such as friends. Adolescents follow the wishes or expectations of their friends solely to avoid punishment, such as the fear of being said to be out of touch or excluded by their friends. At the same time, the influence of information is due to the evidence and information given by their friends (Ardyanti & Tobing, 2017). However, consuming arak Bali in the proper dose can provide good benefits, such as a more relaxed body condition and a calmer mind so that communication becomes smooth with other people (Faizah, 2022). In addition, Balinese people who live in highland areas tend to consume arak Bali before going to work and bed at night to warm their bodies since the air in the highlands is cooler (Dwiantari & Sudiana, 2019).
Balinese people consume arak Bali to weave a social connection to be more valued and received in the environment. This situation is also related to the current term called Fear of Missing Out (FoMO). It is a condition in which a person experiences anxiety after seeing or checking their social media or the excitement done by their colleagues (Przybylski, Murayama, DeHaan, & Gladwell, 2013). With the Balinese people often publishing their mearakan activity, those who are unfamiliar with this activity will start to feel anxious and immediately want to do the same with their closest person. The existence of social influences and product knowledge gained from the experience of consuming arak Bali and FOMO can be the opportunity to influence the purchase intention of Balinese people.
The fact that consuming arak Bali has an essential effect on social engagement in the Balinese community leaves a question on how far the term mearakan can determine someone's social life. However, there is no clear information about the consumer behavior of Balinese people towards arak Bali consumption derived from the previous explanation. Based on the phenomena and the problem in the field, the research aims to examine the effects of social influence, product knowledge, and FOMO on the purchase intention of arak Bali. The research objectives are to determine the effect of social influence and product knowledge on purchase intention and FOMO and the effect of FOMO on purchase intention. The novelty of the research examines a model with a combination of social influence and product knowledge with FOMO as a mediator on purchase intention and arak Bali as object research. The research results make it possible to determine how much influence each factor has on purchase intention directly or indirectly. The research is expected to obtain results that all relationships between variables have a significant effect. Moreover, the research findings will be useful as the model tests the findings regarding social influence, product knowledge, FOMO, and purchase intention. In the marketing practice for arak Bali, they are expected to take advantage of social influence and product knowledge and raise FOMO to increase the purchase intention.
Purchase intention is the motivation to purchase a product with specific characteristics or brands as consideration (Wibisurya, 2018). Based on research conducted by Veronica and Rodhiah (2021), social influence has a positive and significant effect on purchase intention. It is concerned with consumers who have shopped and will be bound and close to a brand. So, there is an emotional connection that can improve consumer behavior in shopping. According to Jacob and Tan (2021), social influence positively and significantly affects purchase intention. However, another previous research finds that social influence has no effect on perceived quality and purchase intention (Teo, Leng, & Phua, 2019). It is found that respondents may focus their attention more on images than text on Instagram. Unlike Facebook, the text is more concerned. So, the influence of social influence may be limited. Hence, the hypothesis is formulated as follows.
H1: Social influence has a positive and significant effect on purchase intention.
Based on research conducted by Ridwan, Solihat, and Trijumansyah (2018), product knowledge has a positive and significant effect on purchase intention. It happens because, according to respondents, product knowledge is important in generating consumers' buying interest. Before consumers determine what product to buy, consumers must know the information or knowledge about the product, function, benefits, or quality. According to Sanita, Kusniawati, and Lestari (2019), attributes of product knowledge on purchase intention provide many developments. Consumers with high or more knowledge will affect consumers' perception and how the knowledge is used in decisionmaking (Widyasari & Haryanto, 2021). However, based on Erida and Rangkuti (2017), product knowledge had no significant effect on purchase intention. Consumers do not have sufficient information according to the type, class, usage situation, maintenance, and problem solution obtained directly from product manufacturers and retailers. Moreover, the discovery and process of obtaining information related to online shopping on the marketplace turn out not only to consumers' personal consumption but is also distributed back to people around, especially promotion information and attractive offers (Christy, 2022). So, the following hypothesis is formulated.
H2: Product knowledge has a positive and significant effect on purchase intention.
According to Carolina and Mahestu (2020), following the development of environmental associations is an important key in forming the selfconcept of adolescents who tend to be FOMO. So, it becomes difficult for them to refuse even though it is dangerous. Adolescents who tend to be FOMO consider this identity a very important part of their lives because they can highlight their uniqueness from others, get them included, and be part of a certain community. Similarly, according to Dewi, Hambali, and Wahyuni (2022), the peer environment provides an important role in individual behavior and development. So, the next hypothesis is formulated as follows.
H3: Social influence has a positive and significant effect on FOMO.
According to Saavedra and Bautista Jr. (2020), products, such as the masstige brand, can potentially influence consumers to share experiences of owning or using clothes. Marketers can trigger FOMO through a sense of exclusivity. Getting opportunities that only a few people have can encourage an increase in one's ego. For example, a masstige clothing brand can create exclusive limited edition products while showing the amount of stock remaining in real time. Meanwhile, it is found by Valdiani and Puspanidra (2020) that there is a message uploaded by Herways_id, which is a content about testimonials. This testimonial message is also still included in the FOMO strategy. Testimonials usually contain positive responses from people who have used Herways_id products. So, the following hypothesis is formulated.
H4: Product knowledge has a positive and significant effect on FOMO.
Based on Good and Hyman (2021), FOMO can affect consumers' purchase intention. In particular, such appeals can strengthen purchase intentions by increasing consumer-anticipated excitement and self-improvement. Moreover, FOMO affects buying interest indirectly through downstream consumers' desires. If the desire of consumers who are influenced by FOMO is strong, it will also have an impact on increasing consumers' buying interest (Van Parijs, 2021). Similarly, according to Yoga, Sistadyani, Fatricia, Yuliant, and Basmantra (2022), a higher level of FoMO will positively increase the purchase intention on Spotify. These results are in accordance with previous research conducted by Good and Hyman (2020) that buying interest is most likely due to FOMO. FOMO-laden appeals can increase purchase intention when individuals anticipate positive outcomes. Therefore, marketers should create advertisements that can attract the audience to buy the product and seal more deals in a limited time. The last hypothesis is as follows.
H5: FOMO has a positive and significant effect on purchase intention.

METHODS
The population in the research is male from millennial and Generation Z at age of 21 to 41 years (based on the generational classification quoted from beresfordresearch.com). They reside in Denpasar city with additional criteria, such as being older than 21 years old, looking for information, and having ever consumed arak Bali. So, it is not clear how many populations of the millennial and Generation Z in Denpasar City fit the criteria mentioned. Using the SEM model, the proper sample size is between 100 to 200 samples. According to Ferdinand (2014), the guideline for sample size depends on the number of indicators times 5 to 10. In the research, the number of samples used is the number of indicators × 6. It is 19×6 = 114 samples.
The data collection technique applies an online survey with a survey platform, Google Forms. Then, it is distributed using social media, such as WhatsApp (personal contact and groups) and Instagram. The questionnaires in the research are given to all populations mentioned before.
The variable is measured by using the indicators from previous research. The indicators of social influence are from Sangadji and Sopiah (2013) and Eryadi and Yulianna (2016). The indicators from Peter and Olson (2010) and Ridwan et al. (2018) explain product knowledge. For FOMO, the indicators are from Przybylski et al. (2013) and Good and Hyman (2021). Last, purchase intention uses a combination indicator from Eryadi and Yulianna (2016) and Resmawa (2017) to measure the variable deeper.
In addition, the research uses a semantic differential scale, ranging from 1 (strongly disagree) to 10 (strongly agree). It evaluates the effects of social influence, product knowledge, FOMO, and purchase intention. The scale of 1-10 (even scale) aims to avoid answers from respondents who tend to choose the middle. Then, the data are analyzed with Structural Equation Model (SEM) with AMOS version 24.

RESULTS AND DISCUSSIONS
The researchers accept 153 respondents who have filled in the questionnaire with three more additional questions or background questions, such as age (above 21 years old), last education, and occupation. Besides background question characteristics, respondents also answer 19 statements. Table 1 (see  Appendices)shows the amount and percentage of 153 respondents' characteristics.
Based on convergent validity with the loading factor approach in Table 2 (see Appendices), all indicators are found to have a loading factor of more than 0,70. The indicators can be classified as having met the convergence validity. The index can be considered valid because they have a loading factor of more than 0,70. So, discriminant validation has been satisfied and can be continued further.
Composite Reliability (CR) results are obtained from all indicators with composite reliability of more than 0,70. The results imply that the indicators are reliable. Meanwhile, the Average Variances Extracted (AVE) results are obtained from all indices with AVE above 0,50. It fulfills the requirements to explain the use of the indicators. Figure 1 shows the SEM model that has been analyzed using the software AMOS 24 and the outputs, such as the loading factor and Goodness of Fit results. Next, Table 3 (see Appendices) shows the Goodness of Fit results from the AMOS 24 program analysis results. Some conditions must be fulfilled first to proceed with hypothesis testing between variables.
Table 3 (see Appendices) indicates that the SEM model cannot be declared fit because it has not met the Goodness of Fit criteria according to the existing cut-off value. Thus, it is necessary to modify the SEM model with steps to correlate between errors based on modification indices with the covariances table of the output results that have been calculated from the AMOS program. Table 4 (see Appendices) shows the modification indices performed on some multiple errors.
After the modification indices are carried out, there is a change in the Goodness of Fit according to Table 5 (see Appendices). It shows the SEM model that has been fit. The GFI, CFI, and TLI criteria have approached the minimum limit. Then, Chi-Square, CMIN/DF, and RMSEA meet the existing cut-off value criteria. Since three criteria meet the requirements of the Goodness of Fit, the next step is hypothesis testing.
After testing the influence between variables, various results have emerged for each hypothesis studied. Table 6 (see Appendices) presents the results of the test of the influence between variables. H1, H2, H3, and H4 imply that each has a significant effect. Meanwhile, in H5, the indicator has no significant effect. Furthermore, the researchers provide further explanations regarding each hypothesis result and support it with the previous research findings.
In H1, it tests the effect of social influence on purchase intention. Table 6 (see Appendices) shows a structural model path that explains the relationship between social influence and purchase intention. It has a significant effect between social influence and purchase intention (t = 1,980 > 1,96) and a significant score (0,048 < 0,05). Therefore, H1 is accepted. The results imply that social influence is one of the factors Figure 1 Result of SEM Analysis with AMOS 24 that can significantly influence the purchase intention of the millennials and Generation Z in Denpasar city. Being local Balinese people is very closely related to their way of life with their social environment in obtaining or generating a sense of interest in arak Bali.
The results of H1 regarding social influence on purchase intention are in line with the findings by Jacob and Tan (2021). It is proven that social influence has a positive and significant effect on purchase intention. Consumers who have shopped and are bound and close to a brand will have an emotional connection that can improve consumer behavior in shopping (Veronica & Rodhiah, 2021). Previous research reveals that social influence in the particular referenced group influences someone's attitude and behavior directly and indirectly (Najib et al., 2022). The influence of social role models such as trust in celebgram and social interaction on ownership jealousy needs to be considered in the influencer marketing on social media to increase consumers' buying interest (Purwanto, 2021). Respondents focus their attention on images rather than text on Instagram. Compared to Facebook, the text is more concerned, so the influence of social influence may be limited (Purwanto, 2021). According to Sambe and Haryanto (2021), consumers believe that buying a virtual item can help them to find new friends and build or even maintain a social relationships between consumers.
In H2, it is about the effect of product knowledge on purchase intention. The findings of the structural path show that there is a significant effect between product knowledge on purchase intention (t = 4,801 > 1,96) with a significant value (0,00 < 0,05). Followed by the result of the analysis, H2 is accepted. The researchers assume that the knowledge spread among the Balinese people, especially the millennials and Generation Z in Denpasar City, aims to obtain benefits according to what is happening in the society, such as gaining recognition from the social environment, gaining calmness and cheerfulness, and obtaining benefits like experiences that can be shared with others.
The hypothesis test results regarding product knowledge on purchase. Intention are similar to the research conducted by Rismawan and Purnami (2017). It proves that product knowledge has a significant positive effect on purchase intention. It means that the higher or better the product knowledge a person owns, the higher the purchase intention will be. It is supported by the fact obtained from the respondents that knowledge of the product to be purchased is important in determining their buying interest because before they decide which product to buy, they must know knowledge about the product itself, its function, benefits, or the good quality of the product (Ridwan et al., 2018). Moreover, another previous research shows a positive effect of product knowledge on purchase intention with a moderate level of connection (Sanita et al., 2019). It shows that product knowledge of purchase intention attributes to the innovation-making of the product. In other words, the more product knowledge the consumers own, the more the purchase intention will be. In addition to previous research, millennials have the potential to purchase a product after receiving information or experiences from others (Sujarwo & Indriani, 2022).
For H3, the findings of the structural path indicate that there is a significant influence between social influence and FOMO (t = 2,926 > 1,96) with a significant value (0,003 < 0,05). These data show that H3 is accepted. The analysis results indicate that the millennials and Generation Z in Denpasar City are indicated to experience FOMO phenomenon that someone does not want to feel left behind regarding information or experiences from their social environment or relationships, group references, and role models. Both generations are concerned with the social influence factor, which significantly affects FOMO in arak Bali.
The results of the hypothesis test of social influence on FOMO are in line with research conducted by Carolina and Mahestu (2020). Following the development of environmental associations is an important key in forming the self-concept of adolescents who tend to be FOMO. It becomes difficult for teenagers to say no for the sake of carrying out their existence, even though it is dangerous. Adolescents who tend to be afraid of missing out believe in such identification as an essential part of their lives since they can highlight their exclusiveness by not making them excluded and being part of a particular community (Carolina & Mahestu, 2020). Reflecting on the research results by Dewi et al. (2022), the social environment also affects FOMO. The stronger the influence of the social environment is, the higher it will be too for FOMO. It shows that the peer environment plays an essential role in individual behavior and development.
In H4, regarding the effect of product knowledge on FOMO, the findings of the structural path show that there is a significant effect (t = 2,184 > 1,96) with a significant value (0,029 < 0,05). Hence, H4 is accepted. The research data show that all knowledge about arak Bali is accepted in Balinese society (especially the millennials and Generation Z). It is to cause or create a fear of being left behind because of the benefits obtained when consuming arak, such as social benefits, psychological benefits, and more value in the midst of the social environment. Product knowledge with a significant effect on FOMO in arak Bali is one of the important findings that explain consumer behavior in the research.
The results of product knowledge on FOMO align with research conducted by Saavedra and Bautista Jr. (2020) that products can influence consumers to share their experiences of owning or using clothes. Developing interventions that target motivational construction by using FOMO marketing appeal is essential to increase consumers' propensity to buy a product. Marketers can trigger FOMO through a sense of exclusivity. Getting opportunities that only a few people have can encourage an increase in one's ego. In a research conducted by Valdiani and Puspanidra (2020), there is a message uploaded containing testimonials. Testimonials usually contain positive responses from people who have used the product, which will lead to a sense of confidence in new consumers of the product.
H5 states the effect of FOMO on purchase intention. The findings of the structural path indicate that there is no significant effect between FOMO and purchase intention (t = 1,239 < 1,96) with a significant value (0,215 > 0,05). H5 is rejected. The research results indicate that FOMO does not significantly affect purchase intention. The researchers assume the behavior of the millennials and Generation Z in Denpasar City has a fear of being left behind. However, it does not cause or affect their interest in arak Bali. It can occur due to several factors. One of which is the more prioritized need. Arak Bali is not a product of necessity but a desire that cannot necessarily be consumed daily, considering the impact that can be obtained if consumed excessively.
The results of this hypothesis are not part in line with previous research because it has no significant effect. However, it still has a positive effect. Reflecting on research conducted by Good and Hyman (2020), buying interest is most likely due to FOMO. FOMOladen appeals can increase purchase intention when individuals anticipate a positive outcome. It plays a psychological trick on customers to take advantage of every opportunity, so they do not miss out on any good deals. Therefore, marketers should create an advertisement that attracts the audience to buy the product and seal more deals in a limited time.
According to Good and Hyman (2021), FOMO can affect consumers' purchase intention. In particular, the attraction can strengthen purchase intentions by increasing consumers' anticipated excitement and self-enhancement or weakening purchase intentions to increase anticipated regret costs. FOMO affects buying interest indirectly through consumers' downstream desires. According to Yoga et al. (2022), the higher the level of FOMO will increase the purchase intention on Spotify (positive effect).

CONCLUSIONS
The research successfully finds the test results of the influence of social influence, product knowledge, FOMO, and purchase intention with SEM. The data obtained from the research show a significant and positive effect of social influence on purchase intention, product knowledge on purchase intention, social influence on FOMO, and product knowledge on FOMO, respectively. Meanwhile, the effect of FOMO on purchase intention has no significant effect. However, it still preserves a positive effect. Using SEM, the effect of product knowledge on purchase intention has the highest score. The research implies unpredictable results. Most of the past studies prove that all variables have a significant and positive effect, but there is a rejected hypothesis in the research. As a result, the research findings can enrich the information and be used as a reference to related topics, such as consumer behavior, social influence, product knowledge, and especially FOMO towards purchase intention, which has no significant effect.
The research can provide several implications that can be applied to business. Based on the results, the importance and significance of product knowledge possessed by consumers are to feel the fear of being left behind that causes the intention to buy arak Bali, especially for the millennials and Generation Z. The entrepreneurs and marketers can explain the benefits obtained when consuming arak Bali. Moreover, the other important thing is that entrepreneurs and marketers must continue to innovate in arak Bali products so that the information and experiences gained by consumers in the future can be used as product knowledge. The product knowledge will be fun and important to remember so that it will provide benefits to arak Bali's business which is engaged in the alcohol industry in Denpasar City. In addition to product knowledge, people cannot ignore the power of social influence to arouse the fear of being left behind in the millennials and Generation Z. The closest person or role model that the consumers follow is also an opportunity to introduce and inform the consumption of arak Bali. Hence, it can create the fear of being left behind and opportunities to generate the consumers' purchase intention.
The research has a limitation regarding SEM, especially the difficulty of gathering a model that properly fits with the process of data obtained from respondents. Moreover, the research only uses one study location, Denpasar City. The study related to the alcohol industry, especially arak Bali, can be wider, such as districts in the north or west of Bali. Besides, the population is also limited to men of millennials and Generation Z, while female has the same opportunity in research related to this topic. It is no doubt that further research can use various samples, and the modeling will be broader, as well as the complexity of variables used, such as the influence of social media, customer awareness, purchase decisions, and others, to analyze consumer behavior.