Rational and Moral Considerations in Organic Coffee Purchase Intention: Evidence from Indonesia

Abstract

The theory of planned behavior is broadly used to study the buying behavior of green products. However, the theory has been criticized for ignoring moral considerations in the purchase decision and not sufficiently explaining the influence of personal factors such as consumers’ product attributes evaluation. This study aims to build a structural model to explain the causality of factors influencing consumer intention to buy organic coffee in Indonesia. Organic coffee has a bright business prospect, but there was only limited evidence of academic studies on organic coffee from a consumer’s perspective. Thus, this study is one of the attempts to explain organic coffee purchase intention from the perspective of rational and moral consideration. The sample of 500 coffee consumers aged 18 years and more who live or do activities (work/study) in Jakarta were recruited using the purposive sampling technique. Structural equation modeling was utilized for data analysis. The results demonstrated that attitude, subjective norms, perceived behavioral control, and values orientation affected purchase intention. The study successfully explains the role of moral considerations in organic coffee purchase intention. These findings suggest managerial implications, including marketing communication and distribution strategy. Future research suggestions were also discussed.

1. Introduction

The circular economy (CE) discourse has become of great interest to academics and practitioners regarding its conceptual and operationalization that can guide businesses to implement sustainable development (Kirchherr et al. 2017). The goal of circular economy (CE) is sustaining economic systems by restraining environmental impacts and reducing resource use (Merli et al. 2018). The underlying core of CE is known as “3R” (reduce, reuse, recycle). Thus, implementing CE principles within green supply chain management can generate visible environmental benefits (Arsova et al. 2021).

Extensive research recently focused on the connection between CE and the bioeconomy (D’Amato et al. 2017) since bioeconomy is considered one part of the CE (Carus and Dammer 2018). Bioeconomy is defined as “the production of renewable biological resources and the conversion of such resources and waste streams into value added products, such as food, feed, biobased products and bioenergy.” (European Commission 2017). The output of such production is known as a bio-based product, defined as “a product that is wholly or partly derived from materials of biological origin, excluding materials embedded in geological formations and/or fossilised.” (Eubia n.d.). There is a growing number of bio-based products available in the market, such as bio-based textiles, wood products, and, bio-polymers (Morone et al. 2021). Regarding this development, growing attention has been placed on the extent of consumers’ intention to purchase these types of products (Reinders et al. 2017).

Within the context of bio food, it promises a safe, healthy, and nutritious food (Carus and Dammer 2018). The nutritional quality of bio food is superior compared to regular food. It contains more antioxidants, a higher level of essential minerals and more nutrients (Pagalea and Uta 2012).

Organic food is an equivalent term for bio food (Pagalea and Uta 2012). Organic food is defined as food that is produced without any man-made substances, such as fertilizers, herbicides, pesticides, and biotechnology (USDA 2012). Agricultural products are certified as organic if they are produced through approved methods that promote resource cycling, ecological balance, and biodiversity without using synthetic fertilizers, antibiotics, and growth hormones (USDA 2015).

Regarding organic food consumption, the market is relatively small and immature. Thus, only a few have limited experience on consuming organic food (Teng and Wang 2015). The leader in organic market shares is Denmark (12.1%). The share is significantly lower for Asian countries, such as Viet Nam (0.2%) and Japan (1.4%) (FiBl and IFOAM 2021). In Indonesia, the organic products share only represents 0.03% of global demand, with a per capita spending of USD 0.06 in 2021 (Organic Trade Association n.d.).

North America and European countries have recently contributed the largest share of organic product consumption. Despite the current high rate of organic food and beverages consumption in North America and European countries, however, the consumption growth in those countries tends to decline over time due to the market maturity. Meanwhile, the Asian market experienced significant growth, approximately 30% (FiBl and IFOAM 2021). Therefore, Asian countries are expected to be the most prospecting market for organic products in the coming years.

Indonesia has been recognized as one of the most potential markets in Asia for organic food and beverages. Along with China and India, Indonesia is expected to grow faster than other countries (FiBl and IFOAM 2021). Organic vegetables and rice are the most organic food consumed by Indonesian consumers (Statista 2022).

For organic beverages, organic coffee is the primary category existing in Indonesia (Organic Trade Association n.d.). Organic coffee has excellent business potential due to the coffee drinking culture in Indonesia. In Indonesia, drinking coffee has become a lifestyle across generation (Utama et al. 2021). During 2015–2019, coffee consumption growth in Indonesia reached 54.3%, compared to the world consumption growth, which was only 9.9% (Foreign Agricultural Service 2019). In addition, Indonesia is one of the biggest organic coffee producers in the world.

However, the domestic sales value of organic coffee is relatively small. In 2019, the sales value only reached USD 1.5 million, compared to the sales value in the United States, which reached USD 552 million in 2016 (Organic Trade Association n.d.). Furthermore, the share of organic coffee sales values only approximately 0.08% of total coffee sales value, reaching USD 1.7 billion in the year 2020 (International Communicafe 2020). Concerning the low consumption of organic coffee, investigating factors that affect consumers’ purchase intention of organic coffee in Indonesia is urgently needed. Therefore, scientific information on rational and moral factors that influence consumer intention to buy organic coffee is needed, which will be valuable for the organic coffee industry, particularly in Indonesia, to increase the effectiveness of its marketing effort.

The theory of planned behavior (TPB) has been widely applied in various research contexts, including research on the purchase intention of organic products (Wibowo et al. 2022). The meta-analysis study by Han and Stoel (2017) concluded that the three predictors of intention in the TPB had a predictive power of 39.7%.

However, several criticisms were raised about the theory, and it is thought to lack the ability to capture the complexity of green purchases (Groening et al. 2018). The TPB is considered only to involve rational considerations of consumers’ purchase decisions. Yet, it has been acknowledged that buying pro-environmental products is not just a matter of rational choice but also of moral consideration (Fornara et al. 2016; Thøgersen 2006). The importance of moral consideration is also significant for buying organic products since the benefit of buying the product is received not only by the consumers but also by other people and nature (Saleki et al. 2019).

Several authors also criticized the causalities of attitude, subjective norms, and intention (Sussman and Gifford 2019). Liska (1984) reported that beliefs might directly influence behavior without being mediated by attitude; thus, the sequence causalities in TPB (and TRA) may not be correct. Several studies based their arguments on the cognitive dissonance theory for questioning the causalities in the TRA and TPB models. According to the cognitive dissonance theory (Festinger 1957), individuals need to maintain consistency between thoughts, feelings, and behavior. When inconsistency exists, it leads individuals to change one of the inconsistent elements to reduce dissonance. In his experiment, Festinger provided evidence that individuals change their attitudes that is inconsistent with their behavior.

For the moral considerations aspect, few studies have considered egocentric, biospheric, and altruistic personal values orientation from Stern et al. (1993) as moral representation (e.g., Wang et al. 2020) that affect consumers’ purchase of green products. Biospheric and altruistic values are believed to have a positive effect, while egocentric values have a negative effect on green purchase intention and decisions.

However, to the best of the author’s knowledge, most studies put values orientation as the direct predictor of green purchase intention (Caniëls et al. 2021; Song and Kim 2018). Our study, however, refers to the value-attitude-behavior theory and we also treats the value orientation as the predictor of purchase attitudes. According to the theory, values are organized into a cognitive belief hierarchy, consisting of global values; domain-specific values, known as personal values orientation (Fulton et al. 1996); and attitude. Within this hierarchy, global values will affect domain-specific values, and domain-specific values will affect attitude.

Thus, this study aims to analyze rational evaluation and moral determinants of organic coffee purchase intention. The results will contribute to the green consumer behavior literature, particularly on the role of rational and moral consideration of coffee organic purchase intention. The organic coffee industry can use the comprehensive and scientific information of the study to increase the marketing effectiveness of the organic coffee market.

2. Theoretical Review

2.1. Theory of Planned Behavior (TPB)

The TPB (Ajzen 1991) is a rational choice model (Groening et al. 2018) that argues that the main predictor of an individual’s behavior is his/her intention to behave. Intention reflects how much someone wants to be willing to take action to achieve the desired goal. An individual’s intentions are determined by three consumer traits: attitudes toward behavior, subjective norms, and perceptions of behavioral control.

Attitude towards behavior refers to “the degree to which a person has a favorable or unfavorable evaluation or appraisal of the behavior in question” (Ajzen 1991, p. 188). Perceived behavioral control refers to a person’s ease or difficulty in performing certain behaviors (Ajzen 1991). Subjective norms are individual beliefs about the behavior desired by people who are considered important to the individual (Ajzen 1991).

The TPB can be considered as the development of the theory of reasoned action (Fishbein and Ajzen 1975). Regarding the limitation of the theory of reasoned action that only depends on internal individual’s traits, Ajzen added perceived behavioral control to represent external factors.

Formerly, the antecedents of intention were measured using unidimensional measurement. Ajzen (2002) then refined the measurement into multi-dimensional ones. The use of multi-dimensional measurements has been included in previous studies (Jellason et al. 2019).

2.2. Values Orientation

The main criticism of the rational choice model is that consumers are considered homo economicus, and not a human (Dietz and Stern 1995). The model perceives buying behavior purely as an economic transaction; a behavior is chosen to maximize satisfaction. However, Dietz and Stern (1995) have demonstrated abundant evidence that the model is insufficient to explain pro-environmental behavior. Rather than being guided by complicated calculation deliberation, consumers tend to use if-the-rules deliberation (Dietz and Stern 1995).

Furthermore, they argued that when the choice is deliberative, consumers evaluate outcomes against their values orientation. In other words, consumers utilize values to select and justify their actions and behavior. Personal values orientation is defined as “a desirable goal that serves as a guiding principle in one’s life.” (Dietz and Stern 1995). Personal values orientation is within the moral domain since values-based behavior has a strong moral element (Dietz and Stern 1995).

Stern et al. (1993) identified three forms of values orientation related to individual concern for the environment: egocentric, biosphere, and altruistic values. These three types of values orientation were rooted from Merchant’s (1992) three classes of ethics: the homocentric, ecocentric, and egocentric. Egocentric values affect the goal of seeking profit for oneself. Altruistic values indicate a primary concern for the welfare of others, and biosphere values represent a primary concern for nature.

2.3. Hypothesis Development

Social pressure, represented by subjective norms in the TPB, may affect the intention of behavior (Fishbein and Ajzen 2011). Others can affect behavior since they have one or more of the following powers: reward power, coercive power, legitimate power, expert power, and referent power (Fishbein and Ajzen 2011). Varshneya et al. (2017) and Liobikienė and Juknys (2016) in their research found that social influence is one of the determinants of organic product purchase intention. Previous studies have proven the effect of subjective norms on organic food purchase intention (Chu 2018; Secapramana and Ang 2019).

Attitude has been used to predict behavior in a wide array of domains. Attitudes are among the most crucial predictors of intentions and behavior (Fishbein and Ajzen 2011). If an action is considered appealing or pleasant (positive attitude), the intention to perform the action is activated.

Intention has been considered the conative aspect of attitude, and it has a strong relationship with the affective aspect of attitude. However, Fishbein and Ajzen (1975) claimed a strong relationship between those two concepts only if the overall favorability between attitude and intention is approximately the same. Furthermore, they suggested that the relationship exists only in a volitional control circumstance. Recent studies have found a positive effect of attitude toward buying behavior on organic product purchase intention (Emekci 2019; Hsu et al. 2019; Najib et al. 2022).

However, the social pressure and attitude themselves are not sufficient. There are external factors that individuals also consider before taking any action. Ajzen (1991) added perceived behavioral control to represent the existence of external factors as another predictor of the intention of behavior. If people believe they have control over behavior, they may form a strong intention to perform the behavior (Ajzen 1991; Fishbein and Ajzen 2011). Strong empirical evidence has been found for the relationship between perceived behavioral control and purchase intention of organic food (Golob et al. 2018; Paul et al. 2016).

Based on those arguments, we propose hypotheses as follows:

H1: 

Attitude towards purchase behavior affect purchase intention.

H2: 

Subjective norms affect purchase intention.

H3: 

Perceived behavioral control affects purchase intention.

Individual values orientation has been considered to have an important role in pro-environmental behavior since they are believed to affect green attitudes, intentions, and behaviors (Steg et al. 2014; Caniëls et al. 2021). Stern et al. (1993) argued that egoistic, biospheric, and altruistic are types of values orientation that determine pro-environmental behavior. Furthermore, they contended that attitude towards a certain behavior is constructed based on the values orientation since it leads the individual to search the relevant information on the consequences of the behavior that will be the basis for attitude formation.

The values, attitude, and behavior relationship has been explained in the values-attitude-behavior (VAB) model (Homer and Kahle 1988). According to this model, values are considered a type of social cognition that has a similar function to attitude, which is to help individuals to adapt to their environment. However, values are more abstract than attitudes, thus representing the basic characteristics of adaptation. These characteristics then, will serve as an architype for the formation of attitude.

Previous studies had found that attitude was determined by altruistic values (Kim and Yun 2019), biospheric values (Shin et al. 2017), and egocentric values (Kim and Yun 2019). Wang et al. (2020) also found that biospheric and altruistic values positively impact green purchase attitudes. Yadav (2016) in his study demonstrated the role of egoistic and altruistic values in determining consumer attitudes toward organic purchases.

Therefore, it is posited:

H4: 

Egoistic values orientation affect attitude.

H5: 

Biospheric values orientation affect attitude.

H6: 

Altruistic values orientation affect attitude.

Individuals need to maintain consistency between their values and behavior. Thus, the more important a value, the goals related to the value become more valuable. It will lead individuals to have a strong intention to perform an act that can reach the goal (Gollwitzer 1999; Bardi and Schwartz 2003). In this line, the cognitive dissonance theory (Festinger 1957) revealed that individual beliefs could affect behavior directly. Similarly, Feather (1992) explained that value would affect the attractiveness of the goal of the behavior, thus motivating him or her to attain the goal; in other words, it develops the intention to perform the behavior since intention captures motivation (Ajzen 1991).

In isolation, previous studies have supported the role of altruistic, biospheric, and egoistic orientation on purchase intention (Caniëls et al. 2021). Rahman and Reynolds (2017) found that egocentric values significantly and positively affect organic wine purchase intentions. Kareklas et al. (2014) found that egocentric and altruistic values simultaneously predict consumers’ organic purchase intention. Wang et al. (2020) have proven the positive effect of altruistic and biospheric values on green purchase intention.

Therefore, we propose the following hypothesis:

H7: 

Egoistic values orientation affect purchase intention.

H8: 

Biospheric values orientation affect purchase intention.

H9: 

Altruistic values orientation affect purchase intention.

The proposed model of this research is in Appendix A.

3. Methodology

3.1. Design

This research was conducted in Jakarta, the capital city of Indonesia. Based on the survey (Safitri and Setiawan 2019), Jakarta is one of the regions with the most prominent coffee consumers. Information collection from respondents was conducted during December 2021–March 2022.

This study used a survey method. The instrument was administered to adult coffee consumers in Jakarta. Using consumers instead of students as respondents will represent the various demographic characteristics of coffee consumers in Jakarta. Five hundred respondents were selected using the purposive sampling technique, and the online survey was employed to gather data from the respondents. The number of samples of 500 respondents met the criteria presented by Hair et al. (2019). Hair et al. (2019) posits that models with large numbers of constructs, some with lower communalities, and/or having fewer than three measured items require a minimum sample size of 500 (p. 633). Thus, our sample size has met the minimum requirement.

Data from respondents were collected from an online survey provider in Jakarta. Participants were selected based on several criteria (e.g., consumers who regularly consume at least 1 cup of coffee a day, were 18 years old and older, familiar with the existence of organic coffee, aware with the existing of organic coffee, considered buying organic coffee in the next purchase, and were domiciled in Jakarta). The survey instrument began with screening questions related to the criteria to select the qualified respondents. Only respondents who met the criteria were allowed to proceed to the rest of the survey.

Data were analyzed using structural equation model (SEM). The goodness of fit analysis on the measurement and structural models was conducted using the chi-square probability index, CMIN/DF, GFI, TLI, CLI, and RMESA. The measurement model and structural model are declared fit if they meet the criteria for chi-squared probability > 0.05, CMIN/DF < 2, GFI > 0.90, TLI and CFI > 0.92, and RMSEA < 0.08 (Hair et al. 2019).

The reliability test was performed with the criteria if the CR value 0.7, the indicator, was declared reliable. To meet the validity criteria, the loading indicator value and AVE value were >0.5 (Hair et al. 2019).

3.2. Measurement

The three predictors of intention were measured using first-order multi-dimensional measurement. Ajzen (2002) introduced this multi-dimensional measurement for the three predictors of intention as the development of the former unidimensional measurement. In their study, Hagger and Chatzisarantis (2006) concluded that all dimensions achieved discriminant validity. However, Rhodes and Blanchard (2006) argued that the first-order model demonstrated better psychometric properties than the higher-order model. Therefore, this study measured attitude, social norms, and perceived behavioral control using the first-order technique.

Attitude towards behavior consisted of instrumental (cognitive) and experiential (affective) aspects, measured by six indicators adapted from Yadav and Pathak (2017). Perceived behavioral control consisted of perceived capacity and perceived autonomy aspects, measured with six indicators adapted from previous studies (Fishbein and Ajzen 2011; Liu et al. 2020).

Furthermore, we also developed a measurement for social norms. The former injunctive and descriptive in subjective norms, which only included the important others (Ajzen 2002), were extended by adding injunctive and descriptive norms from society.

The term others in injunctive subjective norms and descriptive norms in subjective norms constructed by Ajzen (2002) still refer to social pressure from most people who are important to the individual (important others). However, prior studies have demonstrated that social pressure can come from various levels of group: peer group, community, and society (Lahlou 2017, p. 124); or proximal, distal, or societal levels (Park et al. 2009). Based on this argument, several studies have used a more general reference as the source of social pressure, such as thousands of people (Gerber and Rogers 2009) and people (Mabry and Turner 2016).

Therefore, we believe that both specific and general contexts of social pressure will affect consumers’ purchase decisions. Thus, in this study, social norms were measured with four aspects: the two first aspects were specific injunctive and descriptive norms (SI and SD) which refers to injunctive and descriptive in subjective norms. We developed two additional aspects: general injunctive and descriptive norms (GI and GD), representing social pressure from society. Fourteen indicators were used to measure the social norms construct.

Biospheric values were measured using four items from Snelgar (2006) and Wang et al. (2020): harmony with nature, respecting the earth, protecting the environment, and preventing pollution. Altruistic values were measured using four items adapted from previous research (Evans and Ferguson 2014; Snelgar 2006; Tokay Argan and Argan 2017): grace, saving lives, helping others, and creating a better world for the future generation. Egoistic values were measured using five items from prior studies (Snelgar 2006; Vermeir and Verbeke 2008): wealth, influence, public image, my health, and social recognition.

Items of measurement are on Appendix B.

4. Results

4.1. Profile of Respondents

The composition of male and female respondents was relatively balanced: male respondents accounted for 56.8% and women for 43.2%. Most of the respondents were workers (73.8%). The respondents’ educational background concentrated on bachelor’s degrees (72.6%). Regarding income, almost half of the respondents (47.4%) had income in the range of USD 287–USD 400 per month. Regarding coffee consumption, most respondents (81.2%) regularly consumed 1–3 cups of coffee a day. As many as 90% of respondents had been drinking coffee regularly for more than one year. A total of 86% of respondents spent USD 13–USD 26 per month to buy coffee. Among the places to buy coffee, the modern market (34.2% of respondents) and the traditional market (35% of respondents) were the favorites. However, the sample has not been weighted to counteract the deviation between the sample profile and the profile of residents in Jakarta. Thus, the sample characteristics could not represent the population profile. The table of the respondents’ profile is in Appendix C.

4.2. The Results of Model Evaluation

Of the five indices used to assess the goodness of fit of the measurement model, all indexes have values according to the criteria (CMIN/DF = 1.32; GFI = 0.94; TLI = 0.96; CFI = 0.96; and RMSEA = 0.02), so the measurement model showed a good fit. The CR value of each latent variable ranges from 0.80 to 0.85, so it can be stated that the indicators used were reliable. The loading value of the indicator ranges from 0.6 to 0.9, while the AVE value ranges from 0.40 to 0.57 (see Appendix D). Although AVE values for several constructs were less than 0.5 but CR values for all constructs were more than the acceptable level of 0.6, the convergent validity of the construct is still adequate (Fornell and Larcker 1981).

Of the five indices used to assess the goodness of fit structural model, all indexes have values according to the criteria (CMIN/DF = 1.69; GFI = 0.91; TLI = 0.94; CLI = 0.95; and RMSEA = 0.03). The results showed that the structural model has a good fit with the data.

4.3. Hypothesis Testing

The results of the coefficient significance test used as the basis for accepting/rejecting the hypothesis can be seen in

Table 1. The results of hypothesis testing.

Hypotheses	Path	Coefficient	t-Values	p-Values
H1:	ATB → PI	0.79	6.33	***
H2:	SN →PI	0.10	2.33	0.01
H3:	PBC → PI	0.38	5.60	***
H4:	EGO → ATB	0.52	9.58	***
H5	BIO → ATB	0.44	8.51	***
H6:	AL → ATB	−0.11	−2.76	0.01
H7:	EGO → PI	−0.20	−2.33	0.02
H8:	BIO →PI	−0.15	−1.76	0.07
H9:	AL → PI	0.04	0.92	0.35

***: p = 0.000.

:

As seen in Table 1, the results confirm the proposed hypothesis that attitude, subjective norms, and perceived behavioral control influence purchase intention (b = 0.79, b = 0.38, and b = 0.10). Thus, the more positive attitude towards buying organic coffee, the stronger the social pressure, and the higher the perceived control on the buying, the more likely consumers will buy the product.

Furthermore, the attitude was determined by egocentric values (H4: b = 0.52), biospheric values (H5: b = 0.44), and altruistic values (H6: b = −0.11). The results suggest that the more salient egocentric and biospheric values, the more likely consumers will have a positive attitude towards organic coffee purchases. In contrast, the positive effect of altruistic values on attitude was not supported by our data.

In addition, the direct effect of egocentric values on purchase intention was also significant (b = −0.20). However, there was no significant direct effect between biospheric (b = −0.15) and altruistic values (b = 0.04) on purchase intention. These results suggest the existence of a mediating effect, which is discussed in the next section.

4.4. The Results of Mediating Effect of Attitude

The study used attitude as the mediating variable to investigate the effect of egocentric, biospheric, and altruistic values on purchase intentions. Since the direct effect of egocentric on attitude and purchase intention was significant, thus attitude partially mediated the effect of egocentric values on purchase intention. In addition, while the direct effect of biospheric and altruistic values on purchase intention were insignificant, the indirect effect of the two constructs of personal values orientation on purchase intention were significant. Therefore, it indicates that attitude plays a full mediation role in explaining the relationship between biospheric values and purchase intention, as well as in altruistic values and purchase intention.

Based on Hayes (2009), a bootstrapping practice test on the indirect effects was performed. As seen in

Table 2. The results of mediation testing.

Path	Coefficient	p-Values	Results
EGO → ATB → PI	0.416	0.001	Partially mediation
BIO → ATB → PI	0.351	0.001	Full mediation
AL → ATB → PI	−0.08	0.01	Full mediation

, the results of the indirect effect for all the three paths were significant (at p-values 0.001, 0.001, and 0.01). Thus, partial mediation of attitude for the relationship between egocentric and purchase intention and full mediation of the impact of biospheric and altruistic values on purchase intention was statistically proven.

4.5. Discussion

Our findings demonstrate that personal values orientation affect consumers’ attitude toward organic coffee purchase. The influence of egocentric, biospheric, and altruistic values orientation on attitudes confirms the previous studies (Kareklas et al. 2014; Prakash et al. 2019; Wang et al. 2020). These findings show the significance of biospheric and egocentric values in determining attitude. Recently, consumers have been more concerned about chemical and pesticide usage in farming, which may negatively affect health and the natural environment.

It is interesting to find that the positive impact of egocentric values and the negative impact of altruistic values on attitude demonstrates the uniqueness of organic products as green products. In previous studies, egocentric values orientation was proven to have a negative effect on attitude since green products are characterized by premium price and often low quality; thus, it is considered to have a low value. However, in the context of organic products, including organic coffee, the premium price is followed by premium quality attributes, such as better taste and good for health, which is consistent with their personal goal.

Altruistic values have also been proven in prior studies to affect attitude positively (Kim and Yun 2019). The contrariety results may be explained by the value–action gap, which was often found in pro-environmental behavior (Akehurst et al. 2012). The value-action gap refers to the inconsistency between the pro-environmental attitude and actual behavior (Prothero et al. 2011). The premium price of organic products is often considered a barrier for consumers to translate their values into behavior (Caniëls et al. 2021). The similar explanation was also applied for the negative effect of egocentric values on purchase intention.

In addition, the impact of attitude, subjective norms, and perceived behavioral control on purchase intention support prior research (Secapramana and Ang 2019; Yadav and Pathak 2017). The results highlight the importance of consumers’ internal (attitude) and external factors (social pressure and perceived behavioral control) in determining consumer purchase decisions. While attitude has been recognized as a strong predictor of behavioral intention, it has been noticed that external factors may affect consumers’ evaluation of risks and benefits when buying organic foods; it even may prevent consumers from buying organic products (Nguyena et al. 2020).

The indirect effect findings confirm that personal values orientation affects purchase intention indirectly through attitude. Thus, this study affirms the value-attitude-behavior theory (Homer and Kahle 1988). The theory suggests that as the most abstract form of beliefs’ structure that guide individuals to preferred modes of behavior, values play a role as prototypes to shape attitudes and behaviors.

Within this perspective, egocentric and biospheric values provide basic guidance for consumers about the importance of behaving pro-environmentally. This generic belief leads to the formation of a more specific belief; a positive attitude toward buying organic coffee as a green product, which then turns into the intention to buy.

Lastly, our theoretical model has successfully improved the TPB’s ability to explain organic coffee purchase intention behavior. By integrating the theory with personal values constructs, we found the squared multiple correlation values of purchase intention at 0.63—much higher than Han and Stoel (2017) which concluded that the predictive power of intention’s predictor in TPB was at 39.7%. The model also confirmed the important role of moral considerations in purchasing organic products.

Regarding the research results, there are several managerial implications that the organic coffee industry can consider in marketing organic coffee products. First, the organic coffee industry needs to conduct intensive marketing communications to communicate the benefits of organic coffee, especially for health and the environment. This message aligns with consumers’ egocentric (altruistic) and biospheric values orientation. Thus, it will trigger the formation of a positive attitude toward organic coffee purchases.

Second, the organic coffee industry needs to increase consumer perception of their behavioral control; it means perceived barriers to buying organic coffee should be reduced. Marketers should increase the acceptance of the price of organic coffee, which is relatively more expensive than non-organic coffee. Product positioning and marketing communications play an important role. Organic coffee should be set as premium coffee. Furthermore, the communication message should focus on the superiority of organic coffee over other premium coffee in terms of health and environmental benefit.

Besides the price, availability is often seen as a potential barrier. Thus, the organic coffee industry also needs to increase distribution effectiveness to make it easier for consumers to obtain organic coffee, for example, by building a partnership with premium coffee shops or even organic coffee shops. In recent years, premium coffee shops and organic coffee shops started to serve ready-to-drink organic coffee in Indonesia. Other efforts include developing distribution channels such as organic shops, online and offline markets, and display management.

5. Conclusions

The current study examines the effects of attitude, subjective norms, perceived behavioral control, and personal value orientation on coffee organic purchase intention. The findings indicate that the positive effect of attitude and perceived behavioral control on purchase intention have been proven. At the same time, biospheric and egocentric values indirectly affect purchase intention through attitude. However, this study failed to prove the significant effect of altruistic values on attitude and subjective norms to purchase intention.

These results provide support for the worthwhileness of integrating rational and moral bases of behavior into the framework of TPB. Moral consideration seems to be beneficial in understanding and predicting attitudes and purchase intentions to buy organic coffee. In addition, this study discovered the important role of attitude in mediating the effect of moral aspects and purchase intention.

From a practical perspective, this study demonstrates acknowledgment of using a moral aspect in marketing organic coffee. This study implies that one of the reasons consumers buy organic coffee is because they perceive it is morally right. Through product positioning and thorough marketing communication, market penetration for organic coffee can be implemented effectively.

Despite the insight into theory and practice that this study provides, it has some limitations that can provide future research suggestions. This research is limited only to examining purchase intention. Actual behavior was not included in the model due to the limited number of consumers who have actually drunk organic coffee that can be recruited as respondents. Thus, this model cannot explain the role of intention and perceived behavioral control in forming actual behavior, as in the original TPB model. Therefore, further research is recommended to analyze the actual behavior.

Another limitation of this study is the use of a cross-sectional design and self-reported measures instead of an observed behavioral measure. It has been demonstrated that TPB variables account for a larger variance in self-reported than in observed behavior. Thus, further research is advised to conduct an investigation using the experiment method.

Regarding sample size and sampling technique, the number of respondents has reached the minimum requirement. However, it is far below the recommended size, which is ten times the free estimated parameter. This sample size issue might affect the results of the study. It might result in too little statistical power for the test to realistically identify significant results. In addition, the non-probability sampling technique we employed is not generalizable for the population. Further research should consider a higher sample size and adopt a probabilistic sampling method. It is also recommended to conduct studies in other cities of Jakarta to be more confident in generalizing the findings.