In the green product market, eco-labels are important marketing tools to reduce information asymmetry. With the promotion of green products, eco-labeling is becoming more prevalent at both business and communication levels (Ahmad B, et al., 2021; Apipuchayakul & Vassanadumrongdee, 2020). Consumers’ perception and evaluation of eco-labels are particularly related to their knowledge of eco-labels (Naseem et al., 2020). When consumers evaluate eco-labels in a holistic way, this reduces information asymmetry about green products, which leads to more informed decisions. Al-Gasawneh & Al-Adamat (2020) concluded on consumer preferences for certified wood showed that knowledge about eco-labels plays a big role in consumer preferences. The study of Issock et al. (2018) showed how the consumers’ ecological behavior is shaped by their knowledge and trust of eco-labels. The importance of eco-labels has been emphasized by Yue et al. (2021), who suggest that learning about eco-labels will lead people to buy greener products. Consumers are more likely to believe eco-labels are credible and this helps them make better decisions when purchasing products. As Lopes & Veiga (2019) examined the factors affecting organic food, brands and reputations, labels, and the certification bodies’ reputations were deemed integral to establishing the consumers’ trust. Researchers have brought forth the importance of eco-label knowledge in establishing a trust for eco-label products (Irfan et al., 2021a). The arguments above suggest that product values and green trust will be affected by eco-label knowledge. Hence, we proposed following hypotheses:

H1: Consumer’s perceived knowledge of eco-labels has significant impact on purchase intention of energy efficient appliances.

H2: Consumer’s perceived knowledge of eco-labels has significant impact on perceived functional value of energy efficient appliances.

H3: Consumer’s perceived knowledge of eco-labels has significant impact on perceived green trust.

H4: Perceived functional value has significant impact on purchase intention of energy efficient home appliances.

H5: Perceived green trust has significant impact on purchase intention of energy efficient home appliances.

Social responsibility is the most influential factor to the consumption of energy efficient appliances among ethical consumers (Irfan, Elavarasan, et al., 2021b; Jabeen et al., 2019; Song et al., 2019). Generally, consumer social responsibility relates to the willingness of consumers to acknowledge the interests of others, perform their duties as they are expected to, and honor their promises. According to Liao et al. (2020), consumer social responsibility refers to making conscious and deliberate decisions about certain products based on personal moral beliefs. Zhang et al. (2020) discussed that individual belief that it is their responsibility to support energy efficient appliances was conceptualized as the concept of responsibility (Razzaq et al., 2020; Irfan et al., 2021c). As a matter of fact, personal norms define a person’s standards of behavior and ethical beliefs, as well as his or her beliefs about what is right or wrong toward certain objects. Responsibility is the driving force behind moral obligation (Joshi et al., 2019; Ahmad et al., 2020a; Islam et al., 2022). Norms and values affect how people feel responsible for exhibiting prosocial behaviors. Individuals’ attributions of responsibility for their behavior can affect their norms regarding behavior, according to studies (Khan et al., 2021; Nuvvula et al., 2022). An individual’s personal norms become more firmly established if he or she is responsible for a certain behavior. In the same manner, individuals who feel responsible for supporting energy efficient products purchase these products with higher personal standards (Ali et al., 2019).

The attitude of an individual is defined as a preference towards certain objects, ideas, or behaviors. Researchers have hypothesized that attitude is the judgement and evaluation of a particular object based on its abstract associations, and a psychological tendency that has its expression in the degree of favor or disfavor one accords to it (Aguiar-Quintana et al., 2021; Naseem et al., 2021; Razzaq et al., 2021). An individual’s likelihood of purchasing a particular product is defined as his or her purchasing intention. A person’s purchasing intention can be used to measure their likelihood to purchase a certain product. If a consumer has a positive attitude towards a product, the greater the likelihood that he/she will purchase the product (Ahmad, et al., 2021b; Chavez et al., 2020; Razzaq et al., 2020). An individual’s attitude toward a certain product is measured by their positive or negative evaluations thereof (Jamil et al., 2021b; Tanveer et al., 2021). An attitude is defined as a set of beliefs, feelings, and thoughts towards a product, which leads to a specific action to be taken. The attitudes of consumers toward buying energy-efficient products significantly changed their intentions to buy such products (Nguyen et al., 2018). Hence, we proposed following hypotheses:

H6: Consumer’s perceived social responsibility has significant impact on purchase intention of energy efficient home appliances.

H7: Consumer’s perceived social responsibility has significant impact on perceived personal norms towards energy efficient home appliances.

H8: Consumer’s perceived social responsibility has significant impact on consumer’s perceived attitude towards buying energy efficient home appliances.

H9: Perceived personal norms towards energy efficient appliances has significant impact on purchase intention of energy efficient home appliances.

H10: Perceived attitude towards energy efficient appliances has significant impact on purchase intention of energy efficient home appliances.

Consumers need to keep an eye on the functions of eco-friendly products when purchasing them (Abu-Elsamen et al., 2019; Irfan et al., 2020; Wu et al., 2021). Functional values are based on the quality, durability, and cost of the product. It is clear from the preference of environmentally friendly consumers for products such as aerosols, natural ingredients in cosmetics, organic vegetables, and biodegradable products that fair prices play a significant role in their decisions (Gahlot Sarkar et al., 2019; Sarfraz et al., 2021; Xiang et al., 2022). Consumers’ decisions on energy-efficient refrigerators are largely influenced by their price and durability. The functional value of products is strongly associated with purchase intentions and product usage, suggests research. Buying energy-efficient appliances requires a focus on functional values (Akroush et al., 2019; Ahmad et al., 2020b). Hence, we proposed following hypothesis:

H11: Perceived functional value mediates the relationship of perceived knowledge of eco-labels and purchase intention of energy efficient home appliances.

Consumer behavior is fundamentally influenced by trust. The idea is that one party expects another to perform the task. Several studies have found that trust predicts purchase intentions of consumers, including Chen et al. (2015). Guangxia Li G et al. (2021) argues that green trust is one of the critical factors affecting consumers’ purchasing behavior in the context of environmental marketing. Green trust of customers is posited to improve green product purchase intentions by (Khetarpal & Singh, 2020; Yang et al., 2021). It has been observed that consumers’ pro-environmental behavior is positively affected by trust in eco-labels (Li N et al., 2021; Mohsin et al., 2021). Despite this, many studies have revealed that consumers’ trust is sometimes unrelated to their purchase intentions. A study by Helleno et al. (2017) found that Brazilians’ trust in organic foods did not influence their purchasing decisions. Several studies have relied on trust in the decision to buy an environmentally friendly product, but the results have been inconsistent. Green trust can, therefore, influence consumers’ intentions to purchase energy-efficient appliances (Awan et al., 2021a; Sreen et al., 2021). Hence, we proposed following hypothesis:

H12: Perceived green trust mediates the relationship of perceived knowledge of eco-labels and purchase intention of energy efficient home appliances.

Personal norms are the standards of expectation faced by each individual towards a given object and are based on a set of values that are internalized and applied accordingly (Nguyen et al., 2021). Several researchers have proposed that an individual’s beliefs about a behavior shape the way in which he or she responds to that behavior. In other words, attitude toward a particular behavior stems from an individual’s assessment of the behavior’s value (Awan FH. et al., 2021; Siyal et al., 2021). Individuals may be triggered to adopt personal norms by feeling that they must take action in order to protect themselves from negative outcomes (Jamil et al., 2021c). When people have favorable personal norms, i.e., they believe a specific behavior results in a positive outcome, they adopt a positive attitude toward it (Nguyen et al., 2019). Hence, we proposed following hypothesis:

H13: Consumer’s Perceived personal norms mediate the relationship of consumer’s perceived social responsibility and purchase intention of energy efficient home appliances.

A person’s attitude is a “state of mental readiness acquired by experience” affecting responses towards people, objects, and situations. Several studies indicate that consumers’ positive attitude plays a significant role in determining their environmental behavior (Zhang & Tao, 2020; Shimul et al., 2021). Various researchers have concluded that attitude is a significant determinant of energy-efficient appliance purchases. Purchasing intentions among consumers are influenced by their attitudes towards energy-efficient home appliances (Awan et al., 2021c; Joshi et al., 2022). Hence, we proposed following hypothesis:

H14: Consumer’s perceived attitude mediate the relationship of consumer’s perceived social responsibility and purchase intention of energy efficient home appliances (see Figure 1for all relationships).

Before drawing the questionnaire items, we studied and undertook a detailed literature review related to all study variables. A total of 23 items were adapted to create the final questionnaire, and these items were divided into seven sections. First, perceived knowledge of eco-labels was measured with three items adapted from Taufique et al. (2017). Consumer social responsibility was measured with three items adapted from Lee (2009). Perceived functional value was measured with three items adapted from the prior study of Sweeney & Soutar, (2001). Green trust was measured with five items and these items were adapted from the past study of Chen & Chang, (2012). Personal norms were measured with three items adapted from the study of Khare (2015). Attitude towards energy efficient appliances was measured with three items and these items were adapted from the study of Taylor & Todd, (1995). Purchase Intention of energy efficient appliances was measured three items developed by Wang Z et al. (2017). All items were measured on a five-point Likert scale.

In a survey sample, common method bias is a significant problem. Harman’s single-factor test was applied to analyze the common method bias in this research (Podsakoff et al., 2003). As stated by Harman (1976), a single-factor test was established to determine the existence of CMV among constructs. The data revealed that all sample items were classified into 23 distinct factors, with the first component accounting for 39.273 percent of the total variance, less than the proposed criterion of 50 percent. (Appendix Table A1). Furthermore, we applied Smart PLS to conduct a thorough collinearity evaluation test. According to Kock (2015) and many other social science academics, this is a relatively efficient and accurate technique (Zafar et al., 2021). All VIF values are well below the recommended threshold of 5, showing that typical process bias is not a problem in our model (Kock, 2015).

The Smart PLS 3.3.3 software package was employed in this study. PLS-SEM is a multiphase analysis that first evaluates the measurement model and then assesses the structural model (Ringle et al., 2020). The measurement model requirement ensures that only constructs with adequate items loading, convergent validity, composite reliability (CR), and discriminant validity will be considered significant and used in the structural model. The bootstrapping approach evaluates path coefficients and assesses their magnitude in structural model evaluation. Preacher & Hayes (2008) method was employed in terms of mediation evaluation because it is a more specific way for testing mediating effects and is more suited for use with the PLS-SEM methodology (Hayes, 2009; Hair Jr et al., 2014). For data analysis, the PLS-SEM tool has been used in most recent management research studies (Shujahat et al., 2019; Usman Shehzad et al., 2022).

The measurement model was subjected to a first-phase assessment in order to establish the constructs’ reliability and validity Hair et al. (2016a). During the evaluation of the measurement model, two factors were deleted due to low loadings, and the remaining ones were greater than the recommended value of 0.70. Except for PFV3 and PGT5, all questions were included in the final measurement model. The factor loadings in Table 1 and Figure 2 are greater than the recommended threshold of 0.70. Similarly, the AVE and CR of all constructs are equal to or greater than the recommended standards value of 0.50 and 0.70, respectively. This leads to the development of convergent validity and reliability.

Further, the Fornell–Larcker criteria and heterotrait–monotrait (HTMT) ratios are used to validate the discriminant validity of the current research (Hair et al., 2016a). Using the Fornell–Larcker criteria, Table 2 shows that discriminant validity has been confirmed since the top value of variables correlations in each column is the maximum (Hair et al., 2016b). The values of HTMT ratios should be less than 0.85 according to the HTMT ratios criteria; nevertheless, ranges up to 0.90 are allowed (Hair et al., 2016a). As shown in Table 3, all HTMT ratios are below 0.85, indicating that discriminant validity has been verified for the current research model.

The findings show that the R2 and Q2 values for PA are 0.196 (Q2 = 0.135), PFV 0.156 (Q2 = 0.118), PGT 0.164 (Q2 = 0.112), PI 0.499 (Q2 = 0.361), and PPN 0.423 (Q2 = 0.310). The R2 values indicate the model’s in-sample predictive potential (Sarstedt et al., 2014) since they are more than the necessary threshold of 0.10. Also, Q2 values larger than 0 suggest appropriate predictive relevance. Likewise, effect sizes (f2) are computed to determine the amount to which a predicting (exogenous) variable adds to an endogenous variable’s R2 value. The results in Table 4 demonstrate that the effect sizes of the variables in this study range from small to large, confirming the model’s strength (Hair et al., 2016a).

After compulsory evaluation of measurement model, evaluation of structural model test was done in the second phase. Bootstrap resampling technique with 5,000 resamples (Ringle et al., 2020; Gul et al., 2021c) was utilized to establish the significance of direct and indirect paths. Table 5 lists the test outcomes of hypotheses intended for direct and indirect associations (see Figure 3).

First, we assessed the direct relationships before looking at the mediation effects. Results in table 5 revealed that PK was significantly associated with PI (β = 0.116, p = 0.037), PFV (β = 0.395, p < 0.001), and PGT (β = 0.405, p < 0.001). Therefore, H1, H2 and H3 were supported. Similarly, PFV (β = 0.146, p < 0.01) and PGT (β = 0.285, p < 0.01) were also significantly associated with PI and study results supporting H4 and H5. Moreover, results revealed that PCSR was also significantly associated with PI (β = 0.192, p < 0.01), PPN (β = 0.650, p < 0.001), and PA (β = 0.443, p < 0.001). Therefore, findings confirm that H6, H7 and H8 were supported. For H9 and H10, study found PPN (β = -0.116, p = 0.096) was insignificantly asscoiated with PI, Whereas PA (β = 0.282, p < 0.001) was significantly associated with PI. Thus, H9 was rejected and H10 was supported.

To test the mediation effect, we used the bootstrapping the indirect effect method (Preacher & Hayes, 2008) with a 5,000 resample. For H11 and H12, Results revealed that PFV (β = 0.058, p = 0.016) and PGT (β = 0.115, p < 0.01) were significantly mediated the relationship between PK and PI. Therefore, results support H11 and H12. Moreover, For H13 and H14, study found insignificant mediating effects of PPN (β = -0.075, p = 0.095) and significant effects of PA (β = 0.125, p < 0.001) in the relationship beween CSR and PI, which do not support H13, while H14 was not supported.

In contrast to previous studies, this study makes significant contributions. Initially, this study presented an antecedent model for predicting energy-efficient home appliances based on consumer awareness of eco-labels and functional values. It differs from previous studies that have focused primarily on the general understanding of the environment and attitudes toward purchasing energy-efficient appliances (Zhao et al., 2020). The current study focuses primarily on eco-label knowledge. It is a strategic marketing tool to create a positive image about eco-friendly products and to encourage customers to buy them. Additionally, the conceptual model was tested in a developing country (Pakistan). Consumer attitudes toward social responsibility and consumer norms regarding the purchase of eco-friendly products are both important factors in determining whether they will buy eco-friendly products. When eco-label knowledge is combined with consumer perceptions of customer responsibility and consumer norms, the probability of eco-friendly behavior is improved following COVID-19.

Researchers have found that home appliances that are energy-efficient are more likely to be purchased by consumers, and their findings can provide marketing and policy makers with guidelines. Promotion and development of energy-efficient products are largely the responsibility of marketers and policy-makers. It is essential to establish a standard for eco-labeling and to strengthen its supervision in the country. Thus, consumers’ trust in and decisions about buying energy-efficient appliances are largely influenced by the information on ecolabels (Gahlot Sarkar et al., 2019). It is also crucial for companies to help develop and promote energy-efficient home appliances. Technology innovation and improvements in the efficiency of household appliances are among the key functional values that companies should emphasize, in order to attract consumers to value the surprising energy-saving effects of appliances. By providing information on how to use energy-efficient appliances, eco-label designs could help households save money. In turn, it indicates that functional values are holding people back from buying energy-efficient appliances. Consumer confidence will be built through the use of environmental labeling schemes promoted by policymakers and marketers. Consumers’ attitudes and perceptions of consumer’s personal norms are influenced by consumers’ social responsibility, which supports the significance of environmental values (Abu-Elsamen et al., 2019). Consequently, eco-labels are a good way of making sure consumers are informed about how emissions can be reduced by utilizing energy-efficient appliances. By advertising and marketing campaigns, consumers should also be encouraged to save energy by using energy-saving appliances.

We have attempted to present a holistic model in this study. Future research should address some limitations. Results may be limited in their generalizability when self-reported data are used. It may be possible for future researchers to evaluate consumers’ actual consumption of energy-efficient products by using longitudinal data or by using experimental design. In this study, constructs such as self-identity and altruism were included as potential constructs to predict purchase intention, however many other constructs could also be used. Its focus is on a single country i.e., Pakistan. To increase generalizability, future studies need to include more countries.