Factors Inﬂuencing the Purchase Intention for Recycled Products: Integrating Perceived Risk into Value-Belief-Norm Theory

: Recycling used materials is one way to deal with the depletion of natural resources available on earth. Companies increasingly integrate recycled materials into their production processes and transition towards circular business models. However, although the attitude towards sustainable products is positive, consumers still prefer to buy products made from new instead of recycled materials. Empirical research on factors inﬂuencing the purchase intention of recycled products is still limited. This study aims to examine consumers’ individual factors that are important in the decision process to buy recycled products. The Value-Belief-Norm (VBN) theory is explored in the context of recycled product purchase intention. Perceived risk is added to the research model as a moderator that hinders purchase intention. The different inﬂuences are analyzed using partial least squares structural equation modelling with a sample of 177 respondents from Germany. Results indicate that the causal chain of relationships between values, beliefs, and personal norm has a positive inﬂuence on recycled product purchase intention. Perceived risk, on the other hand, has a signiﬁcant negative direct effect on purchase intention but strengthens the relationship between personal norms and purchase intention. Theoretical and managerial implications as well as avenues for further research are discussed.


Introduction
The number of natural resources available on earth is limited and humans consume much more than what can be recovered.In 2021, "Earth Overshoot Day" was on 29 July, indicating that all ecological resources that could be restored within one year were used up five months before the end of the year [1].One way to address the problem of overusing natural resources is integrating recycled materials into the production process of new products.Some of the most common examples of recycling are PET plastic bottles or recycled paper.However, there is a strong shift toward using recycled materials in other areas of consumer goods.For instance, the clothing store chain H&M aims to use 30% recycled materials in their clothes by 2025 [2].This is an example for the efforts companies make to use resources in a sustainable way.However, extant research shows that consumers have a lower willingness to pay for recycled products than for products made from new materials [3], suggesting an increased perceived risk associated with recycled products [4].The important question, besides the scale of integrating recycled materials into production processes, is if consumers value such efforts.This paper explores the antecedents of purchase intention of recycled products.
Recycled products can be classified as green products [5].Previous qualitative and quantitative research has extensively analyzed factors influencing consumers to purchase green products (e.g., [6,7]).In the context of recycled products, Bigliardi et al. [8] proposed an integrative theoretical framework including 20 constructs that influence the purchase intention of recycled products.These authors emphasize the importance of green consumption drivers on the individual level.Research on pro-environmental consumer behavior has extensively focused on studying the effect of subjective norms; however, the influence of personal norms is less considered [9].Whereas people with high subjective norms base their actions on other peoples' opinions, personal norms describe a feeling of personal obligation towards a certain behavior [10].This study focuses on personal norms as a direct lever towards purchase intention.To analyze the effect, the Value-Belief-Norm (VBN) theory [11] is applied.The VBN theory suggests that "green behaviours are more likely to occur when a causal series of variables (i.e., values, beliefs, and personal norms) is present" [12] (p.2).To the author's knowledge, the VBN model has not yet been discussed and empirically tested for the purchase intention of recycled products.Following Kiatkawsin and Han [13], who explored the intention to behave sustainably, this study uses intention as outcome variable in the VBN model.Additionally, previous research found a negative effect of perceived risk on green product purchase intention [14].Apart from the direct effect, the author assumes that perceived risk can potentially alter the relationship between personal norms and purchase intention [15].As an extension to the VBN model, perceived risk is added as a moderator.
The objective of this paper is to empirically test if the VBN model can be applied to the context of recycled product purchase intention.Furthermore, the different effect sizes and their strengths in the causal relationship are explored.As extension of the model, the moderating role of perceived risk is analyzed.This study adds to the existing literature on individual factors influencing consumers' green purchase intention.On the one hand, it extends the knowledge about recycled products and on the other hand, it evaluates how values, beliefs, norms, and perceived risk effect purchase intention.
The paper is structured as follows: Section 2 conceptualizes recycled products and provides an overview of studies analyzing recycled product purchase intention.Furthermore, in this section the research model is established, and hypotheses are developed based on the VBN theory and perceived risk theory.Section 3 provides the research methodology and the measures used for the structural equation model.In Section 4, the study results of the measurement model, the structural model, and the moderating effect are presented.Section 5 discusses the findings and includes theoretical contributions and managerial implications as well as limitations and opportunities for further research.

Recycled Products
Recycling is the practice of reusing materials from used products, instead of unused raw materials, and turning them into new products of the same quality [16].In contrast, upcycling refers to reusing materials to make products of better quality and downcycling creates products of lesser value [17].Apart from conserving the limited natural resources available on earth, recycling has several other substantial environmental benefits including reducing pollution (in water and air), greenhouse gases, and waste [18].Companies integrate the circular economy business model, where instead of following a wasteful "takemake-dispose" mentality, they retain used products from consumers and re-enter them into a loop system of reuse [19].This shift to circular products comes with new product or service characteristics that alter the customer experience [20].Despite the environmental benefits, consumers have a lower willingness-to-pay for recycled products than for the same products made from new materials [3].This shows that it is important to further explore factors that influence consumers to purchase recycled products.So far, only a limited number of studies have been conducted to find triggers, especially on consumers' characteristics, for recycled product purchase intention (

Value-Belief-Norm Theory (VBN)
Individual factors play a central role in the behavioral process of consumers [8]; therefore, this study applies the VBN theory to explore aspects that influence recycled product purchase intention.The VBN theory was developed and first introduced by Stern et al. [11] in the context of social movement support.In this theory, values (altruistic, biospheric, egoistic) impact beliefs (ecological worldview, awareness of consequences and ascription of responsibility).These beliefs influence personal norms, which finally affect behavior.Initially developed to evaluate environmentalism, VBN has been further used to empirically analyze green product purchase behavior such as for cosmetic products [30], sustainable travel mode choice [31], and "alternative fuel vehicles" [32].For the category of recycled products, to the authors' knowledge, the model has not been empirically tested.
Values can be described as a belief which people base their actions on and how they evaluate certain situations [33].In the context of recycled products, there are two relevant values: biospheric and altruistic values [8].Both these values fall under the category of selftranscendence values [34].People with high biospheric values care about the effect of their actions on the nature and the ecosystem, whereas people with high altruistic values put the well-being of others first [35].In the original VBN theory, these two values could not be clearly distinguished from each other, resulting in the use of the biospheric-altruistic value orientation as one variable [36].However, following theoretical explanations, De Groot and Steg [33] provided empirical evidence that biospheric and altruistic values should be analyzed separately.Furthermore, Rahman and Reynolds [37] found that biospheric values are a stronger motivator for pro-environmental purchase intention than altruistic values.Biospheric and altruistic values are predictors of and have a positive influence on environmental concern [38].Environmental concern is a general stance and concern towards protecting the environment [39].Based on extant research that support the positive relationship between self-transcendence values and environmental concern, e.g., [40,41], the following hypotheses are formulated:

Hypothesis 1b (H1b). Altruistic values positively influence environmental concern.
Considering the belief part of the VBN theory, apart from environmental concern, awareness of consequences and ascription of responsibility fall under this category.Awareness of consequences is described as the understanding of individuals about the consequences that environmental problems entail [42].Ascription of responsibility, on the other hand, comprises who is blamed for the consequences of certain actions or situations [11].Personal norms are the self-constructed behavioral expectations that an individual puts on themself in different situations [43].Both, awareness of consequences and ascription of responsibility, are antecedents of personal norms, as they trigger the activation of this mindset [44].The Norm Activation Model [43] postulated a causal chain of relationships where higher awareness of consequences leads to a greater ascription of responsibility, which in turn increases personal norms [45].The dependent variable in this study is purchase intention.According to Sheeran [46] (p.1), "the most immediate and important predictor of a person's behavior is his/her intention to perform it".In the VBN framework, personal norms, when activated, have a direct influence on purchase intention and behavior, e.g., [47,48].Based on this, the following hypotheses are proposed: Hypothesis 2 (H2).Environmental concern positively influences awareness of consequences.

Perceived Risk Theory
Perceived risk, defined as "the expectation of losses associated with purchase" [49] (p.187), is an important factor in the consumer buying and choice process [50].Perceived risk has a long history in consumer research and is typically constructed of six types of risks: social, financial, physical, performance, time-related, and psychological risk [51].All six types of risk can also be present when purchasing recycled products.Table 2 shows perceived risks that could appear in the context of recycled products which are derived from existing literature.Worry or regret that the product was not the right choice [62,63] Extant research shows that consumers perceive green products as more expensive and of lower quality than conventional products [64], resulting in financial and performance risk.Furthermore, Essoussi and Linton [4] found that recycled products with high functional risk result in lower willingness-to-pay.Concerning the relationship of risk and personal norms, De Groot and Steg [65] discovered that perception of risk was a predictor of personal norms in the context of willingness to take actions related to nuclear energy.Additionally, previous research in the context of voluntourism showed that perceived risk acts as a moderator between personal norms and purchase intention that weakens the effect [15].In the present study, the author assumes that perceived risk has a negative influence and is added to the model as a moderator that alters the effect of personal norms on recycled product purchase intention.Based on this discussion, the following hypotheses are formulated: Hypothesis 6 (H6).Perceived risk has a negative influence on purchase intention.

Hypothesis 7 (H7).
Perceived risk negatively moderates the relationship between personal norm and purchase intention.

Measures
To measure consumers' intention to purchase recycled products, measurement scales and items were developed based on extant literature.Each construct was measured with reflective indicators.First, altruistic and biospheric values were assessed following De Groot and Steg [33] using six items adapted from Schwartz [66] value survey.Environmental concern was examined using three items from the short version of the New Environmental Paradigm scale [67] adapted from Stern et al. [11].To examine awareness of consequences, a three-item scale adapted from Ibtissem [68] and Jansson et al. [32] was used.Ascription of responsibility and personal norms were measured with three and four items, respectively, adapted from Ibtissem [68].Perceived risk was assessed using five items adapted from Wang et al. [69].Lastly, purchase intention for recycled products was measured with three items adapted from Lee and Lee [70].All items were measured on a five-point Likert scale ranging from 1, strongly disagree, to 5, strongly agree.Only altruistic and biospheric values were measured on a five-point Likert scale ranging from 1, very unimportant to 5, very important.For a complete list of the constructs, corresponding items and sources see Table 3.

Data Collection and Sample
To evaluate the VBN theory in the context of recycled product purchase intention, an online survey was conducted.The completion of the questionnaire took approximately ten minutes.Respondents were ensured that their answers are handled with confidentiality and anonymity.The survey included a short description of recycled products and four examples for better understanding: clothes from fabric remnants, jeans from denim scraps, notebooks and printer paper from recycled paper, and furniture from recycled waste wood.These examples were given to ensure that the focus of this study is on consumer goods and not on packaging.Data collection was conducted in September 2021.The panel provider Clickworker was used to retain a sample size of 200 respondents from Germany.To ensure survey quality, incomplete datasets and respondents who took less than five minutes to complete the questionnaire were deleted [71].A final sample of 177 respondents remained.Of these respondents, 48% were male and 52% were female, with the majority being in the middle age group 30-49 (62.1%), followed by the 50+ (20.3%) and the 19-29 year-old age group (17.5%).In terms of household income, we saw a fairly even distribution in the sample (13.6% up to 1500€, 11.3% between 1500€ and 2000€, 26% between 2000€ and 3000€, 18.6% between 3000€ and 4000€, 10.2% between 4000€ and 5000€, and 10.7% more than 5000€, with 9.6% not being specified).Most participants' highest degree of education was baccalaureate (28.8%), followed by secondary school (23.7), bachelor's degree (19.2%), diploma (13.0%), master degree (11.3%), and lower secondary school (4.0%).

Statistical Analysis
For the analysis and the calculations of the research model partial least squares structural equation modeling (PLS-SEM) in SmartPLS 3 [72] was used.This variance-based approach, instead of a covariance-based SEM, was chosen for several reasons: First, it allows small sample sizes while still achieving high levels of statistical power [73].Regarding sample size, the '10 times rule' provides that "the sample size should be greater than 10 times the maximum number of inner or outer model links pointing at any latent variable in the model" [74] (p.232).In this model the largest number of paths is five, validating that the sample size is well above the minimum number of participants.Second, PLS-SEM makes no particular assumptions about the data and allows nonnormal data [75].According to the Kolmogorov-Smirnov test, this is the case for this sample.Lastly, PLS-SEM is the preferred method for exploration and prediction [76].Even though the model of this study is based on an established theory (VBN model), this research can be considered exploratory, as it analyzes the influence of perceived risk in the framework.
To test the validity and reliability of constructs with the corresponding items in the measurement model, the PLS-algorithm path weighting scheme with 300 iterations and a stop criterion of 10 −7 [73] was used.For significance testing of the indicator loadings, as well as for path coefficients in the structural model, complete bootstrapping with 5000 subsamples was performed [73].

Measurement Model
All constructs included in the model show one-factor solutions with factor loadings well above 0.5, Cronbach's alphas above 0.7, and explained variances above 50%.Convergent validity was assessed using composite reliability (CR) and average variance extracted (AVE).All CR scores were above the threshold of 0.7 [77] and all AVE values were greater than 0.5 [78].Discriminant validity was determined using the heterotrait-monotrait ratio of correlations (HTMT, Table 4).All values are below the threshold of 0.85 [79].All constructs were successfully validated, and the results are presented in Table 3.To test for common method variance (CMV), a full collinearity test was conducted.The variance inflation factors (VIFs) for the latent variables were all well below 3.3, which suggests that common method bias is not an issue in this model [80].

Structural Model
To empirically test the model, structural equation modelling with SmartPLS 3 was used.The model explains 54.6% of variance in consumers' purchase intention of recycled products (R 2 = 0.546).Both biospheric values (H1a, β = 0.197, p < 0.05) and altruistic values (H1b, β = 0.269, p < 0.05) have a significantly positive effect on environmental concern.Environmental concern has a very strong, positive influence on awareness of consequences (H2, β = 0.570, p < 0.01).The effects of awareness of consequences on ascription of responsibility (H3, β = 0.493, p < 0.01) and ascription of responsibility on personal norms (H4, β = 0.346, p < 0.01) are also significantly positive.The last path in the model also shows a significantly positive influence of personal norms on the dependent variable purchase intention (H5, β = 0.455, p < 0.01).Additionally, perceived risk has a very strong, highly significant, negative influence on purchase intention (H6, β = −0.427,p < 0.01).The results are visualized in Table 5. Regarding the moderation effect, perceived risk is added as a moderator for the relationship of personal norms on purchase intention.A moderator is a third variable that alters the relationship of two other variables by either strengthening, weakening, or even changing the direction of the effect [73].Results show that perceived risk has a large effect (f 2 ≥ 0.025, [73,81]) on the relationship between personal norms and purchase intention, which is significant on a 10% level (t-value = 1.899).This means that higher perceived risk generally strengthens the relationship between personal norms and purchase intention, contradicting H7.

Control Variables
Statistical control variables were included into the analysis to preclude other possible explanations for the findings [82].It was controlled for gender, age, and social desirability.In the context of green product purchase intention, socially desirable responding could be a potential issue [83].To control for social desirability bias, a short, true or false, ten-itemversion of the Crowne & Marlowe [84] Social Desirability Scale adapted from Strahan and Gerbasi [85] was included.For the first five answers, true, and for the last five answers false indicated socially desirable answering.The sum of socially desirable answers was calculated, ranging from 0 to 10, with a higher score indicating social desirability.All three controls did not have a significant effect on the results (see Table 5).

Discussion
This study was designed to get a better understanding of consumers' purchase intention of recycled products and empirically test the influence of values, beliefs, norms, and perceived risk in this context.Results show that the VBN theory, and thus values, beliefs, and norms have strong predictive power for the purchase intention of recycled products.This is consistent with the findings of Kiatkawsin and Han [13] in the context of pro-environmental behavior and Quoquab et al. [30] for green purchase behavior of cosmetics.In general, this model of the VBN theory, including perceived risk, explains a high amount of variance of the purchase intention of recycled products (R 2 = 0.546), which is above the usual 19-35% that this theory explains [86].
Perceived risk, as expected, has a very strong, significantly negative effect on recycled product purchase intention.This is consistent with extant research for other green products, e.g., refurbished products [14].One counterintuitive finding is that perceived risk moderates the relationship between personal norms and purchase intention in a positive way.Previous research in the context of voluntourism shows contrary results, where if the perceived risk increases, the effect between personal norms and intention weakens [15].In the present findings, even though the perceived risk of recycled products has a direct negative influence on purchase intention, it acts like an activator for personal norms.If the perceived risk is high, personal norms has a stronger effect on purchase intention.A possible explanation for this is that the moral obligation to contribute to the protection of the environment outweighs the self-interest of consumers.

Theoretical Contributions
Considering theoretical implications, this study validates the VBN model for a recycling context.Even though the initial VBN model was developed to analyze behavior [11], the present study shows that the theory can also be used with purchase intention as dependent variable.Furthermore, the VBN theory is extended by integrating perceived risk into the model.The results show a significant influence of perceived risk on purchase intention, that should not be neglected in the consumer buying behavior research.Additionally, this study uncovered a significant relationship between perceived risk and the effect of personal norms on purchase intention.In this regard, the findings of the present paper underline the need to include and simultaneously analyze product and individual-related factors in research models [8].This combination would provide a deeper insight into consumers' green product purchase intention.

Managerial Implications
From a managerial point of view, it is crucial to minimize the perceived risk of recycled products, as it is a key factor for increasing sales and thus, for corporate success.Companies should apply different methods to increase trust in their green products, e.g., provide certificates or testimonials, work with influencers who are convinced of a products' worth, and ensure guarantees [87].Additionally, this study can help companies to improve their marketing strategy with content that encourages green consumer behavior.Advertisements should appeal to the awareness of consequences and ascription of responsibility of consumers, as these are important triggers of personal norm [44].If consumers are aware of the consequences that their behavior has and take responsibility for it, they are going to be more likely to buy green products.Lee et al. [88] found negatively framed information to be most successful for consumers' understanding, remembering and in turn performing sustainable behavior.Negatively framed messages convey the negative consequence of not performing a certain behavior, whereas positively framed messages focus on the positive outcome if a certain behavior is performed [89].For recycled product purchases, advertisements should appeal to the negative consequences that wasting natural resources have on the environment.These kinds of advertisements enhance personal norms, which is also validated in the context of recycling-aiding products [90].This is also relevant for institutions or government actions that are promoting green consumption to reach their climate protection goals.Additionally, regulations to oblige companies to use a certain percentage of recycled materials in their products should be implemented.

Limitations and Future Research
There are some limitations to this study that can be used as avenues for further research.First, only a linear causal chain of relationships between values, beliefs, and norms has been tested in this model.Further research might want to test the direct effects of all variables on purchase intention in order to compare effect strengths and find additional relationships in terms of moderators and mediators.This research also focused on purchase intention as the dependent variable; future studies should test the presented framework for actual purchase behavior.Since this questionnaire was solely distributed in Germany, future replications should be conducted in different countries to analyze potential cultural differences, especially on the relationship between values and environmental concern.Additionally, it would be interesting to explore the purchase intention of specific recycled product types, as this research focuses on recycled products in general.Testing up-and downcycling products could potentially also lead to different findings.Furthermore, the different types of risks associated with recycled products need to be analyzed as well as effective methods to reduce these risks.Lastly, there are many more individual, but also product and context related variables [8] and theories influencing purchase intention.These can be integrated into the VBN theory to get a more complete picture of consumers' recycled product purchase intention.Informed Consent Statement: All subjects gave their informed consent for inclusion before they participated in the study.

Figure 1
Figure 1 presents the research model in the context of recycled products.

Figure 1 .
Figure 1.Conceptual VBN Model for Recycled Product Purchase Intention.

Funding:
This research was funded by the Saarland State Research Funding Program, grant number 20/14 WiWi.Institutional Review Board Statement: Not applicable.

Table 1 .
Studies on factors influencing recycled product purchase intention.

Table 2 .
Types of Perceived Risk in Recycled Products.

Table 3 .
Construct Scales and Validity.

Table 5 .
Results of the Structural Analysis.