China having the highest population in the world not surprisingly produces one of the largest if not the largest amounts of solid waste. Promoting the reduction of solid waste sources and recycling to minimizes the negative impact on the environment is the advocated green urban development model. Recycling is “the process of collecting and processing materials that would otherwise be thrown away as trash and turning them into new products”¹. Governments and environmental public welfare organizations have invested many resources in cultivating, publicizing and promoting people’s recycling behavior, such as recycling facilities installed in public places, household waste classification and recycling policies, school education program and ubiquitous public service commercials, and propagandas. Many businesses have also launched second-hand goods recycling deduction plans (such as old clothes recycling plan by H&M) and idle goods circulation market (such as Taobao online idle goods market).

Resource recycling is an important mean to achieve sustainable development. However, the main focus on the work so far has been on factors affecting recycling behavior (e.g., Tonglet et al., 2004; Schultz et al., 2007; Goldstein et al., 2008; Trudel et al., 2016; Wang et al., 2016; Echegaray and Hansstein, 2017; Wan et al., 2017). These studies focused on garbage charges, identity exposure, information presentation, commodity appearance, individual attitudes, level of education, and other factors related to recycling habits. The insights from the these studies can help policymakers educate and persuade the public to participate in recycling activities, though the long term overall effect of these efforts is questionable. The basic assumption underlying such research is that garbage recycling is an effective way to prevent pollution, save energy, and save natural resources (Troschinetz and Mihelcic, 2009; Varotto and Spagnolli, 2017). Nevertheless, an initial sustainable act may lead the individual to perform unsustainable behaviors (Meijers et al., 2013), so peoples’ recycling efforts may increase their levels of future consumption, thus increasing overall resource use. For example, consumers who recycled their used clothes may feel that buying new clothes is more acceptable. If the potential negative consequences of recycling (such as promoting waste) cannot be avoided, it will be hard for policymakers to maximize the beneficial effects of recycling (Ma et al., 2019). Therefore, it is very important to study the subsequent resource saving behavior related to involvement in waste recycling.

The aftermath of recycling behavior has received limited investigation and the questions whether it encourages saving or wasteful behavior remains to be answered. Social psychology studies suggest that humans have motivation for consistent behaviors (Beaman et al., 1983; Burger, 1999; Mullen and Monin, 2016). The theory of self-perception suggests that people will infer their attitudes, beliefs and self-characterizations according to their previous behavior, and then make choices consistent with self-concept. Past environment sustainable behavior leads to the perception that “I am a pro-environmental person,” and this perception promotes the subsequent emergence of similar behavior (Bénabou and Tirole, 2011). Studies have found that moral behavior promotes subsequent moral behavior (Gneezy et al., 2012), and green consumption promotes green consumption or environmental behavior (Cornelissen et al., 2013; Summers et al., 2016). On the other hand, a large amount of literature illustrated just the opposite, i.e., when individuals took sustainable behavior before, they will reduce their sustainable behavior or engage in unsustainable behavior subsequently. For example, after consuming green products, the possibility of purchasing the green products in the future decreases (Mazar and Zhong, 2010). Compared with ordinary cars, hybrid cars drivers use cars more frequently, drive more mileage (Sun and Trudel, 2017), and have more traffic violations and accidents (Woodyard, 2009). Individual support for green funds is lower after garbage recycling (Truelove et al., 2016). When economic incentives were used to encourage household waste recycling, the consumption of electricity in the same household increased (Xu et al., 2018).

A recent study on how recycling behavior affects future consumption comes from Ma et al. (2019) who investigated the negative consequences of recycling behavior. Although previous studies have given conflicting conclusions, their results provide evidence to support the prediction that recycling activities indeed increase future resource consumption. This effect is mediated by two mechanisms, i.e., pride feeling and environmental self-identity, that decrease negative emotions from wasting behaviors. Feelings of pride, as a self-conscious emotion, play an important role in self-regulation. Environmental self-identity, the degree to which individuals regard themselves as environmentalists (Whitmarsh and O’Neill, 2010; Werff et al., 2013, 2014) were found to decrease negative self-attributions associated with wasteful consumption (Bolton et al., 2006; Ma et al., 2019). Furthermore, they indicated that the consideration of future consequences negatively moderates the effects of recycling efforts on pride feeling and environmental self-identity. Based on the above evidence, we tend to presume that individuals’ recycling efforts would reduce their saving behavior. Unfortunately, few studies have examined the relationship between recycling efforts and resource saving behavior. Existing studies have either only focused a specific consumption behavior (such as using paper cups or scrap paper) in an experimental research situation (Catlin and Wang, 2013; Sun and Trudel, 2017) or taken a proxy variable of resource saving, such as average monthly expenditure (Ma et al., 2019), which provided limited insights into the understanding of resource saving behaviors.

In this research, we empirically investigate these issues based on survey data from college students in China, using structural equation modeling. We argue that resource saving behavior is a set of behaviors rather than a single action and saving behaviors could be classified into two types, i.e., costly and costless saving behaviors. From a viewpoint of evolution, the perceptions of the costs of saving behaviors might play a substantial role for different individuals (Poškus, 2017). It is suggested that the self-oriented attitude reflecting personal gain may have different effects on environmental behavior than social-oriented attitude reflecting altruism. Although individuals sometimes show socially desirable behaviors such as carrying a reusable shopping bag and try to improve their social status at the cost of losing resources (costly signaling), at other times, they may be reluctant to pay a cost to save resources, so as to maintain their personal advantages (Gintis et al., 2001; McAndrew, 2002; Millet and Dewitte, 2007; Bereczkei et al., 2010). Therefore, the individual’s resource saving behaviors may not only be affected by the previous recycling behavior in terms of moral self-regulation, but also the costliness of recycling behavior can not be ignored. The current study’s framework are based on the Ma et al. (2019)’s study, and further extends it by examining two types of saving behaviors, which shows different consequences. In addition, we investigated the role of recycling cost when examining the effect of recycling efforts on pride feeling and environmental self-identity.

To investigate the relationship between recycling efforts and subsequent saving behavior, we used Ma et al. (2019) conceptual framework with modification of adding recycling costliness as a moderator. Furthermore, we used a set of saving behaviors to replace the single indicator (i.e., monthly living expenses) used by Ma et al. (2019), and found that the structure of saving behavior show two dimensions. Ma et al. (2019) proposed that recycling efforts can increase peoples’ consumption level, which is mediated by feelings of pride and environmental self-identity that can reduce negative emotions from resource wasteful behaviors. Firstly, as the study is based on the Ma et al. (2019) framework, we wanted to confirm the relationships observed in the original study and explore the relationships between recycling efforts, environmental self-identity, and pride feeling.

Further, we predict that the impact of recycling efforts on saving behavior depends on the costliness of saving behavior, that is, whether it is costly or costliness for individuals. Recycling efforts promote costless saving behavior, but they inhibit costly saving behavior, which is mediated by pride feeling, and environmental self-identity. In addition, we predict that the cost of recycling activities will moderate the effects of recycling efforts. Below, we discuss each concept in more detail and present our hypotheses.

The findings from empirical research consistently suggested that there is a positive correlation between ecological affect referring to the degree of emotionality an individual is attached to environmental issues and ecological behavior (e.g., green purchase; Chan and Lau, 2000). He et al. (2013) found that ecological emotion (such as pride, cherishing) played a mediating role in the positive impact of green cognition (i.e., knowledge about green products) on green behavior. Sun and Trudel (2017) argued negative emotions experienced during resource wasting behavior can be reduced by positive emotions arousing from the following recycling or resource saving activities. Thus, pride feelings are likely to decrease resource saving by reducing the negative emotions of wasteful behavior (Ma et al., 2019). Besides, based on social exchange theory’s rank equilibrium norm (Cropanzano and Mitchell, 2005), individuals’ feeling of pride arousing from their recycling efforts, makes them feel more entitled to make less other environmentally responsible decisions, such as feel comfortable about using more resources (Ma et al., 2019). Therefore, we propose:

Previous work on social cognition based on the concept that individuals have different identities (Reed, 2002; DeMarree et al., 2005) has proved that individuals’ prior behaviors can stimulate a certain self-concept and influence their subsequent behaviors. The moral licensing effect showed that previous moral behavior might activate and promote a positive self-concept, thus allowing consumers to make more of self-indulgent choices later (Khan and Dhar, 2006; Merritt et al., 2010). Following this, an increased feeling of environmental self-identity could act as one of the factors for decreasing negative emotions from wasteful behavior (Bolton and Alba, 2012; Sun and Trudel, 2017). Empirical research in the effect of moral licensing found similar evidence. For example, people who see themselves as typical recyclers are more likely to recycle than those who do not perceive themselves as recyclers (Mannetti et al., 2004), garbage recycling leads to lower green fund support (Truelove et al., 2016). In this research context, recycling efforts also have the potential to activate and confirm environmental self-identity, which can be a “get out of jail free card,” making high levels of consumption more acceptable (Whitmarsh and O’Neill, 2010; Ma et al., 2019). Therefore, engaging in recycling could boost environmental self-identity, which decreased the negative self-attribution associated with wasteful consumption (Ma et al., 2019) and, thus decreased the likelihood of saving behaviors.

Recycling activity is a collection of multiple actions rather than a single behavior, many of which require individuals to accept some costs, including financial costs (such as purchasing recycling equipment), physical costs (going to a specific recycling point), and mental burdens (garbage sorting), etc. Sun and Trudel (2017). Gneezy et al. (2012) proposed that the costliness of initial pro-social behavior is a key moderator of moral consistency between a sequence of behaviors. Costly pro-social behavior signals a pro-social identity, leading to moral consistency of sequential behaviors, whereas costless behavior does not, leading to no moral consistency. In their experiment, compared to the control condition, participants lied significantly less in the costly condition and significantly more in the costless condition. Participants in the costly condition also rated themselves as more helpful and less selfish than participants in either of the other two conditions, and the difference in truth-telling between the costly and the costless conditions was mediated by this self-rating of pro-social identity.

Accordingly, costly recycling behavior is more diagnostic about oneself, leading people to embrace the value indicated by that behavior (Burger, 1999; Gneezy et al., 2012). Individuals who accept the higher cost of recycling, which can prove that they are a person who abides by social norms, thus strengthening their environmental identity. In addition, if the recycling was costly, it means that consumers contributed more efforts for recycling, which will be converted into more moral credits or improve the quality of moral credentials (a mechanism of moral licensing). In the model built by Sun and Trudel (2017), higher recycling cost (for example, putting the recycling bin farther away in the field experiment) would produce stronger positive emotions. Thus, we hypothesize the following:

We constructed a conceptual model to better present the relationships between each constructs (Figure 1).

The subjects of this study were college students majoring in economics and management from Shaanxi University of Technology in China. Respondents who completed the questionnaire were rewarded with money of 10 RMB. A total of 442 questionnaires were sent out during the class and 436 valid sample was obtained. Among them, there were 135 males and 301 females. 58% of the population grew up in rural areas and 42% in urban areas. The M_age was 19.13 (SD = 1.02). Though the females are much more than males in our sample, we included control variables for gender (male/female), and growing background (rural/city), to capture unobservable differences.

We adopt PLS-SEM (partial least squares structural equation modeling) for our data analysis. Using PLS is appropriate because it does not assume normal distributions and allows for analyses with small samples (Hair et al., 2011). Besides, SmartPLS 3.0 can provide consistent results when all constructs are reflective (Dijkstra and Henseler, 2015). If properly used, PLS-SEM can provide more robust estimations of structural models than can covariance-based SEM (Reinartz et al., 2009). Using SmartPLS 3.0 software, a two-stage analytical procedure was applied to analyze the data (Hair et al., 2011). Firstly, we assessed the reliability and validity of the measurement model and then examined the parameters and the explanatory power of the structural model. The significance of the model estimates was based on a bootstrapping procedure with 5,000 samples.

This survey research provides evidence that individuals’ recycling efforts can increase their costless saving behaviors, while decrease their subsequent costly saving behaviors. We demonstrate that the efforts to recycle can influence subsequent resource saving by activating a positive pro-environmental self-identity, which gives individuals a license to take less costly saving behaviors, and at the same time, gives individuals a motive to take more costless saving behaviors. Research based on Construal Level Theory shows consumers save more when the saving goal is construed at a high level and tend to perceive specific goals as more difficult (Ülkümen and Cheema, 2011). Consumers usually don’t care about the costless saving behaviors in their daily life, which means that their psychological distances of these saving activities are distant, so they will construal these saving activities in a more abstract way. But for costly saving behaviors, such as those requiring some physical or mental effort, consumers will be more concerned with them and thus have a closer psychological distance, so they will construal them in a more concrete way. In one word, costliness is a key factor for individuals to make choices that are inconsistent or consistent with their prior recycling behaviors. In addition, we find that recycling cost negatively moderates the relationship between recycling efforts and pride feelings. It is worth noting that we conducted the same research on 224 college students in another university located far away, and the results still support this conclusion, which shows the robustness of the research conclusion.

The original contributions presented in the study are included in the article/supplementary material, further inquiries can be directed to the corresponding author/s.

The studies involving human participants were reviewed and approved by Academic Committee of Center for Behavior and Decision Shaanxi University of Technology. Written informed consent for participation was not required for this study in accordance with the national legislation and the institutional requirements.

SS and SW: conceptualization. YW: data curation. JX: formal analysis. SW and SS: investigation. JX: methodology. SS: project administration. SS and YZ: supervision. YW: writing – original draft. YZ: writing – review and editing. All authors contributed to the article and approved the submitted version.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.