Licensing Effect in Sustainable Charitable Behaviors

Abstract

The theory of licensing effect suggests that consumers tend to perform self-interested or self-indulgent actions after undertaking altruistic behaviors. How do past altruistic experiences affect the willingness of consumers to perform charitable behaviors in the future? Results from an exploratory approach comprising three laboratory studies and one field experiment demonstrate the existence of licensing effect in charitable conditions. We find that consumers are more unwilling to undertake charitable activities when they recall past similar experiences. The donation resources (time/money) do not influence the licensing effect. Two other variables moderate the size of the licensing effect: the way in which the initial charitable behavior is recalled (abstract vs. concrete) and the attribution for initial charitable behavior (collective vs. individual). We find that consumers are more reluctant to carry out charitable behavior when: (1) they recall the concrete details rather than the abstract goal of past activity; (2) consumers are praised for individual efforts rather than collective contribution in past activity. These findings offer new theoretical insights into the licensing effect in consumers’ charitable behaviors and set out practical implications for the sustainability of charitable programs.

1. Introduction

Charitable behavior usually refers to non-profit behavior initiated by individuals or organizations to provide resources to specific social groups [1]. Given that individual donors account for around 67% of annual charitable contributions, it is important for fundraisers to win the favor of individual donors [2]. Most relevant works in the literature first began to study the factors involved in charitable giving at the end of the 1970s: these studies investigated aspects such as the effect of consumer characteristics, forms of donation projects, and features of the hosting enterprises [3,4,5]. During the past year, the spread of COVID-19 has again underlined the importance of public welfare participation during emergencies.

Most of the marketing literature focuses on an individual’s single donation decision. There is therefore a lack of studies on sustainable charitable behavior, except for a few works on longitudinal decisions [6]. However, the purpose of charitable programs is to redistribute resources in order to alleviate inequalities in social development. A single, or several activities cannot fundamentally improve the situation of the recipients and long-term involvement in public welfare should be advocated. This requires deeper investigation of sustainable charitable participation at the level of the individual. Therefore, we decided to investigate the impact of past charitable experiences of individuals on their present charitable decisions. Nowadays, many public welfare projects involve fund-raising, and individual consumers are the main sources of time and money. Here, we aim to determine how to increase the willingness of individual charitable behavior rather than viewing altruistic behavior as “one-time” consumption behavior.

Accordingly, we mainly refer to licensing effect theory to establish our hypotheses. The licensing effect refers to behavioral tendencies wherein people engage in self-interested or even immoral actions after completing ethical or altruistic behaviors [7,8,9]. We use this theory mainly for two reasons: First, charitable behavior is a kind of moral and altruistic behavior that differs from traditional consumption behavior for reasons of self-interest. Licensing effect theory points out the influence of morality and altruism on consumer psychology, which is suitable for the study of charitable behavior. Second, licensing effect theory emphasizes how past decisions affect present behavior, which corresponds to the focus on sustainable charitable behavior.

This research contributes to the literature and theory in two ways. First, the current literature on the licensing effect focuses on how individual moral behavior affects subsequent purchase intentions for certain products [10,11,12]. Few studies investigate the licensing effect when past and present behaviors involve different charitable activities. We find that the licensing effect does exist between charitable behaviors, and certain factors affect the degree of licensing effect, which provides a theoretical reference for future research on sustainable altruistic behavior. Second, we refer to construal level theory to analyze the moderators of licensing effect. We find that the size of the licensing effect is related to the way individuals interpret past public welfare activities: when consumers recall concrete details of past charitable activities, such as the specific individual efforts, they are more reluctant to perform sustainable charitable behaviors.

Our findings also provide practical implications for relevant organizations and enterprises. For government departments or non-profit organizations dedicated to public welfare, it is important to consider how to promote the sustainability of charitable programs because long-term investment is required to effectively alleviate social imbalance. For companies engaged in event marketing related to social responsibility, extant studies show that hosting charitable campaigns can help build a positive social image or promote consumers’ willingness to purchase a company’s products [13,14,15]. Based on our findings, organizers could adjust the publicity strategy of charitable projects to increase consumers’ willingness to donate or volunteer. For example, the organizer could highlight the overall contribution of the whole organization in order to attract new participants.

The article is organized as follows: We first review the literature on licensing effect theory and past studies on donation willingness to develop our hypotheses. Then, we perform one exploratory study (i.e., in-depth interview), three laboratory studies and one field experiment to test the hypotheses; the findings confirm the existence of licensing effect in charitable behaviors and the possible moderators of the licensing effect. The article concludes with a discussion of theoretical contributions, managerial implications, limitations, and directions for future research.

2. Theoretical Background and Hypotheses Development

2.1. Main Theory: Licensing Effect and Charitable Behaviors

Licensing effect (i.e., moral licensing, self-licensing) refers to a psychological phenomenon in which individuals are inclined to be self-interested or self-indulgent and undertake immoral or questionable behaviors after performing socially desirable, altruistic, or moral behaviors [7,9]. For example, consumers may choose unhealthy snacks after they have been on a diet for some time. The previous action, either moral or prosocial, seems to provide a license for the individual’s subsequent dubious behavior.

Researchers first applied the theory of licensing effect to the study of prejudice. Monin and Miller (2001) find that those who had strongly refuted discriminatory statements tended to choose a white person rather than a black person in the following job-hiring task [8]. Subsequently, licensing effect was explored in the context of other fields in psychology and marketing, including consumer’s personal purchase decisions and willingness toward ethical behavior. For example, consumers tend to purchase hedonic goods after they imagine participating in volunteer activities [11]; individuals become more likely to cheat in experimental tasks after purchasing green products [16] or previously performing ethical behaviors [17,18].

The core tenet of licensing effect is that individuals have “conflicting motives” [9]. This refers to the conflict between self-interest and self-regulation, which appears when the person is hesitating about whether to perform the target behavior. Licensing effect usually includes at least two decisions: an initial behavior as the licensing condition (e.g., having participated in a voluntary program), and a following behavior as the target variable (e.g., decision on whether to participate in other voluntary activities). Given that this study and many previous studies on licensing involve a charitable or prosocial situation, it is necessary to clarify the definition of charitable behavior for the purposes of our study. Referring to the definition of charitable giving in Bekkers and Wiepking (2011) [19], we describe individual charitable behavior as the donation of resources (money, goods, or time) to others beyond one’s own family. Because the value of donated goods can also be measured in monetary terms, we only investigate the donation of money and time in this study, which will be discussed in more detail later. Note that charitable behaviors are included in prosocial behavior, yet the latter also consists of other ethical behaviors such as sustainable or green purchase decisions, which will not be our main focus in this study.

Some studies have placed charitable behavior in the initial context and focused on how it affects willingness to purchase certain types of goods such as luxury and green products [10,11,12] or the tendency towards racial prejudice [20]. Other studies have explored an individual’s willingness to undertake charitable behavior as the target variable in examining the licensing effect [11,17,21,22,23]. However, there is still a lack of research on licensing effect in the context of sustainable charitable donations. The previous literature has proven the existence of licensing effect in the moral context of people tending to engage in environmentally unfriendly behavior after charitable donations [24]. We further propose that licensing effect could still exist when the initial behavior and target behavior both involve charitable donations.

Moreover, the initial condition that drives the licensing effect could even be an unfulfilled intention or imagination of one’s prosocial behavior rather than real actions [7,11,25]. Some studies show that participants who merely imagined undertaking volunteering community service later tended to be more self-interested [11,26]. Therefore, we believe that simply imagining future charitable behavior may also affect the present target decision, triggering the licensing effect. These above considerations all lead to the following hypotheses:

H1a. 

Individuals are more reluctant to perform charitable behaviors after recalling past similar experiences.

H1b. 

Individuals are more reluctant to perform charitable behaviors after imagining similar activities in the future.

2.2. Moderators of Licensing Effect: Construal Level Theory

Licensing effect is a relatively new research field that has emerged over the past two decades. Some papers reviewed past studies and proposed different frameworks of relevant moderators. Merritt, Effron, and Monin (2010) [25] focus on the framing of prior behavior and the ambiguity of the target behavior. They suggest that licensing effect is more likely to occur when the individual views prior behavior more as a process rather than a commitment towards the goal, or when the target behavior is more ambiguous in terms of ethics. In a meta-analytic review, Blanken, van de Van, and Zeelenberg (2015) [27] founnd no empirical evidence for potential moderators such as the type of licensing condition (trait vs. action) or type of target behavior (actual or hypothetical). Mullen and Monin (2016) [9] summarize five categories of possible moderators: level of construal, progress vs. commitment, identity relevance, value reflection, and ambiguity. Although it is difficult to find an overarching theory that describes all the moderators of the licensing effect, we recognize that the interpretation of initial behavior and the correlation between initial and target behaviors may influence the level of licensing effect. Construal-level theory (CLT) elucidates different ways of interpreting a behavior; therefore, we introduce this theory to investigate a subset of moderators in this study.

CLT proposes that people have different levels of understanding of an event, influenced by psychological distances such as time, spatial, or social distance. Individuals tend to interpret distant events through a focus on abstract features (i.e., high construal level) and describe low-distance events in terms of concrete aspects (i.e., low construal level) [28,29,30]. When describing an event, people with high construal level pay attention to the purpose and value, while those with low construal level focus on specific context and details [28,31]. For example, reading can be either interpreted as “acquiring knowledge” (high construal level) or “reading the lines of text” (low construal level).

With reference to CLT, we investigate three moderators of licensing effect: consistency of donation resources, way of recalling initial behavior (abstract goal vs. concrete details), and attribution of initial behavior (collective effort vs. individual effort), which will be discussed in the following paragraphs.

2.3. Donation Resources

Money and time are two important limited resources for most consumers, and charitable behaviors involve the transfer of these resources [1]. The donation of time refers to volunteer service to specific social groups such as disabled people and victims of natural disasters. Many researchers have compared consumer’s attitudes towards money and time [32,33,34]. Macdonnell and White (2015) [1] find that consumers usually have a more concrete mindset when considering money than time, although money is no longer viewed as concrete when it is abundant, which implies the perception that money can be changed easily in specific contexts. We speculate that consumers become more abstract-minded when the resource—either time or money—is abundant. On the contrary, when consumers are aware that their resources are insufficient, they have a lower construal level and think more concretely.

Compared to different types of resources, the same type of resource can be more psychologically closer to individuals due to the familiarity. According to CLT, people tend to construe things at a low level when psychological distance is short, which in turn results in a concrete mindset [35,36]. Additionally, an individual considers the feasibility of an activity with a concrete mindset and the values or desirability with an abstract mindset [28,29,35]. Therefore, we predict that when individuals need to donate the same type of resources in both the initial and target charitable activities, they think about the feasibility of decision (e.g., whether I have enough resources to perform the following activity after I spend some resources in the prior one) and become more reluctant to demonstrate subsequent donation behavior. Compared with those who donated time before, consumers who donated money may be more hesitant when they are asked to donate money again. The licensing effect is stronger when the type of resources donated (time or money) is the same in both the prior and present charitable behaviors. Thus, we propose the below hypothesis.

H2. 

When the donated resources (time or money) are consistent in the initial and target decisions, individuals are more reluctant to perform charitable behaviors.

2.4. Recalling Initial Behavior: Abstract vs. Concrete

Here, we focus on the interpretation of initial behavior. The previous literature, although lacking in experimental design, has proposed that the construal of initial behavior is a possible moderator of licensing effect [9,25]. Conway and Peetz (2012) [22] found in experiments that participants who recalled recent moral behaviors were less willing to volunteer than those who recalled temporally distant behaviors. This was because recent behavior activated a more concrete mindset than distant behavior. According to CLT, when individuals have a lower construal level (i.e., think concretely), they consider details and feasibility instead of reflecting on their values and reasons, thus having less self-control and licensing themselves to be more indulgent later [37].

We believe that when individuals recall the initial charitable activity by focusing on details such as the arrangement or specific experience during the event, they are more likely to exhibit licensing effect than those who pay attention to abstract values such as the purpose, reason, or mission of the event. Clot, Grolleau, and Ibanez (2016) [26] found that there was no obvious licensing effect when environmentalists participated in voluntary green activities because they regarded environmental behaviors as their commitment to personal values, and it was easier for them to perform self-disciplined behaviors. When consumers recall the goals and purposes of past charitable behaviors, they may connect their moral behavior with long-term commitment and become more altruistic. Therefore, we predict that:

H3. 

When individuals recall the details (rather than the abstract goals) of past charitable behaviors, they are more reluctant to participate in subsequent charitable activity.

2.5. Attribution of Initial Behavior: Collective vs. Individual

When attending charitable programs, individuals often receive priming messages that call for their contribution to the program. Kouchaki and Jami (2018) [12] find that praising consumers’ personal contributions after charitable activities would lead them to choose hedonic products yet emphasizing the total contribution of the volunteer organization would not cause a strong licensing effect. Other studies have also suggested that attributing past moral behavior to oneself would trigger the licensing effect and result in more self-interested behaviors [11,38].

We believe that the underlying mechanism of licensing also comes into play when the target behavior is charitable. Consistent with CLT, when an individual infers the previous activity to her/his own effort, she/he may think more concretely by recalling peripheral details such as personal experiences and is more likely to license herself/himself. In contrast, when an individual realizes that the contribution of past activity is mainly due to collective effort, she/he may have an abstract mindset and reflect more on the core value and commitment of the charitable organization, and thus be less inclined to engage in self-indulgent behaviors. Therefore, we propose that:

H4. 

When individuals associate the initial activity’s contribution with personal efforts (rather than the effort of a group or organization), they are more reluctant to participate in subsequent charitable activity.

3. Overview of Studies

This article adopts an exploratory approach by combining three laboratory experiments and one field experiment to test the above hypotheses. Studies 2–5 tested hypotheses in an experimental way, and we followed the previous literature when choosing sample size [10]. Based on Cohen and Jacob (1988) [39], we used PASS 15 software to calculate the required sample size. Study 1 supported the formulation of hypotheses in a qualitative way and provided guidance for follow-up experimental design. Study 2–5 tested hypotheses in a quantitative way, and we followed the previous literature in determining sample-size choice [10]. Study 2, which examined our main effect, proved that licensing effect could exist between charitable behaviors (H1a and 1b). Studies 3–5 further tested three possible moderators of the main licensing effect. Study 3 examined the relationship between initial and subsequent charitable behavior and investigated the influence of consistency of the donated resource (H2). Contrary to our prediction, the consistency of resources did not significantly affect the size of licensing effect. Studies 4 and 5 both focused on the understanding of the initial charitable behavior. Study 4 examined how the way of recalling past behavior—more specifically, the level of abstractness of recalling—would affect the licensing effect (H3). Study 5 was conducted in a field setting and investigated the influence of attribution of initial activity (H4).

To test the licensing effect, we compared the control group with groups having licensing conditions. Mullen and Monin (2016) [9] define licensing as the phenomenon wherein “a positive initial behavior yields less positive target behavior than a neutral baseline condition.” In accordance with previous related studies [11], the neutral baseline condition (i.e., control group) in our study did not include a filler task before the target condition in order to avoid losing the participants’ attention. The licensing effect appears when there is a significant difference in the target behavior between the licensing group and the control group. To make sure participants do not associate the target behavior with a prior one, we presented each behavior on a separate page and asked participants to guess the purpose of experiments to rule out invalid samples. All data in the following experiments were organized and analyzed using SPSS software.

4. Study 1: In-Depth Interviews

Study 1 invited 10 interviewees (60% female, mean age = 22.1 years) who had charitable experience before to conduct one on one interviews, each interview lasting around 30 min (see Appendix A’s in-depth interview outline). The specific information of interviewees is shown in

Table 1. List of Interviewee.

No	Name	Age	Education	Charitable Experience
1	Yu	24	Master’s degree	Frequently participating and organizing charitable programs (e.g., voluntary teaching).
2	Yang	23	Master’s degree	Frequently participating in charitable programs (e.g., voluntary clinics).
3	Lu	21	Bachelor’s degree	Occasionally participating in charitable programs (e.g., caring for children).
4	Wang	20	Bachelor’s degree	Occasionally participating in charitable programs (e.g., voluntary teaching).
5	Jing	24	Master’s degree	Frequently participating in charitable programs (e.g., voluntary narrator in museum).
6	Ying	21	Master’s degree	Frequently participating in charitable programs (e.g., donation, helping old people).
7	Min	22	Bachelor’s degree	Occasionally participating in charitable programs (e.g., voluntary teaching).
8	Yuan	20	Bachelor’s degree	Frequently participating and organizing charitable programs (e.g., voluntary teaching).
9	Bo	25	Master’s degree	Frequently participating in charitable programs (e.g., helping disabled children)
10	Guo	21	Bachelor’s degree	Frequently participating and organizing charitable programs (e.g., donating to rural kids)

.

According to the results of in-depth interviews, we found a majority of interviewers are reluctant to participate in similar charitable programs after recalling their charitable experiences, which is consistent with H1a. In terms of the reason, Wang reported that sequential participation could be tiring, and former charitable behavior has already met the requirements of college credits, thus he preferred to put time into other things.

After imagining a money donation scenario, a large proportion of interviewers refused to donate again but were willing to put in time and energy. Min mentioned that she would refuse to join the charitable program if it asked her to do the same thing and devote the same resources. The results partly support H2 that individuals are more reluctant to perform charitable behaviors when donated resources are consistent.

Likewise, most interviewers revealed that they would be more reluctant to participate in charitable programs after recalling details compared to recalling abstract goals, which verifies H3. In explaining the reasons that she preferred abstract goals, Yu remarked that the abstract values of past experiences reminded her of helping people in need and made her feel fulfilled and self-improved.

Based on the interviews, most participants rejected charitable activities that are associated with personal efforts compared to those with organizational efforts, which are consistent with H5. Keywords such as “tiring” and “unfair” are commonly seen in interviewees’ descriptions of their impression towards charitable activities involving more personal efforts. Keywords such as “effective” were usually mentioned when they were describing charitable activities involving more organizational efforts.

It is also interesting to note the interviewees’ views on commercial enterprises that perform charitable behaviors. While most interviewees reported they were aware that companies contributed to charity for commercial reasons, they still thought these companies had a stronger sense of social responsibility and deserved high evaluation. Yang also mentioned that he would be more likely to purchase products from companies engaged in charitable activities. In addition, interviewees also reported that the government could play a better role in charitable activities. For example, Guo suggested the local government should put more emphasis on regulation and publicity because many programs failed to deliver good results due to lack of ability and publicity.

5. Study 2: Main Effect of Licensing in Charitable Behaviors

5.1. Method and Procedure

Study 2 employed a between-participant design with three conditions (recalling past charitable behavior vs. imagining future charitable behavior vs. control). Participants were randomly assigned to one of these three conditions. We distributed the surveys on an online data platform Credamo (credamo.com), which provides similar functions to Amazon’s Mechanical Turk. We recruited 236 participants (58.9% female, mean age = 21.66 years), all undergraduates or graduates from mainland China.

For the recalling group, we first let participants recall a past volunteer experience they had attended and asked them to answer four questions to strengthen their impression. The questions were “Who was the organizer of this charitable program?”, “How much time did you spend on this activity?”, “What was the process of this activity (please explain in detail)?”, and “How did you feel about this activity before and after the experience?” For the imagining group, we first let the participants imagine the following scenario “Your neighborhood committee plans to hold a regular volunteer service next year. Now the community is seeking advice about the type of volunteer service that residents would like to attend.” We then asked participants to express their opinions by answering four questions, namely “Which kind of organizer do you trust for launching community service?”, “What kind of volunteer activity would you like to do?”, “How long would like to spend for each activity?”, and “How frequent do you think the volunteer service should be?” The control group skipped the above process and proceeded directly to the following task.

Then, all participants were told to imagine that “The student union in your university plans to hold a volunteer teaching program during next winter or summer break. The program will take place in the primary and middle schools in underdeveloped regions near Yunnan and Guizhou provinces. The student union is recruiting student volunteers now.” We then asked participants to decide how many days they would like to spend on the volunteer program, which was used to evaluate the willingness to perform subsequent charitable behavior.

5.2. Results and Discussion

A one-way ANOVA revealed significant differences between these three groups (F = 5.348, p = 0.005 < 0.05), as shown in Figure 1. Participants who recalled past charitable behavior (M = 18.69, SD = 16.588) and those who imagined future charitable behavior (M = 19.03, SD = 19.304) both indicated willingness to spend fewer days in the volunteer teaching program than the control group (M = 28.54, SD = 26.969). Both the recalling group and the imagining group were more unwilling to perform subsequent charitable behavior than the control group (t recalling vs. control = 2.746, p recalling vs. control = 0.007 < 0.05, t imagining vs. control = 2.544, p imagining vs. control = 0.012 < 0.05), thus supporting H1a and H1b. Study 2 demonstrates that the licensing effect may occur after individuals recall or imagine charitable behaviors. Next, we tested the moderating roles of resource consistency in Study 3 to determine how it might influence the size of licensing effect in charitable contexts.

6. Study 3: Consistency of Donation Resources

6.1. Method and Procedure

Note that in Study 2, the initial and subsequent behaviors are both volunteer services that involve the allocation of time, while in Study 3, we aim to determine whether the licensing effect still exists when the types of donated resources are inconsistent. Participants in Study 3 were randomly assigned to one of three conditions (recall money donation vs. recall time donation vs. control). The survey was also distributed through the online data platform Credamo; 142 participants were recruited online from mainland China (64.1% female, mean age = 21.08 years).

For the group recalling donation of money, we first asked participants to recall a past money donation program they had experienced. Then we let them answer four questions, which were “Who was the organizer of this charitable program?”, “Who were the recipients of the donation?”, “How much did you donate for this program?”, and “How did you feel about this activity before and after the experience?” For the group recalling time donation, we asked participants the same questions as the group recalling past charitable behavior in Study 2. The control group skipped the above process.

Then, all participants were told to imagine that “The student union in your university plans to hold a donation program for the primary and middle schools in underdeveloped regions near Yunnan and Guizhou provinces…The money would be used for providing free lunches for the children.” We then asked participants to decide how much money they would like to donate to this “free lunch” program. Note that the past and subsequent donated resources are consistent (inconsistent) for the group recalling past money donation (time donation). We chose money donation as the subsequent behavior in this study to increase the robustness of our hypothesis, given that we used the amount of time as the dependent variable in Study 2.

6.2. Results and Discussion

A one-way ANOVA found significant differences between these three groups (F = 2.694, p = 0.071; Figure 2). Participants in the control group (M = 78.26, SD = 66.543) were more willing to donate money than both the group recalling donation of money (M = 53.87, SD = 46.298, t recalling money vs. control = 2.048, p recalling money vs. control = 0.044 < 0.05) and time (M = 56.67, SD = 52.011, t recalling time vs. control = 1.754, p recalling time vs. control = 0.083). The results showed that individuals were unwilling to donate money after recalling past money donation or volunteer work, which means that the licensing effect again occurred when the initial and subsequent behaviors were all charitable behaviors, supporting H1a again. However, there was no significant difference between the groups having prior charitable behaviors (t recalling money vs. recalling time = −0.278, p recalling money vs. recalling time = 0.781). In other words, the resource consistency did not significantly influence the size of the licensing effect. H2 was not supported. We discuss the reasons in more detail in the conclusion section.

7. Study 4: The Way of Recalling Initial Behavior

7.1. Method and Procedure

Study 4 was designed to test whether an individual’s way of recalling initial behavior (abstract vs. concrete) moderates the size of the licensing effect. Participants were randomly assigned to one of three conditions (abstract-mind vs. concrete-mind vs. control). For the abstract-mind (concrete-mind) condition, we let participants recall their past charitable activity abstractly (concretely) in order to test the influence of the way they recalled their initial activity on the licensing effect. The survey was distributed on other online data platforms called Wenjuanxing and Tencent Wenjuan, both similar to Credamo and Mechanical Turk. Two hundred and forty-seven participants (77.7% female, mean age = 27.86 years) were recruited in this study, all undergraduates or graduates from Mainland China.

The abstract-mind group was first asked to recall a past charitable experience and asked to answer three to four questions focusing on the abstract goal of the activity, such as “Who was the organizer of this charitable program?”, “What was the purpose of this activity?”, “Do you think this activity would finally achieve its goal?”, and “What do you think about the purpose of this activity?” For the concrete-mind group, participants were also asked to recall a past charitable experience and answer questions, in detail, about the content of the activity. Questions included “Who was the organizer of this charitable program?”, “How much time did you spend on this activity?”, “What was the detailed process of this activity?”, and “What was the most impressive thing about this activity?” The control group skipped this recalling process and directly proceeded to the following task.

Then, all participants were told to imagine that “The student union in your university plans to hold a regular community volunteer program for each week next semester. The program aims at supporting the mentally disabled people in the community and keeping them company by providing courses targeting at elementary math, arts, and sports.” We then asked participants to decide how many hours per week they would like to spend on the volunteer program, which was used to evaluate the willingness to engage in subsequent charitable behavior.

7.2. Results and Discussion

A one-way ANOVA revealed significant differences between the three conditions in regard to the willingness to volunteer (F = 2.666, p = 0.072 < 0.1). The concrete-mind participants were more reluctant to perform regular community service (M = 4.71, SD = 3.231) than both the abstract-mind condition (M = 5.41, SD = 5.035) and control condition (M = 6.44, SD = 5.751). As shown in Figure 3, there was a significant difference between the concrete-mind and control conditions (t = 2.344, p = 0.021 < 0.05); however, the difference between abstract-mind and control groups was not very significant (t = 1.227, p = 0.222 > 0.1), implying that the participants might not have experienced the licensing effect when they focused on the abstract goal or purpose of past charitable behavior. H3 is thus supported.

8. Study 5: Attribution for Initial Behavior

8.1. Method and Procedure

Study 5 examined whether the attribution for the outcome of the initial behavior (collective effort vs. individual effort) would influence the size of the licensing effect. Participants were randomly assigned to one of three conditions (group-focus vs. individual-focus vs. control). For the individual-focus (group-focus) condition, the participants were led to think about their individual contribution (a volunteer organization’s total contribution) to the charitable activity.

This study was a field experiment that took place during a summer volunteer teaching program held by a university in mainland China. This summer program is viewed as a practice curriculum that is mandatory for all students in a school of this university. On the last day of the volunteer program, we asked each student volunteer to complete an online survey about the program that included our experiment and other questions related to the evaluation and suggestions about the volunteer program; 159 undergraduate students participated in the survey (62.9% female, mean age = 21.1 years). The questions were designed and distributed through the online survey platform Qualtrics (qualtrics.com).

We prepared three different priming messages at the beginning of the survey. In the individual-focus condition, participants were praised for their own efforts and individual contribution. They were also told that their own thoughts and suggestions were valued for improving the performance of future programs, which was said to be the purpose of this survey. In the group-focus condition, participants were informed about the mission and commitment to hold this annual program, and that the purpose of this survey was to help the organizer to improve how future programs were carried out. For the control group, the message was briefer, without excessive emphasis or any highlights; this manipulation is shown in Appendix B.

After reading the message for some time, the participants were allowed to click a “continue” button to answer the remaining questions. Although most questions were irrelevant to the purpose of our study, we included one question in the survey: “For the volunteer teaching program next summer, those student volunteers may need guidance and assistance from previous participants. How much are you willing to participate as assistant for the program next year?” A seven-point scale was used to test the willingness to volunteer next year (1 = absolutely will not participate, 7 = will absolutely participate). Given that the teaching program is held annually and some students do, in practice, participate more than once to volunteer, we believe that the responses reflected the students’ actual thoughts.

8.2. Results and Discussion

A one-way ANOVA revealed differences between the three conditions in regard to the willingness to volunteer next year (F = 2.766, p = 0.066 < 0.1). The individual-focus participants were more reluctant to volunteer next year (M = 3.63, SD = 1.449) than the group-focus participants (M = 4.11, SD = 1.565) and the control group (M = 4.26, SD = 1.359), which indicated that the attribution for past activity would affect the level of licensing. Moreover, as shown in Figure 4, there was a significant difference between the individual-focus and control conditions (t = 2.279, p = 0.025 < 0.05); yet, the difference between group-focus and control conditions was not significant (t = 0.482, p = 0.631 > 0.1). Participants who focused on individual efforts put into past charitable experiences tended to be unwilling to perform more charitable behaviors later, yet there was no licensing effect when individuals concentrated on the mission or joint group effort involved in the initial behavior. H4 is thus supported.

9. Conclusions

9.1. General Discussion

The present research investigates the licensing effect between past and current charitable decisions and explores the possible moderators that affect the level of licensing. Study 1 provides evidence for our four hypotheses in an exploratory study. Study 2 provides empirical evidence for our main hypothesis that the licensing effect may occur in sustainable charitable behaviors. Study 3–5 explores three moderators of the licensing effect under charitable conditions: the consistency of resource type, the abstractness of recalling past behavior, and the attribution of past behavior. Study 4 shows that consumers who focus on individual efforts in past behaviors may license themselves in subsequent charitable decisions. In Study 5, we designed a field experiment and found that consumers who recall past concrete details would experience the licensing effect more easily than abstract-minded individuals.

Study 3 shows that, contrary to our hypothesis, consumers are reluctant to perform the subsequent charitable behavior regardless of the consistency of donated resources. The meta-analytic research of [27] may provide a reason to support our results. They expected that licensing effects would be larger when the initial and target behaviors are in the same domain, yet later found that there was no difference between behaviors in the same domain versus those in a different domain through the meta-analytic review. Similarly, we treated the difference in donated resources as the difference of domains between behaviors, and our results corresponded with those of the meta-analytic study. One of the underlying reasons may be that the domain difference takes effect only when the target behavior is questionable and ambiguously bad [27,40], whereas our research only explores moral charitable behaviors.

9.2. Theoretical Contribution

This research makes contributions to the existing literature on the licensing effect. Previous studies focused more on consumption decisions affected by previous altruistic behavior [10,11,18,41,42] or used charitable behavior as a target condition in the experiments of licensing effect [11,18,22,23,26]. Our study is among the first to investigate the licensing effect between past and current charitable behaviors and proves the existence of licensing in a charitable context, which could be used as a theoretical reference for future research on the licensing effect in prosocial behavior.

This paper also contributes to the theory of consumer psychology by connecting the licensing effect with CLT to further explore the possible moderators of the underlying psychological mechanism. Combining the findings from Study 4 and 5, we find that the individuals are more unwilling to perform charitable behavior when they recall the details of past similar behavior or reflect on individual efforts into past activity. These findings all suggest that the size of the licensing effect increases when consumers have a lower construal level.

We believe that these findings also correspond to the characteristics of charitable programs: most public charitable programs nowadays emphasize individual efforts. A charitable organization is usually established to gather and allocate money from individual donations or to assign tasks to individual volunteers. A sustainable charitable organization strives to unite individual participants by depicting the mission and goals of its programs. Therefore, individual consumers may be more willing to perform sustainable charitable behaviors when they consider abstract concepts such as the organizational mission, the bigger picture of the community, and the future of society rather than specific personal gain and loss. Moreover, we find that the donated resources may not be a moderator of licensing effect; this result provides suggestions for future research.

9.3. Managerial Implications

This research also has important managerial implications for government departments, non-profit organizations, and corporate charitable programs. Public charitable programs held by government sectors and non-profit organizations usually aim to alleviate the imbalances in social development. Therefore, charitable projects are usually designed to be oriented in the long term, and it is important to investigate the sustainability of charitable behaviors.

For commercial enterprises, public welfare projects reflect the good image of enterprises and their corporate social responsibility. The term “cause-related marketing” refers to a company’s decision to donate to specific charitable programs when customers purchase the company’s products or participate in certain campaigns held by the company [43]. Subsequently, more researchers have regarded all kinds of charitable activities organized by companies, or the economic, legal, and moral responsibilities that a company undertakes to meet stakeholders’ expectations, as the broad concept of cause-related marketing [44,45,46]. Many studies have shown that consumers generally have a better impression of enterprises that have carried out charitable programs and are more likely to purchase the company’s products and services accordingly [13,14,15,47]. One recent paper even suggested that consumers tend to view a firm as committed to doing good when it takes a frequency-focused rather than an amount-focused donation strategy [48]. In addition, some studies point out that the implementation of donation programs can also strengthen the relationship between a company and the regional government, evidenced by actions such as achieving subsidies from the government [49]. Therefore, sustainable charitable projects may not only provide a lasting positive reputation for consumers but may also help companies achieve long-term support from governments.

Given that individual consumers are still the main participants in charitable programs, program organizers must think about how to attract new participants and maintain existing volunteers to improve the sustainability of their charitable activities. Most experiments in past studies on donation psychology only measure the participants’ attitude toward a single charitable program, and studies on the impact between multiple donation decisions are lacking. Based on our findings, program organizers could adjust strategies to advertise their programs, such as by emphasizing the overall outstanding contribution or the unique mission of the organization as different from other competitors in order to attract new participants and to build a sense of belonging for the current volunteers. Additionally, if the charitable organization has an established volunteer database, it could promote different kinds of activities to individual volunteers. However, we recommend that the organization avoid promoting too frequently in this context in case licensing effect occurs, regardless of what the donated resources are.

9.4. Limitation and Directions for Future Research

Despite the contributions to the theory and managerial field, our research has several limitations.

Theoretically, we use licensing effect to derive our main hypothesis that people are more reluctant to engage in charitable behavior after recalling and imaging a similar experience. But according to consistency theory, people tend to perform in line with former experience, which suggests they are more willing to participate in charitable activities. Future research could further test when licensing effect works and when consistency effect works. Moreover, although in this study we mainly consider the donated resource type when identifying the possible moderators of the licensing effect, charitable programs could be classified in other ways such as in terms of the duration of the program and the characteristics of recipients and organizers. These factors could be further explored in future studies.

Methodologically, the results should be considered with caution, and there is further need of validation of the findings. On the one hand, an exploratory study can also be designed to be quantitative. In an exploratory study, significant results provide indications of hypotheses to be tested. On the other hand, in our experiments, we let the participants recall previous activities by answering specific questions so as to manipulate their interpretation of past behaviors and use their willingness to participate in the target activity. In future studies, we could further expand the time difference between the initial and subsequent behaviors and conduct longitudinal research on the same group of participants.

With the development of science and technology, charitable projects are no longer defined as merely donating money and products or performing voluntary service. For example, the “Ant Forest” launched by Alipay in China can also be regarded as a charitable project. It transforms consumers’ daily consumption behaviors into visible achievements (e.g., planting virtual trees) and stimulates individuals to conduct sustainable environmental protection behaviors. Furthermore, the Alipay Company performs offline prosocial behaviors such as planting real trees in desert areas in the name of consumers. In the future, charitable programs may no longer appear to be an altruistic behavior that contrasts with traditional consumption, but may be integrated with daily purchase behavior. Other innovative mechanisms are on the rise in the charitable field. The consumer psychology behind these innovative programs warrants further exploration, and more systematic studies should be conducted.