Does resource scarcity promote selfish or selfless behaviours? A nuanced perspective through the lens of nostalgia Open Access

Resource scarcity is prevalent with phenomena like economic recessions, supply chain disruptions and scarce commodities, leading to threats that encourage the conservation of resources (Hosany and Hamilton, 2023; Goldsmith et al., 2021). This need for resource conservation induces a scarcity mindset, with individuals perceiving that specific resources (e.g. income) are insufficient to meet their needs (Roux et al., 2015). In a consumerism context, such perceptions of insufficiency have been linked to consumers making resource allocation decisions that involve trade-offs between benefits to the self versus sharing resources with others (e.g. charitable giving) as a form of pro-social (or selfless – used interchangeably, henceforth) behaviour (Roux et al., 2015; Zhou et al., 2012).

Human beings inherently tend to share resources with others, extending beyond their close family relationships (Aknin et al., 2013; Cui et al., 2022). However, such selfless decisions to allocate resources to others can be suppressed under resource scarcity (Cui et al., 2022). In support of this suppression hypothesis, a vast body of literature shows that under conditions of resource scarcity, consumers focus on resource allocation decisions that benefit the self versus others (Dillard and Shen, 2005; Swain et al., 2006; Markus and Schwartz, 2010; Sharma and Alter, 2012; Gupta and Gentry, 2019). A limited number of studies have also demonstrated that the inherent human tendency to share scarce resources may still exist, although for self-serving purposes, e.g. advancing one’s welfare (Petersen et al., 2014; Roux et al., 2015). In the current work, we extend this body of literature by positing that the decision to suppress other-focused behaviour under resource scarcity conditions may not eventuate under the influence of nostalgia experienced by an individual. In other words, in the presence of resource scarcity, we propose that nostalgia can encourage selfless behaviours.

Past literature defines nostalgia as a longing for the past (Wildschut et al., 2006). Based on previous work, nostalgia is posited to have restorative and social properties. For example, past work shows that nostalgia can buffer against adverse events and threats to the self (Wildschut et al., 2006; Routledge et al., 2008; Zhou et al., 2008). The social properties of nostalgia further posit that people pay attention to the needs of others rather than focusing on their selves (Stephan et al., 2014; Juhl et al., 2021). Thus, when nostalgia is induced in people, it should buffer against any threats associated with scarcity and further motivate people to tune in to the needs of others. Essentially, suppressing other-focused behaviours in the face of resource scarcity should be removed in nostalgic people, resulting in pro-social behaviours. The current research, therefore, inquires how nostalgia can promote selfless behaviours under different situations that promote resource scarcity (e.g. scarce resources, supply chain disruption and pandemic-induced shortages).

We further posit that the combined effect of nostalgia and resource scarcity to guide selfless behaviours will be underpinned by empathy. Past literature defines empathy as an emotion-oriented towards others (Batson et al., 1987). Previous work shows that selfless behaviours under nostalgia can be guided by empathy (Zhou et al., 2012; Stephan et al., 2014). In a related manner, recent work (Cui et al., 2022) demonstrates that empathy can encourage people to share limited resources. Based on these previous findings, we argue that empathy underpins self-less behaviours under the combined influence of nostalgia and scarcity. The current research, therefore, studies how nostalgia encourages selfless acts under resource scarcity, suggesting that the absence of nostalgia drives heightened selfish motivation and allocation of limited resources to the self (vs others).

The choice of nostalgia as the key variable in this work is guided by theoretical reasons (e.g. its psychological and social properties) and addresses the call for more research to understand nostalgia’s role in consumer behaviour (Srivastava et al., 2022; Hartmann and Brunk, 2019). Further, nostalgia is ubiquitous in marketing, from branding and advertising to consumer behaviour (Lasaleta et al., 2014). The current work also addresses a gap in the scarcity literature as extant researchers recommend studying new moderators to understand how selfless behaviours are encouraged under resource scarcity (Cannon et al., 2019; Shi et al., 2020).

We test our hypotheses across different contexts of pro-social behaviour: Study 1 tests intention to donate while Study 2 engages actual donations to charity. The first two studies further delve into the underpinning causal process and adopt a rigorous methodology involving measuring and manipulating the mediator, i.e. empathy. Study 3 introduces a situation that involves an intention to buy regular quantities while shopping during a pandemic-induced scarcity. Finally, study 4 engages a nostalgic appeal from a Chief Executive Officer (CEO) in the face of scarcity induced by supply chain disruption to enforce purchase restrictions for an essential shopping item. Study 4 demonstrates the practical relevance of how managers can engage nostalgia to motivate consumers to share resources with others (i.e. a pro-social act). The conceptual model for this work is encapsulated in Figure 1.

Resource scarcity is sensing or observing a discrepancy between one’s current level of resources versus a higher and desired level (Cannon et al., 2019). A resource, in this regard, is a quantifiable entity (e.g. money) that offers utility to an individual and has the potential to be consumed and depleted (Cannon et al., 2019). An individual, for example, can perceive resource scarcity by comparing his/her income to that of someone better off (Nelson and Morrison, 2005; Tully et al., 2015). Resource scarcity can, thus, be perceived from cues in the general environment (e.g. a recession, financial deprivation, hunger or pandemic) or even induced by marketers through tactics like limited edition products (Radel and Clément-Guillotin, 2012).

The theory of resource scarcity, particularly the self-regulatory model, suggests two coping pathways: scarcity reduction and control restoration (Cannon et al., 2019; Goldsmith et al., 2021; Yang and Zhang, 2022). For example, on the scarcity reduction pathway, a person may purchase a good that is available for a limited time to prevent future scarcity (Cialdini, 2009). On the other hand, control restoration focuses on re-establishing diminished personal control. For example, product unavailability can threaten the loss of personal freedom/autonomy to choose one’s outcome (Brehm, 1966; Brehm and Brehm, 1981). People can restore personal control by making compensatory purchases of products that offer security, such as status products (Cannon et al., 2019). Restoring control can also include non-consumption behaviours (e.g. undertaking creative work) that can enhance personal control by forming a meaningful and structured interpretation of the environment (Krosch and Amodio, 2014; Landau et al., 2015; Mehta and Zhu, 2016).

Based on prior work, resource scarcity has been found to elicit heightened agentic motivation that leads to selfish behaviours (Loewenstein et al., 2001; Dillard and Shen, 2005; Swain et al., 2006; Markus and Schwartz, 2010; Sharma and Alter, 2012; Gupta and Gentry, 2019). For example, Roux et al. (2015) demonstrated that resource scarcity enhanced competitive motives and encouraged selfish resource allocation decisions. Similarly, resource scarcity was found to promote incidental aggressive behaviours that helped to restore personal control (Kristofferson et al., 2017; De Sousa et al., 2018; Shah et al., 2018). For instance, Sharma and Alter (2012) demonstrated that financial deprivation motivated the selfish consumption of scarce goods to make them unavailable to others. In addition, resource scarcity has also been described as a psychological threat (e.g. financial deprivation) that can activate unpleasant feelings such as stress in individuals (Goldsmith et al., 2021; Cannon et al., 2019).

However, limited studies have found that resource scarcity can promote pro-social actions, albeit under limited conditions (Petersen et al., 2014; Roux et al., 2015). For example, Petersen et al. (2014) demonstrated that when people experienced food scarcity like short-term hunger, they acted more greedily (e.g. took more resources from others) while expressing more significant support for social welfare. The increased welfare support in states of hunger was purposely deemed as a “cheap” signal and an attempt to get others to share (Petersen et al., 2014). Similarly, Roux et al. (2015) argued that the intention to donate in the face of scarcity could be driven by self-interest (e.g. advancing one’s welfare to gain status in the eyes of others). Cannon et al. (2019) argued that such benevolent actions are only directed towards similar others, such as ingroup members. More recently, brain studies showed that resource scarcity suppressed selfless behaviours in individuals (Cui et al., 2022). In sum, it seems that under resource scarcity, people predominantly suppressed other-focused behaviour, although some apparent pro-social acts were demonstrated, albeit underpinned by selfish motives. Next, we discuss the literature on nostalgia, an emotion that has important psychological and social properties.

Nostalgia is defined as a sentimental yearning for the past. For example, Holbrook and Schindler (1991, p. 332) define nostalgia as “a longing for or favourable effect towards things from the past”. This yearning for the past has been labelled as nostalgic reverie. It is often associated with significant others, meaningful events, periods, locations, animals or objects (i.e. any memories that are self-relevant to the protagonist; Wildschut et al., 2006; Hepper et al., 2012; Hamilton and Wagner, 2014; Van Tilburg et al., 2019; Khoshghadam et al., 2019). Nostalgia is a common emotion (Wildschut et al., 2006). It occurs frequently and is a universal experience (Wildschut et al., 2006; Garrido, 2018; Hepper et al., 2020). Past research has also linked nostalgia to emotions like self-compassion, pride, guilt, shame, anger and sadness (Van Tilburg et al., 2018). In a related manner, past literature has also discussed the role of empathy in guiding nostalgic behaviours (Zhou et al., 2012; Stephan et al., 2014).

Nostalgia performs important psychological functions such as reducing unpleasant feelings and regulating emotions as a response to, or buffer against, threats experienced to an individual’s self-concept or even in the face of personal distress (Bi et al., 2024; Routledge et al., 2008; Routledge et al., 2012; Van Tilburg et al., 2013; Routledge et al., 2014; Baldwin et al., 2015; Reid et al., 2021). For example, nostalgic people (vs the control group) appeared more calm following distressing events like the loss of a loved one (Reid et al., 2021) or in the face of a threat to the self (Bi et al., 2024). Therefore, nostalgia’s emotional power buffers against threats to the self and stressful events to restore psychological equilibrium (Bi et al., 2024; Reid et al., 2021).

Nostalgia performs important social functions, too. Nostalgia serves an interpersonal function that promotes a focus on others (Turner et al., 2012; Sedikides and Wildschut, 2019). In a related fashion, nostalgia helps to strengthen approach motivation towards others (Stephan et al., 2014). In one study, Stephan et al. (2014) asked participants to reflect on a nostalgic (vs ordinary) event and then to set up two chairs for supposed interaction between themselves and another person. Subjects in whom nostalgia (vs control) had been triggered placed the chairs closer together. Nostalgia also strengthens a sense of acceptance and support and encourages helping behaviours (Juhl et al., 2021). Social mechanisms fostered by nostalgia encourage individuals to perceive others as dependable and supportive, resulting in nostalgic individuals being more comfortable seeking help from others (Juhl et al., 2021). The social properties of nostalgia, therefore, promote a focus on others. Next, we argue that nostalgia can encourage pro-social actions under resource scarcity.

Humans tend to share resources with other people beyond blood ties (Aknin et al., 2013). However, the self-regulatory model of Cannon et al. (2019) argues that resource scarcity induces a self-regulation process that focuses on restoring perceived resource adequacy, which, in turn, explains why individuals become more self-focused. Thus, despite the importance of sharing behaviours, pro-social sharing is suppressed under resource scarcity. Past research demonstrates that resource scarcity drives selfish behaviours (De Sousa et al., 2018; Shah et al., 2018; Roux et al., 2015). However, limited past studies also argue that the inherent natural tendency to share scarce resources can be reserved or even strengthened sometimes, albeit under particular conditions. In support of this, past work suggests that resource scarcity can promote sharing behaviour. However, apparent generosity is still guided by self-serving purposes such as sending “cheap” signals to convince others to share their resources and advance one’s welfare (Petersen et al., 2014; Roux et al., 2015).

We argue that nostalgia can facilitate helping behaviour that is suppressed in the face of resource scarcity. Nostalgia can do this based on its emotion-regulation function and social properties. As argued before, emotion-regulation functions of nostalgia help to deal with adverse events, including threats to the self (Wildschut et al., 2006; Routledge et al., 2008; Zhou et al., 2008; Routledge et al., 2012; Routledge et al., 2014; Reid et al., 2021). Therefore, nostalgia’s emotion-regulation function should help buffer against threats associated with scarcity. The absence of scarcity-related threats (e.g. hunger) has encouraged resource sharing with others (van Dillen et al., 2021; Faber and Häusser, 2022). As such, we hypothesise that the social properties of nostalgia can motivate people to pay attention to the needs of others rather than being entirely self-focused, even in the face of scarcity. Past research shows that when people are socially interdependent, they demonstrate higher motivation to share resources (Faber and Häusser, 2022). Based on this, we posit that:

We further advocate that the selfless acts posited under the combined influence of nostalgia and resource scarcity will be underpinned by empathy. Empathy, an emotion-oriented towards others (Batson et al., 1987), is one possible predictor of attentive capacity to the needs of others (Batson, 1991). For example, past work demonstrates that the empathic process can guide pro-social behaviours, including altruistic sharing, helping and caring (Zaki, 2020). Further, this finding is also substantiated by a large body of work that has previously shown that empathy motivates a focus on others (vs self), thereby leading to helpful behaviours (Mikulincer et al., 2005; Batson et al., 2007; Zhou et al., 2012; Cohen and Hoffner, 2013). Empathy also encourages self-transcendence and a more inclusive attitude towards others (Baldwin et al., 2015). Supporting this, recent research shows that when empathy was induced in people exposed to resource scarcity, it encouraged sharing limited resources with anonymous others (Cui et al., 2022). Thus, empathy can encourage pro-social behaviour, even in resource scarcity.

Empathic processes underpin nostalgic behaviour as well. For example, past studies have shown that nostalgic participants are more likely to donate to charity, with this effect being mediated by empathy (Zhou et al., 2012). Similarly, Stephan et al. (2014) showed that people were more likely to demonstrate empathic behaviour under nostalgia. For example, in a staged interaction, the experimenter dropped a box of pencils in front of participants, with helping behaviour measured by whether participants helped to pick up the pencils. Findings showed that participants in whom nostalgia had been induced were more likely to demonstrate helping behaviour (Stephan et al., 2014). Supporting these previous works (e.g. Zhou et al., 2012; Stephan et al., 2014), recent findings show that nostalgia indeed increased empathy in individuals (Juhl and Biskas, 2023).

Based on the above, empathic processes underpin selfless acts under nostalgia and in the face of resource scarcity. However, these previous works did not study the combined effect of nostalgia and resource scarcity. We extend this logic by suggesting that the joint influence of nostalgia and resource scarcity on selfless behaviours will also be underpinned by empathy. We, therefore, hypothesise:

In our current work, four laboratory experiments were conducted to test our basic premises. Study one involved charitable intentions that encourage selfless behaviour. Study one also tested our mediation hypothesis through empathy and helped to rule out alternate explanations based on the nostalgia and charitable giving literature. Study two engaged a moderation of mediation approach to provide consistent causal support for the mediation mechanism. Study two also used actual donations to incorporate realism. Study three engaged a pandemic driven product purchase decision that should encourage more competitive motives and selfish behaviour. Study three, therefore, tested how scarcity and nostalgia drives selfless behaviour through empathy and competitive orientation. Finally, study four demonstrated how a CEO appeal based on nostalgia can promote selfless behaviour in the face of supply chain driven scarcity. In all studies, our work was conducted among a sample of Master of Business Administration (MBA) students (studies 1, 2 and 3) and the general population (study 4) from Australia. In summary, the range of contexts used in this research either encouraged or discouraged selfless behaviours, thus helping us to test our key hypotheses.

To test our key hypotheses, a two-way nostalgia (present vs control) x resources (scarce vs non-scarce) between-subjects’ laboratory experiment was conducted. In this experiment, 298 MBA students (average age = 43.2 years, female = 48.7%) were recruited from a Pacific Coast university and randomly allocated to one of the four experimental conditions. Participants in each condition completed their responses on a computer while sitting in an individual cubicle, numbered from 1 to 4. The experiment was conducted in small groups.

The study was conducted in three seemingly unrelated parts. The first part involved manipulating nostalgia. Following an established procedure in the literature (Routledge et al., 2011; Zhou et al., 2012), participants in the nostalgic condition were required to list four keywords related to a nostalgic event in their lives. They were then asked to briefly reflect on this event. On the other hand, participants in the control condition were required to briefly reflect on keywords related to an ordinary event in their lives. Two manipulation check items then followed, namely, “Right now, I am feeling quite nostalgic” and “Right now, I am having nostalgic feelings” (1 = strongly disagree, 7 = strongly agree). The two items were based on the literature (Wildschut et al., 2006) and averaged to form an index measure.

In the second part of the study, resource scarcity was manipulated using an established procedure (Roux et al., 2015). In the scarcity condition, participants were asked to recall an episode from the past when they felt like “they did not have enough of something”. Participants in the control condition were asked to describe an episode from their past involving an ordinary grocery shopping experience. Participants were then asked to indicate the extent to which they agreed or disagreed with the following statements “My resources are scarce”, “I don’t have enough resources”, “I need to protect the resources I have” and “I need to acquire more resources” (1 = strongly disagree, 7 = strongly agree). The four items were based on the literature (Roux et al., 2015) and further averaged to form an index of experienced scarcity (Cronbach’s alpha = 0.83).

In the next part, participants completed various other measures related to the nostalgia and charitable-giving literature. Previous research has linked selfless behaviour with concepts such as power and power distance (Rucker et al., 2011; Han et al., 2017). Similarly, nostalgia has been linked to a range of emotions and self-related variables. Based on this literature, several measures were included in the current work to capture self-compassion, pride, self-esteem, self-importance, power distance belief, guilt, shame, fond memories and, finally, the mediator, “empathy” (refer to “control variables” below). Participants then completed standard demographic questions and were thanked for their contribution. This was followed by the third part of the study.

As an individual participant left the laboratory, a confederate posing as an employee from a university-based charity approached the participant. Several confederates were trained and engaged for this purpose. Confederates kept track of a participant’s cubicle number and linked a subject’s responses from the laboratory-based study to the key dependent variable. Using a standardised script, the confederate explained the charity’s purpose and the need for volunteers and donations to support the good cause. Following this short introduction about the charity, the confederate noted the participant’s likelihood of joining the charity and his/her intention to donate. As our measurement involved a realistic approach (Morales et al., 2017), two single items were engaged to measure the dependent variable (see below).

The laboratory-based study, as indicated above, measured several control variables and the key mediator “empathy”. The measures for empathy were based on extant literature (Zhou et al., 2012) and were further embedded within the range of measures for control variables. Empathy was measured using two items: “I feel compassionate” and “I feel sympathetic” towards the “charity-based donation program” (with 1 = strongly disagree and 7 = strongly agree as endpoints). These two items were averaged to form an index. The field data collection that followed the laboratory-based responses captured the key dependent variable. Participants indicated if they were “likely to sign up for the charity” and if they “intend to donate to the charity” using a seven-point Likert scale (1 = strongly disagree and 7 = strongly agree). These items were also based on the extant literature (Zhou et al., 2012). Since these two items were correlated, they were averaged to form an index called “intention towards charity”. The entire range of measures engaged across all the studies can be found in the web appendix.

Manipulation check: The manipulation checks for nostalgia and scarcity were subjected to a two-way analysis of variance (ANOVA) with resource scarcity and nostalgia serving as the independent variables. With the manipulation check for nostalgia, only a significant main effect (F(1, 294) = 569.5, p < 0.001) was found, such that subjects reported higher nostalgia than those in the control condition (M_nostalgia = 4.73 vs M_control = 2.51; p< 0.001). Similarly, results of the two-way ANOVA showed a significant main effect for scarcity (F(1, 294) = 547.3, p < 0.001), such that participants who reported an episode of scarcity (vs a non-scarce condition), registered higher perceived resource scarcity (M_scarce = 4.21 vs M_non-scarce = 2.23, p < 0.001). Based on these findings, we deemed the resource scarcity and nostalgia manipulations successful.

Test of hypotheses: To test H1, we conducted a two-way ANOVA. In this analysis, nostalgia and resource scarcity served as the independent variables, while “intention towards charity” served as the key dependent variable. Findings showed the main effect of nostalgia (p = 0.23) was not significant. However, as posited, the two-way interaction between nostalgia and resource scarcity was significant (F (1,294) = 33.6, p < 0.001, η2 = 0.1). Findings from contrast analyses showed that, in the presence of nostalgia, subjects reported a higher intention towards the charity in the resource-scarce (vs non-scarce) condition (M_scarce = 4.28 vs M_non-scarce = 3.27, t (294) = 5.2, p < 0.001, Cohen’s d = 0.85). In the absence of nostalgia, subjects showed lower intention towards charity in the resource-scarce (vs non-scarce) condition (M_scarce = 3.65 vs M_non-scarce = 4.24, t (294) = −3.0, p = 0.00, Cohen’s d = −0.49). We repeated the analyses with “sadness” as a control (refer to “Control variables” below) and found the two-way interaction was still significant. Thus, H1 was supported.

We further investigated the results for our mediator empathy. Results of ANOVA showed a significant two-way interaction between nostalgia and scarcity (F (1,294) = 57.4, p < 0.001, η2 = 0.16). Follow-up contrast analyses showed that in the presence of nostalgia, people experienced higher empathy in the resource-scarce versus non-scarce condition (M_scarce = 4.37 vs M_non-scarce = 3.44, t (294) = 4.6, p < 0.001, Cohen’s d = 0.73). This reversed in the absence of nostalgia (M_scarce = 3.35 vs M_non-scarce = 4.59, t (294) = −6.1, p < 0.001, Cohen’s d = −1.03).

Control variables: Several control variables were chosen from the literature to rule out alternative explanations. These include variables, such as power and power distance belief, as anchored in the charitable giving literature (e.g. Winterich and Zhang, 2014; Han et al., 2017), were included. Based on past research that links nostalgia to self-related variables, constructs such as self-esteem and self-importance, were also controlled for (Robins et al., 2001). Finally, grounded in the literature examining nostalgia’s relationship with emotions (e.g. Van Tilburg et al., 2018), several other variables, such as self-compassion, pride, guilt, shame, anger and sadness, were also included as control variables. All these variables were subjected to a two-way ANCOVA with nostalgia and resource scarcity as independent variables. The only significant two-way interaction was obtained for the emotion “sadness” (F (1,294) = 7.29, p = 0.01). As explained above, the two-way interaction to test H1 was significant after controlling for sadness. We also tested “sadness” as a possible mediator, as reported below.

To test for H2, we used Hayes (2013) Model 7 using 5,000 bootstrap analyses. Resource scarcity served as the independent variable, with nostalgia serving as the moderator, while “intention towards charity” was the key dependent variable. As we had recorded the subjects’ empathy in our experiment, we tested if the mediator “empathy” carried over the joint influence of resource scarcity and nostalgia to charitable intention.

The findings showed the two-way interaction between resource scarcity and nostalgia significantly affected empathy (β = 2.17, t = 7.58, p = 0.00; LLCI = 1.61, ULCI= 2.74). Furthermore, the mediator empathy had a significant impact on “intention towards charity” (β = 0.22, t = 4.31, p = 0.00; LLCI =0.12, ULCI =0.33). The conditional indirect effect of resource scarcity on intention towards charity through empathy revealed further insights. In the absence of nostalgia, resource scarcity reduced empathy to lower intention towards charity (conditional indirect effect = −0.28, LLCI = −0.44 ULCI = −0.14). On the other hand, in the presence of nostalgia, resource scarcity enhanced empathy to increase intention towards charity (conditional indirect effect = 0.21, LLCI = 0.09 ULCI = 0.36). The index of moderated mediation was 0.49 and deemed significant since the 95% CI did not straddle zero (LLCI = 0.25, ULCI = 0.78). These findings, therefore, supported H2.

Based on the significant two-way ANCOVA results for “sadness”, we tested if this emotion served as an additional mediator. The results showed that the two-way interaction between nostalgia and resource scarcity had a significant impact on “sadness” (β = 1.15, t = 2.7, p = 0.01), with the impact of sadness on intention towards charity being barely significant (β = 0.07, t = 2.01, p = 0.045). However, the indirect effect of sadness in the presence and absence of nostalgia was not significant (index of moderated mediation was not significant since the 95% CI straddled zero).

Theories of nostalgia and resource scarcity predict a countervailing effect on human behaviour. While nostalgia has been predicted to promote a focus on others, resource scarcity has been found to encourage selfishness and even, in some cases, lead to aggressive behaviour. Our theoretical prediction based on both these literature streams was that nostalgia and resource scarcity would jointly promote other-focused selfless behaviour. Furthermore, nostalgia and resource scarcity’s joint influence would encourage kind acts, such as charitable intention, mainly by enhancing empathy.

Our first study, conducted in the context of charity, supported these premises. Firstly, as predicted by H1, subjects who recalled nostalgic events showed higher intention towards charity, even when they perceived resources to be scarce (vs non-scarce). This effect reversed in the absence of nostalgia. Furthermore, in the presence of nostalgia and resource scarcity, people experienced more empathy which motivated higher intention towards charity. Study one also tested several alternative explanations that could drive selfless behaviour based on variables that may influence charitable giving (e.g. power distance) or driven by nostalgia (e.g. emotions, self-esteem). We tested these alternative variables to see if the joint effect of nostalgia and resource scarcity influenced selfless behaviour through these pathways. We found that the underlying process was essentially driven by empathy. The role of empathy carrying over the joint influence of nostalgia and resource scarcity to selfless behaviour was unique, further supporting H2. In summary, Study 1 supported our hypotheses in the context of intention to donate to charity.

In study 1, we engaged a “measurement-of-mediator” design to study the underlying process for moderated mediation hypothesis, H2. In our second study, we engaged “manipulation-of-mediator” design as the extant literature posits that measurement-of-mediator design could be subject to alternate explanations about the causal influence of the mediator on a dependent variable (Zhao et al., 2023; Pirlott and MacKinnon, 2016). A research design that engages both measurement and manipulation of the mediator provides strong support for the mediation mechanism (Zhao et al., 2023; Xu et al., 2021; Pirlott and MacKinnon, 2016). In other words, study 2 manipulates empathy (vs control) first to study the impact of nostalgia and resource scarcity on selfless acts. Based on our theorisation, the control or nostalgia absent condition (when subjects start behaving selfishly under resource scarcity) should start behaving more selflessly when empathy has been primed.

Study 2 engaged a 2 empathy (high vs control) x 2 nostalgia (present vs control) x 2 resources (scarce vs non-scarce) between subjects’ experimental design. 491 MBA students from the same large Pacific Coast university as Study 1 participated in this study (average age = 34.9, female = 52.5%) and were randomly allocated to the experimental conditions (Pirlott and MacKinnon, 2016). Participants were also given $5 in cash to participate in the study and completed the study on a computer sitting individually in a numbered cubicle.

The study was once again conducted in three seemingly unrelated parts. In the first part, empathy was manipulated using an extant technique from the literature (Xu et al., 2021). Participants were randomly allocated to a perspective-taking (vs control) task to manipulate empathy. The perspective-taking task involved participants reading a news post about questionable labour practices from a reputed source and trying to feel its impact on workers’ lives. In the control condition, subjects read the same news post but were exposed to a scenario where they took an objective perspective and remained detached. Following this empathy manipulation, subjects reported their empathy levels by responding to a manipulation check question. In the second part of the study, nostalgia was manipulated using the same technique as study 1, followed by the manipulation check questions. In the third part, subjects were exposed to the same scarcity manipulation, along with manipulation check questions, from the first study. Finally, subjects completed their demographics and were thanked for their participation.

After subjects completed their study and were leaving the premises, they were approached by trained confederates who posed as employees for a university-based charity. Similar to study 1, the confederates explained the purpose of the charity (based on a standardised script) and requested donations. The dependent variable for the second study was the amount of money donated to the charity. Further, similar to study 1, the confederates kept track of an individual’s cubicle number to link the responses from the laboratory experiment to the amount of money donated. The second study, therefore, engaged actual donations as the key dependent variable.

The second study used the same manipulation check questions for the nostalgia and scarcity manipulation reported in the web appendix. The empathy manipulation check question involved asking subjects, “Right now, I feel empathetic” with 1 = strongly disagree and 7 = strongly agree’ (Zhou et al., 2012). As mentioned earlier, the dependent variable was the amount of money donated.

Manipulation check: The manipulation checks for empathy, nostalgia and scarcity were subjected to a three-way ANOVA with our key independent variables. In response to the empathy manipulation, a significant main effect showed that subjects felt more empathetic in the empathy versus control condition (M_empathy = 4.93 vs M_control = 2.55; p < 0.001). A significant main effect also showed that subjects reported to be feeling more nostalgic in the nostalgia versus control condition (M_nostalgia = 5.47 vs M_control = 1.97; p < 0.001). Finally, a significant main effect demonstrated that subjects also reported higher resource scarcity in the scarce versus non-scarce condition (M_scarce = 5.46 vs M_non-scarce = 2.13; p < 0.001). No other significant main or interaction effects were obtained for the three manipulation checks. Based on the results, the manipulations for empathy, nostalgia and scarcity were deemed successful.

Test of hypothesis: The key objective of this study was to replicate results from study 1 while providing further evidence about the causal mechanism for empathy. In line with this, we would expect different results under the empathy versus control conditions. The control condition for empathy should allow us to replicate the results from study 1, i.e. H1. However, under empathy, we believe that nostalgia should still drive selfless acts under resource scarcity (vs non-scarcity). Further, under empathy, the control (i.e. nostalgia absent) condition should behave similarly to the nostalgia present condition and drive selfless acts. Consistent with the manipulation-of-mediator design (Pirlott and MacKinnon, 2016), we would also expect a main effect of empathy on the key dependent variable, (i.e. the donation amount). Based on this theorisation, we should be able to obtain a significant three-way interaction between empathy, nostalgia and scarcity.

To test the above, we ran a three-way between-subjects ANOVA with donation amount as the key dependent variable. As predicted, we obtained a significant main effect for empathy (F (1, 483) = 66.9, p = 0.00, η2 = 0.12). Subjects in the empathy condition were willing to donate more money compared to the control condition (M_empathy = 2.95 vs M_control = 2.22, t (489) = −7.7, p = 0.00). The main effect for nostalgia was not significant (p = 0.81). More importantly, we obtained a significant three-way interaction between the independent variables (F (1, 483) = 5.93, p = 0.02, η2 = 0.01). We engaged contrast analyses to compare the results separately for the empathy versus control conditions.

Similar to H1, under the control condition or when empathy was absent, nostalgia motivated higher payments for subjects in the resource-scarce versus non-scarce condition (M_scarce = 2.41 vs M_non-scarce = 1.93, t (483) = 2.63, p = 0.01, Cohen’s d = 0.47). This reversed when nostalgia was absent (M_scarce = 1.93 vs M_non-scarce = 2.57, t (483) = −3.48, p = 0.00, Cohen’s d = −0.63). The findings were also interesting under the influence of empathy. First, under empathy, nostalgia still drove higher payments in the resource-scarce versus non-scarce condition (M_scarce = 3.37 vs M_non-scarce = 2.59, t (483) = 4.28, p = 0.00, Cohen’s d = 0.78.). However, this time around, the presence of empathy caused subjects in the control condition (e.g. when nostalgia was absent) to donate more in the resource-scarce versus non-scarce condition (M_scarce = 3.22 vs M_non-scarce = 2.66, t (483) = 3.04, p = 0.00, Cohen’s d = 0.56).

Study 1 engaged a measurement-of-mediator design and controlled for several alternate variables to provide support for H1 and H2. Study 2 engaged a manipulation-of-mediator design to show causal impact of the mediator “empathy” on donation amounts. Study 2 (the control condition when empathy was absent) further replicated the findings regarding H1 from study 1. The presence of empathy caused nostalgic subjects to continue with their selfless acts. However, empathy further caused subjects in the nostalgia-absent condition to behave pro-socially. Together, studies 1 and 2 provide consistent support for H1, but more importantly, provide robust causal support regarding our mediation mechanism for empathy.

In Study 3, we test our propositions H1 and H2, by using a scenario that is likely to promote more competitive and aggressive mindsets (e.g. hoarding during a pandemic). Differing from Studies 1 and 2, we used a separate manipulation for resource scarcity and pitched two different mediators – competitive orientation vs empathy – to better understand the underlying processes leading to pro-social behaviour.

A two-way nostalgia (present vs control) x resources (scarce vs non-scarce) between-subjects laboratory experiment was engaged for Study 3. In total, 290 MBA students (average age = 31.7 years, female = 48%) were recruited from a second Pacific Coast university to participate in Study 3. The participants were randomly allocated to one of the four experimental conditions.

Study 3 was conducted in three seemingly unrelated parts. The first part involved manipulating nostalgia vs a control condition using the same technique used in Study 1. Following this procedure, subjects responded to the same manipulation checks and items to measure positive and negative emotions. In Study 1, emotion items were presented after the resource scarcity manipulation to see if they were influenced by the joint effect of resource scarcity and nostalgia. This was done to rule out alternative explanations, for example, if nostalgia triggered different emotions (e.g. pride) that helped them to cope with resource scarcity and encouraged selfless behaviours. In Study 3, we used different items to measure positive and negative emotions. Furthermore, these emotional measures were placed after the nostalgia manipulation to see if nostalgia directly influenced them. Positive emotions were measured with items like “Right now, I am having positive feelings” and “Right now, I feel happy”, while negative emotions were measured with items such as “Right now, I am having negative feelings” and “Right now, I feel unhappy”. These measures were adapted from the existing literature (Van Tilburg et al., 2018).

In the second part, differing from Studies 1 and 2, Study 3 engaged a different manipulation of the resource scarcity (vs non-scarce) condition. This resource scarcity manipulation involved subjects listing three things they could not do if several resources, such as petrol, water, sugar and electricity, were unavailable. In comparison, in the control condition for resource scarcity, subjects mentioned three things they could do with the above resources. This resource scarcity manipulation methodology was adopted from the literature (Roux et al., 2015). Study three subjects were also asked to respond to the same resource scarcity manipulation check items engaged in Study 1.

Finally, in the third part, subjects were given a grocery shopping scenario taking place during the COVID-19 pandemic. The scenario stated that, while shopping, the subjects observed a rush for toilet paper. When the subjects reached the toilet paper aisle, toilet paper rolls were still available, but they had to decide whether they wanted to buy their regular amount or an excess amount to stockpile. After this scenario, subjects reported their key dependent variable, “intention to buy regular quantity”, followed by the mediators engaged for this study: empathy towards fellow shoppers and competitive orientation.

“Intention to buy regular quantity” was measured with four items (1 = strongly disagree, 7 = strongly agree): “I am likely to buy my usual number for regular consumption”, “I will probably buy my usual number for regular consumption” and “I intend to buy my usual number for regular consumption” and a reverse coded item “I will increase the quantity purchased to stock up” (Laroche et al., 2003). The items were averaged to form a scale (Cronbach’s alpha = 0.71).

The mediator – empathy towards other shoppers – was measured by tapping into how subjects felt about the following emotional states with respect to other shoppers: sympathetic, compassionate, soft-hearted and tender (Cronbach’s alpha = 0.73). Finally, competitive orientation was measured across five items (1 = strongly disagree, 7 = strongly agree): “If I do not get the better over other customers, they will get the better of me”, “I would like to get an edge over other customers”, “I can’t stand to lose out to other customers”, “If I am nice, I will lose out to other customers” and “I need to get even with other customers” (Cronbach’s alpha = 0.77). All these measures were again derived from the extant literature (Roux et al., 2015). All the mediating variables were averaged to form an index. Proneness to nostalgia was measured in this study using a single item, “I am prone to feeling nostalgic” (Zhou et al., 2008), which was held as a control variable in our analyses.

Manipulation check: The manipulation checks for nostalgia, scarcity and positive and negative emotions were subjected to a two-way ANOVA with nostalgia and resource scarcity as the independent variables. Findings showed only a main effect for nostalgia (F(1, 286) = 114.1, p < 0.001) such that the subjects reported higher nostalgic feelings than those in the control condition (M_nostalgia = 4.64 vs M_control = 3.22, p < 0.001). Furthermore, no significant differences were found in how subjects experienced positive (M_nostalgia = 4.01 vs M_control = 3.97, p = 0.78) and negative emotions (M_nostalgia = 4.03 vs M_control = 4.0, p = 0.81) across the nostalgic vs control conditions. Similarly, results of the two-way ANOVA showed a main effect (F(1, 286) = 285.5, p < 0.001) such that relative to the non-scarce condition, subjects in the resource scarcity condition perceived resources to be more scarce (M_scarce = 4.81 vs M_non-scarce = 2.87, p < 0.001). Based on these results, the nostalgia and resource scarcity manipulations were deemed to be successful.

Test of hypotheses: To test H1, we once again conducted a two-way ANOVA, using the dependent variables “intention to buy regular quantity”. The main effect for nostalgia was not significant (p = 0.87). However, we found a two-way interaction between nostalgia and resource scarcity (F (1, 286) = 197.7, p < 0.001, η2 = 0.4). Findings from contrast analyses showed that, in the presence of nostalgia, people in the resource scarcity (vs non-scarce) condition indicated a higher intention to buy their regular quantity (M_scarce = 4.72 vs M_non-scarce = 3.18, t (286) = 9.2, p < 0.001, Cohen’s d = 1.68). This effect was reversed in the control condition, suggesting that in the absence of nostalgia, people in the scarce (vs non-scarce) condition reported lower intention to buy their regular quantity (M_scarce = 3.05 vs M_non-scarce = 4.82, t (286) = −10.2, p < 0.001, Cohen’s d = −1.64). The two-way interaction remained significant after controlling for nostalgia proneness (F (1, 285) = 198.3, p < 0.001) using ANCOVA. The observed findings once again supported our H1 about selfless behaviour under the influence of nostalgia and resource scarcity.

Results of ANOVA for our mediator “empathy” showed a significant two-way interaction (F (1, 286) = 324.3, p < 0.001, η2 = 0.5). Findings from contrast analyses showed that in the presence of nostalgia, people in the scarce versus non-scarce condition experienced higher empathy (M_scarce = 4.85 vs M_non-scarce = 2.85, t (286) = 12.6, p < 0.001, Cohen’s d = 2.2). On the other hand, in the absence of nostalgia, people experienced lower empathy in the scarce versus non-scarce condition (M_scarce = 2.82 vs M_non-scarce = 4.86, t (286) = −12.8, p < 0.001, Cohen’s d = −2.0).

We conducted Hayes (2013) Model 7 using 5,000 bootstrap analyses with resource scarcity as the independent variable and nostalgia as moderator, while “intention to buy regular quantity” served as the dependent variable. The two mediators engaged were empathy and competitive orientation.

The findings showed that the two-way interaction between nostalgia and resource scarcity increased empathy (β = 4.04, t = 18.0, p = 0.00) but reduced competitive orientation (β = −1.54, t = −7.67, p = 0.00). Furthermore, both empathy and competitive orientation had a significant influence on “intention to buy regular quantity” (β = 0.65, t = 16.9, p = 0.00 and β = −0.11, t = −2.7, p = 0.00), respectively. A study of the indirect effects showed that nostalgia in the presence (vs absence) of resource scarcity increased empathy and the subsequent intention to buy the regular quantity (conditional indirect effect = 1.31, 95% CI does not straddle zero). The index of moderated mediation (2.64) was significant as the 95% CI did not straddle zero (LLCI = 2.23, ULCI = 3.06). Similarly, the absence of nostalgia and the presence of resource scarcity enhanced competitive orientation to reduce intention to buy regular quantity (conditional indirect effect = −0.09, 95% CI does not straddle zero). The index of moderated mediation (0.17) was once again significant, and the 95% CI did not straddle zero (LLCI = 0.06, ULCI = 0.31). Taken together, this means that the presence of nostalgia and resource scarcity increased empathy to motivate the purchase of regular quantities. The absence of nostalgia and the presence of resource scarcity, on the other hand, enhanced competitive orientation to reduce the purchase of regular quantity.

The findings from Study 3 again supported H1 and H2, albeit with a difference. In Study 3, we employed a different way of manipulating resource scarcity while exposing subjects to a more rigorous test for selfless behaviour. Unlike Study 1, which tested intention to donate, Study 3 tested whether nostalgia and resource scarcity jointly encouraged selfless behaviour in the presence of a situation of resource scarcity induced by a pandemic (i.e. COVID-19). Furthermore, in Study 3, we examined the influence of competitive orientation and empathy to understand the underlying process for selfless behaviour. Overall, the findings supported the notion that nostalgia and resource scarcity jointly encouraged selfless behaviour and that both empathy and competitive orientation underlay this process, albeit in different ways.

In the presence of nostalgia, people showed higher empathy resulting in the purchase of a regular quantity of toilet paper in response to a pandemic-induced scarcity. Furthermore, when nostalgia was absent, the same pandemic-induced scarcity encouraged more competitive orientation and selfish behaviour to reduce the purchase of regular quantities. The presence of nostalgia, therefore, encouraged other-focused behaviour in the face of resource scarcity. The underlying mechanism is guided by empathy. The absence of nostalgia in the face of resource scarcity, on the other hand, is guided by a competitive orientation that benefits the self.

In Study 4, we focused on providing managerial relevance when studying how marketers could simultaneously engage the concepts of nostalgia and resource scarcity to influence selfless behaviour in real life. In line with this, Study 4 engaged a CEO message that was likely to trigger nostalgia followed by a resource scarcity situation, to make an appeal to consumers to exercise purchase restrictions. To be specific, Study 4 addressed the external validity of our conceptual model.

A two-way nostalgia (present vs control) x resources (scarce vs non-scarce) between-subjects’ laboratory experiment was employed for Study 4. In total, 360 participants (average age = 42.6 years, female = 52%) from the general population of a large Pacific Coast city were randomly allocated to one of the four experimental conditions.

Unlike previous studies, Study 4 utilised a CEO message that manipulated nostalgia and resource scarcity with the same stimulus. The CEO message was designed based on a real-life message from the CEO of a chain of retail stores (Coles) in Australia. The message was, however, tweaked to manipulate the key variables, namely, nostalgia and resource scarcity. The CEO message was essentially about a supply shortage leading to resource scarcity and purchase restrictions or, in the alternate version, the supply shortage was handled quickly, leading to no restrictions. In both cases, the CEO message started by triggering nostalgia.

Following the literature (Muehling and Sprott, 2004), the nostalgic and control stimuli were equivalent in all respects (visual format, colour, illustration) except for the opening statement of the CEO message. In the nostalgic version, the CEO message included the statement “How many times did you go shopping at Coles since you were growing up in Australia?” while the non-nostalgic version engaged “How many times do you go shopping at Coles in your daily life in Australia?”. This technique was followed from Muehling and Sprott (2004) who engaged a black and white picture along with nostalgic (vs non nostalgic) cues in the headline. In all four conditions, an old black and white photo of the grocery store was provided, with this photo (the stimulus) located in the web appendix. Following this manipulation, participants replied to the key dependent variable, “intention to buy regular quantity”, using the same items from Study 3. Participants then completed the manipulation check items for resource scarcity and nostalgia. Nostalgia proneness was also measured as a control variable. Finally, participants completed several demographic questions and were thanked for their participation.

We subjected both the nostalgia and scarcity manipulation checks to a two-way ANOVA, with the type of nostalgia and resources as the independent variables. Findings showed a main effect for nostalgia (F(1, 356) = 290.5, p < 0.001) with consumers in the nostalgia condition reported higher nostalgia compared to those in the control condition (M_nostalgia = 4.96 vs M_control = 2.78, p < 0.001). Similarly, for the scarcity manipulation check, only a main effect for scarcity was noted (F(1,356 = 75.66, p < 0.001). Participants perceived that resources were more scarce in the scarcity present condition vs the scarcity absent condition (M_scarce = 4.70 vs M_non-scarce = 3.07, p < 0.001).

More importantly, the results of a two-way ANOVA showed an interaction between nostalgia and resource scarcity (F (1,356) = 158.4, p < 0 0.001, η2 = 0.3). The main effect for nostalgia was not significant (p = 0.76). Contrast analyses showed that participants were more likely to purchase their regular quantity in the face (vs absence) of resource scarcity when the CEO message triggered nostalgia (M_scarce = 4.59 vs M_non-scarce = 3.35, t (356) = 7.03, p < 0.001, Cohen’s d =1.21). This effect reversed in the absence of nostalgia (M_scarce = 3.08 vs M_non-scarce = 4.94, t (356) = −10.8, p < 0.001, Cohen’s d = −1.44). The two-way interaction remained significant after controlling for nostalgia proneness (F (1, 355) = 156.4, p < 0.001) using ANCOVA analysis. These observed findings once again supported H1.

The main objective of study four was to demonstrate how managers can engage nostalgia in real life to influence consumer behaviour. In this study, resource scarcity was manipulated through supply chain disruption, while nostalgia was triggered by reminding consumers about how they grew up with a retail store in Australia. Further, the CEO appeal was also designed to influence consumption restraint given the scarcity situation. Our previous studies demonstrated selfless behaviour under the influence of nostalgia and resource scarcity in a variety of contexts such as charitable intentions, actual donations and buying regular quantities under pandemic-induced resource scarcity. Study four engaged a real-life scenario of supply chain-induced scarcity and an adapted CEO message to show that such selfless behaviour can be actually motivated by managerial actions (e.g. communication from the CEO). Study four, therefore, provides managerially relevant evidence of how nostalgia can be engaged in practice to influence selfless behaviour, especially when resources are perceived to be scarce by consumers.

Past literature shows that resource scarcity can promote self-oriented behaviour with a focus on associated benefits to the self. On the other hand, the nostalgia literature posits that this emotion can indeed have restorative and social functions that can direct other-focused behaviour. In other words, nostalgia has the power to encourage selfless behaviours under scarcity. Based on these forces’ countervailing nature, the current research examined whether scarcity-induced self-interested behaviours can change under the influence of nostalgia. Our basic premise was nostalgia’s psychological and social functions can neutralise the threats associated with scarcity and can further encourage other-focused behaviour through resource sharing, especially in the presence of scarcity. Further, we tested how the combined effect of nostalgia and scarcity can drive selfless behaviours through several mediators that could be specific to a context (e.g. empathy, competitive orientation).

We designed four studies to test the fundamental premises of the current work. Study 1 used the context of supporting a charity, while Study 2 engaged actual donations to charity. Further, both studies 1 and 2 engaged realism by capturing key dependent variables in the field (e.g. confederates posing as charity workers, actual donations; Morales et al., 2017). Study 3 engaged a pandemic-induced, while Study 4 engaged a supply chain-induced scarcity. Through these different contexts, we provided consistent support that nostalgia can encourage selfless behaviours towards others, albeit across a wide range of situations. Further, the different contexts provided an opportunity to test our theory rigorously, as the mechanisms underlying pro-social actions could differ across these situations. For example, while charitable intentions could be driven by empathy, the decision to be kind to other consumers in the face of pandemic-induced scarcity could be guided by a reduced competitive orientation. Further, these contexts also helped to test our theory across different product (toilet paper) and service (charity) categories, which support generalisability.

The selfless behaviours reported in our research were encouraged by enhanced empathy (studies 1 and 2) and increased empathy and reduced competitive orientation (study 3). This provides consistent evidence that nostalgia and scarcity combined, encourage other-focused behaviour underpinned by empathy. The findings that empathy mediates a range of selfless behaviours (e.g. charitable intention, donation, buying regular quantity) provides confidence regarding the causal influence of empathy and that empathy (an other-focused emotion) is conceptually distinctive to our dependent variables that focus on a range of selfless behaviours. Further, studies 1 and 2 together provide robust causal evidence that nostalgia and scarcity jointly influence selfless behaviour through empathy. Extant literature suggests a combined approach (e.g. measurement and manipulation of the mediator) to determine causal mediation (Zhao et al., 2023; Pirlott and MacKinnon, 2016). In this regard, the manipulation-of-mediator design (study 2) rules out explanations regarding alternate causal influences on the dependent variable (Pirlott and MacKinnon, 2016). Such approach is increasingly adopted by recent research to provide more rigorous causal evidence (Zhao et al., 2023). In our case, studies 1 and 2 provide consistent evidence for nostalgia and scarcity’s combined influence through empathy.

In the current work, we also controlled for several variables such as power distance, self-related variables (e.g. self-compassion, self-esteem, self-importance), nostalgia proneness and a range of emotions (e.g. pride, guilt, shame) that could potentially confound our findings. The choice of control variables was mainly guided by the nostalgia and charitable giving literature. In our first study, we tested if nostalgia and scarcity jointly influenced any of these confounding variables and whether this effect carried through to influence our focal dependent variable. Empirical evidence from study 1 shows that none of these confounding variables were jointly influenced by nostalgia and scarcity. Additionally, study 3 also showed that nostalgia did not influence emotions (positive or negative) to drive selfless behaviour. Consistent with our theorisation, we found that nostalgia and scarcity combined influenced selfless behaviour, only through the mediator “empathy”.

The above theorisation about nostalgia and scarcity influencing empathy further rules out other alternate explanations. The literature on nostalgia predicts that it can drive self-continuity, meaning in life, approach orientation, optimism, communal orientation and sense of acceptance (Sedikides and Wildschut, 2018; Sedikides and Wildschut, 2019; Sedikides and Wildschut, 2016; Van Tilburg et al., 2019; Biskas et al., 2019; Routledge et al., 2011). These alternate pathways probably do not explain why nostalgia and scarcity would influence selfless acts through any of these variables. As indicated above, some of these self-related variables have already been ruled out. Further, our nostalgia theorisation suggests selfless acts focused on others. This rules out variables like self-continuity, which still suggests cognitions related to the self (vs others). Further, nostalgia and scarcity influenced empathy (an emotion) to drive selfless acts directed towards others. Some of the alternate pathways (e.g. communal orientation, optimism) suggest a propensity rather than an action tendency to help others. In contrast, empathy as an emotion has been found to drive action tendencies involving selfless acts towards others (Zhou et al., 2012; Cohen and Hoffner, 2013). In sum, our theorisation, engagement of various control variables, robust causal evidence about empathy and consistent empirical findings regarding selfless behaviour predominantly support the role of nostalgia in guiding selfless acts under resource scarcity, mainly through empathy (see Table 1 for a summary of the findings).

Our work addresses important gaps in the literature. As outlined in our introductory section, extant research (Cannon et al., 2019) recommends more work to understand selfish versus selfless behaviours in the face of resource scarcity. Similarly, the role of nostalgia in the context of consumer behaviour is not well understood (Srivastava et al., 2022). Against this backdrop, the current work establishes a link between the two disparate streams of literature. This link is motivated by the existing evidence that scarcity can drive pro-social behaviours under certain situations, although such behaviours can still be guided by agency (e.g. self-interest, such as gaining status in the eyes of others; Petersen et al., 2014; Roux et al., 2015). However, the key question remains as to how and when pro-self (versus selfless) behaviours are demonstrated in the face of resource scarcity. The current research examines the novel role of nostalgia in explaining this behavioural shift, given its underlying psychological and social properties to neutralise threats and foster selfless behaviour focused on others.

Recent literature shows that sharing behaviour is suppressed under resource scarcity, although its inherent in human nature to share resources with others (Cui et al., 2022). The current work builds on this further by showing that nostalgia can indeed restore this suppression through its restorative and social properties. Previous work in the nostalgia literature has demonstrated pro-social behaviour under nostalgic influence (Zhou et al., 2012. The current work extends this line of reasoning by demonstrating that nostalgia can also guide pro-social in the face of resource scarcity. In terms of consumer behaviour, the current work shows that nostalgia-induced pro-social behaviour could be directed towards anyone, with different forms of scarcity at play (e.g. human versus nature induced). The current research, therefore, bridges gaps in understanding selfish vs selfless behaviours under resource scarcity, highlighting nostalgia’s role in promoting selfless behaviours.

The underlying process for selfless acts was mainly guided by empathy. The literature on empathy demonstrates that it can drive pro-social behaviour in nostalgic people (Zhou et al., 2012; Stephan et al., 2014). Similarly, empathy was also found to encourage sharing of limited resources (Cui et al., 2022). The current work posits and demonstrates that the joint influence of nostalgia and resource scarcity leading to pro-social behaviour can be underpinned by empathy as well. The role of empathy also helps to understand why people behave generously in the face of resource scarcity. Previous work has demonstrated apparent generous acts under resource scarcity, only to be guided by selfish motives (Petersen et al., 2014; Roux et al., 2015). Our work shows that nostalgia restores pro-social behaviour in the face of resource scarcity by promoting empathy. The underpinning process explanation is unique to the current work.

This work’s findings have several implications for practice since scarcity is ubiquitous and often experienced in everyday life. Indeed, scarcity can be driven by a range factors such as economic recessions, supply chain disruption, panic buying and market induced. For example, economic recessions are prevalent in our society. Recessions can lead to unemployment and reduced income for those who are employed. With less money to spend, consumers can cut down on their expenses leading to a scarcity mindset. Economic downturns can disrupt supply chains as companies close down. Similarly, natural disasters (e.g. floods, forest fires, earthquakes) can also cause disruption of supply chains. Supply chain disruptions can lead to shortages of raw materials or finished goods, thereby leading to scarcity. In anticipation of harder times (e.g. an ongoing pandemic) consumers can engage in panic buying or stockpiling, which can cause depletion of certain goods (e.g. essentials) more quickly than usual and lead to scarcity. Marketers can further promote scarcity by creating limited edition products. The prevalent scarcity situation suppresses pro-social behaviours, while our work shows that inducing nostalgia can restore resource sharing behaviours.

Therefore, a straightforward implication from the current findings is to promote nostalgia to encourage selfless behaviours. For example, based on our findings, charitable organisations can make promotional appeals that incorporate warm memories for their donors, which is likely to invoke nostalgic moments. Such promotional campaigns can be beneficial in times of scarcity, such as during an economic recession where people are less likely to donate. Similarly, in the context of scarcity that threatens survival and sustenance (e.g. groceries), company executives can make nostalgic appeals to their customers, making them more communally compliant. For example, in the face of COVID-induced panic buying, UK supermarkets emailed customers to reassure them against the need for panic buying (BBC, 2020). The effectiveness of such communication strategies can be further enriched by sharing images from the past to trigger nostalgia among shoppers. Similarly, a hybrid car may address concerns about the limited supply of fossil fuels and further induce nostalgic moments about how our natural environment used to be lush in the past, to encourage buying behaviours.

Nostalgic appeals are likely to encourage kind behaviour towards businesses too. For example, in Australia, a restaurant that operates as a social enterprise was at risk of permanent closure due to COVID related scarcity challenges (Telegramma, 2020). The restaurant set up a crowd-funding page in an attempt to stay afloat and was inundated with generous donations from the community. Based on the restaurant’s social media posts, it appears that most of the donors were people who had great past experiences at the restaurant. In this context, it can be reasoned that the funding appeal could have benefited from the nostalgic reverie of past memories at the restaurant, which in turn, promoted more donations (i.e. selfless behaviours). This is consistent with our findings about nostalgia’s pro-social responses under scarcity.

The current work is not without its limitations, which provide avenues for future research. For instance, we tested the scope of our theory in the context of resource scarcity. This is based on our premise that resource scarcity will induce selfish behaviour, while nostalgia will have a counterbalancing effect by promoting selfless behaviour. However, in the context of marketing, scarcity appeals have been engaged to sell products. Future research can test if our framework holds in the context of demand or supply-related scarcity appeals. In other words, future work can examine the impact of nostalgia in the context of scarcity appeals in product advertisements. Similarly, across our experiments, we studied the underlying mechanism with different mediators, such as empathy and competitive orientation. Future research can study if there are additional ways (e.g. social connection, self-compassion) through which nostalgia can promote selfless behaviours in the context of resource scarcity.

In addition, in the current work, we tested our theory based on state nostalgia and found that trait nostalgia (measured as nostalgia proneness using a single item in studies 3 and 4) does not have an impact on pro-social acts. This is consistent with previous findings that only state (but not trait nostalgia) may have an impact on decisions (Ye et al., 2013). For example, Ye et al. (2013) argued that nostalgia’s effect on creativity was guided by activities aroused by nostalgia rather than the nostalgic propensity of an individual. Our findings support this by showing that selfless acts were indeed guided by state rather than nostalgia proneness. Future research can, therefore, test the conditions under which state vs trait nostalgia can drive selfless behaviours under resource scarcity. In a related manner, future work can also study the impact of collective nostalgia in the face of resource scarcity. The current literature on nostalgia has recognised nostalgia as an individual as well as a collective emotion (Sedikides and Wildschut, 2019). Third, past research has shown that different scarcity types (e.g. quantity vs time) evoked different influences (e.g. Jung and Kellaris, 2004). Future research can, thus, account for a distinction of scarcity types in extending our findings. Future research can also test if variables linked to the nostalgia literature (e.g. communal orientation) can influence the relationships discovered in the current work. Finally, future research can also delve into different ways of manipulating nostalgia (e.g. through music, images, scent, etc.) to supplement our manipulation of recalling nostalgic events.

The supplementary material for this article can be found online.